Application of s100a11 as a biomarker in preparation of products for diagnosis, prevention, treatment of liver transplant rejection

Abstract

The application belongs to the technical field of biological medicine, and provides application of S100A11 as a biomarker in preparation of a product for diagnosing, preventing and treating liver transplant rejection. The product comprises a reagent or kit for detecting the expression level of S100A11 gene. The application screens S100A11 as a key biomarker candidate with important diagnostic potential. Single-cell RNA sequencing and BULK-RNA sequencing are used to confirm that S100A11 is elevated in liver transplant rejection patients and is mainly expressed in mononuclear macrophages. The results of liver transplant patient tissue, liver transplant model mice and in vitro cell experiments are verified, and it is found that the proportion of S100A11+ macrophages in the liver and peripheral blood is increased, and is related to the rejection activity index. The application points out the potential of S100A11 as a biomarker for liver transplant rejection, and provides a basis for developing personalized and effective treatment strategies to improve patient prognosis.

Description

Technical Field

[0001] This invention belongs to the field of biomedical technology, specifically relating to the application of S100A11 as a biomarker in the preparation of products for the diagnosis, prevention, and treatment of liver transplant rejection. Background Technology

[0002] Liver transplantation is the ultimate treatment for various end-stage liver diseases, significantly improving the survival rate of liver disease patients. Although patients routinely take immunosuppressants after surgery to reduce the risk of graft rejection in most cases, a portion of patients still experience post-transplant rejection. Statistics show that approximately 5%-10% of patients experience acute rejection, which progresses to liver failure and has a very high mortality rate, accounting for nearly 10%-20% of post-transplant deaths. Therefore, timely detection and prevention of rejection are crucial for liver transplant patients. However, current diagnosis of rejection still relies on pathological examination after liver biopsy, which is invasive and difficult to monitor long-term. There is an urgent need to find specific biomarkers to guide clinical practice. Furthermore, treatment for rejection is limited to prophylactic or therapeutic immunosuppressants, which have significant side effects and limited efficacy, necessitating the exploration of effective new drugs. Therefore, finding high-quality biomarkers and therapeutic targets is the future direction for reducing liver transplant rejection.

[0003] Liver transplant immune rejection is intricately linked to hepatic immune cells. Different immune cells play a crucial role in post-transplant rejection through complex interactions. The activation and differentiation of immune cells are inseparable from their metabolic status. Multiple metabolic pathways can affect immune cell function, with lipid metabolism playing a significant role in transplant rejection. A high-throughput metabolomics study on hematopoietic stem cell transplantation showed that the proportion of lipid metabolites was highest (46.18%) in recipient plasma samples 90 days post-transplantation. However, the influence and mechanism of lipid metabolism in liver transplant rejection remain unclear. Therefore, it is necessary to identify key lipid metabolism genes affecting liver transplant rejection to guide treatment plans and develop targeted drugs. Summary of the Invention

[0004] This invention provides the application of S100A11 as a biomarker in the preparation of products for the diagnosis, prevention, and treatment of liver transplant rejection.

[0005] The present invention is achieved by the following technical solution: the application of S100A11 as a biomarker in the preparation of products for diagnosing liver transplant rejection.

[0006] The products include reagents or kits for detecting the expression level of the S100A11 gene.

[0007] The reagents or kits include primers or probes for detecting S100A11 mRNA.

[0008] The reagents or kits include those for detecting the expression level of S100A11 in mononuclear macrophages in the liver and peripheral blood.

[0009] The mononuclear macrophages showed high expression of S100A11, increased ACSL1 expression, CD8+ T cell infiltration, liver transplant rejection, and a positive correlation between S100A11 expression and rejection response.

[0010] This invention also provides the application of the S100A11 gene or protein as a target in the preparation of drugs for the prevention and / or treatment of liver transplant rejection.

[0011] This invention also provides the use of S100A11 inhibitors in the preparation of medicaments for the prevention and / or treatment of liver transplant rejection.

[0012] This invention also provides the application of the S100A11 gene or protein in screening drugs for the treatment of liver transplant rejection.

[0013] This invention uses single-cell transcriptomic data from liver tissue and blood after human liver transplantation to study lipid metabolism. It observes lipid metabolism status after liver transplantation and identifies cell types associated with lipid metabolism. Key genes related to lipid metabolism after liver transplantation are screened to clarify their diagnostic value for rejection. The expression changes of key genes in specific cell types are studied in the rejection group to explore how these key genes influence rejection status by regulating lipid metabolism.

[0014] This invention analyzes a single-cell RNA sequencing dataset from the NGDC Genome Sequence Archive on liver transplantation. Five algorithms—AUCell, UCell, AddModuleScore, ssGSEA, and Singscore—were used to calculate cellular lipid metabolism scores, identifying 21 characteristic lipid metabolism genes after liver transplantation. Subsequently, five machine learning algorithms—gradient boosting, extreme gradient boosting, random forest, minimum absolute contraction and selection operator, and Boruta—were used to screen for S100A11 as a key biomarker candidate with significant diagnostic potential. Single-cell RNA sequencing and BULK-RNA sequencing confirmed that S100A11 was elevated in liver transplant rejection patients, primarily expressed in monocytes and macrophages. Subsequent pseudo-time series analysis revealed a gradual decrease in S100A11 levels in monocytes and macrophages, and cell communication showed significant activation of S100A11+ interactions with other cells. The results were validated using liver transplant patient tissues, liver transplant model mice, and in vitro cell experiments, revealing an increased proportion of S100A11+ macrophages in the liver and peripheral blood, correlated with the rejection activity index. This invention highlights the potential of S100A11 as a biomarker for liver transplant rejection, providing a foundation for developing personalized and effective treatment strategies to improve patient outcomes. Attached Figure Description

