https://orcid.org/0009-0000-7039-8366
Figures
Abstract
Background
Acute lung injury (ALI) involves the release of growth factors and inflammatory mediators from damaged pulmonary tissues, fostering endothelial cell proliferation, migration, and vascular lumen formation, thereby driving pathological angiogenesis. Macrophages contribute to angiogenesis and vascular homeostasis, but their dysregulation in pathological states worsens vascular dysfunction. This study aims to identify macrophage-associated angiogenesis-related genes as novel diagnostic biomarkers and therapeutic targets for sepsis-associated ALI (SALI).
Methods
Transcriptomic datasets from the GEO database were analyzed using differential expression profiling and weighted gene co-expression network analysis (WGCNA) to identify candidate genes. These candidates were compared with macrophage- and angiogenesis-related gene sets from GENECARDS for functional prioritization. Three machine learning algorithms (LASSO regression, random forest, and SVM) were employed to refine predictive biomarkers, followed by immune infiltration analysis (via CIBERSORT) to assess correlations with immune subsets. Single-cell RNA sequencing and RT-PCR were used for spatial validation of gene expression.
Results
Two macrophage-associated angiogenesis-related genes, Fcer1g (FCER1G) and St3gal1 (ST3GAL1), were identified as key biomarkers. Both genes showed significant upregulation in the training cohort (p < 0.001) and independent validation sets (p < 0.05), with robust diagnostic accuracy (AUC > 0.85). Immune correlation analysis indicated strong positive associations with macrophage infiltration (p < 0.01), particularly M2-polarized subsets. scRNA-seq confirmed their predominant expression in macrophage clusters, with increased activity in SALI tissues (log2FC > 2.0, p < 0.001).
Conclusions
In mouse in vivo studies, Fcer1g and St3gal1 were shown to precisely mediate intricate macrophage-endothelial cell interactions via glycoimmune signaling pathways at the molecular level. This interaction finely modulates endothelial cell activation and drives angiogenic remodeling, critically impacting SALI progression. Given the physiological and pathological parallels between mice and humans, our findings offer a theoretical underpinning for subsequent human – oriented research. Moving forward, efforts should focus on verifying the expression patterns, action mechanisms, and diagnostic/therapeutic potential of these genes in relation to human SALI – associated signatures.
Citation: Liu L, Luo P, Wang M, Yu S, Xiao J (2026) Fcer1g and St3gal1: Macrophage-associated angiogenesis biomarkers and therapeutic targets in sepsis-induced acute lung injury. PLoS One 21(2): e0343449. https://doi.org/10.1371/journal.pone.0343449
Editor: Tomasz W. Kaminski, Versiti Blood Research Institute, UNITED STATES OF AMERICA
Received: September 6, 2025; Accepted: February 4, 2026; Published: February 27, 2026
Copyright: © 2026 Liu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: Comprehensive gene expression data was searched from the Database of the National Center for Biotechnology Information in the United States (https://www.ncbi.nlm.nih.gov/geo/)The bioinformatics data for this study were sourced from public databases, with strict adherence to their usage guidelines (GSE15739, GSE23767, GSE40180, GSE52474, GSE226807, GSE60088, GSE130936, GSE1871, GSE2411, GSE32707, GSE207651). Cell - experiment data are within the paper and its Supporting Information files.
Funding: This work was supported by Scientific Research and Innovation Fund of the First Affiliated Hospital of Harbin Medical University(2024M13), Heilongjiang Jizhong Medical Assistance Charity Foundation(jzkyht2024-001) and Harbin Medical University Youth Fund（2025）.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Acute Lung Injury (ALI) and its more severe form, Acute Respiratory Distress Syndrome (ARDS), constitute a life – threatening clinical syndrome. This syndrome is characterized by widespread pulmonary inflammation, disruption of the alveolar – capillary membrane, and the presence of protein – rich alveolar edema. These pathophysiological changes often culminate in acute respiratory failure. Among the various etiological factors, sepsis stands out as the most prevalent trigger, being responsible for more than 40% of ALI/ARDS cases [1]. The key pathological feature of this syndrome is increased vascular permeability, which results in the accumulation of fluid in the interstitial and alveolar spaces, impaired gas exchange, and progressive hypoxemia [2].
Angiogenesis, the process of forming new microvasculature from pre – existing blood vessels, plays a dual role in the progression of ALI. Endothelial cells (ECs) are central regulators of angiogenesis, orchestrating processes such as proliferation, migration, and the formation of tubular networks [3,4]. In the context of ALI, injured pulmonary tissues release a variety of mediators, including vascular endothelial growth factor (VEGF) and inflammatory cytokines. These mediators can stimulate both physiological revascularization and pathological hyperpermeability [3,5]. While moderate angiogenesis is beneficial as it supports tissue repair by restoring the supply of oxygen and nutrients [5,6], excessive neovascularization can have detrimental effects. It can exacerbate alveolar edema due to persistent vascular leakage and promote fibrotic remodeling through abnormal interactions between endothelial cells and mesenchymal cells [7,8].
In the pathophysiological process of ALI/ARDS, the role of macrophages, as key immune cells, cannot be overlooked. Macrophages exhibit high plasticity and multifunctionality, enabling them to adjust their phenotypes and functions according to changes in the local microenvironment [9]. In the early stage of ALI/ARDS, macrophages primarily play a pro – inflammatory role, releasing a large amount of inflammatory mediators and exacerbating the pulmonary inflammatory response [10]. However, as the disease progresses, macrophages gradually transform into an anti – inflammatory and reparative phenotype, participating in tissue repair and regeneration processes [11]. Notably, macrophages can not only secrete various pro – angiogenic factors, such as VEGF and Platelet - Derived Growth Factor (PDGF), directly promoting angiogenesis [12] but also regulate the proliferation, migration, and tubular formation ability of endothelial cells through direct interactions with them [13]. Therefore, macrophages likely play a central role in driving angiogenesis in ALI/ARDS. An in – depth study of the mechanism of macrophage – driven angiogenesis will help us gain a more comprehensive understanding of the pathophysiological process of ALI/ARDS and develop new therapeutic strategies.
In addition, angiogenesis markers hold significant potential value in the clinical management of ARDS. Multiple studies have demonstrated that the levels of angiogenesis – related factors such as VEGF in the bronchoalveolar lavage fluid or serum of ARDS patients are closely related to the severity and prognosis of the disease [14,15]. For example, an elevation in VEGF levels may indicate a worsening of the patient's condition or a poor prognosis [16]. Therefore, identifying and monitoring angiogenesis markers not only assists in more accurately assessing the condition of ARDS patients but also provide important references for formulating individualized treatment plans.
Despite the progress made in understanding ALI/ARDS, the regulatory relationship between macrophages and angiogenesis pathways in Sepsis - associated Acute Lung Injury (SALI) has not been fully elucidated. In this study, we employed a combined approach of transcriptomic profiling and single – cell RNA sequencing (Single – cell RNA sequencing, scRNA - seq) to map out the molecular network related to macrophage – driven angiogenesis in SALI. By applying machine learning algorithms, our objective was to identify key genes that govern the interaction between macrophages and endothelial cells. Based on these findings, we propose novel diagnostic biomarkers and potential therapeutic targets for this life – threatening disease, with the aim of improving the clinical prognosis of ARDS patients.
