Molecular markers for colorectal cancer prognosis evaluation and application thereof

Abstract

The application discloses a molecular marker for colorectal cancer prognosis evaluation and application thereof, and the molecular marker comprises TINAGL1 and / or MMP9. + CAFs and MMP9 + The two key cell subpopulations of macrophages have clear functional characteristics. Compared with the analysis method based on whole tissue transcriptome sequencing in the prior art, the application overcomes the problem of specific signal dilution caused by cell type mixing, and can accurately capture the molecular characteristics of the two low-abundance but high-activity subpopulations.

Description

Technical Field

[0001] This invention relates to the field of biomedical technology, specifically to a molecular biomarker for prognostic assessment of colorectal cancer and its application. Background Technology

[0002] Colorectal cancer (CRC) is one of the most common malignant tumors of the digestive tract. Epidemiological data shows that the incidence of CRC is second only to lung cancer and breast cancer, with a higher incidence in men than women, posing a serious challenge to socioeconomic and public health. The high mortality rate of patients with advanced CRC is mainly due to rapid tumor progression and distant metastasis. Metastatic colorectal cancer (mCRC) most commonly metastasizes to the liver, and patients with liver metastases have a poor prognosis, with a 5-year survival rate of only 10%. In recent years, immune checkpoint blockade (ICB) has made breakthroughs in the treatment of various tumors. However, for colorectal cancer, only about 4% of patients with high microsatellite instability (MSI-H) are sensitive to immunotherapy due to their high tumor mutational burden (TMB) and abundant immune cell infiltration; while the vast majority of metastatic colorectal cancers are microsatellite stable (MSS), exhibiting extremely low immune response rates. Therefore, in-depth analysis of the immunosuppressive microenvironment of tumor metastases is of great clinical significance for breaking down immunotherapy resistance in MSS-type colorectal cancer.

[0003] The tumor microenvironment (TME) is a complex ecosystem composed of immune cells, stromal cells, and their secreted cytokines, chemokines, and metabolites. Among these, cancer-associated fibroblasts (CAFs) are the most abundant and functionally critical stromal cell population in the TME. CAFs secrete cytokines, remodel the extracellular matrix (ECM), activate downstream signaling pathways, and induce epithelial-mesenchymal transition (EMT), promoting tumor invasion, metastasis, and drug resistance. While strategies targeting CAFs (such as eliminating specific CAF subsets or blocking their secreted factors) are theoretically feasible, the high heterogeneity of CAFs in origin and phenotype makes it difficult for single-target strategies to achieve ideal clinical efficacy. The development of single-cell transcriptome sequencing (scRNA-seq) technology has provided technical support for resolving CAF heterogeneity at single-cell resolution, making it possible to identify specific CAF core functional subgroups that truly drive malignant tumor progression.

[0004] Macrophages are among the most abundant types of immune cells. Macrophages infiltrating tumor tissue are defined as TAMs, which polarize into classically activated M1 and alternatively activated M2 types under different stimuli. M1 macrophages primarily kill tumor cells through cytotoxicity, while M2 macrophages suppress T cell-mediated immune responses and promote tumor progression. In CRC, macrophages play a crucial role in tumor growth, metastasis, immunotherapy, and angiogenesis, and have become important therapeutic targets. Current targeting strategies include macrophage clearance, blocking their recruitment, inhibiting key chemokines, and promoting M1 polarization.

[0005] It is noteworthy that current clinical research primarily focuses on targeting single cell types, with less attention paid to the interaction mechanisms between different cell populations within the tumor microenvironment (TME). Studies have shown that specific cancer cell efflux cells (CAFs) interact with macrophage subsets. CAFs can induce macrophage polarization towards a pro-tumor phenotype by secreting chemokines and immunomodulatory molecules, while immunosuppressive macrophages can activate CAFs, forming a positive feedback loop. This synergistic network may be an important mechanism leading to immunotherapy resistance and persistent tumor progression. However, the specific interaction mechanisms between CAFs and macrophages in CRC and liver metastases are not fully elucidated. Therefore, there is an urgent need to identify specific CAFs and macrophage subsets with synergistic pro-cancer effects in the colorectal cancer microenvironment and to reveal their precise signaling communication networks (such as the ligand-receptor axis), thereby providing novel multidimensional molecular markers for prognostic assessment of colorectal cancer and offering new dual therapeutic targets for breaking the immunosuppressive loop in the TME. Summary of the Invention

[0006] The main objective of this invention is to propose a molecular biomarker for prognostic assessment of colorectal cancer and its application. This invention aims to address the problem that traditional transcriptome sequencing can only obtain the average gene expression level of the whole tissue and cannot distinguish the differences between different cell types. Based on the joint analysis of specific cell subpopulations rather than the overall tissue expression level, this invention can more accurately reflect the true state of the TME, thereby improving the specificity and reliability of prognostic assessment.

[0007] To achieve the above objectives, the present invention proposes a molecular biomarker for prognostic assessment of colorectal cancer, the molecular biomarker comprising TINAGL1 and / or MMP9.

[0008] Preferably, TINAGL1 is derived from a fibroblast subpopulation in the tumor microenvironment, and MMP9 is derived from a macrophage subpopulation in the tumor microenvironment.

