Bioinformatics analysisIn this study, we downloaded the raw gene expression data from the GSE28829, GSE41571, GSE43292, GSE179828, GSE145200, and GSE3037 datasets obtained from the GEO database (https://www.ncbi.nlm.nih.gov/geo/). We merged and preprocessed the raw data using the sva package in R software (R-project.org), which included background adjustment, normalization, and logarithmic transformation, using a robust multiarray averaging method. The synthetic datasets were normalized using “limma” and “pheatmap” in R. Differentially expressed genes were screened and clustered with P < 0.05 and |fold change| ≥ 1. Differentially expressed genes in the GSE28829 dataset were screened and clustered. Late plaque samples in GSE28829, ruptured plaque samples in GSE41571, and unstable plaque samples in GSE43292 were classified into the unstable plaque group, while early plaque samples, stable plaque samples, and macroscopically intact tissues in each of the above datasets were classified into the stable plaque group. Samples from the GSE179828 dataset were classified into the stimulus group of NETs and the control group of HUVECs and NTs. The NETs samples in the GSE145200 dataset were classified as NETs pos in the positive group and NETs neg in the negative group. The GSE3037 dataset was derived from human peripheral blood neutrophils. Control individuals (Patient control) were categorized as the control group, and HMGB1-treated patients (Patient HMGB1 treated) and LPS-treated patients (Patient LPS treated) were categorized as the experimental group. Gene Ontology (GO) enrichment analysis of enrichment of GO terms, including biological process (BP), cellular component (CC), and molecular function (MF) terms, and KEGG pathway enrichment analysis were performed using the cluster analysis package in R. A p.adjust value < 0.05 and q value < 0.2 were considered to indicate significant differences. An xCell score based on the xCell analysis method in R was used to assess the enrichment of each cell type. Weighted gene co-expression network analysis (WGCNA) was performed using the WGCNA package in R to identify co-expression networks and cluster genes into different modules. Genes in the top 75% of the normalized variance were selected from the dataset for the next experiment. We selected the soft threshold power (β) based on the scale-free topology criterion using the selection soft threshold function. Then, a multistep network construction method was used to identify gene co-expression modules. Each module had at least 100 genes. A threshold of 0.6 was set for the merged modules to plot the inter-module correlations using the dynamic cut-tree method. Data on immune-infiltrating cells that differed significantly among all samples according to the xCell calculation were combined. The samples are clustered, and the correlation between the model feature matrix and the sample information matrix was calculated. The genes associated with neutrophils were identified and downloaded from the GeneCards (https://www.genecards.org/) website. After the intersection of deg&purple&neutrophils, 57 genes were imported into the STING database to construct a PPI network with a minimum interaction score of 0.4. Cytoscape (v3.9.1) was then utilized to construct a visual PPI network and identify hub genes. The top 10 genes in GSE43292 were screened to validate and evaluate the diagnostic and discriminative values of the genes using subject work characteristic (ROC) curve analysis. Finally, the GSEA method was used to identify the enriched marker gene set pathways (p-value < 0.05, FDR < 0.05) between the screened unstable plaques and stable plaques, and the marker gene sets used were obtained from the Molecular Signatures Database (MSigDB).Patient characteristics and sample collectionA total of 30 patients with carotid artery stenosis due to carotid atherosclerotic plaque and 10 normal volunteers were enrolled at the Second Affiliated Hospital of Harbin Medical University from August 2020 to November 2022. Fresh venous blood samples from