Methods
Data sources
A comprehensive search for gene expression data was conducted in the database of the National Center for Biotechnology Information (NCBI) (https://www.ncbi.nlm.nih.gov/geo/) to identify datasets relevant to sepsis - associated acute lung injury (SALI). Ultimately, 11 gene datasets (GSE15739, GSE23767, GSE40180, GSE52474, GSE226807, GSE60088, GSE130936, GSE1871, GSE2411, GSE32707, GSE207651) were identified and collected for subsequent analysis. The bioinformatics data in this study were from public databases. We strictly followed their usage terms during data download and use. These databases had proper ethical reviews on data handling, ensuring data legality. Thus, no extra ethical approval or informed consent from data providers was needed.
Among them, GSE15739, GSE23767, GSE40180, GSE52474, and GSE226807 (acute lung injury caused by systemic infection) served as the training set. GSE60088 and GSE130936 comprised mouse lung tissue samples with acute lung injury induced by systemic infection (extrapulmonary SALI validation set). GSE1871 and GSE2411 included mouse lung tissue samples with acute lung injury induced by local pulmonary infection (intrapulmonary SALI validation set). GSE32707 represented blood samples from human SALI patients (human SALI validation set). GSE207651 was single – cell RNA sequencing (scRNA - seq) data from mouse lung tissues with SALI.
Data processing
For data preprocessing, rows (genes) and columns (samples) with a missing value proportion exceeding 50% were initially removed from the expression matrix. Subsequently, the impute.knn function from the “impute” R package was employed to estimate the remaining missing values. The “impute” R package is a valuable tool for handling missing data in biological datasets, offering various imputation methods. The impute.knn function utilizes the k-nearest neighbors algorithm to estimate missing values based on the similarity of gene expression patterns among samples. Additionally, a log2 transformation was applied to the data to facilitate subsequent analyses.
To integrate multiple datasets, the “inSilicoMerging” R package was first utilized. This package is specifically designed to combine gene expression datasets from different sources, enabling integrated analyses. Moreover, the method developed by Johnson WE et al. was adopted to eliminate batch effects, ensuring data consistency across different batches or experiments [17]. Batch effects can introduce confounding factors in gene expression data analysis, and this method helps correct for these variations, enhancing data comparability. The data analysis in this step was carried out in March 2025.
Identification of differentially expressed genes and functional enrichment analysis
Differential analysis was performed using the limma R package (version 3.40.6) to identify differentially expressed genes (DEGs) between the lung tissues of the SALI group and the sham-operated (Sham) group. The limma R package is a widely adopted tool for analyzing gene expression microarray and RNA-seq data, providing a robust and flexible framework for linear modeling and differential expression analysis. Genes were considered to exhibit significant differential expression when the fold change exceeded 1.5 and the P-value was less than 0.05.
Gene Ontology(GO) analysis and Kyoto Encyclopedia of Genes and Genomes(KEGG) analysis were conducted using the DAVID database (http://david.abcc.ncifcrf.gov/) to identify enriched pathways associated with the gene set [18]. The DAVID database is a web-based bioinformatics resource that offers a comprehensive suite of functional annotation tools for interpreting the biological significance of large gene lists. It aids in identifying over-represented biological processes, molecular functions, cellular components, and signaling pathways within a given gene set.
WGCNA
Firstly, we calculated the Median Absolute Deviation (MAD) for each gene based on the gene expression profiles and excluded the top 50% of genes with the lowest MAD values. Subsequently, we utilized the goodSamplesGenes function from the R package WGCNA to remove outlier genes and samples. Following this, we employed WGCNA to construct a scale-free co-expression network. Specifically, we initially computed Pearson's correlation matrices and applied the average linkage method for all pairwise genes. Then, a weighted adjacency matrix was constructed using a power function A_mn = |C_mn|^β (where C_mn represents the Pearson's correlation between Gene_m and Gene_n; A_mn represents the adjacency between Gene_m and Gene_n). The parameter β served as a soft-thresholding parameter, emphasizing strong correlations between genes while penalizing weak ones. After selecting a power of 3, the adjacency matrix was transformed into a topological overlap matrix (TOM), which measures the network connectivity of a gene, defined as the sum of its adjacency with all other genes for network generation. The corresponding dissimilarity (1-TOM) was also calculated. To classify genes with similar expression profiles into gene modules, average linkage hierarchical clustering was performed based on the TOM-derived dissimilarity measure, with a minimum size (gene group) of 5 for the gene dendrogram. To further analyze the modules, we calculated the dissimilarity of module eigengenes, selected a cut-off line for the module dendrogram, and merged some modules. Furthermore, we merged modules with a distance less than 0.25, ultimately resulting in 16 co-expression modules. We calculated the gene significance (GS) based on the correlation with the gene expression profile. Using a cutoff criterion of |GS| > 0.1, we identified 1,386 genes with high connectivity within the clinically significant module [19].
Machine learning-based target gene screening
To further identify target genes, three machine learning algorithms were employed: the Least Absolute Shrinkage and Selection Operator (LASSO), the Random Forest (RF) algorithm, and the Support Vector Machine Recursive Feature Elimination (SVM-RFE) algorithm [20]. LASSO is a regression analysis method that performs both variable selection and regularization, enabling the selection of a relevant feature subset from a large number of variables. The Random Forest algorithm is an ensemble learning method that constructs multiple decision trees and combines their predictions to enhance model accuracy and robustness. SVM-RFE is a feature selection algorithm based on support vector machines that recursively eliminates the least important features to identify the most relevant ones. Genes that overlapped in the results of these three algorithms were selected as the target genes [21]. The analysis was conducted in March 2025.
Immune infiltration analysis
For immune infiltration analysis, the ImmuCellAI tool (https://guolab.wchscu.cn/) was utilized [22]. ImmuCellAI is an online tool specifically designed to evaluate the abundance of immune cells in biological samples. It can assess the abundance of 24 immune cell types in human samples and 36 immune cell types in mouse samples. The tool uses gene expression data to estimate the proportion of different immune cell types in the sample, providing insights into the immune microenvironment. The immune cell abundances obtained from both human and mouse samples were visualized, and the correlations between the target genes and these immune cells were examined. The analysis was carried out in March 2025.
scRNA-seq
In this study, we employed scRNA-seq data analysis methods to conduct a systematic analysis of the GSE207651 dataset. Initially, data preprocessing was carried out, encompassing data loading, merging, and quality control to eliminate low-quality cells and specific genes. Low-quality cells were excluded if they expressed fewer than 300 or more than 6000 genes, as well as cells where the proportion of unique molecular identifiers (UMIs) from the mitochondrial genome exceeded 15%. Subsequently, data normalization was performed, and batch effects were eliminated to ensure the consistency and comparability of the data [23]. Ultimately, 17,792 cells and 20,479 genes were retained for further analysis.