[0009] Matrix metalloproteinases (MMPs) are calcium- and zinc-dependent endopeptidases that primarily participate in the degradation and remodeling of the extracellular matrix (ECM). MMP9, a core member of the MMP family, enhances the invasiveness and metastatic potential of tumor cells by degrading the ECM and promoting angiogenesis. Furthermore, the adhesin / MMP9 axis activates CAFs, promoting ECM deposition and the formation of a malignant microenvironment, ultimately driving tumor progression. This chapter uses scRNA-seq data analysis to classify macrophages into several functional subsets. Among them, MMP9... + Macs were significantly enriched in primary CRC lesions and liver metastases, but expressed at low levels in normal tissues, suggesting they may be a key pro-tumor macrophage subset mediating CRC progression and metastasis. GSEA showed MMP9... +Macs were significantly upregulated in pro-metabolism / metabolic pathways such as interferon-α response, EMT, and oxidative phosphorylation, while significantly downregulated in pro-inflammatory pathways such as TNFα / NF-κB and inflammatory responses. This characteristic is consistent with the transcriptional pattern of typical M2 macrophages. Cell communication analysis revealed TINAGL1... + CAFs and MMP9 + Inter-mac cell communication was frequent and of high intensity. Clinical prognostic analysis showed that TINAGL1... + CAFs and MMP9 + When Macs were highly expressed in combination, CRC patients had the shortest survival time, suggesting that the two have a synergistic pro-cancer effect in disease progression.

[0010] This invention also proposes the application of reagents for detecting the expression levels of molecular markers TINAGL1 and / or MMP9 in samples in the preparation of products for assessing the prognosis of colorectal cancer and the risk of colorectal cancer liver metastasis.

[0011] Preferably, the assessment is to evaluate the risk of colorectal cancer liver metastasis or the prognosis of colorectal cancer patients based on the expression level or enrichment status of TINAGL1 and / or MMP9 in the biological sample of the test subject.

[0012] Preferably, when TINAGL1 and MMP9 are both highly expressed or highly enriched in the sample, it indicates the worst prognosis for colorectal cancer patients.

[0013] This invention also proposes the application of TINAGL1 inhibitors in the preparation of drugs for the prevention or treatment of colorectal cancer and colorectal cancer liver metastases.

[0014] Preferably, the TINAGL1 inhibitor inhibits tumor cell proliferation, migration, invasion, or epithelial-mesenchymal transition by reducing the expression of TINAGL1 in tumor-associated fibroblasts.

[0015] Preferably, the TINAGL1 inhibitor blocks TINAGL1. + CAFs and MMP9 + Signal communication between macrophages.

[0016] Preferably, the TINAGL1 inhibitor is selected from shRNA or siRNA that targets the TINAGL1 gene.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) Based on single-cell transcriptome sequencing technology, this invention systematically analyzed the cellular heterogeneity of the tumor microenvironment (TME) of primary colorectal cancer lesions and liver metastases at single-cell resolution, and accurately identified TINAGL1. + CAFs and MMP9+ Two key cellular subpopulations of macrophages with well-defined functional characteristics. Compared with existing analysis methods based on whole-tissue transcriptome sequencing, this invention overcomes the problem of specific signal dilution caused by cell type mixing, and can accurately capture the molecular characteristics of the above two low-abundance but highly active subpopulations. Specifically, TINAGL1 + CAFs were identified as an early initiation subpopulation of the pseudo-chronological developmental trajectory of CAFs, and their infiltration rate in tumor tissues and liver metastases was significantly higher than that in normal tissues; MMP9 + Macrophages were identified as a pro-cancer macrophage subset that specifically enriches in tumors and metastases and exhibits typical M2 transcriptional characteristics. The precise identification of this subset provides a clear cell biology basis for subsequent intervention studies targeting specific cell subsets, while significantly improving the precision and specificity of TME cell analysis.

[0018] (2) This invention comprehensively utilizes the BayesPrism deconvolution algorithm, CellChat cell communication analysis, and multiplex immunofluorescence (mIF) tissue spatial colocalization technology to systematically construct TINAGL1. + CAFs and MMP9 + This study elucidates the cell communication network between macrophages and identifies the MIF-CD74 / CD44 ligand-receptor signaling axis as the core pathway mediating the functional interactions between these two cell types. In vitro co-culture experiments further confirmed that knocking down TINAGL1 expression in CAFs significantly reduced the protein levels of MMP9, CD74, CD44, and CXCR4 in macrophages; while the addition of exogenous recombinant MIF protein (rMIF) effectively reversed these effects, verifying the functional dependence of this signaling axis. Compared to existing intervention strategies targeting single cell types or single molecules, the CAFs-macrophage positive feedback interaction loop and its key signaling node (TINAGL1 / MIF-CD74 / CD44 axis) revealed in this invention provide a novel dual intervention target for simultaneously blocking the synergistic pro-cancer effects of stromal cells and immune cells, possessing potential application value in overcoming resistance to existing immunotherapies.

[0019] (3) This invention systematically validates TINAGL1 based on multiple independent clinical cohorts (including TCGA-COAD, TCGA-READ, and GEO cohorts GSE14333, GSE17536, GSE17537, and GSE39582). + CAFs and MMP9 +Significant correlations were found between macrophage expression levels and their combined expression characteristics and survival outcomes in colorectal cancer patients. Results confirmed that patients in the TINAGL1 high-expression group had significantly shorter disease-free survival (DFS) and overall survival (OS) than those in the low-expression group; MMP9 high expression was also significantly associated with poorer DFS and OS; and the worst prognosis was observed when both were highly expressed, suggesting a synergistic prognostic value. Compared to existing prognostic assessment methods based on overall tissue gene expression levels, the prognostic assessment system established in this invention is based on the precise molecular characteristics and interaction states of specific cell subpopulations within the tumor immune microenvironment (TME), which can more realistically reflect the actual functional state of the tumor immune microenvironment. This significantly improves the specificity, accuracy, and clinical reliability of prognostic assessment, providing a new basis for individualized prognostic stratification and treatment decisions for colorectal cancer patients. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 The images show the results of single-cell atlas analysis of CRC and liver metastasis tissues in this invention; where A: high-quality single-cell screening results; B: UMAP clustering visualization; C: cell type annotation results; D: characteristic gene expression analysis; E: characteristic gene expression of each cell type; F: the top 10 characteristic genes with the most significant differential expression in each cell type; G and H: distribution ratio of each cell type in different tissues.

