Spp1+Carmil1 + and Spp1+Cav1 + macrophage new subgroups, screening method and application in RA diagnosis and treatment

Abstract

The invention belongs to the technical field of biological medicine, and particularly relates to a new Spp1 + Carmil1 + and Spp1 + Cav1 + macrophage subgroup, a screening method and application in diagnosis and treatment of rheumatoid arthritis (RA). Aiming at the function limitation of a traditional M1 / M2 classification model, the invention proves the following new subgroup pathological interaction mechanism: osteoclast (ACP5 +) specifically releases TNFSF11 (RANKL), and is co-localized with Spp1 + Carmil1 + and Spp1 + Cav1 + subgroup space; the Spp1 + Carmil1 + subgroup secretes IL-18, and space interaction exists between the Spp1 + Carmil1 + subgroup and Treg (Foxp3 +) and Th17 (Il17a +) cells. The invention provides a novel cell marker and a targeted treatment strategy for accurate diagnosis and treatment of RA.

Description

Technical Field

[0001] This invention belongs to the field of biomedical technology, and in particular relates to a new macrophage subpopulation, screening method and application. Background Technology

[0002] Rheumatoid arthritis (RA) is a common, chronic, and highly disabling autoimmune disease characterized by persistent synovitis, progressive destruction of articular cartilage and bone, ultimately leading to joint deformity and loss of function. Although biologics targeting pro-inflammatory factors (such as TNF-α, IL-6, and IL-17) and small-molecule targeted drugs (such as JAK inhibitors) have achieved significant clinical efficacy, a considerable proportion of patients (approximately 30%-40%) still experience poor response, relapse after discontinuation of treatment, or develop drug resistance. This indicates that the core pathogenesis of RA, especially the key cellular and molecular mechanisms driving persistent synovial inflammation and bone destruction, remains incompletely understood.

[0003] The synovial microenvironment in rheumatoid arthritis (RA) patients is a highly heterogeneous and complex niche, containing various immune cells (such as T cells and macrophages) and non-immune cells (such as osteoclasts (OCs) and fibroblast-like synovial cells (FLSs)). Among them, macrophages are considered the core drivers of RA synovial inflammation and tissue destruction, forming a key hub in the microenvironment signaling network.

[0004] Recent studies have revealed a complex multi-level signaling regulatory network: Organisms (OCs) enhance the pro-inflammatory properties of myeloid cells through signals such as RANKL; FLSs recruit and activate myeloid cells, including macrophages, by secreting chemokines (such as CCL2 and CXCL12) and pro-inflammatory factors; activated myeloid cells (with macrophages at their core) then profoundly affect the stability of Treg cell function and the differentiation of Th17 cells by secreting cytokines such as IL-23, IL-1β, and IL-6, leading to a disruption of the Treg / Th17 immune balance (manifested as decreased Treg function / Th17 amplification and the release of the pro-inflammatory factor IL-17); this, in turn, enhances the inflammatory activity of macrophages through IL-17 and other factors. Simultaneously, OCs and macrophages synergize at the forefront of bone destruction, driving a vicious cycle of inflammation and bone destruction. Multiple pieces of evidence suggest that the sustained activation of this network is a core driving force behind the progression of rheumatoid arthritis (RA).

[0005] Macrophages, as key nodes in this complex network, exhibit significant heterogeneity and plasticity in their function. However, current research on them is mainly based on an oversimplified M1 (pro-inflammatory) / M2 (anti-inflammatory / repair) bipolar differentiation model. Increasing evidence suggests that this binary model has serious limitations: 1) it fails to reflect the true heterogeneity (continuous lineage / transitional state) and functional diversity of macrophages in the synovium of RA; 2) it neglects the highly dynamic responses of macrophages to complex microenvironment signals (such as RANKL and IL-6 released by OCs) and the resulting phenotypic plasticity changes; 3) it is difficult to elucidate the specific molecular mechanisms of the dynamic interactions between macrophages and OCs / FLSs (e.g., OCs activate the macrophage NF-κB pathway through RANKL, and FLSs maintain macrophage activity through GM-CSF) and the synergistic regulation of Treg / Th17 cell plasticity differentiation. These limitations hinder the precise definition of macrophage functional subsets driving the core aspects of RA pathology.

[0006] The development of single-cell RNA sequencing (scRNA-seq) technology has provided a new perspective for understanding this complexity, enabling the mapping of cellular atlases and the identification of functional heterogeneity at the single-cell level. For example, previous studies have used this technology to identify a subset of macrophages expressing the bone-associated protein SPP1 (Osteopontin) in the synovium of rheumatoid arthritis (RA), a molecule considered highly associated with areas of bone destruction. However, existing research: 1) on Spp1... + There is a lack of consensus on whether macrophages constitute a functionally specific and independent subset; 2) there has been a failure to further refine this potentially important subset based on other specific markers (such as identifying the existence of subsets), and its heterogeneity has also been unclear; 3) these Spp1 cells have not been included in the study. + Macrophages or their potential subsets are precisely positioned at specific nodes in the aforementioned key pathological network of RA (OCs / FLSs-Treg / Th17 network and the vicious cycle of inflammation-bone destruction driven by it), and their specific mechanisms of action in specific pathological processes (such as receiving recruitment of specific cell subsets, regulating Treg / Th17 plasticity, and mediating bone destruction) are systematically elucidated (especially the specific regulated pathways and ligand-receptor pairs). For the core cell communication events driving the entire network, there is a lack of in-depth identification and functional validation of the precise ligand-receptor regulatory axes between specific macrophage subsets and other key partner cells (such as osteoclasts, fibroblasts, and Treg / Th17 cells).

[0007] The lack of understanding of the aforementioned core cell subpopulations (especially the new functionally specific macrophage subpopulations) and their interaction networks directly leads to: insufficient in-depth understanding of the driving mechanisms of key pathological links in RA; and difficulty in developing novel, efficient, and precise intervention strategies targeting specific pathogenic cell subpopulations or blocking specific pathogenic signaling axes.

[0008] Therefore, breaking through the limitations of traditional understanding, discovering and precisely screening novel, function-specific macrophage subsets (defined based on expression analysis of specific molecular combinations) at single-cell precision, capable of receiving key recruitment signals from osteoclasts and fibroblasts in synovial tissue and playing a core regulatory role in the forefront of bone destruction in the synovial microenvironment of rheumatoid arthritis (especially precisely regulating the plastic differentiation function of key regulatory cells of immune homeostasis (such as Treg / Th17 cells), and elucidating in depth the spatial distribution characteristics, potential dynamic transformation patterns, and key molecular mechanisms driving disease progression of these specific subsets in disease states, is of vital significance for deconstructing the core driving network of RA pathology and developing targeted therapeutic drugs. Summary of the Invention

[0009] To address the technical problems in existing technologies, such as the high heterogeneity of synovial macrophages in RA making it difficult to analyze their specific functional subpopulations, the lack of molecular markers specific to pathological processes, and insufficient understanding of the precise regulatory network mechanisms of different macrophage subpopulations in RA, especially in the bone destruction microenvironment, this invention provides two new functionally specific macrophage subpopulations, their screening methods, and their potential applications in the diagnosis and treatment of RA.

[0010] In a first aspect, the present invention provides a new subpopulation of macrophages present in the pathological microenvironment of rheumatoid arthritis (RA), characterized in that:

[0011] Spp1 + Carmil1 + Macrophage subsets: characterized by the simultaneous co-expression of macrophage marker CD68, bone-related protein SPP1, and adaptor protein CARMIL1 in RA synovial tissue; Spp1 + Cav1 + Macrophage subsets: characterized by the simultaneous co-expression of macrophage markers CD68, bone-associated protein SPP1, and caveolin CAV1 in RA synovial tissue.