[0015] Figure 1 The figure shows the single-cell transcriptome characteristics after liver transplantation. Figure a shows the distribution of 143,164 single cells after liver transplantation, divided into 22 subpopulations. Figure b shows the cell types after liver transplantation, divided into 12 groups. Figure c shows the cell distribution in the non-rejection and rejection groups. Figure d shows the expression of characteristic markers for different cell types. Figure e shows the percentage of cell types in the non-rejection and rejection groups. Figure f shows the percentage of cell types in different samples. Figure 2 This figure shows the expression of characteristic genes after liver transplantation. In the figure: a) distribution of different samples after liver transplantation; b) distribution of cell types from blood and liver samples; c) bubble chart showing the expression of characteristic genes in different subpopulations; d) violin plot showing the expression of characteristic genes in different subpopulations; e) UMAP plot showing the expression of characteristic genes in different subpopulations; f) the proportion of each cell type in each sample. Figure 3 This study assesses the correlation between lipid metabolism and liver transplantation. Figure a shows the correlation between different cell types and lipid metabolism in single-cell data after liver transplantation, assessed using five methods: AUCell, UCell, AddModuleScore, ssGSEA, and Singscore. Figure b shows the distribution of the overall lipid metabolism correlation score in single-cell data after liver transplantation. Figure c compares lipid metabolism scores in the non-rejection and rejection groups, using the Wilcoxon test. Figure d compares lipid metabolism scores in different cell types between the non-rejection and rejection groups, using the Wilcoxon test. *** represents p < 0.001, ns represents p > 0.05. Figure e shows the distribution of cells divided into upregulated and downregulated lipid metabolism groups based on the mean lipid metabolism score, and the distribution of these groups in single-cell data after liver transplantation. Figure f shows differentially expressed genes between the upregulated and downregulated groups; yellow dots represent highly expressed genes in the upregulated group, purple dots represent highly expressed genes in the downregulated group, and gray dots represent genes with no significant difference in expression between the two groups. The figure highlights 21 genes highly expressed in the upregulated group. g represents genes significantly associated with lipid metabolism-AUCell scores selected using correlation analysis. The left figure shows the top 300 genes related to the disease, and the right figure shows the correlation coefficients of the top 130 genes. h represents the intersection of the top 300 correlated genes and the top 542 differentially expressed genes in the upregulated group, yielding a characteristic gene set of 21 genes. i represents the KEGG enrichment analysis of the characteristic genes. Figure 4The figures show the results of different lipid metabolism scores after liver transplantation. Figure a shows the distribution of lipid metabolism scores after liver transplantation assessed by five methods: AUCell, UCell, AddModuleScore, ssGSEA, and Singscore. Figure b shows the difference in lipid metabolism scores between the non-rejection and rejection groups after liver transplantation assessed by the five methods. Wilcoxon test was used. Figure c shows the difference in lipid metabolism scores between the non-rejection and rejection groups after liver transplantation assessed by the five methods: AUCell, UCell, AddModuleScore, ssGSEA, and Singscore. Wilcoxon test was used. ns represents p>0.05, * represents p<0.01-0.05, ** represents p<0.001-0.01, and *** represents p<0.001. Figure 5 This refers to machine learning algorithms; in the figure: a) the relationship between Bernoullievability and iteration in gradient boosting machines; b) the relationship between the number of trees and error rate in random forests. The x-axis represents the number of decision trees, and the y-axis represents the error rate; c) the importance of features calculated using the Boruta algorithm; d) the relevance of the optimal target gene selected by five machine learning algorithms. Figure 6The optimal target gene S100A11 was screened using various machine learning algorithms, and its validity was validated in transcriptomic and single-cell transcriptomic data. In the figures: a) relative importance of feature genes calculated by gradient boosting machine; b) gain value of feature genes by XGBoost; c) importance of feature genes calculated by Random forest; d) results of feature gene screening by LASSON; e) optimal target gene selected by five machine learning algorithms; f) comparison of S100A11 expression in non-rejection and rejection groups in GSE145780 liver transplant data. Wilcoxon test was used, *** represents p<0.001; g) ROC curves show the diagnostic value of S100A11 in GSE145780; h) expression of S100A11 in different cell types in single-cell sequencing after liver transplantation. The violin plot shows S100A11 expression levels, and the heatmap shows the standardized z-score. *** represents p<0.001. i represents the distribution of S100A11 expression in single-cell sequencing after liver transplantation. j represents the distribution density of S100A11-highly-expressing cells in single-cell sequencing after liver transplantation. k represents the expression of S100A11 in mononuclear macrophages in the non-rejection and rejection groups. RF, Random forest; GBM, Gradient boosting machine; DT, Decision Trees. Wilcoxon test was used, *** represents p<0.001. ROC, Receiver operating characteristic. AUC, The area under the ROC curve; Figure 7 The figures show the co-expression of S100A11 and marker genes. In the figures: a) The top image shows the distribution of S100A11 in monocytes and macrophages. The bottom image uses color blocks: S100A11 is orange, marker genes are purple, and co-expression is pink. b) The top image shows the distribution of the M1 macrophage marker gene TNF in monocytes and macrophages. The bottom image shows the co-expression and distribution of S100A11 and TNF in monocytes and macrophages. c) The top image shows the distribution of the M2 macrophage marker gene MRC1 in monocytes and macrophages. The bottom image shows the co-expression and distribution of S100A11 and MRC1 in monocytes and macrophages. d) The top image shows the distribution of the Kupffer cell marker gene CLEC4F in monocytes and macrophages. The bottom image shows the co-expression and distribution of S100A11 and CLEC4F in monocytes and macrophages. Figure 8The figures show the characteristics of S100A11+ monocytes / macrophages. In the figures: a) GO analysis of S100A11+Mono / Mφ, showing the top 15 pathways. b) GSEA analysis of S100A11+Mono / Mφ using the GO dataset, showing two pathways highly relevant to this study. c) KEGG analysis of S100A11+Mono / Mφ, showing the top 15 pathways. d) GSEA analysis of S100A11+Mono / Mφ using the KEGG dataset, showing two pathways highly relevant to this study. e) Changes in other genes after simulated S100A11 knockdown in monocytes / macrophages. f) The intersection of genes highly expressed in S100A11+Mono / Mφ, differentially expressed genes after S100A11 knockdown in monocytes / macrophages, and lipid metabolism genes yields one gene. g represents the expression of ACSL1 in S100A11+Mono / Mφ and S100A11-Mono / Mφ cells in monocytes / macrophages. Wilcoxon test was used; *** represents p<0.001. Figure 9 This section presents a pseudo-time series analysis of S100A11- and S100A11+ mononuclear macrophages. In the figure: a) After liver transplantation, all Mono / Mφ values ​​are divided into 5 subsets based on the pseudo-time series of two nodes. b) In the pseudo-time series analysis, the later the pseudo-time series, the lighter the color. c) The distribution of Mono / Mφ values ​​representing S100A11- and S100A11+ in the pseudo-time series. d) The dynamic expression of S100A11 in the pseudo-time series analysis. e) The dynamic expression of ACSL1 in the pseudo-time series analysis. f) The dynamic expression of lipid metabolism scores in the pseudo-time series analysis. g) The dynamic expression of S100A11 in the non-rejection and rejection groups in the pseudo-time series analysis. h) The dynamic expression of ACSL1 in the non-rejection and rejection groups in the pseudo-time series analysis. i) The dynamic expression of lipid metabolism scores in the non-rejection and rejection groups in the pseudo-time series analysis. Figure 10 This section presents a pseudo-time series analysis of S100A11- and S100A11+ monocytes / macrophages. In the figure: a) shows the dynamic changes of three genes—JDP2, CSF1R, and MAFB—representing the original state of monocytes / macrophages along the pseudo-time series. b) is a dendrogram showing the expression of S100A11 in different subpopulations of monocytes / macrophages. c) is a dendrogram showing the expression of ACSL1 in different subpopulations of monocytes / macrophages. d) is a dendrogram showing the expression of lipid metabolism scores in different subpopulations of monocytes / macrophages. e) A pseudo-time series heatmap was created by searching for differentially expressed genes and extracting the 100 genes with the greatest differences. Figure 11Cell communication analysis of S100A11- and S100A11+ monocytes / macrophages; in the figure: a) the quantity and intensity of cell communication among all cell types after liver transplantation. b) the quantity and intensity of communication between S100A11- monocytes / macrophages and other cell types. c) the quantity and intensity of communication between S100A11+ monocytes / macrophages and other cell types. d) the receptor-ligand interaction between S100A11- and S100A11+ monocytes / macrophages and other cell types. e) the relationship between outgoing signaling strength and incoming signaling strength for all cell types. f) a comparison of cell communication quantity and intensity between the non-rejection and rejection groups using Circleplots. Red indicates that the quantity / intensity of this communication is higher in the rejection group than in the non-rejection group, and blue indicates that the quantity / intensity of this communication is higher in the non-rejection group than in the rejection group. g) a heatmap showing the comparison of cell communication quantity and intensity between the non-rejection and rejection groups. Red indicates that the number / intensity of this communication is higher in the rejection group than in the non-rejection group, and blue indicates that the number / intensity of this communication is higher in the non-rejection group than in the rejection group. Figure 12 Cell communication analysis of S100A11- and S100A11+ monocytes / macrophages; Figure: a) is a chord plot showing the ligand-receptor relationship between S100A11- and S100A11+ monocytes / macrophages and other cells. S100A11- monocytes / macrophages are shown in red, and S100A11+ monocytes / macrophages are shown in blue. b) is a heatmap showing outgoing signaling patterns and incoming signaling patterns of different cell types; Figure 13The proportion of S100A11+ macrophages in the liver and peripheral blood of liver transplant rejection patients is elevated and associated with rejection. Figures show: a) HE staining of human liver tissue (200×); b) Comparison of Liver Rejection Activity Index (RAI) scores between the acute rejection group (R, n=10) and the non-rejection group (NR, n=10); c) Multiplex immunofluorescence (mIF) staining of human liver tissue showing S100A11+ macrophages (CD68-green, S100A11-red, DAPI-blue); d) Proportion of macrophages (CD68+) in total cells in human liver tissue; e) Proportion of S100A11+ macrophages (S100A11+ CD68+) in total macrophages in human liver tissue; f) Correlation analysis between the proportion of S100A11+ macrophages in total macrophages and RAI score; g) Flow cytometry analysis of the proportion of S100A11+ macrophages in human peripheral blood PBMCs. h represents the percentage of S100A11+ macrophages in human peripheral blood PBMCs. i represents the correlation between the proportion of S100A11+ macrophages in human peripheral blood PBMCs and the RAI score. j represents the correlation between the proportion of S100A11+ macrophages in human peripheral blood PBMCs and the proportion of S100A11+ macrophages in total macrophages in human liver tissue. Figure 14 This study investigates the correlation between the proportion of S100A11+ macrophages in the liver and peripheral blood of liver transplant patients and liver function levels and CD8+ T cell infiltration. Figures show: a) Comparison of liver function (ALT, AST, TBIL) in non-rejected and rejected liver transplant patients; b) Correlation analysis between the proportion of S100A11+ macrophages in human liver tissue and liver function; c) Gating strategy for S100A11+ macrophages in human peripheral blood PBMCs detected by flow cytometry; d) Correlation analysis between the proportion of S100A11+ macrophages in human peripheral blood PBMCs and liver function; e) CD8+ T cells (CD8-yellow, DAPI-blue) shown by multiplex immunofluorescence (mIF) staining in human liver tissue; f) The proportion of CD8+ T cells (CD8+) in human liver tissue; g) Correlation analysis between the proportion of S100A11+ macrophages in human liver tissue and the proportion of CD8+ T cells in human liver tissue. Figure 15The increased proportion of S100A11+ macrophages in the liver and peripheral blood of liver transplantation model mice is associated with rejection. In the figure: a) HE staining of mouse liver tissue (200×). b) Comparison of liver rejection activity index (RAI) scores between the liver transplantation rejection group (R, n=5) and the non-rejection group (NR, n=5). c) Multiplex immunofluorescence (mIF) staining of mouse liver tissue showing S100A11+ macrophages (F4 / 80-green, S100A11-red, DAPI-blue). d) Proportion of macrophages (F4 / 80+) in mouse liver tissue. e) Proportion of S100A11+ macrophages (S100A11+F4 / 80+) in mouse liver tissue. f) Correlation analysis between the proportion of S100A11+ macrophages in mouse liver tissue and RAI score. g represents the proportion of S100A11+ macrophages in mouse liver and peripheral blood PBMCs as determined by flow cytometry. h and i represent the percentage statistics of S100A11+ macrophages in mouse liver and peripheral blood PBMCs. j and k represent the correlation analysis between the proportion of S100A11+ macrophages in mouse liver and peripheral blood PBMCs and the RAI score. l and m represent the correlation analysis between the proportion of S100A11+ macrophages in mouse peripheral blood PBMCs and the proportion of S100A11+ macrophages in mouse liver tissue by flow cytometry and multiplex immunofluorescence staining. Figure 16 This study investigated the correlation between the proportion of S100A11+ macrophages in the liver of a liver transplant model mouse and lipid metabolism and CD8+ T cell infiltration. Figure a shows the lipid metabolism activity in the liver of the liver transplant model mouse detected by the FAO kit. Figure b shows the correlation analysis between lipid metabolism activity and the proportion of S100A11+ macrophages in mouse liver. Figure c shows CD8+ T cells (CD8-yellow, DAPI-blue) in mouse liver tissue multiplex immunofluorescence (mIF) staining. Figure d shows the proportion of CD8+ T cells (CD8+) in mouse liver tissue mIF. Figure e shows the correlation analysis between the proportion of S100A11+ macrophages and the proportion of CD8+ T cells in mouse liver tissue mIF. Figure f shows the proportion of CD8+ T cells in mouse liver detected by flow cytometry. Figure g shows the percentage of CD8+ T cells in mouse liver. Figure h shows the correlation analysis between the proportion of S100A11+ macrophages and the proportion of CD8+ T cells in mouse liver flow cytometry. Figure 17 The figure shows the efficiency of S100A11 knockdown detection; in the figure: a) is the relative mRNA expression levels of S100A11 and ACSL1 detected by qRT-PCR; b) is the expression level of S100A11 detected by flow cytometry. Figure 18S100A11 knockdown inhibits lipid metabolism activity and CD8+ T cell activation in BMDM. In the figure: a) Western blot bands and grayscale quantitative analysis of the relative expression levels of S100A11 and ACSL1 proteins. b) Detection of lipid metabolism activity in the two BMDM groups using the FAO kit. c) Flow cytometry analysis of the expression of BMDM antigen-presenting related molecules. d) Comparison of Granzyme B and IFN-γ secretion levels after co-culturing CD8+ T cells with the two BMDM groups. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains, and all materials publicly cited herein and cited by them are incorporated herein by reference.