[0022] Figure 2 The figures show the results of the fibroblast infiltration level and CAF subpopulation analysis of this invention; where A: curve showing the relationship between fibroblast infiltration level and CRC patient survival; B: clustering diagram of CAF subpopulations; C: expression characteristics of marker genes in each CAF subpopulation; D: diagram of the top 10 differentially expressed characteristic genes in each CAF subpopulation; E: TINAGL1. + Distribution of CAFs in different tissues; F: represents the distribution of TINAGL1 in TINAGL1 + Expression level map in CAFs; G: Diagram showing the difference in infiltration among cell types in the high and low expression groups of TINAGL1; H: Diagram showing the expression of TINAGL1 in CAFs by mIF staining.

[0023] Figure 3The diagram shows the developmental trajectory and functional enrichment analysis results of CAFs subpopulations in this invention; where A: is the developmental trajectory diagram of CAFs; B and C: are the distribution of each CAFs subpopulation in the developmental trajectory; and D: is the KEGG enrichment analysis diagram of each CAFs subpopulation.

[0024] Figure 4 The diagram shows the effect of TINAGL1 knockdown on HCT116 cell function in this invention; where A: RT-qPCR verification of TINAGL1 knockdown efficiency; B: CCK-8 assay for cell proliferation; CE: Transwell assay for cell migration and invasion; F: Western blot for EMT-related protein expression.

[0025] Figure 5 This is a graph showing the expression and prognostic value analysis of TINAGL1 in CRC according to the present invention; wherein, A and B: expression graphs of TINAGL1 in pan-cancer types; C and D: expression and quantitative analysis graphs of TINAGL1 in CRC tissue detected by IHC; E: expression graph of TINAGL1 in CRC tissue in the TCGA database; FH: survival curves of patients in the high and low expression groups of TINAGL1.

[0026] Figure 6 This is a correlation analysis diagram of CAFs and myeloid cells in this invention; where A: is a correlation characteristic diagram of CAFs and various cell types; B: For TINAGL1 + Scatter plot of correlation between CAFs and myeloid cells; C: Circos plot of communication between CAFs and myeloid cells.

[0027] Figure 7 For the analysis of macrophage subsets and MMP9 in this invention + Macs functional analysis diagram; where A: myeloid cell cluster diagram; B: macrophage subset cluster diagram; C and D: UMAP diagrams of macrophage subset marker gene expression; E: heatmap of macrophage subset characteristic gene expression; F: MMP9 + Distribution of Macs in different organizations; G: MMP9 in MMP9 + Horizontal graph in Macs; H: for MMP9 + Macs GSEA analysis plot; I, J: MMP9 expression plots in tumor tissues of various tumor types; K, L: survival curves of patients with high and low MMP9 expression groups.

[0028] Figure 8 This invention is TINAGL1 + CAFs and MMP9 +Correlation and combined prognostic analysis of Macs; where A: heatmap of correlation between CAFs and various macrophage subsets; B: TINAGL1 + CAFs and MMP9 + C: Correlation scatter plot of Macs; D: Circos plot of communication between CAFs and macrophage subsets; E: TINAGL1 + CAFs and MMP9 + Survival curves of patients with co-expression of Macs.

[0029] Figure 9 This invention is TINAGL1 + CAFs and MMP9 + The diagram shows the signaling pathway analysis between Macs; where A: is the outward communication pattern analysis diagram of CAFs; B: is the MIF-related ligand-receptor interaction diagram between CAFs and macrophages; C: is the ligand-receptor interaction diagram of the collagen signaling pathway; D and E: are the communication probability analysis diagrams of the MIF signaling pathway; and F: is the ligand-receptor interaction diagram of the FN1 signaling pathway.

[0030] Figure 10 This invention is TINAGL1 + CAFs and MMP9 + Spatial colocalization analysis of Macs (mIF staining map); A: Comparison of mIF staining of Tumor and Normal tissues in patient #1; B: mIF staining map of CRLM tissue in patient #1; C: Verification map of mIF staining of Tumor tissues in patients #2, #3, and #4.

[0031] Figure 11 This diagram illustrates the effect of sh-TINAGL1 of the present invention on the expression of macrophage-related proteins.

[0032] Figure 12 This diagram illustrates the role of the MIF signal axis in the intervention experiment verification of this invention.

[0033] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0034] 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. Where specific conditions are not specified in the embodiments, conventional conditions or conditions recommended by the manufacturer shall apply. Reagents or instruments whose manufacturers are not specified are all conventional products that can be purchased commercially. Furthermore, the technical solutions of the various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed by the present invention. 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.

[0035] The technical solution of the present invention will be further described in detail below with reference to specific embodiments and accompanying drawings. It should be understood that the following embodiments are only used to explain the present invention and are not intended to limit the present invention.

[0036] Example 1 Experimental Materials I. Cell Lines Primary CAF cell lines (isolated and cultured from CRC liver metastasis tissue to the 3rd generation), HCT116 colorectal cancer cell line, and THP-1 mononuclear cell line (used to induce macrophages).

[0037] II. Tissue Samples Paraffin-embedded tissue samples were collected from patients with colorectal cancer after surgery, including primary tissue samples from 11 CRC patients, liver metastasis tissue samples from 1 patient, and normal tissue samples from 4 patients. All samples were confirmed by pathological diagnosis, and informed consent was obtained from the patients and approved by the Ethics Committee of Tangshan People's Hospital (Ethics Approval Numbers: RMYY-LLKS-2025031 and RMYY-LLKS-2026011).

