IntroductionSepsis is a life-threatening complex clinical syndrome caused by immune imbalance following severe infection, involving damage to multiple organ systems1. Host immune dysregulation is a central mechanism in the development of sepsis and an important feature to distinguish sepsis from simple infection. It specifically includes an early overwhelming inflammatory response and acute cascade-like cytokine-mediated damage to tissues and the immune system, as well as later immune paralysis leading to worsening primary infections or secondary infections, which is related to the massive consumption of inflammatory factors and immune system dysfunction in early sepsis2,3. Accompanying immune dysregulation is multiple organ dysfunction syndrome (MODS), involving acute respiratory distress syndrome (ARDS), acute lung injury (ALI), acute kidney injury (AKI), and myocardial injury, which can lead to severe adverse outcomes such as metabolic acidosis, disseminated intravascular coagulation (DIC), and septic shock4. In recent years, early prevention, dynamic monitoring, and the implementation of sepsis management protocols have been widely promoted and applied, achieving certain results in clinical practice related to sepsis diagnosis and treatment5. However, the complexity of sepsis in terms of etiology, pathological mechanisms, and regulatory targets, as well as individual differences in clinical manifestations, population aging, and antibiotic misuse, have increased the difficulty of treating sepsis. As a result, the mortality rate of sepsis patients has not significantly improved6,7. Therefore, in-depth exploration of the mechanisms of sepsis and identification of core targets related to immune dysregulation are important research directions for the development of specific treatment strategies for sepsis.The development of sepsis involves both innate and adaptive immunity. Macrophages, as crucial immune cells and antigen-presenting cells within the mononuclear phagocyte system, play a significant role in both innate and adaptive immunity, and are highly heterogeneous and plastic. Originating from the bone marrow, macrophages are particularly abundant in the liver, spleen, gastrointestinal tract, upper respiratory tract, and brain. Their production and differentiation are regulated by various factors, including colony-stimulating factor 1 (CSF-1) and granulocyte-macrophage colony-stimulating factor (GM-CSF)8. Macrophages can secrete inflammatory cytokines and possess strong phagocytic and bactericidal capabilities, playing vital roles in tissue development, host defense, and maintaining internal homeostasis. On one hand, macrophages can recognize danger signals such as bacterial components in the microenvironment through pathogen-associated molecular patterns (PAMPs), thereby initiating innate immune responses. On the other hand, under the interaction of free radicals, cytokines, chemokines, and metabolites, macrophages can undergo polarization and transform into diverse phenotypic spectra, thereby modulating the host’s immune response to adapt to different microenvironmental changes9. Macrophage polarization is a dynamic process, and different stimuli can induce distinct functional phenotypes. Immune homeostasis can only be maintained when macrophages of various functional states remain in dynamic balance10. Among the numerous macrophage polarization phenotypes, M1 (classically activated macrophages) and M2 (alternatively activated macrophages) represent two extremes, closely related to the progression and prognosis of sepsis11,12. Current research has revealed that the mutual conversion between M1 and M2 macrophages is associated with multiple signaling pathways, including JAK/STAT, TLR-4/NF-κB, IRF5-IRF4, NF-κB-PPARγ, AP1-CREB, AP1-PPARγ, and Notch. The imbalance in their conversion significantly impacts the development and prognosis of sepsis13,14,15. Therefore, clarifying the regulatory mechanisms of macrophage polarization at different stages of sepsis and achieving a relative balance between M1 and M2 macrophages is an urgent priority for treating sepsis and reducing its complications.Given the crucial regulatory role of macrophage polarization in sepsis, this study aims to identify core immune regulatory genes in sepsis through RNA sequencing and bioinformatics analysis. Additionally, single-cell sequencing will be employed to locate the core target genes within specific cell lineages, providing a basis for selecting appropriate cell carriers in subsequent cellular experiments. Furthermore, the polarization status of macrophages will be assessed using flow cytometry, and the expression levels of related inflammatory cytokines will be validated using ELISA. These approaches will further elucidate the impact of core target genes on macrophage polarization and the expression of downstream inflammatory factors. Ultimately, this study identified the core immune regulatory gene ELANE, which promotes M1 macrophage polarization and upregulates the expression of inflammatory cytokines IL-1β and TNF-α, closely associated with poor prognosis in sepsis patients. A detailed flowchart of the study is presented in Fig. 1.Fig. 1Schematic of the three-pronged strategy: (A) Bulk RNA-seq identifies sepsis-associated DEGs and core targets; (B) scRNA-seq maps cellular localization; (C) Public datasets validate targets and clinical relevance. Results converge