[0018] Equivalent technologies of the specific embodiments described herein that are readily apparent to those skilled in the art through routine experimentation are included in this application.

[0019] Unless otherwise specified, the experimental methods used in the following examples are conventional methods. Unless otherwise specified, the instruments and equipment used in the following examples are all standard laboratory instruments and equipment; unless otherwise specified, the experimental materials used in the following examples were all purchased from regular biochemical reagent stores.

[0020] I. Experimental Methods Study Subjects and Grouping: This study included 20 liver transplant patients who underwent liver biopsy at Beijing Friendship Hospital, Capital Medical University, between January and December 2025 (Table 1). Based on the pathological results of the liver biopsy, patients were divided into an acute rejection group (R group = 10) and a non-rejection group (NR group = 10). The study was approved by the Bioethics Committee of Beijing Friendship Hospital, Capital Medical University (2023-P2-222-02). All participants signed informed consent forms, agreeing to the use of their liver tissue, blood samples, and clinical data for research and publication.

[0021] Table 1: Clinical characteristics of the subjects In the table: Cases 1–10 are the non-rejection group; Cases 11–20 are the rejection group. HBV, Hepatitis B virus; DDLT, cadaveric liver transplantation; LDLT, living donor liver transplantation; IS, Immunosuppression; TRL, Tacrolimus; MMF, Mycophenolate Mofetil; MP, Methylprednisolone; ALT, Alanine aminotransferase; AST, Aspartate aminotransferase; TBIL, Total Bilirubin; RAI, Rejection Activity Index.

[0022] 1. Data Quality Control and Processing: The single-cell transcriptome sequencing (scRNA-seq) datasets HRA002091 and HRA007802 after liver transplantation were downloaded from the NGDC Genome Sequence Archive (https: / / ngdc.cncb.ac.cn / gsa-human / ), including 15 liver tissue samples (7 rejected, 6 non-rejected) and 4 blood samples (2 rejected, 2 non-rejected). Seurat (5.3.0) was used to read and analyze the cells in each sample. Cells expressing >500 genes and those with <25% mitochondrial gene proportions were retained. The Seurat anchor integration method was then used to integrate the data, yielding a total of 143,164 cells for subsequent analysis. The FindVariableFeatures function was then used to identify hypervariable genes. Dimensionality reduction and clustering were then performed using the FindNeighbors (dimension set to 1:23) and FindClusters (resolution set to 0.5) functions in Seurat. Finally, dimensionality reduction is performed using uniform manifold approximation and projection (UMAP).

[0023] 2. Subpopulation Segmentation and Annotation: For cell subpopulation annotation, the FindAllMarkers function in the Seurat package is first used to identify highly expressed genes specific to each cell subpopulation (parameters: min.pct=0.25, Thresh.use=0.25). A set of typical marker genes is generated based on the CellMarker database (http: / / bio-bigdata.hrbmu.edu.cn / CellMarker), and cell types are annotated according to this genome (Table 2). The annotation results are imported into the metadata of the Seurat object and visualized using Seurat built-in functions and ggplot(3.5.2).

[0024] Table 2: Cell Type Annotation 3. Lipid Metabolism Scoring: First, a lipid metabolism-related gene set was constructed. Hallmark gene sets, KEGG Medicus gene sets, KEGG Legacy gene sets, and Reactome gene sets from the MSigDB database were merged. Gene sets whose names contained "FATTY_ACID", "LIPID", "CHOLESTEROL", "TRIGLYCERIDE", "PHOSPHOLIPID", "SPHINGOLIPID", "LIPOPROTEIN", "OXIDATION", "LIPOGENESIS", "STEATOSIS", "PPAR", and "BILE_ACID" were merged. After removing duplicate genes, the lipid metabolism-related gene set of this invention was obtained, containing 1447 genes. Using this lipid metabolism-related gene set as a reference, five algorithms—AUCell (1.28.0), UCell (2.10.1), AddModuleScore (seurat built-in function), ssGSEA (GSVA 2.0.7), and Singscore (1.26.0)—were used to evaluate lipid metabolism relevance at the single-cell level. It is worth noting that since the lipid metabolism-related gene set constructed in this invention includes both lipid synthesis and breakdown, the lipid metabolism score represents the relevance of lipid metabolism, but not the intensity of lipid synthesis and breakdown. Based on the median lipid metabolism score, cells were divided into UP and DOWN groups. FindMarkers was used to identify differentially expressed genes in the two groups, and differentially expressed genes highly expressed in the UP group were extracted. Correlation analysis was then performed to identify genes closely related to lipid metabolism. The intersection of correlated genes and differentially expressed genes was defined as characteristic genes.