[0012] The existence and phenotypic characteristics of the above subgroups were confirmed by multiplex immunofluorescence colocalization technology (using specific antibody combinations to verify the co-expression of CD68, SPP1 and CARMIL1 or CAV1 in cellular spatial locations).

[0013] Secondly, the present invention provides a method for screening the aforementioned Spp1. + Carmil1 + With Spp1 + Cav1 + The method for identifying new macrophage subpopulations employs a multi-level, progressive analysis workflow to pinpoint target subpopulations, specifically including:

[0014] Data integration and dimensionality reduction clustering: Quality control (number of genes detected 200-7500, mitochondrial genes account for <10%, and erythrocyte genes account for less than 3%) and batch effect correction were performed on single-cell RNA sequencing data from RA synovium. Macrophage clusters were initially divided through dimensionality reduction clustering.

[0015] Initial screening of disease-related clusters: Analysis revealed macrophage screening clusters (such as MΦ_2, MΦ_5) with a significantly increased proportion in the RA disease state, and identified that the interaction strength of these clusters with key cells such as osteoclasts is specifically enhanced in the disease.

[0016] Breakthrough in core biomarker identification and classification: SPP1 was identified as the common core biomarker of the disease-related initial screening cluster; CARMIL1 and CAV1 were identified as characteristic biomarkers to distinguish the two target subgroups through differential expression analysis; it was confirmed that the expression characteristics of the target subgroups (MΦ_2 / MΦ_5) deviated significantly from the traditional M1 / M2 polarization classification system.

[0017] Experimental validation of loop closure: Multiplex immunofluorescence analysis was performed using specific antibody combinations based on CD68, SPP1, CARMIL1 or CD68, SPP1, CAV1 to detect spatially co-localized Cd68 in RA synovial tissue. + Spp1 + Carmil1 + Or Cd68 + Spp1 + Cav1 + Triple-positive cells were used as the experimental standard for determining the target subpopulation.

[0018] In the identification and classification breakthroughs of core biomarkers, the M1 scoring gene set includes: Cd40, Il1b, Nos2, Itgax, Ccl2, Ccl5, Il18, Il6, Cd274, Tnf, Ccr7, Fcgr3, Cdh1, Il12a, Il12b, Irf5, Cd14, Msr1, Fcgr1, Cd68, Cd80, Cd86, Ly6c1, Mertk, Stat1; the M2 scoring gene set includes: Chil 3,Mrc1,Clec10a,Il4,Fcgr3,Cd200,Cdh1,Arg1,Tgfb1,Il10,Cd163,Ccl22,Itgam,Adgre1,Csf1r,Cd14,Msr1,Cd209e,Fcer1a,Irf4,Ly6c1,Retnla, use the AddModuleScore algorithm to calculate the score, and the deviation standard is: |M1-M2 score difference| < 0.1.

[0019] Thirdly, this invention reveals the functional role and regulatory network basis of the new macrophage subset in RA: (I) Core functional mechanism verified by experiments

[0020] Osteoclast-mediated regulation of macrophage subset activation: Acp5 was confirmed by multiplex immunofluorescence co-localization. + TNFSF11 ligand (RANKL) secreted by osteoclasts and Spp1 + Carmil1 + and Spp1 + Cav1 + Significant spatial correlations were observed in the spatial distribution of macrophage subsets, suggesting a potential role for osteoclast-derived TNFSF11 signaling in driving the activation of these macrophage subsets.

[0021] Regulation of T cell plasticity by macrophage subsets: Spp1 confirmed by multiplex immunofluorescence colocalization + Carmil1 + Macrophage subsets and IL-18 protein, Foxp3 + Treg cells and Il17a + Th17 cells exhibit spatial co-localization relationships in the synovial microenvironment of joints, suggesting that this subpopulation may participate in regulating the plastic differentiation process of Treg cells into Th17 cells through the IL-18 signaling pathway.

[0022] (II) Regulatory network prediction based on single-cell transcriptome analysis:

[0023] Cell-cell interaction-driven model: CellChat interaction network analysis of single-cell RNA sequencing data from RA synovial membranes revealed that osteoclasts interact with the aforementioned Spp1 cells. + There are strong interactions among macrophage subsets mediated by TNFSF11 (RANKL) ligands (the receptor is TNFRSF11A / RANK);

[0024] Spp1 + Carmil1 + Macrophage subsets and Treg / Th17 cell populations showed a tendency to interact mediated by IL-18 ligand and its receptor complex (IL18R1+IL18RAP).

[0025] Coordinated Regulation Network Extension: Fmod + Fibroblast subsets may participate in the regulation of Spp1 through the MIF signaling pathway (ligand MIF interacting with receptors CD74 / CD44 or CD74 / CXCR4). + Recruitment of macrophages;

[0026] Spp1 +Macrophage subsets may affect the functional status of Treg cells through the complement C3 signaling pathway (ligand C3 and receptor ITGAX / ITGB2).

[0027] (III) Calculation and inference of subgroup dynamic transformation relationships:

[0028] Based on pseudo-temporal analysis of single-cell transcriptome, predictive analysis of Spp1 + Carmil1 + Macrophage subsets exist towards Spp1 + Cav1 + The potential trajectory of subpopulation transformation; during the transformation process, Spp1 + Cav1 + Subpopulations may lose the ability to regulate T cell plasticity through IL-18 signaling, but retain the ability to interact with Treg cells through C3 signaling.

[0029] Fourthly: Providing the potential use of the aforementioned new macrophage subsets in the development of RA therapeutics, based on the Spp1. + Carmil1 + Spatial association between macrophage subsets and IL-18, and the relationship between osteoclast-derived TNFSF11 ligands and Spp1. + Spatial correlation of macrophage subsets, focusing on the following intervention strategies:

[0030] (1) Inhibit Spp1 + Carmil1 + Activity of IL-18 ligands in macrophage subsets;

[0031] (2) Blocking osteoclasts from binding to Spp1 via TNFSF11 ligand + Activation of macrophage subsets.

[0032] The fifth aspect: providing a drug screening method for RA based on the aforementioned new macrophage subsets, establishing an in vitro efficacy evaluation system, and detecting indicators including:

[0033] (1)Spp1 + Carmil1 + Changes in IL-18 protein expression in macrophage subsets (immunofluorescence / flow cytometry);

[0034] (2) Osteoclasts and Spp1 + Changes in the probability of interactions between macrophage subsets mediated by TNFSF11 ligand (single-cell RNA sequencing + CellChat algorithm).

[0035] The sixth aspect: providing the use of the aforementioned new macrophage subset in the preparation of a RA therapeutic drug, wherein the active ingredient of the drug is:

[0036] (1) Substances that target IL-18 ligands or their signaling pathways;

[0037] (2) Substances that target TNFSF11 ligands or their signaling pathways.

[0038] The seventh aspect: providing core targets for RA treatment, namely:

[0039] (1)Spp1 + Carmil1 + IL-18 ligands in macrophage subsets;

[0040] (2) TNFSF11 ligands derived from osteoclasts.

[0041] The beneficial effects of this invention are as follows:

[0042] (1) Defining novel pathological targets and screening methods: Using single-cell omics integrated analysis technology, a progressive screening method for RA synovial macrophage subsets was established to accurately screen out the pathology-related core subsets (Spp1). + Carmil1 + / Spp1 + Cav1 + ) and its molecular marker combination (CD68 / SPP1 / CARMIL1 or CAV1), breaking through the traditional M1 / M2 classification system and solving the problem of heterogeneity analysis;

[0043] (2) Revealing the target regulation mechanism: Discovering Acp5 + osteoclasts via RANKL→SPP1 + Macrophage axon mediates bone destruction, and Spp1 + Carmil1 + The subgroups fill a gap in the mechanism by driving a dual-pathway interaction network of inflammatory activation through the IL-18→Th17 / Treg transformation axis.