[0038] Example 2 Experimental Method I. Data Acquisition and Filtering 1. scRNA-seq data scRNA-seq data of primary colorectal cancer lesions and liver metastases were obtained from the Gene Expression Omnibus (GEO) database of the National Center for Biotechnology Information (NCBI) (https: / / www.ncbi.nlm.nih.gov / geo / ). The corresponding gene expression profile serial numbers are GSE231559 and GSE221575, respectively. The data includes samples of primary colorectal cancer tissue, normal tissue, liver metastases, and normal liver metastases, totaling 31 samples from 19 patients.

[0039] 2. CRC transcriptome data Raw read counts and related clinical information, including patient gender, age, clinicopathological stage, survival status, and survival time, were obtained from the UCSC Xena website (https: / / xenabrowser.net / datapages / ). Four high-throughput sequencing datasets from the CRC were obtained from the GEO database, with corresponding gene expression profile serial numbers GSE17536 (n=177), GSE17537 (n=55), GSE39582 (n=556), and GSE14333 (n=290). All samples selected from these databases were "Homosapiens" and of "RNA" type. The sampling and patient inclusion criteria for both datasets met the requirements of this study, and specific sample information was provided.

[0040] 3. Gene expression differential analysis tools Using the online tool TIMER (https: / / cistrome.shinyapps.io / timer / ), we comprehensively analyzed the expression differences of TINAGL1 and MMP9 in different tumor and normal tissues.

[0041] The nucleotide sequence of the TINAGL1 gene is shown in SEQ ID NO:1 (corresponding to NCBI RefSeq accession number NM_022164.3); the nucleotide sequence of the MMP9 gene is shown in SEQ ID NO:2 (corresponding to NCBI RefSeq accession number NM_004994.3).

[0042] 4. Cell marker information Cell population markers were defined by combining annotation information from the CellMarker database (http: / / xteam.xbio.top / CellMarker / index.jsp) and classic specific markers from published CRC-related studies. T cells were defined using CD3D and CD3E, myeloid cells using LYZ, epithelial cells using EPCAM, fibroblasts using COL3A1 and CXCL12, endothelial cells using ENG, B cells using CD79A, plasma cells using MZB1, tumor cells using TYMS, and NK cells using NKG7.

[0043] II. scRNA-seq data analysis and CAFs subset identification 1. CAF subpopulation screening and identification Dimensionality reduction and clustering of fibroblasts were performed, and batch correction was performed using the Harmony algorithm. The FindAllMarkers function was then used to screen for specific high-expression genes in each subpopulation. The screening threshold was set as log-fold change (logFC) > 0.25 and corrected P < 0.05. Based on the CellMarker database and published colorectal cancer-related studies, four functional subpopulations of CAFs were obtained.

[0044] 2. Functional enrichment analysis of CAF subpopulations KEGG enrichment analysis was performed on four CAF subgroups to identify the signaling pathways enriched in each subgroup and to explore their functional differences.

[0045] 3. Developmental trajectory analysis of CAF subpopulations We conducted pseudo-temporal analysis on CAFs to explore the developmental relationships and dynamic evolution of CAF subtypes.

[0046] 4. Cell localization verification The spatial colocalization relationship between CAFs and macrophage subsets was observed in a tissue microarray (TMA) using multiplex immunofluorescence (mIF) technology.

[0047] III. Survival Analysis To explore the potential value of target genes in the clinical diagnosis and prognostic assessment of colorectal cancer, a survival prognostic analysis was performed on transcriptome data. The specific steps are as follows: 1. Research cohorts included: including the cancer genome atlas-colon adenocarcinoma (TCGA-COAD) and cancer genome atlas-rectum adenocarcinoma (TCGA-READ) cohorts from the TCGA CRC cohort, as well as the GEO cohorts (GSE14333, GSE17536, GSE17537, and GSE39582).

[0048] 2. Patient grouping: Based on the expression level of the target gene in each cohort, patients were divided into high expression group and low expression group with the cut-off value.

[0049] 3. Survival analysis: The Surv() function of the Survival R package is used to construct survival objects, and the Survdiff() function is used to perform the Log-Rank test to assess the survival difference between patients in the high and low expression groups; the ggsurvplot() function of the Survminer R package is used to plot Kaplan-Meier survival curves to visually display the differences between groups in overall survival (OS) and disease-free survival (DFS).

[0050] 4. Statistical criteria: P<0.05 is considered statistically significant. Genes that are significantly associated with patient prognosis are screened to clarify their value as potential prognostic biomarkers for CRC. IV. Tissue Sample Preparation and Staining

[0051] 1. Sample collection: Paraffin tissue samples were collected from colorectal cancer patients after surgery, including primary tumor tissue, normal tissue and liver metastasis tissue.

[0052] 2. Fabrication of tissue microarrays: (1) Perform pathological diagnosis on the HE-stained tissue sections and circle the required tissue locations; (2) Prepare suitable blank recipient wax blocks using a tissue embedding machine, and then extract the pathological wax block cores from the selected sites using a tissue array instrument according to the array design, and arrange them regularly on the blank recipient wax blocks. (3) Heat the wax block in a constant temperature oven at 52°C to fuse it, so that the tissue core is tightly connected to the recipient wax block; (4) Use a fully automatic tissue slicer to trim the paraffin block at a feed rate of 20 micrometers / revolution until 80% of the tissue core is fully exposed; (5) Use a fully automatic tissue slicer to slice the tissue array block at a feed speed of 4 micrometers / revolution, and mount the slices on imported glass slides that have been treated to prevent detachment; (6) Place the array slices in a 60°C oven and bake for 16 hours to complete the tissue chip preparation. After numbering, store in a 4°C refrigerator for later use.