Next, high-variable genes were selected using functions such as FindVariableFeatures and FindClusters. Principal Component Analysis（PCA） and Uniform Manifold Approximation and Projection（UMAP） were applied for dimensionality reduction, followed by clustering analysis to reveal differences among cell types. PCA is a statistical technique that reduces data dimensionality while preserving as much variance as possible, and UMAP is a non-linear dimensionality reduction method that better preserves the local structure of the data. Additionally, the FindAllMarkers function was used to identify marker genes for each cluster, and cell annotation was performed based on known information from CellMarker 2.0 [24] and PanglaoDB [25]. CellMarker 2.0 and PanglaoDB are databases that provide information on cell type-specific marker genes, which can be used to annotate cells in scRNA-seq data. Finally, the cellchat package was employed to analyze cell-cell interactions, and target genes were selected for validation of their expression levels. The data analysis in this step was completed in March 2025.
Cell culture and RT-PCR
The RAW264.7 cells were procured from Saiou Biotechnology Co., Ltd. (China). These cells were cultured in DMEM medium supplemented with 1% penicillin-streptomycin solution and 10% fetal bovine serum, and incubated in a humidified incubator at 37°C with 5% carbon dioxide. The cell experiment part was exempted from ethical review by the Ethics Committee of the First Affiliated Hospital of Harbin Medical University. As it used routine cell lines without involving human tissues or direct impact on human subjects, no ethical review or informed consent was required.
In the cell experiment, Sham group cells were exposed to an equivalent concentration of PBS-containing culture medium for the same duration (3 hours) as the experimental group cells exposed to lipopolysaccharide (LPS). This design ensures that the only difference between the Sham group and the experimental group is the presence or absence of the SALI-inducing factor, allowing for a more accurate assessment of the effects of SALI on gene expression, immune cell infiltration, etc.
Total RNA was extracted using an RNA extraction kit (TransGen Biotech, China), and the RNA samples were subsequently reverse-transcribed into complementary deoxyribonucleic acid (cDNA). The expression levels of candidate genes were measured using SYBR Green on a real-time fluorescent quantitative PCR system (Applied Biosystems, USA). (Ethical approval for this cell experiment was exempted, as approved by the Ethics Committee of the First Affiliated Hospital of Harbin Medical University.) The primer sequences used in this experiment were as follows: for Fcer1g (forward primer, F: 5’-ATCTCAGCCGTGATCTTGTTCT-3’; reverse primer, R: 5’-ACCATACAAAAACAGGACAGCAT-3’) and for St3gal1 (forward primer, F: 5’-TCCAACACGGGAGTACCCA-3’; reverse primer, R: 5’-GCTGGTCGAACCAATATGATACC-3’).
Statistic analysis
All calculations were performed were performed using R software and Sangerbox 3.0（http://www.sangerbox.com [26]. RT-PCR was performed using Graphpad prism software for statistical analysis and mapping, and t-test was used for inter group data analysis and comparison. A P-value of less than 0.05 was considered statistically significant.
Results
Identification and functional enrichment analysis of DEGs in SALI
Five datasets pertaining to SALI were integrated, and batch effects were eliminated. Following batch effect correction, the five datasets were well-integrated, as evidenced by PCA results, which revealed a clear separation between the SALI group and the Sham group (Fig 1A - D). Differential analysis between the SALI and Sham groups was conducted on the integrated dataset using the limma package, identifying 994 DEGs, including 480 upregulated genes and 514 downregulated genes (Fig 1E). A heatmap was generated to illustrate the expression levels of the top 20 genes with the highest degrees of upregulation and downregulation in lung tissues (Fig 1F).
(A-C) Integration and batch effect removal of gene expression profiles from five Acute lung injury (ALI) datasets (GSE15739, GSE23767, GSE40180, GSE52474, GSE226807) associated with systemic infection-induced ALI. (D) Principal Component Analysis（PCA） analysis of the integrated dataset. (E-F) Differential analysis of the integrated dataset revealed 480 upregulated and 514 downregulated genes. The volcano plot illustrates the distribution of genes, while the heatmap displays the expression of the top 20 upregulated and downregulated genes in the SALI and Sham groups.
Further functional enrichment analysis of the DEGs was performed using the DAVID database. GO analysis revealed that the DEGs were involved in biological processes such as immune response, inflammatory response, and angiogenesis (Fig 2A). Additionally, extracellular components and cytokine interactions were closely associated with SALI, indicating that intercellular interactions play a pivotal role in the pathogenesis of SALI (Fig 2B and C). KEGG pathway analysis demonstrated that the DEGs were significantly enriched in pathways related to IL17, tumor necrosis factor (TNF), Toll-like receptors, and nuclear factor-κB (NF-κB) signaling, all of which are crucial in inflammatory responses (Fig 2D).
(A)Biological Process (BP) in Gene Ontology(GO) analysis; (B) Cellular Component (CC) in GO analysis; (C) Molecular Function (MF) in GO analysis; (D) Kyoto Encyclopedia of Genes and Genomes(KEGG) analysis.
WGCNA
We employed WGCNA to identify genes highly associated with SALI. First, we calculated the MAD for each gene based on the gene expression profiles and excluded the top 50% of genes with the lowest MAD values. Subsequently, the goodSamplesGenes function was utilized to eliminate outlier genes and samples. By selecting an appropriate soft – thresholding power (β = 3, R² = 0.87), genes were clustered into 16 modules (Fig 3A–D). Among these modules, the blue and orange modules exhibited the highest correlation with the clinical traits of SALI.
(A) Removal of outlier samples; (B) Selection of an appropriate soft threshold (β = 3, 0.87); (C-D) Construction of gene co-expression modules, resulting in a total of 16 modules; (E-F) Correlation analysis of modules with clinical features of SALI, where the blue and orange modules showed the highest correlation with SALI; (H-I) Screening of core genes from the blue and orange modules followed by Gene Ontology(GO) and Kyoto Encyclopedia of Genes and Genomes(KEGG) analysis, revealing that these genes are enriched in pathways related to inflammation, immunity, and angiogenesis.
The GS was calculated based on the correlation with gene expression profiles. Using a cutoff criterion of |GS| > 0.1, a total of 1386 genes were ultimately identified from these two modules (Fig 3E–G). Functional enrichment analysis revealed that these genes were also closely related to immune response, inflammatory response, and angiogenesis (Fig 3H–I).
These findings further corroborate that inflammation and angiogenesis may play pivotal roles in the onset and progression of acute lung injury.
Identification of Macrophage-Associated Angiogenesis-Related Genes(MARG) in SALI
We have discovered that immune response, inflammatory response, and angiogenesis are all involved in the onset and progression of SALI. Moreover, inflammation and angiogenesis are closely interrelated, and macrophages play a pivotal role in the immune and inflammatory processes of ALI.