[0025] 4. Machine Learning: Five machine learning algorithms were used for feature selection and key gene identification: Gradient Boosting Machine (GBM), Extreme Gradient Boosting (XGBoost), Random Forest (RF), Least Absolute Shrinkage and Selection Operator (LASSO), and Boruta. Genes consistently identified by all algorithms were considered the best feature genes. GBM was implemented using gbm (2.2.2), and the optimal number of trees was determined before selecting feature genes. XGBoost was implemented using xgboost (1.7.11.1), using 70% of the original data for training and 30% for validation, using the average information gain from splitting to find key genes. RF was implemented using ranger (0.17.0), selecting the number of trees with the lowest error rate based on the error rate curve for calculation. LASSO was implemented using glmnet (4.1.10), with a lambda value of 1 standard error selected to strictly control variables. Boruta was implemented using Boruta (9.0.0), with a maximum run count of 500. Finally, genes were categorized into Confirmed, Rejected, and Tentative based on importance, and only Confirmed genes were selected. The target gene was obtained by taking the intersection of all genes selected through machine learning.

[0026] 5. Target Gene Validation: The liver transplant transcriptome data (RNA-seq) dataset GSE145780 was downloaded from the GEO database (https: / / www.ncbi.nlm.nih.gov / geo / ). The dataset includes 129 non-rejection cases and 37 rejection cases. The Wilcoxon test was used to compare target gene expression in the rejection and non-rejection groups. Receiver operating characteristic (ROC) curves of the target gene were plotted, and the corresponding area under the curve (AUC) value was calculated. Then, the same single-cell transcriptome data was used to detect the expression level and distribution density of the target gene in different cell types. Similarly, the Wilcoxon test was used in the specified cell types to compare target gene expression in the rejection and non-rejection groups.

[0027] 6. Enrichment analysis: The gene set enrichment analysis was performed using enrichGO, enrichKEGG, gseGO, and gseKEGG in clusterProfiler (4.14.6). A pvalue < 0.05 indicates a gene set with significant expression.

[0028] 7. Simulated Gene Knockout in Single Cells: scTenifoldKnk (1.0.2) was used to simulate changes in the expression of other genes after knocking down genes in single-cell sequencing. Visualization was performed using ggplot, where genes with an absolute Z-value > 2 and p.adj < 0.01 were labeled with their names. The sign of the Z-value does not represent the level of downstream gene expression, but rather the magnitude of the change before and after knockdown.

[0029] 8. Pseudo-time series analysis: Pseudo-time series analysis was performed using the Monocle package (2.34.0). Genes meeting the mean expression criterion ≥0.05 and actual dispersion ≥0.8 expected dispersion were retained for subsequent trajectory analysis. The data were then sorted after dimensionality reduction using the DDRTree method. Monocle built-in functions and ggplot were used for plotting.

[0030] 9. Cell Communication: Cells were categorized into negative and positive cells based on whether target gene expression was greater than 0. CellChat (1.6.1) was used to study communication between negative and positive cells and other cell types, focusing on the "secretion signaling" pattern. Cell-cell communication networks were aggregated after calculating potential ligand-receptor interactions. Cell communication was visualized using built-in CellChat functions. Communication between negative and positive cells and other cell types was then studied in both non-rejection and rejection groups, and visualized after comparison.

[0031] 10. Hematoxylin-eosin (HE) staining: Liver tissue was fixed in formalin, embedded in paraffin, and sectioned (2 μm) for HE staining. Histological grading was performed according to the Banff protocol based on the HE staining results. The Rejection Activity Index (RAI) was scored based on the degree of venous endothelial inflammation, bile duct inflammation, and portal vein inflammation, with a total score of 9 points. A RAI score > 3 was considered an acute rejection reaction.

[0032] 11. Multiplex Immunofluorescence (mIF) Staining: Paraffin-embedded tissue sections were baked at 60°C for 2 hours, then dewaxed and hydrated. Antigens were then repaired at 95°C for 20 minutes in pH 6.0 buffer. Human liver tissue was stained sequentially with CD68 (1:1000, 76437, CST) (4°C overnight), CD8 (1:2000, SOB2307, STARTER) (room temperature 60 minutes), and S100A11 (1:200, CL647-10237, Proteintech) (4°C overnight) primary antibodies. Mouse liver tissue was stained sequentially with F4 / 80 (1:800, 70076S, CST) (4°C overnight), CD8 (1:200, ab217344, abcam) (room temperature 60 minutes), and S100A11 (1:200, CL647-10237, Proteintech) (4°C overnight) primary antibodies. Enzyme-labeled anti-rabbit IgG polymer (Opal IHC assay kit; Akoya Biosciences) was used as the secondary antibody, and TSA chromogenic solution was used for staining. Sections were sealed with DAPI anti-fluorescence quenching mounting medium (PO131, Beyotime, China) and stored at 4°C protected from light. Images were acquired using the Vectra Polaris multispectral imaging platform (Akoya Biosciences) and analyzed using Inform 2.4.8 software (Akoya Biosciences).

[0033] 12. Flow cytometry detection: The detection is performed using a BD FACSAriaTM III flow cytometer (BD Biosciences, USA). For the identification of human peripheral blood macrophages, FITC-conjugated anti-CD45 antibody (1:200, 560976, BD), PE-conjugated anti-CD11b antibody (1:200, 101207, BioLegend), APC / Cy7-conjugated anti-CD14 antibody (1:200, 560270, BD), BV421-conjugated anti-CD68 antibody (1:200, 864943, BD) and APC-conjugated anti-S100A11 antibody (1:200, CL647-10237, Proteintech) are used. For the identification of macrophages from mouse liver and peripheral blood, APC / Cy7-conjugated anti-CD45 antibody (1:200, 103154, BioLegend), BV421-conjugated anti-CD11b antibody (1:200, 3044006, Invitrogen), PerCP / Cy5.5-conjugated anti-F4 / 80 antibody (1:200, 123127, BioLegend) and APC-conjugated anti-S100A11 antibody (1:200, CL647-10237, Proteintech) are used. For the identification of mouse liver T cells, PerCP / Cy5.5-conjugated anti-CD3 antibody (1:200, 100218, BioLegend), FITC-conjugated anti-CD4 antibody (1:200, 100406, BioLegend) and AF700-conjugated anti-CD8 antibody (1:200, 100730, BioLegend) are used. The Live / DeadTM Fixable Aqua dead cell staining kit is used to distinguish dead cells.

[0034] 13. Construction of mouse liver transplantation model: 7-8 week-old SPF grade male C3H and C57 mice are selected (purchased from SPF (Beijing) Biotechnology Co., Ltd., License No.: SYXK (Beijing) 20170010). In the non-rejection group (NR group), C3H mice are used as both donors and recipients; in the rejection group (R group), C57 mice are used as donors and C3H mice as recipients, and the mouse liver transplantation model is established via the "double-cuff" technique. The present invention is approved by the Laboratory Animal Welfare and Ethics Committee of the First Hospital of Shanxi Medical University (DWLL-2025-010). All mice are kept in the SPF-grade animal facility of the Laboratory Animal Center of Shanxi Medical University, and tissue samples are collected on day 14 after surgery.

[0035] 14. Cell Preparation: Primary bone marrow cells from C57 mice were isolated and cultured for 7 days in high-glucose DMEM complete medium containing M-CSF to obtain mature bone marrow-derived macrophages (BMDMs). Mature BMDMs were seeded and transfected using Lipofectamine 3000 transfection reagent combined with small interfering RNA (siRNA) targeting S100A11 (si-S100A11 group). A negative siRNA control group (si-NC group) was also established. Six hours after transfection, the medium was replaced with standard complete medium, and cell samples were collected after another 48 hours of culturing.

[0036] Spleens from C57 mice were isolated, ground, and filtered through a 70 μm mesh to prepare a single-cell suspension. Red blood cells were removed using red blood cell lysis buffer. CD8+ T cells from the spleen were isolated and purified using a mouse CD8a+ T cell sorting kit (130-104-075, Miltenyi Chemicals, Germany).