[0053] 3. Staining: (1) HE staining: Take out a tissue chip for HE staining to determine whether the tissue chip is qualified and whether the sampling site is accurate.

[0054] (2) Multiplex immunofluorescence histochemical staining (TINAGL1): ① Baking: Place the tissue chip in a 63℃ oven and bake it with wax for one hour; ② Dewaxing: Dewaxing is completed in a fully automatic dyeing machine. The dewaxing time is as follows: 100% alcohol, 2 tanks 7 minutes per cylinder 1 tank of 90% alcohol 5min 1 tank of 80% alcohol 5min 1 tank of 70% alcohol 5min ③ Antigen retrieval: Dilute 10x retrieval solution to 1x working solution, microwave on high for 3 minutes to boil, then place a glass slide in the microwave, microwave on low for 18 minutes (ensure the tissue is immersed in the liquid throughout the process), and allow to cool naturally at room temperature; ④ Sealing: Remove the slide, wash it with TBST, place it in a humidified chamber, add blocking buffer, and incubate for 10 minutes; ⑤ Primary antibody incubation: Remove blocking buffer, place in a humidified chamber, add primary antibody, incubate at room temperature for 1 hour, and wash slides with TBST. ⑥ Secondary antibody incubation: Remove the slide, place it in a humidified chamber, add the secondary antibody, incubate at room temperature for 10 min, wash the slide with TBST, Opal dye development: Remove the slide, place it in a humidified chamber, add Opal dye dilution solution (1:100 dilution), incubate at room temperature for 10 min, wash the slide with TBST; ⑦ Microwave treatment: Repeat step 3) and then clean the slides with TBST (to remove primary and secondary antibodies); ⑧ Counterstaining: Repeat steps 4)-8) until all indicators are marked; ⑨ DAPI staining and mounting: Take out the slide, place it in a humidified chamber, add DAPI working solution, incubate at room temperature for 10 min, wash the slide with TBST, take out the slide, add fluorescent anti-quenching mounting medium and mount.

[0055] 4. Spectral scanning and cell counting The spectral characteristics of each fluorophore were determined by single antigen staining and captured using a multispectral fluorescence microscope, which recorded images every 10 nm using full emission spectroscopy. Positive cells were counted using ImageJ analysis software after setting a threshold. Three fields of view were randomly selected for each sample, and the average value was taken.

[0056] 5. Immunohistochemistry (IHC) (1) Baking: Place the paraffin slices on a baking machine at 65℃ and heat for 1 hour to melt the paraffin on the surface of the slices.

[0057] (2) Dewaxing and hydration: The baked slices were soaked in xylene and anhydrous ethanol, 95% ethanol, 85% ethanol and 75% ethanol in turn for 5 min each.

[0058] (3) Add 1×PBS buffer, place the slide box on a shaker, centrifuge at 15 rpm for 5 min, twice.

[0059] (4) Antigen retrieval: Incubate with 3% hydrogen peroxide solution for 10 min to block endogenous peroxidase activity in the sample. Wash with 1×PBS, place the slide box on a shaker, centrifuge at 15 rpm for 5 min twice. Immerse the slides in antigen retrieval EDTA solution and heat to 100℃ and boil for 5 min, then cool naturally to room temperature.

[0060] (5) Blocking: Block the sections in 3% goat serum for 30 min.

[0061] (6) Primary antibody incubation: Add primary antibody (TINAGL1, 1:100 dilution) and incubate overnight at 4°C.

[0062] (7) Secondary antibody incubation: The next day, the sections were washed three times with distilled water, and horseradish peroxidase (HRP) labeled secondary antibody (1:100 dilution) was added and incubated at 37°C for 30 min. The reaction was terminated after staining with diaminobenzidine (DAB) as a chromogenic agent for 20 s. (8) Counterstaining: Hematoxylin sections were counterstained for 20 s, rinsed with water and blued back for 10 min. The sections were cleared sequentially with graded alcohol and xylene, and then mounted. The results were confirmed by two experienced pathologists using an optical microscope. V. Cell resuscitation, passage, and cryopreservation

[0063] 1. Cell resuscitation Following the principle of slow freezing and rapid thawing, the cryovials containing CAFs were removed from the liquid nitrogen container, wrapped in sterile gloves, and quickly placed in a 37°C constant temperature water bath. The cryovials were gently agitated to allow the cryopreservation solution to thaw completely within 1 minute. The cells were then centrifuged at 37°C and 1000 rpm for 4 minutes. The cryovials were transferred to a laminar flow hood, the supernatant was discarded, and 1 mL of DMEM / F12 medium (containing 20% ​​FBS + 1% PS) was added to resuspend the cells. The resuspended cell suspension was then transferred to a T25 cell culture flask, 4 mL of cell culture medium was added, and the flask was gently agitated to ensure even cell distribution. The flask was then incubated at 37°C in a 5% CO2 cell culture incubator for at least 12 hours. After observing cell adhesion and growth under a microscope, the culture medium was replaced. When the cell density reached 80%–90%, subsequent passage or transfection procedures were performed.

[0064] 2. Cell Passage: Culture cells in a 37℃, 5% CO2 cell culture incubator. Daily observation of cell status and density is necessary, with timely replacement of the culture medium. When the cell density reaches 80%–90% under a microscope, pass the cells. In a clean bench, discard the culture medium, wash twice with 1× phosphate-buffered saline (PBS), add 1 mL of 0.05% trypsin for digestion, and incubate at 37℃ for 1–2 min. When the cells gradually become rounder and the gaps between them increase, but they are not completely detached from the flask wall, add 2 mL of culture medium to stop digestion and gently pipette the cells to completely free them. Pass the cell suspension at a 1:3 ratio.