Based on this, we obtained a set of macrophage – related genes (ARGs) and a set of angiogenesis – related genes (MRGs) from the GENECARDS database. By taking the intersection of these two gene sets, we identified 463 MARGs. Then, we took the intersection of these 463 genes with the DEGs and the genes identified through WGCNA. Ultimately, we obtained 43 overlapping genes. Compared with the Sham group, most of these genes exhibited higher expression levels in the SALI group (Fig 4A–B). GO and KEGG analyses of these 43 genes revealed that they remained closely associated with inflammatory and immune processes as well as relevant pathways (Fig 4C).
A set of macrophage-associated genes (ARG) and a set of angiogenesis-related genes (MRG) were obtained from GEENCARDS. These gene sets were intersected with differentially expressed gene and genes identified from WGCNA, resulting in a total of 43 MARG. (B) A heatmap shows the expression levels of these 43 genes in the sepsis – associated ALI group and the Sham group. (C) Gene Ontology and Kyoto Encyclopedia of Genes and Genomes analysis of the 43 genes.
Target gene screening based on machine learning
Machine learning approaches were employed to further precisely identify MARGs in SALI. The LASSO algorithm identified 7 key genes. The RF algorithm screened out the top 10 genes based on their importance rankings. Additionally, the Support Vector Machine (SVM) algorithm distinguished 13 genes. By taking the intersection of these gene lists, we ultimately obtained two target genes: Fcer1g and St3gal1 (Fig 5).
(A) Identification of 7 feature genes using the the Least Absolute Shrinkage and Selection Operator method; (B-C) Selection of the top 10 feature genes using the Random Forest method; (D) Identification of 13 feature genes using the Support Vector Machine method; (E-G) Receiver Operating Characteristic(ROC) curves of three machine learning methods; (H) Overlap of the three machine learning algorithms, resulting in two target genes: Fcer1g and St3gal1.
Validation of target genes
We proceeded to conduct further validation on the two ultimately identified target genes. Initially, it was observed that in the training set, the expression levels of Fcer1g and St3gal1 were elevated. The Receiver Operating Characteristic (ROC) curves demonstrated the favorable predictive performance of these two genes (Fig 6A).
(A) Expression and Receiver Operating Characteristic (ROC) curves of Fcer1g and St3gal1 in the training set. (B) Validation of Fcer1g and St3gal1 in an extra-pulmonary sepsis - associated acute lung injury (SALI) validation set, constructed by combining and batch-corrected GSE60088 and GSE130936, with expression and ROC curves shown. (C) Validation of Fcer1g and St3gal1 in an intra-pulmonary SALI validation set, constructed by combining and batch-corrected GSE1871 and GSE2411, with expression and ROC curves shown. (D) Validation of FCER1G and ST3GAL1 in human SALI blood samples (human SALI validation set, GSE32707), with expression and ROC curves shown.
To further verify the stability of the results, we carried out multi – faceted validation. Firstly, in the context of direct lung injury (extrapulmonary SALI), there was an increase in the expression levels of Fcer1g and St3gal1. The ROC curves indicated the excellent predictive ability of these two genes (Fig 6B). The same outcome was also obtained in the case of indirect lung injury (extrapulmonary SALI) (Fig 6C).
Finally, upon analyzing human samples from sepsis – induced ARDS, we found that in the blood of patients with sepsis – induced ARDS, the expression levels of FCER1G and ST3GAL1 were also elevated. The ROC curves revealed that these two genes still exhibited satisfactory predictive performance (Fig 6D).
Immune infiltration
We utilized the ImmuCellAI online platform to investigate immune cell infiltration in both lung tissues of mice with SALI and blood samples from affected patients. In mouse lung tissues, a marked reduction was observed in the quantities of B cells, NK cells, T cells, and plasma cells. Conversely, there was an increase in the numbers of macrophages, monocytes, and neutrophils, with both M1 and M2 macrophage subtypes showing elevated counts (Fig 7A).
(A) Immune cell infiltration in mouse lung tissues (training set) from SALI. (B) Relationship between Fcer1g, St3gal1, and immune cells. (C) Immune cell infiltration in human blood samples from SALI. 1: SALI group, 2: Sham group. (D-E) Correlation between Fcer1g, St3gal1, and Macrophages in both mouse lung tissues and human blood.
In human blood samples, the pattern of immune cell infiltration resembled that observed in mouse lung tissues, albeit with a noted decrease in monocyte numbers (Fig 7C). Further analysis revealed a close association between Fcer1g and St3gal1 and immune cells (Fig 7B). Notably, both genes exhibited a significant positive correlation with macrophages (Fig 7D–E).
scRNA-seq
GSE207651 comprises scRNA-seq data derived from lung tissues of mice with SALI and Sham. Following normalization, dimensionality reduction, and other preprocessing steps, the cells were clustered into 16 distinct clusters (Fig 8A). Based on the expression of marker genes, 11 cell subtypes were identified, including fibroblasts, macrophages, neutrophils, endothelial cells, T cells, smooth muscle cells, NK cells, epithelial cells, mesothelial cells, and B cells (Fig 8B). Compared to the sham group, the proportions of fibroblasts, macrophages, neutrophils, and mesothelial cells were increased in the SALI group, while the proportions of endothelial cells, T cells, B cells, smooth muscle cells, NK cells, and pericytes were decreased (Fig 8C–D). Further analysis revealed a complex regulatory network among various cell types in the lung tissue, with close interactions between immune cells and lung resident cells (Fig 8E–F). This mutual regulation between macrophages and endothelial cells suggests that crosstalk between inflammation and angiogenesis may be a crucial mechanism underlying the pathogenesis of SALI.
(A-B) Cells within the lung tissue were co-clustered into 16 clusters, which were identified as 11 distinct cell types. (C-D) Comparison of the distribution and differences in cell types between the sepsis - associated acute lung injury (SALI) and Sham groups. (E) Interaction analysis revealing the crosstalk among the 11 cell types. (F) Specific analysis of the interaction between Macrophages and Endothelial cells.
The expression of Fcer1g and St3gal1 in various types of cells.
The study revealed significant differences in the expression patterns of Fcer1g and St3gal1 across various cell types. Specifically, the expression level of Fcer1g was notably higher than that of St3gal1. Fcer1g was predominantly expressed in macrophages, neutrophils, and NK cells, whereas St3gal1 was mainly expressed in endothelial cells, mesothelial cells, and to a lesser extent, in macrophages (Fig 9A–C). The expression of Fcer1g was found to be upregulated in macrophages and neutrophils, but downregulated in NK cells. Conversely, the expression of St3gal1 was elevated in endothelial cells, mesothelial cells, and macrophages (Fig 9D-G).
(A-C) Expression profiles of Fcer1g and St3gal1 across various cell types. (D-E) Differential expression of Fcer1g between the sepsis - associated acute lung injury (SALI) group and the Sham group. (F-G) Differential expression of St3gal1 between the SALI group and the Sham group.