[0044] (3) Targeting diagnostic and therapeutic applications: Based on the above findings, two new intervention targets are directly provided—SPP1. + Macrophage IL-18 signaling and Acp5 + RANKL signaling in osteoclasts, along with the establishment of a targeted drug screening system, will promote precision diagnosis and treatment of rheumatoid arthritis (RA). Attached Figure Description

[0045] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments of this application will be briefly introduced below. The drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0046] Figure 1 For Spp1 + Carmil1 + and Spp1 + Cav1 + Flowchart of the construction method for new macrophage subpopulations, screening methods, and their application in RA diagnosis and treatment.

[0047] Figure 2 This is a bubble chart showing the expression patterns of specific marker gene populations in different immune cell types. The size of the bubbles represents the expression ratio, and the color represents the average expression level.

[0048] Figure 3 UMAP expression map of key marker genes in different immune cell types. Red areas indicate high expression of the gene in the corresponding cell type, and gray areas indicate low or no expression.

[0049] Figure 4 UMAP visualization of aggregated scRNA-seq data from 15,196 synovial immune cells. Eight cell populations were identified, including B cells, dendritic cells (DCs), macrophages, monocytes, neutrophils, mast cells, NK cells, and T cells. Clusters were named based on their specific gene expression patterns and color coding.

[0050] Figure 5 A bubble diagram showing the expression patterns of specific marker gene populations in different non-immune cell types. Bubble size represents the expression ratio, and color represents the average expression level.

[0051] Figure 6 This is a UMAP expression map of key marker genes in different non-immune cell types. Red areas indicate high expression of the gene in the corresponding cell type, while gray areas indicate low or no expression.

[0052] Figure 7 UMAP visualization of aggregated scRNA-seq data from 10,549 non-immune synovial cells. Six cell populations were identified, including fibroblasts, chondrocytes, osteoclasts, endothelial cells, parietal cells, and myocytes. Clusters were named based on their specific gene expression patterns and color coding.

[0053] Figure 8This is a UMAP dimensionality reduction map of the overall cell types, with different colors representing different cell types. Cell types are clustered in two-dimensional space based on their gene expression characteristics.

[0054] Figure 9 The stacked bar chart shows the proportions of various immune cells in the control group and the disease group.

[0055] Figure 10 The stacked bar chart shows the proportion of various non-immune cells in the control group and the disease group.

[0056] Figure 11 The grouped bar charts show the differences in the proportion of each immune cell type between the control group and the disease group. The marking criteria are as follows: p_adj<0.001 is “***”; 0.001≤p_adj<0.01 is “**”; 0.01≤p_adj<0.05 is “*”; p_adj≥0.05 is “ns”.

[0057] Figure 12 The grouped bar charts show the differences in the proportion of each non-immune cell type between the control group and the disease group. The marking criteria are as follows: p_adj<0.001 is “***”; 0.001≤p_adj<0.01 is “**”; 0.01≤p_adj<0.05 is “*”; p_adj≥0.05 is “ns”.

[0058] Figure 13 The UMAP diagram illustrates the different subpopulations of myeloid cells, including clusters of monocytes, different macrophage subpopulations, and dendritic cells. Each color represents a myeloid cell subpopulation.

[0059] Figure 14 The stacked bar chart shows the differences in the proportions of different myeloid cell subsets in the control and disease groups.

[0060] Figure 15 The grouped bar charts show the differences in the proportion of different myeloid cell types in the control group and the disease group. The significance criteria were: p_adj<0.001 as "***"; 0.001≤p_adj<0.01 as "**"; 0.01≤p_adj<0.05 as "*"; p_adj≥0.05 as "ns".

[0061] Figure 16 The UMAP diagram shows the different subpopulations of fibroblasts, with each color representing a different fibroblast subpopulation.

[0062] Figure 17 The bar chart shows the proportions of different fibroblast subpopulations in the HC and DG groups.

[0063] Figure 18The odds ratio (OR) analysis of fibroblast subsets in the HC group and the DG group is presented.

[0064] Figure 19 Left panel: Network diagram showing changes in the total number of cell communication events, illustrating the changes in communication events in the disease group relative to the control group. Right panel: Network diagram showing changes in the total cell communication intensity, illustrating the changes in communication intensity in the disease group relative to the control group.

[0065] Figure 20 Left panel: Overall heatmap showing the change in the number of cell communication events in the disease group relative to the control group. Right panel: Overall heatmap showing the change in cell communication intensity in the disease group relative to the control group.

[0066] Figure 21 Left: Scatter plot of overall cell signaling patterns in the control group; Right: Scatter plot of overall cell signaling patterns in the disease group.

[0067] Figure 22 A network diagram of cell communication analysis between myeloid cell subpopulations and other cell types, with the number of communication events shown on the left and the communication intensity shown on the right.

[0068] Figure 23 A heatmap of cell communication analysis between myeloid cell subpopulations and other cell types, showing the number of communication events on the left and the communication intensity on the right.

[0069] Figure 24 The cell communication network diagram shows the upregulation (red) and downregulation (blue) of the interaction strength between cell types in the disease group relative to the healthy group.

[0070] Figure 25 The cell communication heatmap shows the upregulation (red) and downregulation (blue) of the interaction strength between cell types in the disease group relative to the healthy group.

[0071] Figure 26 The scatter plot illustrates the signal transduction characteristics of different cell types in the control group (HC) and the disease group (DG).

[0072] Figure 27 The scatter plot shows the M1 and M2 scores for each cell in all macrophage subsets.

[0073] Figure 28 Box plots show the differences in M1-M2 fractions for each cell in each macrophage subset.

[0074] Figure 29 The scatter plot shows the mean M1 and M2 scores for each macrophage subset.

[0075] Figure 30The heatmap shows the differences in the distribution of mean M1 / M2 fractions among the various macrophage subsets.

[0076] Figure 31 The heatmap shows the differential gene expression among different subpopulations of myeloid cells.

[0077] Figure 32 Volcano plots illustrate differential gene expression among different myeloid cell subpopulations. The volcano plots for each subpopulation show the up- and down-regulation states of genes.

[0078] Figure 33 UMAP expression map of different macrophage subsets.

[0079] Figure 34 Gene expression density maps show the expression distribution of key genes in macrophage subsets.

[0080] Figure 35 The expression ratio of the Spp1 gene in different macrophage subsets.

[0081] Figure 36 Mean expression level of Spp1 gene in different macrophage subsets.

[0082] Figure 37 Overall arthritis score: The total score of both hind claws in the CIA group (n=4) was significantly higher than that in the healthy control group (n=4) ("****" p<0.0001).

[0083] Figure 38 Pathological manifestations of left hind claw arthritis: significant unilateral inflammation was observed in the CIA group ("***" p<0.001).

[0084] Figure 39 Pathological severity of the right hind claw: The right claw in the CIA group showed symmetrical arthritis progression ("****" p<0.0001).

[0085] Figure 40 Multiplex immunofluorescence validation of Spp1 + Carmil1 + The presence of macrophage subsets in synovial tissue and their disease-related enrichment.

[0086] Figure 41 Multiplex immunofluorescence validation of Spp1 + Cav1 + The presence of macrophage subsets in synovial tissue and their disease-related enrichment.

[0087] Figure 42 Fluorescence intensity distribution curves show the spatial co-localization characteristics of CD68 and SPP1 in the disease group.

[0088] Figure 43Fluorescence intensity distribution curves show the spatial co-localization characteristics of CD68 and CARMIL1 in the disease group.

[0089] Figure 44 Fluorescence intensity distribution curves show the spatial co-localization characteristics of SPP1 and CARMIL1 in the disease group.

[0090] Figure 45 Fluorescence intensity distribution curves show the spatial co-localization characteristics of CD68 and SPP1 in the disease group.