patients on the first day of hospitalization and venous blood from healthy volunteers were collected, and plasma was collected after centrifugation and frozen at -80 °C. Twenty-nine of the patients were treated with carotid endarterectomy (CEA). Atherosclerotic plaques were obtained during the procedure, and the intact plaques were cut into two sections. One section was preserved in 4% paraformaldehyde for pathological staining, immunohistochemistry and immunofluorescence, and the remaining section was frozen in liquid nitrogen for RNA and protein extraction.Histopathologic examination and classification of carotid plaques in patientsHematoxylin (H8070, Solarbio, China), eosin (A600190, Sangon, China), oil red O (O0625, Sigma, USA), and Masson trichrome (p8330, 71019360, Sinopharm, China) were used to stain the tissue sections. Carotid plaque staining results were observed and photographed using an Olympus BX53 microscope. Plaques were classified as stable or unstable according to clinical and histologic criteria17 (Supplementary Table 1). Classification of the carotid plaque was performed by two independent investigators. The validation experiment was approved by the local ethics committee (YJSKY2022-149), and consent was obtained from all patients to participate in the study.Validation with reverse transcription-quantitative polymerase chain reaction (RT‒qPCR)A fluorescence quantitative PCR instrument (Exicycler 96, BIONEER, Korea) was used for the validation of four candidate hub genes (LY86, ITGB2, CCR1, and CSF1R) by RT‒qPCR. Total RNA was extracted from 10–50 mg of plaques using TRIpure (RP1001, BioTeke, Beijing, China) and reverse transcribed into complementary DNA (cDNA) using BeyoRT II M-MLV reverse transcriptase (D7160L, Biyuntian, Shanghai) according to the instructions. cDNA was amplified using SYBR Green 2x Taq PCR MasterMix (PC1150, Solarbio, Beijing, China) and 0.4 μmol of each primer pair (Anhui General Bio Co., Ltd., China). β-actin was used as an internal control, and the expression of the candidate genes was calculated by the 2−ΔΔCt method. The sequences of the primers used for the RT‒qPCR analysis are shown in Supplementary Table 2.ImmunohistochemistryImmunohistochemistry was used to assess the expression of MD-1 (LY86). Primary antibodies against MD-1 (1:50, A6185, ABclonal, China) were incubated overnight at 4 °C. Sections were incubated with horseradish peroxidase-labeled secondary antibodies and then stained with DAB. Images were taken at 40× to calculate the MD-1-positive areas in the plaques. The integrated optical density (IOD) of the target genes and the total tissue area were measured using Image-Pro Plus 6.0 software (IPP 6.0, Media Control Sciences, USA). The gene expression intensity was expressed as the IOD per unit area.ELISAThe plasma MD-1 levels of 30 patients and 10 healthy volunteers were determined using a commercial enzyme-linked immunosorbent assay (ELISA) MD-1 Kit (ZC-55707, Shanghai Thrive Biotechnology Co., Ltd., China) according to the manufacturer’s instructions, with three replicate wells for each sample. A human NETs kit (mm-2479H1) was used to determine the level of NETs produced by neutrophils induced in vitro, and a murine NETs kit (MM-1183M2), murine MD-1 kit (MM-47021M2), murine ICAM1 kit (MM-0183M2), murine VCAM1 kit (MM-0129M2), murine MMP14 kit (MM-46773M2), mouse VEGFA kit (MM-44452M2) and mouse IL6 kit (MM-0163M2, Jiangsu Enzyme Immunity Industry Co., Ltd., China) were used to determine the expression levels of the proteins in mouse plasma.Isolation of human peripheral blood neutrophilsAfter providing informed consent, fresh blood was drawn from healthy volunteers (males and females, 20–30 years old) by venipuncture. The research protocol was approved by the ethics committee of Harbin Medical University. Neutrophils were isolated using a human peripheral blood neutrophil isolation solution kit (P9040, Solarbio, Beijing, China) and then resuspended in RPMI 1640 medium supplemented with or without 2% fetal bovine serum (FBS, SA101, CellMax, Lanzhou, China). The