to functional validation.Full size imageMaterials and methodsResearch objectivesVenous peripheral blood samples were collected from sepsis patients (sepsis group, n = 20) admitted to the Intensive Care Unit of the Emergency Department at the Affiliated Hospital of Southwest Medical University between January 2019 and December 2020, as well as from healthy volunteers (normal control group, n = 10) during the same period. The inclusion criteria for the sepsis group were as follows: (1) Diagnosis met the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) jointly released by the Society of Critical Care Medicine and the European Society of Intensive Care Medicine in 2016, indicating the presence of infection and resulting organ dysfunction (SOFA score ≥ 2)16; (2) Age ≥ 18 years; (3) The patient or their legal representative consented to participate in the study. Exclusion criteria included: (1) History of severe organ dysfunction; (2) History of malignant tumors or mental illnesses; (3) Pregnant or lactating women; (4) History of hematological diseases, immune system disorders, or immunodeficiency; (5) Lack of clinical data. Clinical information and laboratory test results from routine blood tests, liver function, and kidney function were collected for the included participants, including gender, age, white blood cell count, neutrophil count, monocyte count, lymphocyte count, direct bilirubin, total bilirubin, urea, and creatinine. Data analysis was performed using GraphPad Prism 9.5, employing unpaired t-tests and calculating the mean values for each parameter. This study complies with the Declaration of Helsinki and medical ethical standards and has been reviewed and approved by the Ethics Committee of the Affiliated Hospital of Southwest Medical University (Ethics Approval Number: ky2018029; Clinical Trial Registration Number: ChiCTR1900021261; Registration Date: 4/2/2019). The study adheres to the Declaration of Helsinki.RNA sequencingRNA was extracted using the Trizol method (Invitrogen, Carlsbad, CA, USA)17. Total RNA was extracted from whole peripheral blood (PAXgene tubes) with globin mRNA reduction to minimize erythrocyte interference. Briefly, Trizol was added to the cell samples and thoroughly mixed, followed by high-speed centrifugation at 4 °C for 5 min. The supernatant containing RNA was transferred using a pipette to a new EP tube containing 0.3 mL of chloroform-isoamyl alcohol (24:1), mixed, and centrifuged again. After centrifugation, the mixture separated into three layers, and the upper aqueous phase containing RNA was transferred to a new tube with an equal volume of isopropanol for RNA precipitation. The RNA precipitate was washed with 75% ethanol. Finally, the dried RNA precipitate was dissolved in DEPC water, and the quality and quantity of RNA were analyzed using the Agilent 2100 Bioanalyzer (Thermo Fisher Scientific, MA, USA). Subsequently, DNA library construction was performed, which included the following steps: (1) Removal of oligonucleotides and rRNA from total RNA using targeted specific oligonucleotides and ribonuclease H reagents, followed by purification of mRNA using solid-phase reversible immobilization (SPRI) beads; (2) The purified mRNA was fragmented into small pieces under high-temperature conditions in a buffer containing divalent cations; (3) Using random hexamer primers, the fragmented mRNA was reverse-transcribed into first-strand cDNA with reverse transcriptase, followed by synthesis of the second-strand cDNA with RNase H and DNA polymerase; (4) The size distribution of cDNA fragments was analyzed using the Agilent 2100 Bioanalyzer, and the cDNA fragments were amplified by PCR. The amplified cDNA was further purified using Ampure XP Beads.Quality control of the sequencing dataBased on the BGISEQ-500/MGISEQ-2000 system18, the raw cDNA data were filtered using SOAPnuke (v1.5.3) software developed by BGI19. The filtered clean reads were saved in FASTQ format for subsequent analysis using second-generation high-throughput sequencing technology. The obtained transcriptomic RNA sequencing matrix data were submitted to the iDEP online platform (http://bioinformatics.sdstate.edu/idep/) for normalization and quality control20. Parameter settings included Homo Sapiens and Read counts data. The overall levels of RNA expression of different samples were assessed using boxplots and density distribution plots to confirm data homogeneity and comparability. Additionally, principal component analysis (PCA) was performed to compress and reduce the dimensionality of the sequencing data, thereby excluding outlier samples. RNA sequencing datasets for this study are available in the China National GeneBank DataBase (CNGBdb) at https://db.cngb.org/. Accession codes: CNP0002611.Screening for differentially expressed genesThe DESeq2 differential analysis method models gene count data in RNA sequencing based on the negative binomial distribution principle to identify expression differences of genes across samples21. This method accounts for the dispersion and heterogeneity in RNA sequencing data caused by batch effects or sample variability, making it a classic and highly efficient tool for sequencing data analysis. In this study, DESeq2 was applied to screen DEGs between the sepsis group and the normal control group using peripheral blood sequencing data. The screening thresholds were set as a false discovery rate (FDR)