[0091] Figure 46 Fluorescence intensity distribution curves show the spatial co-localization characteristics of CD68 and CAV1 in the disease group.

[0092] Figure 47 Fluorescence intensity distribution curves show the spatial co-localization characteristics of SPP1 and CAV1 in the disease group.

[0093] Figure 48 The relative fluorescence intensity of CARMIL1 in the disease group (DG) was significantly higher than that in the control group (T test: "*" p<0.05). Error bars represent standard errors (SEM), and asterisks indicate statistical differences.

[0094] Figure 49 The relative fluorescence intensity of CAV1 in the disease group (DG) was significantly higher than that in the control group (T test: "**" p<0.01). Error bars represent standard errors (SEM), and asterisks indicate statistical differences.

[0095] Figure 50 The heatmap shows the differential gene expression among different fibroblast subpopulations.

[0096] Figure 51 The bubble chart shows the Acp5 levels in the HC and DG groups. + osteoclasts and Fmod + Specific ligand-receptor pairs between fibroblast subsets and macrophage subsets MΦ2 and MΦ5.

[0097] Figure 52 Multiplex immunofluorescence validation of osteoclasts (ACP5) + ) specifically releases TNFSF11 (RANKL) and binds to Spp1 + Carmil1 + Subgroup spatial co-positioning.

[0098] Figure 53 Multiplex immunofluorescence validation of osteoclasts (ACP5) + ) specifically releases TNFSF11 (RANKL) and binds to Spp1 + Cav1 + Subgroup spatial co-positioning.

[0099] Figure 54 The relative fluorescence intensity of ACP5 in the disease group (DG) was significantly higher than that in the control group (T test: "****" p<0.0001). Error bars represent standard errors (SEM), and asterisks indicate statistical differences.

[0100] Figure 55 The relative fluorescence intensity of TNFSF11 in the disease group (DG) was significantly higher than that in the control group (T test: "****" p<0.0001, error bars represent standard errors (SEM), and asterisks indicate statistical differences).

[0101] Figure 56 UMAP dimensionality reduction clustering analysis of T cell subsets (grouped by cluster).

[0102] Figure 57 UMAP dimensionality reduction clustering analysis of T cell subsets (grouped by cell type).

[0103] Figure 58 Expression of specific marker genes for T cell subsets.

[0104] Figure 59 Stacked bar charts show the differences in the proportions of different T cell subsets in the control and disease groups.

[0105] Figure 60 The grouped bar charts show the differences in the proportions of different T cell subsets in the control group and the disease group. The significance criteria are: p_adj<0.001 is “***”; 0.001≤p_adj<0.01 is “**”; 0.01≤p_adj<0.05 is “*”; p_adj≥0.05 is “ns”.

[0106] Figure 61 The cell communication heatmap shows the upregulation (red) and downregulation (blue) of the interaction strength between cell types in the disease group relative to the healthy group.

[0107] Figure 62 The bubble diagram shows the specific ligand-receptor pairs of Spp1+ macrophages and various T cell subsets in the HC and DG groups.

[0108] Figure 63 Spp1 + Carmil1 + Macrophages send signaling molecules to various T cell subsets that are upregulated in the disease group.

[0109] Figure 64 Spp1 + Carmil1 + Macrophages send receptor pairs to various T cell subsets that are upregulated in the disease group.

[0110] Figure 65 Spp1 + Cav1 + Macrophages send signaling molecules to various T cell subsets that are upregulated in the disease group.

[0111] Figure 66 Spp1 + Cav1 + Macrophages send receptor pairs to various T cell subsets that are upregulated in the disease group.

[0112] Figure 67 and Figure 68 Spp1 + Carmil1 + and Spp1 + Cav1 + Pseudo-temporal analysis of macrophages.

[0113] Figure 69 and Figure 70 Pseudo-temporal analysis of Treg and Th17 cells.

[0114] Figure 71 Multiplex immunofluorescence assay of Spp1 in the bone-synovial junction region + Carmil1 + Subgroups secrete IL-18, and are associated with Treg(Foxp3) + ) and Th17(Il17a + Cells exhibit spatial interactions, and this region contains Cav1 + Low expression.

[0115] Figure 72 The relative fluorescence intensity of IL18 in the disease group (DG) was significantly higher than that in the control group (T test: "*" p<0.05). Error bars represent standard errors (SEM), and asterisks indicate statistical differences.

[0116] Figure 73 The relative fluorescence intensity of IL17A in the disease group (DG) was significantly higher than that in the control group (T test: "*" p<0.05). Error bars represent standard errors (SEM), and asterisks indicate statistical differences.

[0117] Figure 74 The ratio of relative fluorescence intensity of IL-17A to FOXP3 in the disease group (DG) was significantly higher than that in the healthy control group (HC group) (T test: "**" p<0.05). Detailed Implementation

[0118] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions of the present invention will be further described below with reference to the accompanying drawings and embodiments. The following embodiments are for illustrative purposes only and should not be considered as limiting the scope of the present invention. Figure 1 For Spp1 + Carmil1 + and Spp1 + Cav1 + Flowchart of the construction method for new macrophage subpopulations, screening methods, and their application in RA diagnosis and treatment.

[0119] Example 1: Spp1 + Carmil1 + Macrophages and Spp1 + Cav1 + The screening and validation of new macrophage subpopulations were carried out using version R-4.3.2 in this embodiment.

[0120] Step 1: Data Source and Compliance

[0121] (1) This invention is based on mouse synovial single-cell transcriptome data from the NCBI GEO public database (accession numbers: GSE184609 and GSE192504).

[0122] (2) Data acquisition complied with the terms (https: / / www.ncbi.nlm.nih.gov / geo / info / disclaimer.html). The dataset selection criteria met the following conditions: clearly grouped healthy control group (n=11) and disease group (n=6); both used 10x Genomics platform single-cell RNA sequencing.

[0123] Step 2: Data Preprocessing and Batch Correction

[0124] (1) Quality control was performed using Seurat software (v5.0), including cell filtering based on preset thresholds: the number of gene detections was between 200 and 7500 to exclude low-information cells (such as cell debris or dead cells) and multi-cell capture noise; the proportion of mitochondrial genes was less than 10% to eliminate interference from apoptotic cells (a high proportion of mitochondrial genes indicates poor cell condition), with mitochondrial gene identification based on prefix 'mt-' pattern matching; the proportion of erythrocyte genes was less than 3% to exclude erythrocyte contamination, with the erythrocyte gene list including Hba-a1, Hba-a2, Hbb-b1, Hbb-b2, Hbb-bs, Hbb-bt, Hbb-bh1, Hbb-y, Hbe1, and Hba-x, sourced from the ImmGen database; and the UMI count was less than 100,000 to remove abnormally high-expression cells (possibly due to technical noise or multi-cell capture). After quality control, 23,567 effective cells were retained, covering 20,010 genes.

[0125] (2) Data integration was completed by processing the merged dataset of 17 samples using the merge function. Subsequently, a standardization process was performed, which included: a normalization step using the LogNormalize method with the scale.factor parameter set to 10000; a hypervariable gene screening step using the vst method with the nfeatures parameter set to 3000; a standardization step using the ScaleData function on all genes; a PCA dimensionality reduction step using hypervariable genes as input features; and a batch correction step implemented using the RunHarmony function, where the group.by.vars parameter was set to orig.ident to group the samples, and the dims.use parameter was set to dimensions 1 to 30 for dimensionality reduction, thereby eliminating batch effects and ensuring data consistency.

[0126] Step 3: Construction of a single-cell heterogeneity map of synovial membrane

[0127] (1) Cell clustering and dimensionality reduction: The FindNeighbors function is used to construct a K-nearest neighbor graph (KNN graph) based on the first 20 principal components (parameter dims = 1:20); Louvain algorithm clustering is performed by the FindClusters function (parameter resolution = 0.8); and nonlinear dimensionality reduction visualization is performed by the RunUMAP function.