neutrophil viability was >95%, as assessed by the Trypan blue dye exclusion test, and the neutrophil purity was >95%, as assessed by Giemsa staining.Quantification of NETs releasesFresh human neutrophils (1 × 105 cells per well) were inoculated into 6-well plates and 6-well plates precoated with 0.001% polylysine L and incubated with different concentrations of recombinant human MD-1 (novoprotein, Suzhou, China) in the presence of 2% FBS. The concentrations of MD-1 were determined according to the results of the previous assays as 25, 50, 75, 100, and 125 ng/ml, and the specified continuous induction times were 1, 2, 4, 6, 8, 10, and 12 h at 37 °C and 5% CO2. A 1:1 mixture of deionized water and dimethyl sulfoxide (DMSO) was used as a blank control, and phorbol myristate18 (PMA, 30 nM, P3189, Sigma, USA) and lipopolysaccharide19 (LPS, 30 ng/mL, Sigma, USA) were used as positive controls. After induction, the solution in each well was mixed by gentle blowing and collected by centrifugation (200 × g, 5 min), and the supernatant was collected to determine the NETs concentration by ELISA. After reaching the appropriate induction concentration, the cells were pretreated with selected inhibitors (1 μM TAK-242, 10 μM ST2825, 0.5 μM IRAK1-4 INH, 10 μM C25-140, 10 μM SB203580, 10 μM U0126, and 10 μM DPI; Apexbio, USA) for 1–4 h before being assayed again for the determination of the NETs concentration.Hoechst stainingAfter the above treatment, liquid was aspirated from the coated 6-well plates, the plates were stained with a Hoechst 33342 Staining Kit (C0003, Beyotime, China), and the structure of the NETs in five random fields of view was observed and photographed using a fluorescence microscope.ImmunofluorescenceSamples from the neutrophil inhibitor assay group, clinical plaque tissue, and animal tissue described above were processed, fixed in 4% paraformaldehyde for 15 min, and permeabilized with Triton X-100 (A2576, ALPHABIO, Tianjin, China) for 30 min at room temperature. Nonspecifically bound sites were blocked with 50% goat serum in PBS at 37 °C for 30 min. Neutrophil samples were incubated overnight at 4 °C with a murine anti-myeloperoxidase antibody (1:100, ab90810, Abcam, USA), and tissue and animal samples were incubated with an anti-CD31 antibody (1:50, 66065-2-Ig, Proteintech, Wuhan, China), followed by incubation with an Alexa Fluor 555-conjugated secondary antibody (1:500, Beyotime, Shanghai, China). DNA was labeled with an anti-fluorescence burst sealer containing DAPI (50 μL, AL739A, ALPHABIO, Tianjin, China). Images were acquired using fluorescence microscopy. Tissue and animal samples were observed and photographed under a confocal microscope, and the microvessel density (MVD) was calculated. The calculation of the MVD required three randomly selected high-magnification (20×) fields of view from three different sections of the tissue samples, and CD31-positive luminal or nonluminal ECs in the field of view were observed and counted. The microvessel density, expressed as the average number of microvessels per field of view, was quantified by two individuals who were unaware of the experimental design.Co-immunoprecipitationTo determine the interaction between MD-1 and TLR4, we performed co-immunoprecipitation (Co-IP). Temporally, neutrophils were incubated with recombinant human MD-1 (75 ng/mL) for 60 min at 37 °C in 5% CO2. After incubation, the cells were washed twice with PBS, and total cellular proteins were extracted using Western and IP Cell Lysis Buffer (P0013, Beyotime, Shanghai, China) containing protease inhibitors, followed by centrifugation at 14,000 × g for 5 min at 4 °C. Experiments were performed using an immunoprecipitation kit (Protein A magnetic bead assay, P2175S, Beyotime, Shanghai, China) according to the instructions. Briefly, 50 μg/ml of mouse anti-MD-1 antibody (SANTA CRUZ BIOTECHNOLOGY, INC., USA) and 50 μg/ml of mouse anti-TLR4 antibody (Proteintech, Wuhan, China) were conjugated to Protein A magnetic beads and separated from the magnetic beads. Then, the