[0037] Different groups of BMDMs (si-NC group and si-S100A11 group) were co-cultured with CD8+ T cells in X-VIVO complete medium. After 3 days, CD8+ T cells were collected for flow cytometry to evaluate their secretory function.

[0038] 15. Western blotting: Protein extraction from BMDMs. S100A11 polyclonal primary antibody (10237-1-AP; proteintech; USA) and ACSL1 polyclonal primary antibody (13989-1-AP; proteintech; USA) were diluted to 1:3000 using primary antibody diluent (Abcam, UK). The membrane was then incubated with horseradish peroxidase (HRP)-labeled goat anti-rabbit polyclonal secondary antibody. Protein bands were visualized using a Bio-Rad gel imaging system (USA).

[0039] 16. Real-time quantitative PCR: Total RNA was extracted from BMDMs samples using Trizol reagent, and 2 μg of RNA was reverse transcribed into cDNA. A 20 μl reaction system was prepared using 2×M5 HiPer real-time quantitative PCR premixed reagent (low Rox) and S100A11 and ACSL1 primers, with β-actin as an internal control. The relative mRNA expression level was calculated using the 2-ΔΔCt method, where ΔCt = Ct (target gene) - Ct (internal control gene).

[0040] 17. Lipid metabolism activity assay: Mouse liver tissue and BMDMs samples were treated with Fatty Acid Oxidation (FAO) Colorimetric Assay Kit (E-BC-K784-M; Elabscience; USA), and the absorbance value at 450 nm wavelength was detected by microplate reader to reflect the lipid metabolism level.

[0041] 18. Statistics: All bioinformatics statistical analyses and visualizations were performed using R (4.4.3), with Wilcoxon rank-sum test, Pearson correlation analysis, and logrank test. Other data were analyzed using GraphPad Prism 9 software. Continuous variable data are expressed as mean ± standard deviation. T-tests were used for comparisons between two groups, and one-way ANOVA was performed followed by Tukey's post-hoc multiple comparison test for comparisons among multiple groups. Pearson correlation analysis was used to assess the correlation between data points. A p-value less than 0.05 was considered statistically significant (ns, p ≥ 0.05; *, p < 0.05; **, p < 0.01; ***, p < 0.001).

[0042] II. Experimental Results 1. Distribution of different cell types after liver transplantation: After quality control and merging of single-cell data, the UMAP plot showed uniform cell distribution in different samples, successfully correcting for batch effects. Figure 2 a). Based on differences in transcriptome expression, 143,164 cells were divided into 22 clusters ( Figure 1 a). The 22 clusters showed significant differences in the expression of characteristic genes, suggesting the cell type to which they belonged ( Figure 2 c). Based on characteristic genes, cells are classified into 12 types: mast cells, hepatic stellate cells, plasma cells, bile duct epithelial cells, hepatocytes, dendritic cells, neutrophils, endothelial cells, B cells, monocytes / macrophages, NK cells (natural killer cells), and T cells (…). Figure 1 b). Bubble chart ( Figure 1 d) Violin illustration ( Figure 2 d) and UMAP diagram ( Figure 2 e) Confirmed accuracy of cell type annotation. Since the data included both blood and liver samples, UMAP plots were used to observe differences in cell distribution between samples from different sources. It was observed that liver samples contained every cell type, while blood samples showed a higher concentration of monocytes / macrophages and neutrophils. Figure 2 b). The percentage of each sample in different cell types is shown in the figure. Figure 2 f, B2-5 are blood samples, and the rest are liver samples. Next, we analyzed the differences in cell type distribution between the non-rejection group and the rejection group ( Figure 1c). The proportion plot shows a significant difference in the proportion of cell types between the two groups, especially in mononuclear macrophages (MBCs). Figure 1 e). The proportion of cell types varied considerably among the samples. Figure 1 f). In summary, the proportion of mononuclear macrophages in the rejection group was significantly higher than that in the non-rejection group.

[0043] 2. Monocytes and macrophages are target cells for studying lipid metabolism after liver transplantation. Identifying characteristic genes of lipid metabolism after liver transplantation: To investigate lipid metabolism after liver transplantation, multiple algorithms were used to calculate the lipid metabolism score for each cell, including AUCell, UCell, AddModuleScore, ssGSEA, and Singscore. Figure 3 a). The summarized bubble chart shows that all five algorithms indicated higher lipid metabolism scores for mononuclear macrophages and hepatocytes after liver transplantation, with a higher proportion of cells having a score >1. This result was also confirmed in the UMAP plot. Figure 4 a). Next, the mean lipid metabolism score obtained from the five algorithms for each cell was defined as its final lipid metabolism score. It was observed that the lipid metabolism scores of monocytes / macrophages and hepatocytes were significantly higher than those of other cells. Figure 3 b). Comparing lipid metabolism scores between the non-rejection and rejection groups, it was found that the rejection group had a higher lipid metabolism score. Figure 3 c). The lipid metabolism scores of the five algorithms also showed higher scores in the rejection group. Figure 4 (b) This demonstrates the stability of the results. Subsequent comparisons of lipid metabolism scores between the non-rejection and rejection groups across different cell types revealed a significant difference in lipid metabolism scores between the two groups in monocytes / macrophages, while no difference was observed in hepatocytes. Figure 3 d). Lipid metabolism scores from five algorithms confirmed that the differences in monocytes and macrophages between the two groups were relatively stable. Figure 4 c). Because monocytes and macrophages have higher lipid metabolism scores, and the lipid metabolism scores of the rejection group are significantly higher than those of the non-rejection group, we designated monocytes and macrophages as target cells for studying lipid metabolism after liver transplantation.

[0044] All cells were divided into two groups based on the mean lipid metabolism score: the UP group had a lipid metabolism score greater than the mean of 0.7605494, and the rest were in the DOWN group. The UMAP plot shows the distribution of the two groups. Figure 3 (e) As shown in the figure, most mononuclear macrophages are in the UP group, while very few are in the DOWN group, and other cell types do not show such a significant distribution pattern. Next, differentially expressed genes between the UP and DOWN groups were identified, and 542 differentially expressed genes highly expressed in the UP group were selected to represent the lipid metabolism regulated in the UP group. Figure 3f), the figure marks the 21 genes highly expressed in the upregulated group. Based on correlation, the top 300 genes related to lipid metabolism were selected to form highly correlated lipid metabolism pathways, and the correlation coefficients of the top 130 genes are also shown. Figure 3 g). The intersection of the two gene sets representing UP-regulated lipid metabolism and highly correlated lipid metabolism pathways yields a gene set of 21 genes, which constitutes the final lipid metabolism characteristic genes after liver transplantation. Figure 3 h). Target genes in Figure 3 The genes were marked in f. Finally, KEGG enrichment analysis revealed that the characteristic genes were highly correlated with functions such as alcohol metabolism and cell adhesion. Figure 3 i).

[0045] 3. S100A11 is the most critical lipid metabolism-related gene. In the exclusion group, S100A11 expression was upregulated in Mono / Mφ: the relative importance of characteristic genes was calculated using GBM, with an optimal number of trees (1500) selected for calculation. Figure 5 a) Genes with relative importance greater than the average were selected as target genes, and a total of 6 genes were screened out. Figure 6 a). XGBoost was used to calculate the gain of the feature genes, and genes with a gain > 0.08 were selected as target genes. A total of 5 genes were screened out ( Figure 6 b). The importance of feature genes is calculated using RF, selecting 600 trees in a random forest with the lowest error rate for calculation. Figure 5 b) Selecting genes with importance > 0.010 as target genes, a total of 3 genes were screened out ( Figure 6 c). Using the LASSO regression algorithm, 18 key genes with the lowest binomial bias were identified. Figure 6 d). The Boruta algorithm converged early after 15 iterations, identifying 21 feature genes that were all confirmed genes. Figure 5 c). After merging the results of all five algorithms, the target gene S100A11 was determined. Figure 6 e), the gene sets selected by each machine learning algorithm are shown in Table 3. The correlation between the selection results of different machine learning algorithms was observed. Figure 5 (d) It can be seen that LASSO and Boruta have the highest similarity, with a similarity of 0.857; while the gradient boosting algorithm has the lowest similarity with Boruta, at 0.095. In summary, multiple machine learning algorithms have identified S100A11 as the most critical lipid metabolism-related gene.