[0128] (2) Immune cell subset identification: Based on the ImmGenData and MouseRNAseqData reference databases, initial annotation was performed automatically using the SingleR package. Manual verification criteria: ≥80% of cells in the target cell population must express at least one of the marker genes listed in Table 1 for that subset, and the average expression value (log-normalized) of that gene in the target population must be ≥1.0, while the average expression value in the non-target population must be ≤0.5. The following eight types of immune cells were identified: macrophages (marker genes: Cd14, Cd68, Itgam, Adgre1), T cells (marker genes: Cd3e, Cd3d, Trbc2, Bcl11b, Il7r, Cd4), B cells (marker genes: Cd19, Cd79b, Cd79a, Cd37, Pax5), dendritic cells (marker genes: Itgax, Cd74, Irf8, Flt3), monocytes (marker genes: Ly6c2, Ccr2, Cd14, Fcgr3), neutrophils (marker genes: S100a8, S100a9), mast cells (marker genes: Kit, Cpa3), and natural killer cells (marker genes: Tbx21, Klrc2, Klrc1, Klrb1c, Nkg7). The expression of marker genes in each subgroup and the cell annotation results are presented using bubble charts and UMAP dimensionality reduction (see appendix for details). Figures 2 to 4 ).

[0129] (3) Identification of non-immune cell subsets: Using the same automated annotation (SingleR package) and manual verification process (standard as in step 3(2) of Example 1), the following 6 types of non-immune cells were identified: fibroblasts (marker genes: Thy1, Col5a2, Fn1, Lgals3, Vim), osteoclasts (marker genes: Ctsk, Acp5), chondrocytes (marker genes: Acan, Comp), endothelial cells (marker genes: Pecam1, Cdh5), myocytes (marker genes: Ttn, Myh1, Myh4), and parietal cells (marker genes: Acta2, Myh11, Tagln, Pdgfrb, Rgs5). The expression of marker genes in each subset and the cell annotation results were displayed using bubble charts and UMAP dimensionality reduction (see Appendix for details). Figures 5 to 7 ).

[0130] (4) Panoramic Atlas Integration: The above 14 cell subpopulations (8 immune cell subpopulations + 6 non-immune cell subpopulations) were integrated to construct a single-cell reference atlas of the synovium. All annotation results are stored in a metadata file (see appendix for details). Figure 8 ).

[0131] Table 1. References for Overall Cell Map and Marker Genes

[0132]

[0133]

[0134] Step 4: Initial screening of RA-related target cell populations

[0135] (1) Screening for abnormal proportions

[0136] Extract metadata from the Seurat object, and count cells by grouping them according to control / disease (HC_DG) and cell type: Calculate the percentage of each cell type in each group: Freq = (count / total number of cells in the sample) × 100; plot the results using a stacked bar chart (see appendix for details). Figures 9 to 10 ) and grouped bar chart (see appendix for details) Figures 11 to 12 Visualization revealed that, except for mast cells, the proportions of other immune cell types in the disease group were significantly higher than those in the control group, and the proportion of osteoclasts in the disease group was also significantly higher than that in the control group. Statistical analysis was performed using the Wilcoxon rank-sum test (Mann-Whitney U test).

[0137] Further dimensionality reduction and clustering analysis were performed on myeloid cells from the synovial tissue of mice in both the disease and control groups. UMAP dimensionality reduction (principal component count = 20) and Louvain clustering algorithm (resolution parameter = 1.2) were used to classify classical monocytes, dendritic cells, and eight different macrophage subpopulations, defined as MΦ_0, MΦ_1, MΦ_2, MΦ_3, MΦ_5, MΦ_6, MΦ_7, and MΦ_9, respectively (see appendix for details). Figure 13 Proportional analysis revealed that the proportions of MΦ_2 and MΦ_5 were significantly higher in the disease group than in the control group (see appendix for details). Figures 14 to 15 The above statistical analysis was performed using the Wilcoxon rank-sum test (Mann-Whitney U test).

[0138] To further explore the heterogeneity of fibroblasts, we first performed dimensionality reduction and cluster analysis using UMAP (principal component count = 20) and Louvain's algorithm for clustering (resolution parameter = 1.2), dividing them into 10 distinct subpopulations (Fi0-Fi9) (see appendix for details). Figure 16 Among them, the proportions of Fi0, Fi1, Fi4, and Fi9 subsets increased in the disease group. Freq = (count / total number of cells in the sample) × 100 (see appendix for details). Figure 17Odds ratio (OR) analysis was used to quantify the difference in the proportion of subsets between the disease group and the control group. Key parameters included: statistical model: Fisher's exact test (two-sided); significance threshold: p < 0.05 after Benjamini-Hochberg correction; OR effect threshold: > 1.5 (enrichment) or < 0.5 (reduction). Odds ratio analysis revealed that the proportions of the Fi1 and Fi9 subsets were significantly upregulated in the disease group (see appendix for details). Figure 18 ).

[0139] (2) Interaction Network Enhancement Verification

[0140] Cell communication networks in healthy controls (HC) and disease groups (DG) were modeled using the CellChat software package. The specific steps are as follows: ① Normalized gene expression matrices and cell type annotation information were extracted from preprocessed Seurat objects; ② Communication analysis was performed based on the "Secreted Signaling" pathway in the CellChatDB.mouse database; ③ Parameters were set: raw.use = TRUE (using raw data) was set when calculating intercellular communication probabilities, and cell populations with fewer than 3 cells were filtered (min.cells = 3); ④ Protein-protein interaction (PPI) mapping was performed on ligand-receptor interactions to improve prediction accuracy; ⑤ Communication strength at the signaling pathway level was inferred using computeCommunProbPathway; ⑥ The aggregateNet function was used to integrate the cell type-level communication network.

[0141] Intergroup communication network comparison: ① Overall interaction strength comparison: Calculation using the compareInteractions function revealed a significant increase in the number of interaction events in the disease group compared to the control group (p < 0.001, Wilcoxon rank-sum test); ② Key interaction changes: In the disease group, osteoclasts and fibroblasts, acting as signal senders (especially the Fi1 fibroblast subset), significantly enhanced the intensity of signal transmission to myeloid cells (especially macrophage subsets MΦ_2 and MΦ_5) (see appendix for details). Figures 19 to 26 ).

[0142] Step 5: Double-locking of marker genes

[0143] (1) The macrophage subset exhibiting significantly enhanced cell communication with osteoclasts and Fi1 fibroblasts was verified to deviate from the traditional polarization axis using the M1 / M2 polarization score. The M1 scoring gene set included: Cd40, Il1b, Nos2, Itgax, Ccl2, Ccl5, Il18, Il6, Cd274, Tnf, Ccr7, Fcgr3, Cdh1, Il12a, Il12b, Irf5, Cd14, Msr1, Fcgr1, Cd68, Cd80, Cd86, Ly6c1, Mertk, Stat1; the M2 scoring gene set included: Chil3, Mrc1, Clec10a, Il4, Fcgr3, Cd200, Cdh1, Arg1, Tgfb1, Il10, Cd163, Ccl22, Itgam, Adgre1, Csf1r, Cd14, Msr1, Cd209e, Fcer1a, Irf4, Ly6c1, Retnla, use the AddModuleScore algorithm to calculate the score, and the deviation standard is: |M1-M2 score difference| < 0.05.

[0144] Table 2 M1 / M2 polarization scores

[0145]

[0146] The results show that both MΦ_2 and MΦ_5 subgroups obtained from the initial screening in step 4 are significantly deviated from the traditional M1 / M2 polarization axis (see appendix for details). Figure 27-30 ).