[0065] 3. Cell cryopreservation Cells in the logarithmic growth phase were collected, and in a clean bench, the culture medium was discarded. The cells were washed twice with 1×PBS, digested according to the cell passage method described above, and a cell suspension was prepared. The cells were centrifuged at 1000 rpm for 4 min, and the supernatant was discarded. The cells were resuspended in cryopreservation solution, aspirated, and placed in cryovials. The cryovials were sealed with sealing film, and labeled with the cell name and cryopreservation date. The cryopreservation sequence was 4℃ for 30 min, -20℃ for 30 min, and -80℃ for 30 min. Finally, the cells were transferred to liquid nitrogen for long-term storage.

[0066] VI. Protein and Gene Expression Related Detection Experiments 1. Western blot assay (1) Preparation and concentration detection of cell protein samples: Discard the culture medium in the well plate, wash twice with 1×PBS, add RIPA lysis buffer, shake at 4℃ for 30 min, transfer the cells to centrifuge tubes after repeated pipetting, centrifuge at 12000rpm for 15 min at 4℃, and collect the supernatant to obtain protein samples. Take a portion of the protein samples for BCA protein concentration detection. Perform protein quantification according to the BCA kit instructions: Prepare BCA working solution by taking solution A and solution B at a ratio of 50:1, add 200 μL of working solution to a 96-well plate, add 2 μL of protein sample to each well and add 18 μL of 1×PBS, set three replicates for each group, incubate at 37℃ for 30 min, cool to room temperature, and measure the absorbance standard curve and absorbance value of the sample for each well at a wavelength of 562 nm.

[0067] (2) Protein denaturation: Adjust all sample concentrations to be consistent using RIPA lysis buffer, add 1 / 4 volume of 5× protein loading buffer, incubate at 100℃ for 5 min to denature the protein, centrifuge at 12000 rpm for 5 min, and take the supernatant for electrophoresis analysis.

[0068] (3) Preparation of 10% polyacrylamide gel: Take 4.835 mL of double distilled water (DDW), 2.475 mL of 40% acrylamide, 100 μL of 10% sodium dodecyl sulfate (SDS), 4.0 μL of N,N,N',N'-tetramethylethylenediamine (TEMED), 2.5 mL of Tris(Hydroxymethyl)aminomethane hydrochloride (MTris-Cl) (pH 8.8), and 100 μL of 10% ammonium persulfate to prepare 10% polyacrylamide gel. After mixing, quickly pour the gel to 2 / 3 of the total height of the glass plate. Add 1 mL of water-saturated n-butanol on top of the gel to ensure the upper layer of the gel is flat, and let it stand until the gel solidifies.

[0069] (4) Protein loading and electrophoresis: Load an appropriate amount of protein sample, electrophore at 90 V for 30 min to concentrate the protein, then adjust to 120 V until the target protein is effectively separated and then stop electrophoresis.

[0070] (5) Transfer: After electrophoresis, remove the gel and place it in a sandwich clamp for transfer. Place the gel at the negative electrode and the PVDF membrane at the positive electrode. Immerse the membrane in 1× transfer solution and transfer it in a 350 mA constant flow water bath for 1.5 h to transfer the protein in the gel to the PVDF membrane to form an imprint.

[0071] (6) Blocking: After the transfer is completed, place the membrane in 5% skim milk powder and shake at room temperature for 2 h. Then place the membrane in 1×TBST solution and shake and rinse for 5 min, for a total of 4 times.

[0072] (7) Primary antibody incubation: Cut the membrane according to the protein imprinting sites, add diluted primary antibodies at the following ratios: epithelial cadherin (E-cadherin) (1:1000), neural cadherin (N-cadherin) (1:1000), and vimentin (1:1000). Incubate overnight at 4°C with shaking. Place the membrane in 1×TBST solution and wash with shaking for 5 min, repeating 4 times.

[0073] (8) Secondary antibody incubation: Place the membrane in the corresponding secondary antibody (HRP-labeled goat anti-rabbit antibody), incubate at room temperature for 1.5 h, place in 1×TBST solution, rinse for 5 min, repeat 4 times.

[0074] (9) Development and results display: Place the membrane in the prepared ultrasensitive ECL chemiluminescence reagent for 30 s, immediately place the membrane in the exposure box, expose the photosensitive film in the dark room for 1 min, then perform development and take pictures, and use ImageJ software to analyze the protein bands and calculate the gray value.

[0075] 2. Cell counting kit-8 (CCK-8) experiment HCT116 cells in logarithmic growth phase were seeded into 96-well plates (2000 cells / well), and the supernatant of TINAGL1-knocked CAFs was added. After 48 h of culture, 10 μL of CCK-8 solution was added to each well, and the cells were cultured for another 2 h. The absorbance of each well was detected by enzyme-linked immunosorbent assay (ELISA) at a wavelength of 450 nm. The experiment was independently repeated three times.

[0076] 3. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) (1) RNA extraction from cells: The procedure was performed on an ice plate. After transfection, the culture medium was completely discarded, the cells were washed with 1×PBS, the PBS was aspirated, and 1 mL of Trizol (pre-cooled at 4℃) was added to the cell culture flask. The cells were shaken on a shaker for 15 min to ensure full contact between the cell surface and the Trizol. The cells were thoroughly resuspended and the lysis buffer was transferred to a new 1.5 mL EP tube. 200 μL of pre-cooled chloroform was added to each 1 mL of Trizol, and the mixture was inverted to mix thoroughly. The tube was allowed to stand for 5 min. The tube was centrifuged at 12000 rpm at 4℃ for 15 min, and the supernatant was carefully aspirated into a new EP tube. 1 mL of pre-cooled anhydrous ethanol (1:5) was added, the mixture was mixed, and the tube was incubated at -80℃ for at least 30 min. The tube was centrifuged at 12000 rpm at 4℃ for 30 min to precipitate the RNA. The supernatant was discarded, 1 mL of 75% ethanol (pre-cooled) was added, and the tube was centrifuged at 12000 rpm at 4℃ for 10 min. Discard the supernatant and let stand for 5 min (to allow the ethanol to dry). Add 20 µl of diethyl pyrocarbonate (DEPC) water to each tube to dissolve the RNA. Store the RNA solution at -80°C.