[0046] Table 3: Gene sets selected by each machine learning algorithm The expression of S100A11 was verified by selecting the transcriptome GSE145780 after liver transplantation. The results showed that the expression level of S100A11 was significantly increased in the rejection group. Figure 6 f). The area under the ROC curve is as high as 0.948, indicating that S100A11 has high diagnostic value (f). Figure 6 g). S100A11 expression was also validated using single-cell transcriptome data after liver transplantation. Violin plots show S100A11 expression levels in different cell types, and heatmaps show the normalized z-score of S100A11 expression. It is evident that S100A11 expression was highest in mononuclear macrophages. Figure 6 h). UMAP also observed increased S100A11 expression in monocytes and macrophages (h). Figure 6 i). Furthermore, the cell distribution density of S100A11+ cells was observed. S100A11+ cells were defined as cells with S100A11 expression greater than 0. Density maps showed a higher density of S100A11+ cells among mononuclear macrophages. Figure 6 j). Next, the expression of S100A11 in monocytes and macrophages was compared between the non-rejection and rejection groups. The results showed a significant difference between the two groups, with increased S100A11 expression in the rejection group (j). Figure 6 k). It was verified that S100A11 was upregulated in the rejection group Mono / Mφ.

[0047] 4. S100A11+Mono / Mφ characteristics: To identify the characteristics of S100A11+Mono / Mφ, the co-expression of S100A11 and markers from M1, M2, and Kupffer cells was first observed in monocytes and macrophages. Figure 7 The results showed that S100A11+Mono / Mφ was co-expressed with the three markers relatively little, therefore S100A11+Mono / Mφ cannot be simply defined as one of them. Next, enrichment analysis was performed based on the differentially expressed genes in S100A11+Mono / Mφ to explore the function of S100A11+Mono / Mφ. Figure 8 (ad). The study found that S100A11+Mono / Mφ exhibits better antigen-presenting ability and T-cell regulation, modulating immune and inflammatory responses and being associated with graft rejection. This suggests that S100A11+Mono / Mφ may regulate immune and inflammatory responses by enhancing its own antigen-presenting ability to activate T cells, ultimately inducing post-transplant rejection. Subsequent investigations explored the downstream genes of S100A11+Mono / Mφ.

[0048] After virtual knockout of S100A11 in mononuclear macrophages based on single-cell sequencing data, changes in the expression levels of multiple genes were observed. Figure 8 e). Then, the intersection of the genes altered before and after knockdown, the genes highly expressed by S100A11+Mono / Mφ, and lipid metabolism genes was taken, finally yielding one gene, ACSL1 ( Figure 8f). ACSL1 is a key gene regulating cellular lipid metabolism, ranking second among genes altered before and after S100A11 knockdown, and its expression level is higher in S100A11+Mono / Mφ than in S100A11-Mono / Mφ. Figure 8 g). Therefore, we infer that monocytes and macrophages affect lipid metabolism by increasing ACSL1 expression through high expression of S100A11, ultimately leading to transplant rejection.

[0049] 5. Gradual increase of S100A11 in monocytes and macrophages during pseudo-time progression: Pseudo-time series analysis was performed on monocytes and macrophages extracted after liver transplantation, with 2 nodes and a total of 5 states ( Figure 9 a). Subsequently, based on the prompts, the three genes JDP2, CSF1R, and MAFB of monocytes in their primitive state underwent dynamic changes during pseudotime progression. Figure 10 a) State 4 is designated as the initial state, and state 1 as the final state. The pseudo-timeline of monocytes / macrophages is shown below. Figure 9 b. We then divided monocytes / macrophages into S100A11- and S100A11+ monocytes / macrophages based on whether S100A11 expression was greater than 0. Similarly, we observed that S100A11+ monocytes / macrophages were more concentrated in the terminal phase, while S100A11- monocytes / macrophages were more concentrated in the initiation phase. Figure 9 c). The time series plot shows that the expression level of S100A11 generally increases and then decreases over time. Figure 9 d), ACSL1 changes are consistent with S100A11 ( Figure 9 e), the lipid metabolism score shows a wavy pattern over time (e), Figure 9 f). The dendrogram suggests that cells with high S100A11 expression are mostly concentrated in the terminal stage (f). Figure 10 b), cells with high ACSL1 expression levels are also mostly concentrated in the terminal stage ( Figure 10 c), cells with high lipid metabolism scores are mostly concentrated in the terminal stage ( Figure 10 d). Then we compared the expression of S100A11 over time in the non-rejection and rejection groups, and found that the trend of S100A11 expression in both groups was consistent, showing an overall increase. The overall expression of S100A11 in monocytes and macrophages in the rejection group was higher than that in the non-rejection group. Figure 9 g). The trend of ACSL1 variation in both groups is consistent with that of S100A11 ( Figure 9 h). Next, the expression of lipid metabolism scores over time was compared between the non-rejection and rejection groups. It was found that the lipid metabolism scores in both groups showed a consistent upward trend, with the overall expression of lipid metabolism scores in monocytes and macrophages in the rejection group being higher than that in the non-rejection group. Figure 9i). Finally, differentially expressed genes were identified along the pseudo-timeline and divided into 5 modules based on their changes. Genes in the module containing S100A11 all increased from lowest to highest expression along the pseudo-timeline. Figure 10 e).

[0050] 6. S100A11+ monocyte-macrophage cell communication was significantly enhanced: To explore the effect of S100A11- / + on monocyte-macrophage activity, the number and intensity of interactions between S100A11- / + monocyte-macrophages and other cells were investigated. Figure 11 a). To observe more clearly, S100A11- / + monocytes and macrophages were isolated to observe their cell communication. Figure 11 (bc) It can be seen that, in terms of both the quantity and intensity of communication, S100A11+ monocytes / macrophages are higher than S100A11- monocytes / macrophages. (bubble chart) Figure 11 d) Chord diagram ( Figure 12 a) This image illustrates the ligand-receptor interactions of S100A11- / + monocytes / macrophages with other cells. Enhanced communication between S100A11+ monocytes / macrophages is visible in the figure. A heatmap is then created to further illustrate the specific ligand-receptor expression patterns of each cell's output and input signaling modes. Figure 12 b). Furthermore, compared to S100A11- monocytes and macrophages, S100A11+ monocytes and macrophages exhibited stronger outward and afferent interactions, indicating that S100A11+ monocytes and macrophages are more actively involved in cell communication. Figure 11 e). In summary, S100A11+ monocyte-macrophage cell activity was significantly activated. Cell communication in the non-rejection and rejection groups was then calculated and compared, and a comparative cell communication network diagram was drawn. Figure 11 f) and heatmap ( Figure 11 g). In the figure, red indicates that this communication is enhanced in the rejection group, and blue indicates that this communication is enhanced in the non-rejection group. The figure shows that the number and intensity of communication between S100A11+ monocytes / macrophages and T cells are significantly increased in the rejection group, suggesting that S100A11+ monocytes / macrophages may play a role through interaction with T cells.

[0051] 7. Increased proportion of S100A11+ macrophages in the liver and peripheral blood of liver transplant rejection patients, which is associated with rejection: To verify the above analysis results, liver tissue from liver transplant patients was stained with hematoxylin and eosin (HE). The results showed that, compared with the NR group, the R group had a large number of inflammatory cell infiltrations, ballooning degeneration, necrosis, hepatocellular edema, bile duct injury, and endothelial injury in the portal area of ​​the liver. Figure 13 a). The Liver Rejection Activity Index (RAI) score showed that the R group score was significantly higher than the NR group ( Figure 13 b). This is consistent with the liver function levels of liver transplant patients. Figure 14a). Subsequently, mIF staining was used to assess macrophage infiltration in the liver. Compared with the NR group, macrophage infiltration was increased in the R group, and the proportion of S100A11+ macrophages was significantly increased (a). Figure 13 ce). By analyzing the correlation between the proportion of S100A11+ macrophages in the liver and RAI score and liver function, it can be seen that the two are positively correlated. Figure 13 f, Figure 14 b). Simultaneously, flow cytometry was used to detect the proportion of S100A11+ macrophages in peripheral blood PBMCs of liver transplant patients, and its correlation with RAI score and liver function was analyzed, all of which were consistent with the liver results. Figure 14 cd Figure 13 The infiltration of CD8+ T cells in the liver was also examined. mIF staining results showed that CD8+ T cell infiltration in the R group was significantly higher than that in the NR group, and the infiltration levels of S100A11+ macrophages were positively correlated with those of CD8+ T cells. Figure 14 e.g., further analysis of the correlation between peripheral blood PBMCs and the proportion of S100A11+ macrophages in the liver of liver transplant patients revealed a positive correlation between the two. Figure 13 (j) These results suggest that the proportion of S100A11+ macrophages in the liver and peripheral blood of liver transplant rejection patients is synchronously increased, and both can reflect the degree of liver rejection.