[0147] (2) Next, the gene expression differences among myeloid cell subpopulations were analyzed. Initial screening (Seurat algorithm): Differentially expressed genes (DEGs) analysis was performed using Seurat (v5.0). FindAllMarkers was used to identify differentially expressed genes among different cell subpopulations, and statistical analysis was performed using the Wilcoxon rank-sum test. The screening criteria were set as expression in at least 25% of cells (min.pct = 0.25) and log2(foldchange) ≥ 0.25. High-confidence screening (custom validation function): Corrected p-value < 0.05; core parameters: mean log2FC > 0.5 (enhancing the magnitude of difference), specificity ratio > 1.5 (ensuring subpopulation specificity), expression difference value > 0.1 (excluding background fluctuations), target population expression ratio > 15% (pct.1 > 0.15). Potential differentially expressed genes were captured through wide-threshold initial screening, and high-confidence screening identified genes related to high-confidence subpopulations. The results revealed the characteristic gene expression patterns of each subpopulation. Based on this, we further analyzed the genes that were highly upregulated and downregulated among the subpopulations. Using these specific molecular markers, we tentatively named the MΦ_2 subpopulation Carmil1. +Macrophage subsets, the MΦ_5 subset was named Cav1. + Macrophage subsets (see appendix for details) Figures 31-32 ).

[0148] Next, calculate Carmil1 separately. + Macrophage subsets and Cav1 + Macrophage subset gene specificity score = (pct.1 - max_pct_other) × avg_log2FC × log10(pct.1 / (max_pct_other + 0.001) + 1) (where max_pct_other is the highest expression proportion of other myeloid subsets); genes that meet any of the following criteria are retained: corrected p-value < 0.05, log2FC > 0.5, expression difference > 0.1, specificity proportion > 15%; genes highly expressed outside the target macrophage subset are excluded: maximum expression proportion < 20% (max_pct_non_mac < 0.2), average expression proportion < 10%.

[0149] (mean_pct_non_mac < 0.1), differential expression intensity > 30% (diff_to_non_mac > 0.3). Final selection also met the following criteria: in Carmil1 + Macrophage subsets and Cav1 + Macrophage subsets showed significant expression (p<0.05 & log2FC>0.3) but low expression in non-target macrophage subsets (under the same conditions). We observed that the Spp1 gene, one of the upregulated genes mentioned above, was expressed in Carmil1. + Macrophage subsets and Cav1 + Carmil1 was simultaneously highly expressed in macrophage subsets. + Macrophage subsets and Cav1 + The macrophage subsets were named Spp1. + Carmil1 + Subgroups and Spp1 + Cav1 + Subgroups (see appendix for details) Figure 33-36 ).

[0150] Step 6: CIA mouse modeling

[0151] (1) Preparation of CII emulsion for main drugs (for immunization injection)

[0152] Ingredients: Chicken type II collagen (CII) solution (2 mg / mL) and complete Freund's adjuvant (CFA) are mixed in equal volumes (1:1). Preparation steps: ① Mix the CII solution and CFA on ice; ② Emulsify using a high-speed stirrer, initially mixing at low speed, then switching to high speed for thorough stirring until a stable water-in-oil emulsion is formed (test standard: the emulsion does not spread when dropped into water); ③ The final CII concentration is 1 mg / mL. This preparation should be prepared immediately before use and should be avoided from storage. Precautions: The emulsification process requires strict temperature control (operation on ice) to prevent protein denaturation.

[0153] (2) Establishment of CIA mouse model

[0154] Experimental animals: Mice (C57BL / 6J mice), with 4 mice in the model group and 4 mice in the control group. Immunization schedule: On day 0 (primary immunization), CII emulsion was injected intradermally into the base of the tail of each mouse at a dose of 0.1 mL / mouse; on day 21 (boost immunization), the same dose of CII emulsion (0.1 mL / mouse) was injected intradermally again into the base of the tail. Successful model assessment: Confirmed by the Arthritis Index (AI) score (see Table 3).

[0155] (3) Arthritis Index (AI) score

[0156] Assessment time point: Day 28 post-primary immunization. Assessment method: Observers: Two independent off-site personnel were used to avoid subjective bias. Scoring criteria: The degree of erythema and swelling on the limbs (forelimbs and hindlimbs) of each mouse was scored. Detailed criteria are as follows:

[0157] Table 3 Arthritis Index Scoring Criteria

[0158]

[0159] Total score calculation: AI score = sum of limb scores, maximum score 16 points. Data integration: The average of the scores from two observers is taken as the final AI score. Model success criteria: AI score > 2 points is considered a successful model; if scores are inconsistent, they need to be reviewed or outlier data needs to be removed (see appendix for details). Figures 37-39 ).

[0160] (4) Experimental grouping and treatment

[0161] Group design: Group A (normal blank control group, n=4): Mice were not induced with CIA, but only treated by gavage. 0.2 mL of 0.9% sodium chloride solution (physiological saline) was administered by gavage at 9:00 AM daily. Group B (CIA model group, n=4): Mice underwent the above CIA induction, and after model establishment, 0.2 mL of 0.9% physiological saline was administered by gavage at 9:00 AM daily.

[0162] Step 7: Validation by multiplex immunofluorescence analysis

[0163] (1) Tissue sample preparation

[0164] ① Sample source: Knee joint tissue of collagen-induced arthritis (CIA) model mice (n=2) and normal controls (n=2). ② Section preparation: Tissue type: paraffin-embedded sections (thickness: 4μm); Instrument: Thermo HM325 microtome.

[0165] ③ Dewaxing process: Place the slides with the attached sections at an angle into the baking tray and bake at 60℃ for 2-4 hours. Dewax and remove xylene in the following sequence: three consecutive 5-minute cycles of xylene (the number of cycles can be increased depending on the dewaxing situation) → two 5-minute cycles of 100% ethanol → one 5-minute cycle of 95% ethanol → one 5-minute cycle of 70% ethanol → one 5-minute cycle of distilled water. (2)

[0167] Table 4 Antibody Combinations and Parameters

[0168]

[0169] (3) Procedure and method of multiplex immunofluorescence assay

[0170] ① Antigen retrieval: Place the slide in retrieval solution and microwave for retrieval. Then remove the slide and cool to room temperature, then rinse with tap water for 5 minutes. ② Use a hydrophobic pen to draw circles around the slide, 0.5 mm away, to prevent loss during subsequent solution addition. ③ Rinse three times with PBST buffer, 5 minutes each time. ④ Inactivate endogenous peroxidase: Add 3% H2O2 methanol solution to the slide and incubate at room temperature. Then rinse three times with PBST. ⑤ Blocking: Use immunohistochemical blocking solution or the blocking solution provided with the kit to block non-specific binding sites at room temperature. ⑥ Primary antibody incubation: Dilute and incubate the antibody according to the antibody instructions, incubating at room temperature or overnight at 4°C. ⑦ Rinse three times with PBST buffer, 5 minutes each time, to remove unbound primary antibody. ⑧ Secondary antibody incubation: Add secondary antibody according to the instructions and incubate at room temperature. ⑨ Rinse three times with PBST buffer, 5 minutes each time, to remove unbound secondary antibody. ⑩ Fluorescent dye incubation: Dilute the dye and signal amplification solution according to the ratio and add them to the slide. Wash three times with PBST buffer for 5 minutes each time to remove fluorescent dye. Antibody elution: Method (1): Add antibody elution buffer and wash three times with PBST buffer. Method (2): Perform antigen retrieval according to step ①, cool to room temperature, and wash three times with PBST buffer. Repeat step ①- Steps ②-④ do not need to be repeated until all indicators are stained. DAPI: Add DAPI staining solution and incubate at room temperature. Wash three times with PBST buffer for 5 minutes each time to remove DAPI. Add an anti-fluorescence quencher, cover with a coverslip, and you can start scanning.