[0077] (2) Reverse transcription to obtain cDNA: Add 3 μL Total RNA, 1 μL primer (100 pmol / μL), and DEPC water to a RNase-free centrifuge tube to a final volume of 13.5 μL. Mix well, incubate briefly for 5 s, and heat at 42℃ for 2 min. Add 4 μL 5× buffer, 1 μL dNTPs (10 mM), 0.5 μL RNasin (40 U / μL), and 1 μL M-MLV (200 U / μL) to the reaction tube. Mix well, incubate briefly for 5 s, and extend at 42℃ for 1 h. Heat at 70℃ for 5 s. Obtain cDNA and store at -20℃.

[0078] (3) RT-qPCR detection: A qPCR reaction system was prepared according to the following formula: 10 μL 2×SYBR Premix Ex Taq, 1 μL Fwd Primer (5 pmol / μL), 1 μL Reverse Primer (5 pmol / μL), 1 μL cDNA, and DDW to a final volume of 20 μL. Each sample was tested in triplicate. After mixing and momentary infiltration, the samples were detected on a PCR instrument. The cycling temperatures were: 94℃ for 4 min, 94℃ for 30 s, 58℃ for 30 s, and 72℃ for 30 s for 40 cycles. The fold change in relative gene expression was calculated using the 2-ΔΔct method, as shown in the following formula: Folds—Folds representing the relative change in gene expression levels Ct—Cycle threshold △Ct = Ct (target gene) - Ct (internal reference gene) ΔΔCt=(Ct1-Ct2)-(Ct3-Ct4) Ct1: Critical cycle number of sh-TINAGL1 Ct2: Critical cycle number of sh-TINAGL1 β-actin Ct3: Critical cycle number of sh-NC Ct4: Critical cycle number of sh-NC β-actin VII. Transwell Migration and Invasion Experiments Dilute the matrix gel with serum-free RPMI-1640 medium at a ratio of 5:1, and add 50 μL to the upper chamber with an 8 μm pore size. Incubate at 37°C for 2 h. Add lentivirally infected CAFs to DMEM medium containing 10% FBS and place in the lower chamber. Then process the cells in the upper chamber. Discard the culture medium completely after digestion of HCT116 cells, wash once with 1×PBS, resuspend in serum-free DMEM medium, and then add to the upper chamber with or without matrix gel (invasion assay). Note that 10 μL of fibronectin (0.5 mg / mL) should be coated on the bottom of each Transwell chamber and air-dried in a laminar flow hood. Carefully place the upper chamber seeded with HCT116 cells (using the matrix gel upper chamber for invasion assay and the unmajor upper chamber for migration assay) into the lower chamber containing CAFs and co-incubate at 37°C and 5% CO2 for 48 h. After co-culture, remove the chambers, wipe the cells out with a cotton swab, wash with 1×PBS, fix the cells with a methanol:acetic acid (3:1) mixture for 30 min, stain with crystal violet for 15 min, wash thoroughly, and place the chambers in a 24-well plate. Observe the cells under an inverted microscope, and randomly select 3 fields of view to take pictures and calculate the cell count.

[0079] 8. Co-culture of CAFs with tumor cells (HCT116) 1. Migration and invasion ability detection: HCT116 cells were seeded in the upper chamber with an 8 μm pore size, and CAFs from the sh-TINAGL1 group and sh-NC group were seeded in the lower chamber, respectively. The upper chamber was filled with serum-free DMEM medium containing 1% PS, and the lower chamber was filled with DMEM medium containing 1% PS and 20% FBS. After co-culturing for 24 h, the migration and invasion ability of HCT116 cells were detected.

[0080] 2. Protein expression detection: HCT116 cells were seeded in the upper chamber with a 0.4 μm pore size, and CAFs from the sh-TINAGL1 group and sh-NC group were seeded in the lower chamber, respectively. The upper chamber was cultured in serum-free DMEM medium containing 1% PS, and the lower chamber was cultured in DMEM medium containing 1% PS and 20% FBS. The cells were cultured together for 24 h, and then the protein levels of E-cadherin, N-cadherin and Vimentin in HCT116 cells were detected.

[0081] IX. Co-culture experiment of CAFs and macrophages Protein expression detection: Macrophages were seeded in the upper chamber of a 0.4 μm pore size Transwell, while CAFs from the sh-TINAGL1 and sh-NC groups were seeded in the lower chamber, respectively. Both chambers were co-cultured in complete culture medium for 24 h. Subsequently, the protein levels of MMP9, CD74, CXCR4, and CD44 in macrophages were detected using the same methods as Western blot experiments.

[0082] Intervention experiment: In the sh-TINAGL1 co-culture system, the sh-TINAGL1+MIF inhibitor group and the sh-TINAGL1+ISO-1+recombinant MIF protein group were set up. After co-culturing for 24 h, the protein levels were measured again to verify the role of the MIF signaling axis.

[0083] The protein levels of MMP9, CD74, CXCR4, and CD44 in macrophages after co-culture were detected by adding ISO-1 and recombinant macrophage migration inhibitory factor (rMIF).