Expression of Fcer1g and St3gal1 in Macrophages
We observed an elevation in the expression of both Fcer1g and St3gal1 in the lungs of subjects with SALI. Further research revealed that both genes are expressed in macrophages, and their expression levels increase during acute lung injury. We further validated the expression of Fcer1g and St3gal1 in macrophages using PCR. After stimulating RAW264.7 cells with LPS and PBS, we found that, compared with the Sham group (PBS-stimulated), the expression levels of both Fcer1g and St3gal1 were significantly elevated in the LPS-treated group (Fig 10).
Using lipopolysaccharide (LPS) to stimulate RAW264.7 cells, the results showed that after LPS treatment, the expression of Fcer1g (A) and St3gal1 (B) was significantly increased.
Discussions
The research analysis indicates that inflammation, immune response, and angiogenesis are closely associated with the onset and progression of SALI. The pathophysiological mechanisms underlying SALI are highly complex, involving the activation and dysregulation of multiple overlapping and interacting pathways. These pathways encompass various aspects, including pulmonary and systemic injury, inflammatory processes, and coagulation cascades, among others [27]. Consequently, there is likely to be an interplay among immune response, inflammation, and angiogenesis, which in turn influences the occurrence and progression of SALI.
In our investigation, two macrophage-associated angiogenesis-related genes, Fcer1g and St3gal1, were identified through computational methods such as machine learning. This computational approach enabled us to screen a large amount of data and identify potential candidate genes that might be involved in the pathogenesis of SALI. However, it is crucial to distinguish between computational inference and experimental causality. Computational analysis can only suggest associations between genes and diseases based on statistical patterns in the data, but it cannot prove a causal relationship. Experimental validation is essential to confirm the causal roles of Fcer1g and St3gal1 in SALI. We further validated the expression of Fcer1g and St3gal1 in macrophages using PCR. After stimulating RAW264.7 cells with LPS and PBS, we found that, compared with the Sham group (PBS-stimulated), the expression levels of both Fcer1g and St3gal1 were significantly elevated in the LPS-treated group
Macrophages represent critical immune cells within the lungs, with the capacity to recognize and eliminate foreign pathogens and necrotic cells. Currently, two well – characterized primary phenotypes of macrophages have been identified: the inflammatory or classically activated (M1) macrophages and the healing or alternatively activated (M2) macrophages [28]. M1 macrophages play a pivotal role in inflammation and immune responses. They produce high levels of inflammatory mediators such as IL – 1β, IL – 6, and TNF – α through the activation of the NF - κB signaling pathway [29]. These cytokines can bind to their respective receptors on endothelial cells, such as IL – 1 receptor (IL - 1R), IL – 6 receptor (IL - 6R), and TNF – α receptor (TNFR), triggering a series of downstream signaling events that lead to increased vascular permeability, recruitment of more immune cells, and promotion of angiogenesis under certain conditions [30]. However, an excessive activation of M1 macrophages can trigger uncontrolled inflammatory responses, ultimately exacerbating tissue damage [31–33].
On the other hand, M2 macrophages primarily secrete high levels of anti – inflammatory factors, including IL – 10 and TGF – β. IL – 10 can bind to its receptor on immune cells and endothelial cells, inhibiting the production of pro – inflammatory cytokines and promoting an anti – inflammatory microenvironment [34]. TGF – β, through the TGF – β/Smad signaling pathway, plays a crucial role in tissue repair and fibrosis. It can stimulate the proliferation and differentiation of fibroblasts and promote the deposition of extracellular matrix components [35]. Nevertheless, an over – activation of M2 macrophages can result in aberrant tissue repair processes, eventually contributing to the development of lung fibrosis [36,37].
Macrophages also play a substantial role in angiogenesis. They promote angiogenesis and maintain vascular stability through the secretion of growth factors, proteases, and other factors. However, under pathological conditions, macrophages may exert adverse effects on angiogenesis. Studies have shown that M1 macrophages can inhibit VEGF – mediated angiogenesis, while M2 macrophages promote endothelial cell migration and angiogenesis through direct interactions with endothelial cells [38]. This mechanism may be the underlying cause of the abnormal repair processes in the lungs during the later stages of ALI, leading to the onset of lung fibrosis [39–41].
Therefore, a comprehensive exploration of the mechanisms and interactions between macrophages and angiogenesis holds significant potential for the diagnosis and treatment of SALI.
In our investigation, two macrophage-associated angiogenesis-related genes, Fcer1g and St3gal1, were identified through methods such as machine learning. Notably, we observed elevated expression levels of these genes in both lung tissues and blood samples obtained from patients with SALI. scRNA - seq and RT – PCR analyses further revealed that both Fcer1g and St3gal1 are expressed within macrophages and exhibit a positive correlation with macrophage abundance.
The protein encoded by Fcer1g is the high – affinity immunoglobulin ε receptor γ chain precursor, which serves as an essential component of the IgG high – affinity receptor [42]. Fcer1g has garnered significant attention in the field of oncology due to its widespread upregulation across most tumor types. This upregulation is often associated with poor prognosis, heightened cellular proliferation, and increased metastatic potential. Moreover, Fcer1g is intimately intertwined with the tumor microenvironment and tumor immunology. It is predominantly expressed in monocytes/macrophages within the tumor microenvironment, where it likely plays a pivotal role in modulating immune responses and tumor progression [43]. Beyond its involvement in tumors, Fcer1g may also be a key player in other immune – related diseases. For instance, in group 3 innate lymphoid cells (ILC3s), Fcer1g is highly expressed and functions to stabilize the abundance of membrane receptors, such as CD16 and NKp46. This stabilization, in turn, promotes the secretion of IL - 17A cytokine, which is crucial for anti – infectious immunity by enhancing the host's defense capabilities against pathogens [44]. In the context of pathological conditions like tumors and autoimmune diseases, macrophage infiltration and function are tightly regulated. During these processes, the expression level of Fcer1g may undergo dynamic changes, thereby influencing macrophage recruitment and activity. In clear cell renal cell carcinoma (ccRCC), for example, high expression of Fcer1g has been correlated with increased macrophage infiltration. This overexpression contributes to an unfavorable prognosis by modulating tumor immunity, potentially through the regulation of macrophage polarization and cytokine secretion [45].
Furthermore, Fcer1g has been implicated in a range of other diseases, including chronic rhinosinusitis, asthma, renal diseases, cardiovascular disorders, and neurological conditions [46–49]. Its involvement in these diverse pathologies suggests that Fcer1g may have a broad impact on immune cell function and the regulation of inflammatory responses. In addition, Fcer1g may also play a role in fibrotic diseases. Aberrant expression of Fcer1g has been observed in both liver and kidney fibrosis, indicating that it could potentially contribute to the fibrotic process indirectly by modulating the function of the immune system. For instance, Fcer1g may influence the activation and polarization of macrophages, which are known to play a critical role in fibrosis development [50,51]. Moreover, Fcer1g may impact endothelial cell function, which has been linked to coagulation disorders associated with acute pancreatitis [52]. The interaction between immune cells and endothelial cells is a complex process that is crucial for maintaining vascular homeostasis. Fcer1g may participate in this interplay by modulating the signaling pathways between immune cells and endothelial cells, thereby influencing endothelial cell function and contributing to angiogenesis processes. This, in turn, may play a significant role in the pathogenesis of various diseases, including SALI.