samples were incubated with Protein A magnetic beads conjugated with antibodies or normal IgG, and the samples were magnetically separated and washed after SDS‒PAGE. Proteins transferred to PVDF membranes were imputed using mouse anti-MD-1 antibody and mouse anti-TLR4 antibody and then incubated with HRP-coupled secondary antibody (1:5000, ZSGB-BIO, Beijing, China). The presence of MD-1 and TLR4 in the immunoprecipitates was detected by HRP-coupled goat anti-mouse IgG secondary antibody (1:5000, AS003, ABclonal, Wuhan, China).Duolink proximity ligation assayTo detect the interaction between MD-1 and TLR4, we used the Duolink® In Situ Proximity Ligation Assay Kit (DUO92101, Sigma, USA). Freshly isolated neutrophils from carotid plaque patients and healthy volunteers were inoculated in 24-well plates, in which healthy volunteer neutrophils were incubated with recombinant human MD-1 (50 ng/ml and 75 ng/mL) and MD-1 (75 ng/ml)&TAK-242 (1 μM) for 60 min at 37 °C with 5% CO2. After treatment, neutrophils from each group were fixed with 4% paraformaldehyde for 15 min at room temperature, treated with 0.5% Triton X-100 for 30 min, and blocked with 1x blocking solution for 60 min at 37 °C. The cells were then incubated with two primary antibodies against MD-1 and TLR4 for 3 h at 37 °C. After washing, the cells were incubated with the mixed PLA probe for 60 min at 37 °C. This pair of secondary antibodies produces a signal only when the two probes are in close proximity (<40 nm). After washing again, the cells were ligated, amplified, and finally assayed separately. Fluorescence images were acquired under a fluorescence microscope.Cell cultureHUVECs (ATCC, Manassas, USA) and HAECs (HTX23972, Haodi Huatuo Biotechnology Co., Ltd., Shenzhen, China) were incubated at 37 °C under 5% CO2 in DMEM-F12 or endothelial cell medium (HTX23972, Zhong Qiao Xin Zhou Biotechnology Co., Ltd., Shanghai, China) supplemented with 10% FBS and 100 nmol/L penicillin/streptomycin (C0222, Beyotime, Shanghai, China) for 5 days. All assays were performed using cells from generation 6 and below.Generation, isolation, and preparation of NETs and NETs culture mediaPurified neutrophils (1 × 106/well) were seeded in 6-well plates, stimulated with 50 ng/ml MD-1 or 75 ng/ml MD-1, and incubated at 37 °C and 5% CO2 for 4 h. The medium was then gently removed, leaving the NETs and neutrophils attached to the plate. Precooled PBS without calcium and magnesium was added to the eluted NETs and neutrophils. The liquid in the 6-well plates was collected and centrifuged at 450 × g at 4 °C for 10 min, after which the supernatant was collected. The supernatant was collected again by centrifugation at 15,000 × g for 15 min at 4 °C and its concentration was determined. Preparations were cultured in DMEMF12 medium containing 2% FBS containing low and elevated concentrations of NETs (as described above), NETs, and DNase I (100 U/mL, 10104159001, Sigma, USA) or ddH2O and DMSO. Cultures were incubated at 37 °C and 5% CO2 for 48 h, after which DNase I was used to pretreat the NETs for 1 h at 37 °C before subsequent assays.Protein extraction and Western blot analysisNeutrophils (5 × 106/ml) were treated with recombinant human MD-1 (75 ng/mL) and maintained at 37 °C with 5% CO2 for 15 min, 30 min, 45 min, or 1 h. The cells were also induced by the addition of MD-1 (75 ng/mL) for 3 h after the grouping of each of the previously mentioned inhibitors. HUVECs and HAECs were treated with medium containing NETs, NETs, and BAY11-7082 (5 μM, HY-13453, MedChemExpress, Shanghai, China) or ddH2O, DMSO, and BAY11-7082. Carotid plaque tissues, neutrophils, HUVECs, and HAECs were washed twice with cold PBS and lysed with RIPA buffer (P0013B, BAY11, Shanghai, China) containing protease inhibitor (K1007, Apexbio, USA) and phosphatase inhibitor (K1015, Apexbio, USA). The total protein concentration was determined using a BCA Protein Assay Kit (P0010S, Biyuntian, Shanghai). Equal amounts of the proteins were separated by SDS‒PAGE and transferred to a PVDF membrane. After blocking (P0252, Biyun