[0052] 8. Increased proportion of S100A11+ macrophages in the liver and peripheral blood of liver transplant model mice, associated with rejection: Liver transplant rejection and non-rejection models were established in mice. HE staining and RAI scoring of mouse liver tissue showed that group R mice exhibited extensive inflammatory cell infiltration, ballooning degeneration, necrosis, hepatocyte edema, bile duct injury, and endothelial damage in the portal area of ​​the liver, with significantly increased RAI scores. Figure 15 ab). Macrophage infiltration in mouse liver was assessed by mIF staining. Macrophage infiltration was increased in group R, and the proportion of S100A11+ macrophages was significantly increased (ab). Figure 15 Analysis showed that the proportion of S100A11+ macrophages in mouse liver was positively correlated with RAI score. Figure 15 f). Simultaneously, flow cytometry was used to detect S100A11+ macrophages in mouse liver and peripheral blood PBMCs. In both tissues, the proportion of S100A11+ macrophages in the R group was significantly higher than that in the NR group. Figure 15 The proportion of S100A11+ macrophages in mouse liver and peripheral blood PBMCs was positively correlated with RAI scores. Figure 15 (jk). Comparing the lipid metabolism levels in the livers of the two groups of mice, the FAO activity in the R group was higher than that in the NR group, and was positively correlated with the proportion of S100A11+ macrophages in the liver (jk). Figure 16ab). CD8+ T cell infiltration in mouse liver was also examined. mIF staining and flow cytometry results showed that CD8+ T cell infiltration in group R was significantly higher than in group NR, and the infiltration levels of S100A11+ macrophages were positively correlated with those of CD8+ T cells. Figure 16 Further analysis of the correlation between peripheral blood PBMCs and the proportion of S100A11+ macrophages in the liver of liver-transplanted mice revealed a positive correlation between the two. Figure 15 These results suggest that the proportions of S100A11+ macrophages in the liver and peripheral blood of liver transplant rejection model mice are simultaneously increased, and both can reflect the degree of liver rejection.

[0053] 9. S100A11 knockdown inhibits lipid metabolism activity and CD8+ T activation in BMDM: Mouse bone marrow-derived macrophages (BMDM) were induced in vitro, and S100A11 was knocked down. The gene knockdown efficiency was verified by qRT-PCR, flow cytometry, and Western blot, while ACSL1 expression was also detected. Results showed that compared with the negative control group (si-NC), the mRNA transcription level and protein expression level of S100A11 in the S100A11 knockdown group (si-S100A11) were significantly decreased, indicating a targeted knockdown effect; the mRNA and protein expression levels of ACSL1 were also synchronously reduced. Figure 17 ab, Figure 18 a). To further investigate the effect of S100A11 on lipid metabolism, the FAO kit was used to detect the activity of two BMDM groups. The results showed that compared with the si-NC group, the FAO activity of the si-S100A11 group was significantly decreased ( Figure 18 (b) suggests that S100A11 deficiency may inhibit the lipid metabolism activity of BMDM by reducing ACSL1 expression. Subsequent analysis of BMDM antigen presentation-related molecules revealed significantly decreased CD86 expression and no significant change in MHC-II expression in the si-S100A11 group. Figure 18 c). Co-culturing BMDM with CD8+ T cells showed that the activation effect of BMDM on CD8+ T cells was inhibited in the si-S100A11 group, and the secretion levels of IFN-γ and Granzyme B in CD8+ T cells of the si-S100A11 group were significantly inhibited compared with the si-NC group. Figure 18 d).

[0054] Liver transplant rejection significantly reduces patient prognosis and quality of life, imposing a heavy burden on patients and public health. Therefore, researching the specific mechanisms of transplant rejection and implementing interventions is currently a hot topic. The activation and differentiation status of immune cells directly lead to post-transplant rejection and tolerance, and the function of immune cells is affected by their own metabolic status. Lipid metabolism plays a significant role in regulating immune cell function, but its specific role in liver transplant rejection remains unclear. Therefore, this invention investigates the relationship between post-liver transplant lipid metabolism levels and rejection and identifies key genes. First, multiple lipid metabolism scoring algorithms were used to assess post-liver transplant lipid metabolism levels, and differentially expressed genes were screened after dividing the groups into upregulated and downregulated groups based on the lipid metabolism scores. After taking the intersection of the differentially expressed genes and the lipid metabolism gene set, 21 characteristic genes highly correlated with lipid metabolism after liver transplantation were obtained. Subsequently, multiple machine learning algorithms were used to screen the intersection of the characteristic genes based on their diagnostic efficacy against rejection, resulting in the key gene S100A11. S100A11 was found to be elevated in the rejection group in both conventional and single-cell transcriptomic analyses, with the highest expression level observed in monocytes and macrophages. Therefore, S100A11+ monocytes and macrophages were considered core cells in liver transplant rejection. Enrichment analysis, pseudo-temporal analysis, and cell communication were then used to observe the characteristics of S100A11+ monocytes and macrophages. These analyses revealed that S100A11+ monocytes and macrophages exhibited stronger antigen-gathering capacity, were more inclined towards terminal cells, and showed significantly enhanced interactions with other cells. Further comparison of cell communication between rejection and non-rejection groups revealed that the interaction strength between S100A11+ monocytes and macrophages and T cells was significantly enhanced under rejection conditions, suggesting that S100A11+ monocytes and macrophages may induce rejection by activating T cells. Finally, the expression differences of S100A11+ monocytes and macrophages in the non-rejection and rejection groups were verified in both humans and mice. The correlation between S100A11+ monocytes and macrophages in peripheral blood and liver was analyzed, and the effectiveness of S100A11+ monocytes and macrophages in assessing liver transplant rejection was validated. This study provides new insights into lipid metabolism regulation in liver transplant rejection and offers new directions for future research and intervention strategies.

[0055] The liver is rich in various innate and acquired immune cells, forming a unique immune microenvironment. This invention, through dimensionality reduction and clustering of single-cell transcriptional data after liver transplantation, observed multiple coexisting cell types in the liver, confirming that the liver is a major immune organ. The hepatic immune microenvironment plays a crucial role in rejection and tolerance after liver transplantation; therefore, studying the functional status of immune cells and the dynamic interactions between cells is key to treating transplant rejection. Changes in the activation, differentiation, and proliferation of immune cells are central to building a tolerable environment; however, the mechanisms triggering these changes are complex, and key intervention points have not yet been identified. Recent studies have found that changes in immune cell function are often accompanied by alterations in metabolic pathways, and these metabolic changes significantly affect the rejection response. Currently, the main metabolic pathways of immune cells involve glycolysis, the pentose phosphate pathway (PPP), the tricarboxylic acid cycle (TCA), fatty acid oxidation (FAO), fatty acid synthesis, and amino acid metabolism. Among these, lipid metabolism is currently a research hotspot, and its role in liver transplant rejection still needs further investigation.