The protein encoded by St3gal1 is a type II membrane protein belonging to the glycosyltransferase family 29. It functions primarily as a catalyst, transferring sialic acid from CMP – sialic acid to galactose – containing substrates, thereby participating in the glycosylation modification of proteins. This glycosylation modification has a profound impact on protein function and stability, influencing processes such as protein folding, cell – cell recognition, and signal transduction [53]. Studies have demonstrated that St3gal1 is associated with the inhibition of CD8 T cell cytotoxicity induced by α2,3 – sialylation in tumor - associated macrophages (TAMs). In the tumor microenvironment, the upregulation of St3gal1 in tumor cells promotes the polarization of M2 - like macrophages. These M2 - like macrophages, in turn, produce higher levels of IL – 6 and IL – 10, which can suppress anti – tumor immune responses and contribute to tumor progression by influencing the tumor immune microenvironment [54]. Furthermore, St3gal1 silencing has been shown to inhibit TGF - β1 signaling and angiogenesis. In human umbilical vein endothelial cells, St3gal1 silencing suppresses the expression of angiogenesis – related genes and reduces the activation of Smad2 and Smad3, which are key signaling molecules in the TGF - β1 pathway [55]. This suggests that St3gal1 may participate in the signaling between endothelial cells and other cells, such as immune cells and tumor cells, by modulating the sialylation level of glycoproteins. By doing so, it indirectly influences the role of endothelial cells in processes such as inflammatory responses, immune reactions, and tumor metastasis, thereby affecting angiogenesis.
Despite the valuable insights gained from mouse models, it is essential to acknowledge the significant species differences between mice and humans. The immune systems of mice and humans have distinct characteristics in terms of immune cell populations, cytokine profiles, and signaling pathways. For example, the proportion and function of different macrophage subsets may vary between the two species. In mice, the M1/M2 macrophage polarization is more clearly defined, while in humans, there is a more heterogeneous population of macrophages with intermediate phenotypes [56]. Moreover, the cytokine – receptor interactions and downstream signaling pathways may also differ. An earlier review discussed the similarities and differences between mouse and human macrophages, and raised the possibility that human tissue-resident macrophages might be more reliant on bone marrow-derived monocytes than those in mice [57]. These differences could potentially lead to discrepancies in the expression patterns and functional roles of genes such as Fcer1g and St3gal1 between mice and humans. Therefore, the results obtained from mouse models may not directly translate to human SALI, necessitating further validation in human – derived samples.
This study has several limitations that need to be addressed. Firstly, the dataset predominantly utilized for this research was sourced from mouse models. Although mouse models have been widely used in biomedical research, they may not fully recapitulate the complexities of human diseases. The absence of human lung tissue samples in our study is a major limitation. Human lung tissues can provide more relevant information about the gene expression patterns and cellular interactions in the context of SALI. Without human samples, we are unable to determine whether the observed changes in Fcer1g and St3gal1 expression are truly representative of the human disease state. Secondly, although we have implemented rigorous data pre – processing steps to mitigate batch effects among different datasets, the relatively small sample size remains a concern. Batch effects can arise from various sources, such as differences in experimental conditions, sample collection methods, or data processing pipelines. Even with careful normalization and correction techniques, there is a possibility that some residual errors may persist, especially when dealing with a limited number of samples. A larger sample size would provide more statistical power and reduce the impact of random variations, thereby enhancing the reliability and reproducibility of our findings. Furthermore, the absence of clinical data from human subjects in the dataset represents a critical limitation. Clinical data, including information on disease severity, clinical outcomes, and patient demographics, are crucial for determining the clinical relevance of Fcer1g and St3gal1 in SALI. Without this information, we are unable to assess whether the observed changes in gene expression are associated with specific clinical phenotypes or outcomes in human patients. For instance, it remains unclear whether high expression levels of Fcer1g and St3gal1 correlate with more severe disease progression, poorer treatment responses, or increased mortality rates in SALI patients. Lastly, this study primarily relied on bioinformatics methods to explore the correlation between Fcer1g, St3gal1, and SALI. While bioinformatics approaches can provide valuable insights into gene – disease associations and potential molecular mechanisms, they lack the experimental validation necessary to confirm the specific roles of these genes in the pathogenesis of SALI. In vitro and in vivo experiments are essential to elucidate the precise molecular pathways through which Fcer1g and St3gal1 interact with macrophages and endothelial cells, and how these interactions ultimately influence endothelial cell function and angiogenesis.
In summary, our study has illuminated the pivotal role of the crosstalk between inflammation and angiogenesis in the pathogenesis of SALI. Through comprehensive bioinformatics analysis, we have identified two macrophage – associated angiogenesis – related genes, Fcer1g and St3gal1, which may participate in the signaling communication between macrophages and endothelial cells. These genes have the potential to influence endothelial cell function and, consequently, modulate angiogenesis, thereby playing a crucial role in the onset and progression of SALI by contributing to SALI-associated signatures.
Given the potential significance of Fcer1g and St3gal1 in relation to SALI, they may serve as promising biomarkers for disease diagnosis, prognosis, and monitoring, reflecting SALI-associated signatures rather than being specific solely to SALI. Elevated expression levels of these genes in lung tissues or blood samples could potentially indicate the presence of SALI or predict disease severity and clinical outcomes, aligning with SALI-associated molecular patterns. Moreover, targeting these genes or their downstream signaling pathways may offer novel therapeutic strategies for the treatment of SALI. For instance, inhibitors of Fcer1g or St3gal1 could be developed to disrupt the pro – inflammatory and pro – angiogenic signals in SALI, thereby attenuating tissue damage and promoting lung repair.
To further validate our findings and advance the understanding of the roles of Fcer1g and St3gal1 in the context of SALI-associated signatures, we plan to conduct additional experiments in the future. These experiments will include in vitro studies using human lung-derived cell lines and primary cells, as well as in vivo studies using animal models that more closely mimic the conditions associated with SALI in humans. By combining bioinformatics approaches with experimental validation, we aim to provide a more comprehensive and in-depth understanding of the molecular mechanisms underlying SALI-associated signatures, ultimately paving the way for the development of more effective diagnostic and therapeutic interventions for this devastating disease.