[0171] (4) Combination of core detection indicators

[0172] Target subgroup: Spp1 + Carmil1 + Macrophages (marker combination: CD68) + SPP1 + CARMIL1); Spp1 + Cav1 + Macrophages (marker combination: CD68) + SPP1 + CAV1)

[0173] (5) Tissue imaging and analysis equipment

[0174] Panoramic section scanning was performed using a 3D HISTECH Pannoramic MIDI fully automated microscopy scanning system (model: PMIDI-FL), and multispectral analysis was performed using an AKOYA Vectra Polaris microscopy system. TM Multiple fluorescence imaging system (software version: AKOYAInform 2.6) was used for microscopic observation, and a Carl Zeiss microscope (model: Axioscope A1) was used.

[0175] (6) Fluorescence verification results

[0176] ① Cellular localization characteristics: CD68 + Macrophages with SPP1 / CARMIL1 or CD68 + Macrophages exhibit extensive co-localization with SPP1 / CAV1 (see appendix for details). Figures 40-41 ); ② Colocalization patterns: CD68-SPP1, CD68-CARMIL1, SPP1-CARMIL1; CD68-SPP1, CD68-CAV1, SPP1-CAV1 fluorescence signals exhibit synchronous fluctuations in spatial distribution (see appendix for details). Figure 42-47 ③ Fluorescence intensity verification: The fluorescence intensity of CARMIL1 and CAV1 in the disease group was significantly higher than that in the healthy control group (HC) (see appendix for details). Figures 48-49 ).

[0177] Example 2: Fmod + Fibroblasts / Acp5 + Osteoclast-Spp1 + Analysis of the interaction mechanism of new macrophage subsets

[0178] Step 1: Single-cell sequencing analysis of Fmod + Fibroblasts / Acp5 + Osteoclast-Spp1 + Macrophage New Subpopulation Interaction Network (1) In Example 1, we identified the Fi1 fibroblast subpopulation and directed it towards Spp1 + The signaling intensity of the new macrophage subsets was significantly enhanced. Next, differentially expressed genes among fibroblast subsets were analyzed. Initial screening (Seurat algorithm): Differentially expressed gene analysis was performed using Seurat (v5.0). FindAllMarkers was used to identify differentially expressed genes among different cell subsets, and statistical analysis was performed using the Wilcoxon rank-sum test. The screening criteria were set as expression in at least 25% of cells (min.pct = 0.25) and log2(foldchange) ≥ 0.25. High-confidence screening (custom validation function): Corrected p-value < 0.05; core parameters: avg_log2FC > 0.5 (enhancing differential amplitude), specificity ratio > 1.5 (ensuring subset specificity), expression differential value > 0.1 (excluding background fluctuations), target population expression ratio > 15% (pct.1 > 0.15). Based on specific molecular markers, we named the Fi1 fibroblast subset Fmod. + Fibroblasts (see appendix for details) Figure 50 ).

[0179] (2) Algorithm basis: CellChat package. Input data: Seurat object output from Example 1 (extracting Fmod). + Fibroblasts / Acp5 + osteoclasts and Spp1 + (New macrophage subsets). Core parameter settings: Ligand-receptor database: CellChatDB.mouse (limited to the "secretion signaling" pathway).

[0180] (3) Interaction network: Fmod + Fibroblasts → Spp1 + New macrophage subsets: Significantly upregulated pathways: MIF signaling pathway (Mif-Cd74 / Cd44); Acp5 + osteoclasts → Spp1 + New macrophage subsets: Significantly upregulated pathway: RANKL signaling pathway (Tnfsf11-Tnfrsf11a) (see appendix for details) Figure 51 ).

[0181] Step 2: Multiplex immunofluorescence validation of Spp1+Carmil1+ and Spp1+Cav1+ macrophage subsets receiving activation of Acp5+ osteoclasts via TNFSF11 signaling

[0182] (1) The preparation of tissue samples and experimental methods are the same as in Example 1. (2)

[0184] Table 5 Antibody Combinations and Parameters

[0185]

[0186] (3) In a collagen-induced arthritis (CIA) disease model, multiplex immunofluorescence assays revealed that ACP5 was present in the bone-synovial junction. + TNFSF11 ligand (RANKL) secreted by osteoclasts and Carmil1 + Macrophages form subcellular spatial colocalization. ACP5 in the synovial matrix. + osteoclasts and Cav1 + Macrophages establish a ternary colocalization complex via TNFSF11 (see appendix for details). Figures 52-53 ).

[0187] (4) Fluorescence intensity verification: The fluorescence intensity of ACP5 and TNFSF11 in the disease group was significantly higher than that in the healthy control group (HC) (see appendix for details). Figures 54-55 ).

[0188] Example 3: Spp1 + New macrophage subsets regulate Foxp3 + Treg cells / Il17a + Step 1 in elucidating the mechanism of Th17 cell plasticity differentiation: Construction of a heterogeneity map of T cell subsets

[0189] T cell subset dimensionality reduction clustering and identification using UMAP: ① Input data: T cell subset expression matrix (T cells extracted from the Seurat object in Example 1). ② Algorithm parameters: Clustering based on the Louvain algorithm (FindClusters, resolution = 1.2), the first 20 principal components (dims = 1:20), and different T cell subsets annotated with different marker genes (see appendix for details). Figures 56-58 ).

[0190] Table 6. References for T cell subset marker genes

[0191]

[0192] Step 2: Analysis of differences in proportions between groups

[0193] Metadata was extracted from T cell Seurat objects, and cell counts were grouped by control / disease (HC_DG) and cell type. The percentage of each cell type was calculated: Freq = (count / total number of cells in the sample) × 100; a stacked bar chart was then generated (see appendix for details). Figure 59 ) and grouped bar chart (see appendix for details) Figure 60 Visualization was performed. The proportions of T cell subsets (including Treg, CD8+ T, Th17, Th2, and NK cells) in the disease group were all increased to varying degrees compared to the control group. Statistical analysis was conducted using the Wilcoxon rank-sum test (Mann-Whitney U test).

[0194] Step 3: Single-cell sequencing analysis of Spp1 + A new macrophage subset—Foxp3 + Treg cells / Il17a + Th17 Cell Interaction Network (1) Algorithm Basis: CellChat package. Input Data: Seurat object output from Example 1 (including Spp1) + (New macrophage subsets and various T cell subsets). Core parameter settings: Ligand-receptor database: CellChatDB.mouse (limited to the "secretion signaling" pathway).

[0195] (2) Interaction Network: Spp1 + Carmil1 + Macrophage subsets and Treg / Th17 cell populations showed a tendency for interaction mediated by IL-18 ligand and its receptor complex (IL18R1+IL18RAP); Spp1 + Macrophage subsets may affect Treg cell functional status through the complement C3 signaling pathway (ligand C3 and receptor ITGAX / ITGB2) (see appendix for details). Figures 61-66 ).

[0196] Step 4: Revealing Foxp3 time-series trajectories + Treg cells / Il17a + Th17 cell plasticity differentiation and Spp1 + Carmil1 + Macrophage subsets to Spp1 + Cav1 + Subgroup transformation

[0197] (1) Data preparation: Data on target macrophage subsets and target T cell subsets were extracted from Seurat objects in Example 1.

[0198] (2) Single-cell dataset construction: Create a standardized data structure and set key parameters. Expression level detection threshold: 0.5 (to filter low-confidence signals); Data distribution model: negative binomial distribution (suitable for single-cell count data).

[0199] (3) Data preprocessing: ① Eliminate technical bias by size factor correction; ② Screen effective genes based on gene expression dispersion (average expression level ≥ 0.1); ③ Estimate gene expression fluctuation model.