[0056] This invention revealed that lipid metabolism scoring showed the strongest lipid metabolism activity in macrophages after liver transplantation, with higher activity in the rejection group than in the non-rejection group, suggesting that lipid metabolism plays a dominant role in macrophage-induced rejection. Macrophages switch between pro-inflammatory M1 and anti-inflammatory M2 phenotypes to perform different functions. In the rejection group, M1 cells primarily exerted the immune rejection effect; therefore, promoting the conversion of M1 to M2 is crucial for building immune tolerance. Previous studies have confirmed that macrophage phenotype conversion is regulated by fatty acid oxidation (FAO), cholesterol homeostasis, and phospholipid metabolism. Enhanced FAO within macrophages can promote macrophage polarization towards the M2 phenotype through pathways such as mitochondrial oxidative phosphorylation. Simultaneously, macrophage polarization is also regulated by CD36, a core regulatory receptor for lipid uptake. High expression of CD36 can maintain M2 function by promoting fatty acid uptake, leading to fatty acid accumulation in cells, while CD36 deficiency can polarize macrophages to the M1 phenotype. Furthermore, cholesterol metabolism also affects macrophage phenotype conversion; blocking cholesterol efflux and maintaining high cholesterol levels can promote macrophage conversion to M2. Not only does lipid metabolism within macrophages affect macrophage phenotypic transformation, but extracellular lipid stimulation can also lead to an increase in the pro-inflammatory phenotype of M1 macrophages and a decrease in the anti-inflammatory phenotype of M2 macrophages, exacerbating allogeneic transplant rejection. The mechanism by which lipids induce M1-M2 macrophage phenotypic transformation remains unclear, but it may be related to changes in cholesterol homeostasis-related protein ABCA1 and endoplasmic reticulum stress-induced inositol-demanding enzyme 1α (IRE1α). It is evident that controlling lipid metabolism has a significant impact on macrophages, but how macrophages enhance rejection responses through lipid metabolism after transplantation requires further investigation.

[0057] This invention identified 21 genes highly associated with lipid metabolism after liver transplantation and then used various machine learning algorithms to screen for S100A11, the gene most effective in diagnosing rejection and non-rejection. The high expression of S100A11 in rejection was then validated on other datasets. S100A11, as a calcium-binding protein, can regulate intracellular gene expression and mediate long-range biological effects. Currently, most research on S100A11 and its relationship with lipid metabolism focuses on non-alcoholic fatty liver disease (NAFLD), while research on other liver diseases is relatively limited.

[0058] This invention, after identifying the target gene S100A11, analyzed its expression in different cell types. It found that S100A11 expression was highest in macrophages, and there was a significant difference in expression between the rejection and non-rejection groups. Finally, flow cytometry and immunofluorescence were used to verify the expression of S100A11 in macrophages in human and mouse liver transplant specimens, confirming the above analysis. Furthermore, analysis showed a positive correlation between S100A11 expression and liver transplant rejection scores, indicating that S100A11 can serve as a diagnostic biomarker for liver transplant rejection. Simultaneously, a positive correlation was found between S100A11 expression in peripheral blood and liver macrophages; therefore, S100A11 expression in peripheral blood macrophages can serve as a potential biomarker for liver transplant rejection.

[0059] Changes in macrophage lipid metabolism originate from the induction and resolution of inflammation. S100A11 simultaneously plays a dual role in both inflammation and lipid metabolism; therefore, it can be inferred that the effect of S100A11 on macrophage status is the result of the combined effects of inflammation and lipid metabolism. Previous studies have found that S100A11 may be a key mediator for the maintenance of cancer stem cells, and cancer cells can induce macrophages to transform into a pro-tumor phenotype by secreting S100A11. Bile duct cells can also release vesicles containing S100A11 to induce bone marrow-derived macrophages to express pro-inflammatory cytokines. S100A11 in macrophages can release SP3 from radionuclides, promoting the binding of SP3 to the STAT6 promoter, thereby enhancing STAT6 expression and M2a macrophage activation. The expression of S100A11 in macrophages during liver transplant rejection was first discovered and verified in this invention; however, how S100A11+ monocytes and macrophages exert their rejection effect remains unclear.

[0060] First, it was determined whether S100A11+Mono / Mφ could be defined as a traditional M1, M2, or Kupffer cell type. Co-localization of S100A11 with M1, M2, and Kupffer marker genes revealed that these genes were not expressed simultaneously. Therefore, S100A11+Mono / Mφ is inconsistent with previously known macrophage types, necessitating functional analysis. Enrichment analysis showed that S100A11+Mono / Mφ exhibited stronger antigen-presenting capabilities and participated in anti-inflammatory and immune responses. Pseudo-temporal analysis indicated that S100A11+Mono / Mφ was more inclined towards the terminal stage of macrophage development, thus inferring that S100A11+Mono / Mφ represents an activated monocyte-macrophage. Subsequently, through virtual gene knockout and analysis of highly expressed genes in S100A11+Mono / Mφ, ACSL1 was identified as a downstream gene of S100A11. ACSL1, a key enzyme in lipid metabolism, participates in crucial processes and plays a central role in macrophage lipid metabolism, providing a reference for research on lipid dysregulation and rejection after liver transplantation. ACSL1 can regulate macrophage lipid homeostasis and promote pro-inflammatory responses; silencing ACSL1 can inhibit lipid accumulation in macrophages and promote their M2 polarization. Considering the high expression of S100A11 in macrophages of liver transplant rejection patients, it is speculated that ACSL1 may be involved in post-transplant lipid dysregulation and rejection-related inflammation. ACSL1 is an important regulatory target for metabolic diseases; targeting and inhibiting it can alleviate the progression of various diseases, possibly through mechanisms that promote oxidative phosphorylation and inhibit related signaling axes. As a downstream gene of S100A11, ACSL1 is closely related to lipid dysregulation and rejection after liver transplantation. It is speculated that S100A11 may affect macrophage function and rejection responses by regulating ACSL1, making it a potential intervention target.

[0061] To explore whether S100A11+ monocytes / macrophages directly cause immune rejection or act on other cell types, cell communication analysis was conducted to observe the interaction between S100A11+ monocytes / macrophages and other cells. Cell communication revealed significantly enhanced activity of S100A11+ monocytes / macrophages and T cells in the rejection group. Furthermore, S100A11+ Mono / Mφ enrichment analysis suggested its activation of T cell activity. Since the prevailing view is that liver transplant rejection is mostly a T cell-mediated rejection response, it is inferred that S100A11+ monocytes / macrophages may exert a pro-rejection effect by activating T cells. In human and mouse liver transplant specimens, flow cytometry and immunofluorescence techniques confirmed the correlation between CD8+ T cell infiltration and S100A11+ monocytes / macrophages. Currently, most macrophage studies in liver transplant rejection focus on Kupffer cells, and it is generally accepted that Kupffer functions through T cells. Post-transplantation Kupffer cells can promote T cell apoptosis and build immune tolerance, a process that occurs via the factor-associated suicide / factor-associated suicide ligand (Fas / FasL) pathway. Kupffer cells can affect the function of different T cell types. Kupffer cells can directly act on CD8+ T cells, the direct execution cells of rejection, promoting apoptosis of CD8+ T cells to alleviate rejection. Simultaneously, Kupffer cells can indirectly inhibit CD8+ T activity by stimulating Treg cells to secrete IL-10.

[0062] This invention investigated the correlation between immune rejection and lipid metabolism after liver transplantation at the single-cell level. Using various lipid metabolism scores, we identified monocytes / macrophages as the cell type with the strongest lipid metabolism activity after liver transplantation. We then used multiple machine learning algorithms to identify the elevated target gene S100A11 in the rejection group, which was validated in RNA-seq data. Further analysis using pseudo-time series and cell communication revealed that S100A11+ monocytes / macrophages were more likely to be in a terminal state and exhibited enhanced interactions with other cell types. The interaction strength between S100A11+ monocytes / macrophages and T cells was significantly enhanced in the rejection group, suggesting that S100A11+ monocytes / macrophages may lead to immune rejection after liver transplantation by activating T cells. These results highlight the potential of S100A11 as a biomarker for liver transplant rejection, providing a foundation for developing personalized and effective treatment strategies to improve patient outcomes.

[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. Application of S100A11 as a biomarker in the preparation of products for diagnosing liver transplant rejection.

2. The application according to claim 1, characterized in that: The products include reagents or kits for detecting the expression level of the S100A11 gene.

3. The application according to claim 2, characterized in that: The reagents or kits include primers or probes for detecting S100A11 mRNA.

4. The application according to claim 2, characterized in that: The reagents or kits include those for detecting the expression level of S100A11 in mononuclear macrophages in the liver and peripheral blood.

5. The application according to claim 4, characterized in that: The mononuclear macrophages showed high expression of S100A11, increased ACSL1 expression, CD8+ T cell infiltration, liver transplant rejection, and a positive correlation between S100A11 expression and rejection response.

6. Application of the S100A11 gene or protein as a target in the preparation of drugs for the prevention and / or treatment of liver transplant rejection.

7. Application of S100A11 inhibitors in the preparation of drugs for the prevention and / or treatment of liver transplant rejection.

8. Application of S100A11 gene or protein in screening drugs for treating liver transplant rejection.

Images