Tian, Shanghai, China), each sample membrane was incubated with primary antibodies against various candidate proteins, such as MD-1 (1:100, sc-390613, Santa Cruz Biotechnology, USA), CCR1 (1:2000, abs105357, Absin, Shanghai, China), p-IRAK1 (1:500, bs3192R), p-IRAK4 (1:500, bs4080R, Beijing Boaoxian Biotechnology Co. ab180747), IRAK4 (1:1000, ab32511), p38 MAPK (1:1000, ab170099), p-p38 MAPK (1:1000, ab195049), ERK1/2 (1:10000, ab184699), p-ERK1/2 (1:10000 ab278538, Abcam, USA), CSF1R (1:500, A3019), ITGB2 (1:500, A19012), ICAM1 (1:2000, A22596), VCAM1 (1:100, A0279), MMP14 (1:500, A0067), VEGFA (1:500, A5708), IL6 (1:500, A22222), TLR4 (1:500, A0007), NF-κB (1:2000, A22331), p-NF-κB (1:500, AP0838), and β-actin (1:100000, AC026, ABclonal, Wuhan, China) were incubated at 4 °C overnight. Then, a specific HRP-conjugated secondary antibody (1:10000, ZB-5301, Zhongshi Jinqiao, Beijing, China) was used. Immunoreactive bands were detected with an enhanced chemiluminescence (ECL) substrate (BL520A, Biosharp, Hefei, China). Quantitative blotting was performed with ImageJ.Cell proliferation and colony formation assayA cell counting kit-8 (CCK-8, K1018, Apexbio, USA) was used to determine the relative cell growth at different time intervals (4, 8, 12, 24, and 48 h). The absorbance (optical density at 450 nm) of each 96-well plate was measured using an enzyme marker. In the colony formation assay, HUVECs resuspended in the various media mixtures described above were seeded in 6-well plates (200 cells/well) and subjected to the indicated treatments for 8 days. The cell colonies were fixed with 4% formaldehyde and stained with 0.1% crystal violet for 10 min, after which the colonies were counted. All assays were repeated three times.Cell migration assayCell migration was measured using a two-chamber Transwell migration assay as previously described. The lower chamber (24-well plate) was filled with 600 μL of DMEM-F12 medium containing 10% FBS. The upper chamber was filled with 200 μL of each group of medium prepared as described above, and 1 × 104 HUVECs were seeded onto 8 μm pore size membranes (Selection 14341, Selection, Hefei, China). Unmigrated cells were swabbed from the upper chamber after 4, 8, 12, 24 and 48 h. The migrating cells were fixed with 4% paraformaldehyde and stained with crystal violet dye. The average number of migrating cells was counted using an inverted microscope at 100×.Scratch wound-healing assayHUVECs (5 × 105 cells/well) were inoculated in 6-well plates and incubated at 37 °C with 5% CO2. After the cells formed a fused monolayer, a sterile P1000 tip was used to produce a scratch in the center of the cell layer. The detached cells were gently rinsed off with PBS. Subsequently, the cells were treated with each of the media mixtures described above and imaged after 4, 8, 12, 24, and 48 h. ImageJ was used to estimate cell migration distances.Tube formationThe matrix gel (356234, BD-Pharmingen, USA) was diluted with DMEM-F12 at a 1:1 ratio and added to 48-well plates. The dishes were placed in an incubation chamber at 37 °C for 30 min to polymerize the matrix gel. HUVECs and HAECs (1 × 104 cells/well) were resuspended in the various media mixtures described above, seeded into matrix gel-coated wells, and incubated in growth-supplement-free medium at 37 °C for 6 h. The samples were observed using an inverted microscope, and the tube formation was quantified using ImageJ.Establishment of animal models and assay validationEight-week-old male, 19–20 g purebred ApoE−/− mice (n = 40, Beijing Vital River Laboratory Animal Technology Co., Ltd., China) were randomly divided into experimental (A–C) and control (D) groups of 10 mice each, and the experimental groups were randomly divided into three subgroups, including the A: MD-1 group (75 ng*0.072/g/mouse), calculated at 72 ml/kg of blood per mouse; B: MD-1&DNase I (50 μg/mouse) group; C: MD-1&TNP-470 (20 mg/kg, HY-101932, MedChemExpress, China) group; and D: Control group. All mice were housed in a specific pathogen-free facility and given weekly