[0200] (4) Screening of key driver genes: A dual-filtering strategy based on dispersion was used to calculate gene expression fluctuation characteristics. Screening criteria: baseline expression level ≥ 0.1; measured dispersion ≥ theoretical dispersion.

[0201] (5) Trajectory Construction and Calculation: Dimensionality reduction was performed using the DDRTree algorithm (maximum number of components = 2). Calculation and assignment of: cell differentiation state (State) and pseudotime score (Pseudotime).

[0202] (6) Calculation and inference of subgroup dynamic transformation relationships:

[0203] Based on pseudo-temporal analysis of single-cell transcriptome, predictive analysis of Spp1 + Carmil1 + Macrophage subsets exist towards Spp1 + Cav1 + The potential trajectory of subpopulation transformation; during the transformation process, Spp1 + Cav1 + Subpopulations may lose the ability to regulate T cell plasticity through IL-18 signaling, but retain the ability to interact with Treg cells through C3 signaling. Foxp3 + Treg cells / Il17a + Th17 cells exhibit plastic differentiation (see appendix for details). Figures 67-70 ).

[0204] Step 5: Multiplex immunofluorescence verification of Spp1 + Carmil1 + Subpopulations drive Treg cells to differentiate into Th17 cells through IL-18 signaling.

[0205] (1) The preparation of tissue samples and experimental methods are the same as in Example 1. (2)

[0207] Table 7 Antibody Combinations and Parameters

[0208]

[0209] (3) In a collagen-induced arthritis (CIA) disease model, multiplex immunofluorescence assays revealed that Carmil1 was present in the osteosynovial junction. + IL18 ligand secreted by macrophages and Foxp3 + Treg cells / Il17a + Th17 cells exhibit subcellular spatial colocalization, and this region shows low expression of Cav1. + Macrophages (see appendix for details) Figure 71 ).

[0210] (4) Fluorescence intensity verification: The fluorescence intensity of IL-18 and IL-17A in the disease group was significantly higher than that in the healthy control group (HC), and the relative fluorescence intensity ratio of IL-17A to FOXP3 in the disease group (DG) was significantly higher than that in the healthy control group (HC group) (see Appendix for details). Figures 72-74 ).

Claims

1. A novel subset of RA-associated macrophages, characterized in that, This subgroup satisfies any of the following characteristics: (1)Spp1 + Carmil1 + Macrophage subsets: co-express macrophage markers CD68, bone-associated protein SPP1, and adaptor protein CARMIL1; (2)Spp1 + Cav1 + Macrophage subsets: co-express macrophage markers CD68, bone-associated protein SPP1, and caveolin CAV1.

2. The new macrophage subpopulation according to claim 1, characterized in that, Acp5 + Osteoclasts regulate Spp1 via TNFSF11 signaling + Carmil1 + And / or activation of Spp1+Cav1+ macrophage subsets.

3. The new macrophage subpopulation according to claim 1, characterized in that, Spp1 + Carmil1 + Subpopulations drive Treg cells to differentiate into Th17 cells via IL-18 signaling.

4. A method for screening RA-associated macrophage subsets, characterized in that, This screening method is used to screen for new macrophage subpopulations as described in claim 1, including: Data integration, quality control and RunHarmony batch calibration of RA synovial single-cell RNA sequencing data; The target cluster was identified, and macrophage subsets with significantly increased proportions and significantly enhanced cell communication with osteoclasts were screened in the established disease group. A breakthrough in classification was achieved by verifying, through M1 / M2 polarization scoring, that a macrophage subset with significantly enhanced cell communication with osteoclasts deviated from the traditional polarization axis; Biomarker identification: Differential expression analysis confirmed SPP1 as Spp1. + Carmil1 + Subgroups and Spp1 + Cav1 + Subgroup common marker, CARMIL1 is Spp1 + Carmil1 + Subgroup marker, CAV1 is Spp1 + Cav1 + Subgroup markers; Experimental verification showed that multiplex immunofluorescence was performed using a combination of CD68 / SPP1 / CARMIL1 or CD68 / SPP1 / CAV1 antibodies, with spatially co-localized triple-positive cells as the criterion.

5. The screening method according to claim 4, characterized in that, In the classification breakthrough step, the M1 scoring gene set includes: Cd40, Il1b, Nos2, Itgax, Ccl2, Ccl5, Il18, Il6, Cd274, Tnf, Ccr7, Fcgr3, Cdh1, Il12a, Il12b, Irf5, Cd14, Msr1, Fcgr1, Cd68, Cd80, Cd86, Ly6c1, Mertk, Stat1; the M2 scoring gene set includes: Chil3, Mr c1,Clec10a,Il4,Fcgr3,Cd200,Cdh1,Arg1,Tgfb1,Il10,Cd163,Ccl22,Itgam,Adgre1,Csf1r,Cd14,Msr1,Cd209e,Fcer1a,Irf4,Ly6c1,Retnla, use the AddModuleScore algorithm to calculate the score, and the deviation standard is: |M1-M2 score difference| < 0.

1.

6. A method for non-therapeutic RA pathological regulation, characterized in that, Regulation methods include: (1)Spp1 + Carmil1 + and Spp1 + Cav1 + Macrophage subsets receive Acp5 + Osteoclasts are activated via TNFSF11 signaling, and their interaction was verified by co-localization of ACP5 / TNFSF11 / CARMIL1 or ACP5 / TNFSF11 / CAV1 antibodies. (2) Activated Spp1 + Carmil1 + The subset drives Treg cells to differentiate into Th17 cells through IL-18 signaling, and its function was verified by co-localization of CARMIL1 / IL-18 / FOXP3 / IL-17A antibodies.

7. A kit for assessing the pathological status of RA, characterized in that, The kit includes: a primary antibody, which comprises anti-CD68 antibody, anti-SPP1 antibody, anti-CARMIL1 antibody, and anti-CAV1 antibody; The kit uses multiplex immunofluorescence technology to detect Spp1, which co-expresses CD68, SPP1, and CARMIL1, in synovial tissue samples from subjects. + Carmil1 + Macrophage subsets and Spp1 cells that co-express CD68, SPP1, and CAV1 + Cav1 + Spatial location and relative number of macrophage subsets.

8. A method for in vitro screening of RA drugs, characterized in that, Includes the following steps: Establish or use cells derived from RA synovial tissue or in vitro models established using said cells, said cells or in vitro models containing the Spp1 as described in claim 1. + Carmil1 + Macrophage subsets and / or Spp1 + Cav1 + Macrophage subsets; The candidate compound was applied to the in vitro model; Detection indicators: (a) the Spp1 mentioned above + Carmil1 + (a) Changes in the expression levels of cell-associated IL-18 protein in macrophage subsets; (b) the Spp1 + Carmil1 + Macrophage subsets and / or Spp1 + Cav1 + Macrophage subsets and Acp5 + Changes in the strength or probability of intercellular interactions mediated by TNFSF11 ligand in the osteoclast interaction network.

9. The Spp1 according to claim 1 + Carmil1 + Macrophage subsets and / or Spp1 + Cav1 + The use of macrophage subsets in the preparation of drugs for treating RA, characterized in that, The drug has any one of the following functions: (1) Specific intervention of the Spp1 + Carmil1 + The activity or function of macrophage subsets, particularly the activity of the IL-18 signaling pathway that targets and inhibits the secretion of these subsets; (2) Blocking Acp5 + TNFSF11 ligands produced or secreted by osteoclasts target the Spp1 + Carmil1 + Macrophage subsets and / or Spp1 + Cav1 + Activation of macrophage subsets.

10. The use according to claim 9, characterized in that: (1) The drug contains a substance that targets the IL-18 signaling pathway; (2) The drug contains a substance that targets the TNFSF11 ligand or its receptor signaling pathway.

Images