References
- 1. Incorrect Language and List in the Supplement. JAMA. 2016;316(3):350. pmid:27434458
- 2. Bos LDJ, Ware LB. Acute respiratory distress syndrome: causes, pathophysiology, and phenotypes. Lancet. 2022;400(10358):1145–56. pmid:36070787
- 3. Henke C, Fiegel V, Peterson M, Wick M, Knighton D, McCarthy J, et al. Identification and partial characterization of angiogenesis bioactivity in the lower respiratory tract after acute lung injury. J Clin Invest. 1991;88(4):1386–95. pmid:1717512
- 4. Leng L, Cao R, Ma J, Mou D, Zhu Y, Li W, et al. Pathological features of COVID-19-associated lung injury: a preliminary proteomics report based on clinical samples. Signal Transduct Target Ther. 2020;5(1):240. pmid:33060566
- 5. Thébaud B, Ladha F, Michelakis ED, Sawicka M, Thurston G, Eaton F, et al. Vascular endothelial growth factor gene therapy increases survival, promotes lung angiogenesis, and prevents alveolar damage in hyperoxia-induced lung injury: evidence that angiogenesis participates in alveolarization. Circulation. 2005;112(16):2477–86. pmid:16230500
- 6. Huang L-T, Chou H-C, Chen C-M. Roxadustat attenuates hyperoxia-induced lung injury by upregulating proangiogenic factors in newborn mice. Pediatr Neonatol. 2021;62(4):369–78. pmid:33865748
- 7. Kou X, Sun Y, Li S, Bian W, Liu Z, Zhang D, et al. Pharmacology Study of the Multiple Angiogenesis Inhibitor RC28-E on Anti-Fibrosis in a Chemically Induced Lung Injury Model. Biomolecules. 2019;9(11):644. pmid:31652997
- 8. Garantziotis S, Zudaire E, Trempus CS, Hollingsworth JW, Jiang D, Lancaster LH, et al. Serum inter-alpha-trypsin inhibitor and matrix hyaluronan promote angiogenesis in fibrotic lung injury. Am J Respir Crit Care Med. 2008;178(9):939–47. pmid:18703791
- 9. Gordon S, Martinez FO. Alternative activation of macrophages: mechanism and functions. Immunity. 2010;32(5):593–604. pmid:20510870
- 10. Herold S, Mayer K, Lohmeyer J. Acute lung injury: how macrophages orchestrate resolution of inflammation and tissue repair. Front Immunol. 2011;2:65. pmid:22566854
- 11. Wynn TA, Chawla A, Pollard JW. Macrophage biology in development, homeostasis and disease. Nature. 2013;496(7446):445–55. pmid:23619691
- 12. Ferrara N, Gerber H-P, LeCouter J. The biology of VEGF and its receptors. Nat Med. 2003;9(6):669–76. pmid:12778165
- 13. Wang Z, Wu Y, Wang J, Zhang C, Yan H, Zhu M, et al. Effect of Resveratrol on Modulation of Endothelial Cells and Macrophages for Rapid Vascular Regeneration from Electrospun Poly(ε-caprolactone) Scaffolds. ACS Appl Mater Interfaces. 2017;9(23):19541–51. pmid:28539044
- 14. Amano H, Mastui Y, Ito Y, Shibata Y, Betto T, Eshima K, et al. The role of vascular endothelial growth factor receptor 1 tyrosine kinase signaling in bleomycin-induced pulmonary fibrosis. Biomed Pharmacother. 2019;117:109067. pmid:31176171
- 15. Abadie Y, Bregeon F, Papazian L, Lange F, Chailley-Heu B, Thomas P, et al. Decreased VEGF concentration in lung tissue and vascular injury during ARDS. Eur Respir J. 2005;25(1):139–46. pmid:15640335
- 16. Hattori K, Matsuda N, Hattori Y. Vascular hyperpermeable molecules potentially contributing to the development of pulmonary edema in sepsis-associated ARDS. Nihon Yakurigaku Zasshi. 2022;157(4):226–31. pmid:35781449
- 17. Johnson WE, Li C, Rabinovic A. Adjusting batch effects in microarray expression data using empirical Bayes methods. Biostatistics. 2007;8(1):118–27. pmid:16632515
- 18. Jamesdaniel S, Hu B, Kermany MH, Jiang H, Ding D, Coling D, et al. Noise induced changes in the expression of p38/MAPK signaling proteins in the sensory epithelium of the inner ear. J Proteomics. 2011;75(2):410–24. pmid:21871588
- 19. Tang J, Kong D, Cui Q, Wang K, Zhang D, Gong Y, et al. Prognostic Genes of Breast Cancer Identified by Gene Co-expression Network Analysis. Front Oncol. 2018;8:374. pmid:30254986
- 20. Yang Y-Y, Gao Z-X, Mao Z-H, Liu D-W, Liu Z-S, Wu P. Identification of ULK1 as a novel mitophagy-related gene in diabetic nephropathy. Front Endocrinol (Lausanne). 2023;13:1079465. pmid:36743936
- 21. Liu L, Wang M, Yu S. Identification of Common Angiogenesis Marker Genes in Chronic Lung Diseases and Their Relationship with Immune Infiltration Based on Bioinformatics Approaches. Biomedicines. 2025;13(2):331. pmid:40002743
- 22. Miao Y-R, Xia M, Luo M, Luo T, Yang M, Guo A-Y. ImmuCellAI-mouse: a tool for comprehensive prediction of mouse immune cell abundance and immune microenvironment depiction. Bioinformatics. 2022;38(3):785–91. pmid:34636837
- 23. Korsunsky I, Millard N, Fan J, Slowikowski K, Zhang F, Wei K, et al. Fast, sensitive and accurate integration of single-cell data with Harmony. Nat Methods. 2019;16(12):1289–96. pmid:31740819
- 24. Hu C, Li T, Xu Y, Zhang X, Li F, Bai J, et al. CellMarker 2.0: an updated database of manually curated cell markers in human/mouse and web tools based on scRNA-seq data. Nucleic Acids Res. 2023;51(D1):D870–6. pmid:36300619
- 25. Franzén O, Gan L-M, Björkegren JLM. PanglaoDB: a web server for exploration of mouse and human single-cell RNA sequencing data. Database (Oxford). 2019;2019:baz046. pmid:30951143
- 26. Shen W, Song Z, Zhong X, Huang M, Shen D, Gao P, et al. Sangerbox: A comprehensive, interaction-friendly clinical bioinformatics analysis platform. Imeta. 2022;1(3):e36. pmid:38868713
- 27. Bellani G, Laffey JG, Pham T, Fan E, Brochard L, Esteban A, et al. Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. JAMA. 2016;315(8):788–800. pmid:26903337
- 28. Chen X, Tang J, Shuai W, Meng J, Feng J, Han Z. Macrophage polarization and its role in the pathogenesis of acute lung injury/acute respiratory distress syndrome. Inflamm Res. 2020;69(9):883–95. pmid:32647933