intraperitoneal injections for 16 weeks. A mouse atherosclerotic plaque model was generated using a Western diet containing 0.15% cholesterol and 21% fat (H10141, Beijing Huafukang Laboratory Animal Company, China). Assay procedures and protocols were reviewed and approved by the Animal Investigation Ethics Committee of the Second Affiliated Hospital of Harbin Medical University (YJSDW2022-116) and were performed in accordance with the Guidelines for the Care and Use of Laboratory Animals of the National Institutes of Health, China. At week 17, blood was taken from the apical part of the heart after sacrifice with excess sodium pentobarbital and centrifuged to retain the plasma, which was frozen and stored at −80 °C as extra samples. The carotid arteries were carefully isolated and cut and the obtained tissues were immersed in 4% paraformaldehyde overnight, paraffin-embedded, and divided into 4 μm paraffin sections for subsequent pathological staining, immunohistochemistry, and immunofluorescence assays (all as described previously).Depletion of neutrophilsWe further confirmed that MD-1 plays an important role in regulating neutrophils. We constructed a neutrophil depletion model based on the in vivo injection of InVivoPlus anti-mouse Ly6G (BP0075-1-5MG, BioXCell, New Hampshire, USA). The MD-1 dose in both the Ly6G group and the MD-1 + Ly6G group was the same as that used previously, and the Ly6G dose was 5 μg/g/mouse/week (divided into three injections every other day) for 16 weeks. The feeding, collection, and experimental methods were the same as before.Flow cytometryFor the neutrophil depletion model group of mice, approximately 150 μl of peripheral blood was taken from the tail vein of mice and the erythrocytes were lysed, resuspended in cell staining buffer (420201), and incubated at room temperature with the addition of purified anti-mouse CD16/32 antibody (1.0 µg/106 cells/100 µl, 101301) for 30 min. Then, each sample was incubated with FITC anti-mouse/human CD11b antibody (0.25 µg/106 cells/100 µl, 101205) and APC anti-mouse Ly-6G antibody (0.06 µg/106 cells/100 µl, 164506, BioLegend, San Diego, USA) for 40 min at room temperature protected from light. Finally, after washing and resuspension, the CD11b+ and Ly6G+ neutrophils in each sample were quantified via nanoflow analysis. The data were acquired using Apogee and analyzed using FlowJo X.Single-cell RNA-seq data analysisExperimental single-cell data on carotid atherosclerotic plaques were obtained from the GSE155514 and GSE224273 datasets, and GSE131778 was used as the validation. The R package Seurat (ver. 5.0.1) was used to filter each single-cell sample and retain genes expressed in at least 3 cells and cells with at least 300 genes expressed. To reduce the effect of cell cycle genes, the cell cycle genes were scored in each cell, and the data were normalized. UMAP and t-SNE were used for cell clustering while significant differentially expressed genes in each cluster were calculated and the annotation of cell types was completed by CellMarker (http://xteam.xbio.top/CellMarker/index.jsp) and classical highly expressed genes common to all types of cells documented in the literature. Pseudotime analysis of the relevant cell types was performed using the R package Monocle 2. Intercellular communication analysis was performed using the R package CellChat to identify and predict interactions between different cell types. Transcription factor activity analysis was performed with the R package SCENIC.Statistical analysisAll the statistical analyses were performed using R software (ver. 4.1.3) or SPSS statistical software (ver. 22.0, IBM SPSS Inc., Chicago, IL, USA), or GraphPad Prism 9.0 (La Jolla, CA, USA). The data are shown as the mean ± standard deviation. One-way ANOVA followed by Tukey’s multiple comparison test was used for multiple groups. Student’s t test was used for comparisons between two groups. Significance levels are expressed as follows: ns P value ≥ 0.05, *P value < 0.05, **P value < 0.01, ***P value < 0.001, ****P value < 0.0001.