- 29. Li Z, Li X, Shen R, Wang Y, Yu J, Pan C, et al. Interleukin-38 Ameliorates Atherosclerosis by Inhibiting Macrophage M1-like Polarization and Apoptosis. Biomolecules. 2025;15(12):1741. pmid:41463394
- 30. Pober JS, Sessa WC. Evolving functions of endothelial cells in inflammation. Nat Rev Immunol. 2007;7(10):803–15. pmid:17893694
- 31. Xu Y, Zhang C, Cai D, Zhu R, Cao Y. Exosomal miR-155-5p drives widespread macrophage M1 polarization in hypervirulent Klebsiella pneumoniae-induced acute lung injury via the MSK1/p38-MAPK axis. Cell Mol Biol Lett. 2023;28(1):92. pmid:37953267
- 32. Jiao Y, Zhang T, Zhang C, Ji H, Tong X, Xia R, et al. Exosomal miR-30d-5p of neutrophils induces M1 macrophage polarization and primes macrophage pyroptosis in sepsis-related acute lung injury. Crit Care. 2021;25(1):356. pmid:34641966
- 33. Grommes J, Soehnlein O. Contribution of neutrophils to acute lung injury. Mol Med. 2011;17(3–4):293–307. pmid:21046059
- 34. Moore KW, de Waal Malefyt R, Coffman RL, O’Garra A. Interleukin-10 and the interleukin-10 receptor. Annu Rev Immunol. 2001;19:683–765. pmid:11244051
- 35. Massagué J. TGF-beta signal transduction. Annu Rev Biochem. 1998;67:753–91. pmid:9759503
- 36. Li C, Liu J, Zhang C, Cao L, Zou F, Zhang Z. Dihydroquercetin (DHQ) ameliorates LPS-induced acute lung injury by regulating macrophage M2 polarization through IRF4/miR-132-3p/FBXW7 axis. Pulm Pharmacol Ther. 2023;83:102249. pmid:37648017
- 37. Zhou Y, Peng H, Sun H, Peng X, Tang C, Gan Y, et al. Chitinase 3-like 1 suppresses injury and promotes fibroproliferative responses in Mammalian lung fibrosis. Sci Transl Med. 2014;6(240):240ra76. pmid:24920662
- 38. Tylek T, Wong J, Vaughan AE, Spiller KL. Biomaterial-mediated intracellular control of macrophages for cell therapy in pro-inflammatory and pro-fibrotic conditions. Biomaterials. 2024;308:122545. pmid:38547831
- 39. Yin L, Fan Z, Liu P. Anemoside A3 activates TLR4-dependent M1-phenotype macrophage polarization to represses breast tumor growth and angiogenesis. Toxicol Appl Pharmacol. 2021 Dec 1;432:115755.
- 40. Corrigendum. J Cell Mol Med. 2022;26(5):1727–8. pmid:35253381
- 41. Zhang J, Muri J, Fitzgerald G, Gorski T, Gianni-Barrera R, Masschelein E, et al. Endothelial Lactate Controls Muscle Regeneration from Ischemia by Inducing M2-like Macrophage Polarization. Cell Metab. 2020;31(6):1136-1153.e7. pmid:32492393
- 42. Küster H, Thompson H, Kinet JP. Characterization and expression of the gene for the human Fc receptor gamma subunit. Definition of a new gene family. J Biol Chem. 1990;265(11):6448–52. pmid:2138619
- 43. Yang R, Chen Z, Liang L, Ao S, Zhang J, Chang Z, et al. Fc Fragment of IgE Receptor Ig (FCER1G) acts as a key gene involved in cancer immune infiltration and tumour microenvironment. Immunology. 2023;168(2):302–19. pmid:36054819
- 44. Huang C, Zhu W, Li Q, Lei Y, Chen X, Liu S, et al. Antibody Fc-receptor FcεR1γ stabilizes cell surface receptors in group 3 innate lymphoid cells and promotes anti-infection immunity. Nat Commun. 2024;15(1):5981. pmid:39013884
- 45. Dong K, Chen W, Pan X, Wang H, Sun Y, Qian C, et al. FCER1G positively relates to macrophage infiltration in clear cell renal cell carcinoma and contributes to unfavorable prognosis by regulating tumor immunity. BMC Cancer. 2022;22(1):140. pmid:35120484
- 46. Zhu Y, Sun X, Tan S, Luo C, Zhou J, Zhang S, et al. M2 macrophage-related gene signature in chronic rhinosinusitis with nasal polyps. Front Immunol. 2022;13:1047930. pmid:36466903
- 47. Gu Y, Zhang X, Li H, Wang R, Jin C, Wang J, et al. Novel subsets of peripheral immune cells associated with promoting stroke recovery in mice. CNS Neurosci Ther. 2024;30(4):e14518. pmid:37905680
- 48. Shao X, Shi Y, Wang Y, Zhang L, Bai P, Wang J, et al. Single-Cell Sequencing Reveals the Expression of Immune-Related Genes in Macrophages of Diabetic Kidney Disease. Inflammation. 2024;47(1):227–43. pmid:37777674
- 49. Yu M, Eckart MR, Morgan AA, Mukai K, Butte AJ, Tsai M, et al. Identification of an IFN-γ/mast cell axis in a mouse model of chronic asthma. J Clin Invest. 2011;121(8):3133–43. pmid:21737883
- 50. Sung P-S, Kim C-M, Cha J-H, Park J-Y, Yu Y-S, Wang H-J, et al. A Unique Immune-Related Gene Signature Represents Advanced Liver Fibrosis and Reveals Potential Therapeutic Targets. Biomedicines. 2022;10(1):180. pmid:35052861
- 51. Yuan Y, Xiong X, Li L, Luo P. Novel targets in renal fibrosis based on bioinformatic analysis. Front Genet. 2022;13:1046854. pmid:36523757
- 52. Hu S, Lin T, Chen Y, Guo Y, Sun X, Shi L, et al. NLRC4-mediated pyroptosis was involved in coagulation disorders of acute pancreatitis. J Gene Med. 2024;26(4):e3683. pmid:38571451
- 53. Lin W-D, Fan T-C, Hung J-T, Yeo H-L, Wang S-H, Kuo C-W, et al. Sialylation of CD55 by ST3GAL1 Facilitates Immune Evasion in Cancer. Cancer Immunol Res. 2021;9(1):113–22. pmid:33177111
- 54. Zou Y, Guo S, Liao Y, Chen W, Chen Z, Chen J, et al. Ceramide metabolism-related prognostic signature and immunosuppressive function of ST3GAL1 in osteosarcoma. Transl Oncol. 2024;40:101840. pmid:38029509
- 55. Yeo HL, Fan T-C, Lin R-J, Yu J-C, Liao G-S, Chen ES-W, et al. Sialylation of vasorin by ST3Gal1 facilitates TGF-β1-mediated tumor angiogenesis and progression. Int J Cancer. 2019;144(8):1996–2007. pmid:30252131
- 56. Fischer V, Ignatius A, Schmidt-Bleek K, Duda G, Haffner-Luntzer M. Using artificial intelligence-based software for an unbiased discrimination of immune cell subtypes in the fracture hematoma and bone marrow of non-osteoporotic and osteoporotic mice. PLoS One. 2025;20(4):e0322542. pmid:40299871
- 57. Gallerand A, Han J, Ivanov S, Randolph GJ. Mouse and human macrophages and their roles in cardiovascular health and disease. Nat Cardiovasc Res. 2024;3(12):1424–37. pmid:39604762