[0084] 10. Data Analysis 1. Correlation analysis Reference expression profiles were generated from scRNA-seq data in the GEO database, and the BayesPrism algorithm was used to deconvolve the corresponding CRC batch RNA sequencing data. The top 2000 highly variable genes in each cell type were selected as feature labels. The proportion of each cell type was obtained based on deconvolution, and the correlation coefficient between each cell type was calculated using Spearman analysis. Gene expression profile data of CRC tissues were obtained from the TCGA database. The infiltration level of each cell in the dataset was calculated based on the marker genes defined for each cell type in the single-cell sequencing dataset, and the correlation coefficient between the two cell types was calculated using Spearman analysis. The correlation coefficient (r) ranges from -1 to 1, with the absolute value closer to -1 or 1 indicating a stronger correlation. r>0 indicates a positive correlation, and r<0 indicates a negative correlation. P<0.05 is considered statistically significant.

[0085] 2. Intercellular communication analysis We used the CellChat R package to infer cell-cell interactions and construct ligand-receptor regulatory networks. To ensure the reliability of the analysis, we filtered out ligand-receptor pairs with low expression levels in specific cell populations (ligand-receptor pairs expressed in fewer than 10 cells). The netVisual function was used to visualize the strength and patterns of cell-cell interactions. To assess complex multi-ligand-receptor pathways, the netAnalysis_contribution function was used to calculate and display the relative contributions of ligand-receptor pairs to the overall signaling pathway. Non-negative matrix factorization (NFF) combined with the identifyCommunicationPatterns function was used to identify core cell communication patterns and determine the dominant signaling pathways in specific cell populations. Furthermore, network centrality scores were calculated using netAnalysis_centrality and analyzed using netAnalysis_signalingRole_negative.

[0086] 3. Statistical Analysis Data analysis was performed using SPSS 22.0 software, and graph plotting was performed using GraphPad Prism 8.0 software. All cell experiments were repeated three times. Quantitative data were expressed as mean ± standard deviation. Independent samples t-tests were used to compare data between two groups. All data followed a normal distribution. One-way ANOVA was used to analyze expression differences among multiple groups. Survival curves were plotted using the Kaplan-Meier method, and the Log-rank test was used to analyze differences in survival rates among groups. P < 0.05 was considered statistically significant. All tests were two-tailed.

[0087] Example 3 Experimental Results 1. Single-cell atlas analysis of tumor tissues from patients with colorectal cancer and liver metastases To investigate the cellular composition characteristics of the microenvironment in CRC and CRLM, this study integrated two publicly available single-cell transcriptome datasets (GSE221575 and GSE231559) from the GEO public database. After integrating the two datasets and undergoing rigorous quality control, low-quality cells with fewer than 300 or more than 6000 gene counts and mitochondrial gene expression exceeding 10% were removed. The Harmony algorithm was used to eliminate batch effects, perform dimensionality reduction, and clustering, ultimately selecting 60,515 high-quality single cells from the tissues for subsequent analysis. Figure 1 A). Using the UMAP algorithm to visualize cell clusters, a total of 17 cell clusters with significant heterogeneity were identified. Figure 1B). Based on the Seurat software, combined with the CellMarker database and classic specific markers reported in published CRC-related literature, the above 17 cell clusters were annotated and finally classified into 9 major cell types ( Figure 1 C). Characteristic gene expression analysis showed that all nine cell types exhibited their own clear and relatively specific classical marker expression patterns (C). Figure 1 D). Among them, T cells highly express CD3D, B cells highly express CD79A, myeloid cells highly express LYZ, endothelial cells highly express PECAM1, epithelial cells highly express KRT19, plasma cells highly express MZB1, fibroblasts highly express COL3A1, NK cells highly express NKG7, while tumor cells specifically express TYMS ( Figure 1 E). Figure 1 F shows the top 10 genes with the most significant differential expression across various cell types. Further analysis revealed significant compositional differences and invasive heterogeneity among these nine cell types in primary CRC tumor tissue, normal tissue, and liver metastases. The distribution range and proportion of different cell clusters varied significantly across tissues. Notably, the proportion of fibroblasts was significantly higher in the tumor and liver metastasis groups than in the normal group (…). Figure 1 (G, H) indicates that fibroblasts may be associated with the development and progression of CRC and liver metastasis. The dynamic changes in this cellular composition reflect the differences in TME at different stages of CRC progression and CRLM.

[0088] In summary, this study constructed single-cell transcriptome atlases of CRC and CRLM, systematically revealing the composition and distribution characteristics of different cell types in the TME under different tissue conditions, providing important data support for in-depth analysis of the role of TME in the occurrence, development and liver metastasis of CRC.

[0089] 2. Fibroblast infiltration level and subtype heterogeneity To further evaluate the clinical significance of fibroblasts in CRC progression, this study performed a survival analysis based on the CRC cohort from the TCGA dataset. Results showed that compared to the low-infiltration group, patients in the high-infiltration group had significantly shorter overall survival (…). Figure 2A) This suggests that the enrichment of fibroblasts is closely related to poor patient prognosis. The extracellular matrix (TME) is considered a core driver regulating the progression and metastasis of CRC, with its complex cellular composition and signaling network continuously reshaping the functional state of stromal cells. Fibroblasts, as the most abundant stromal cells in the extracellular matrix, primarily participate in extracellular matrix synthesis and tissue repair under physiological conditions; however, in the TME of CRC, normal fibroblasts are continuously activated by tumor cells, inflammatory factors, and various paracrine signals, transforming into highly heterogeneous and plastic CAFs. Although previous studies have confirmed the pro-cancer potential of CAFs, due to the complexity of the TME, the functional differences among different CAF subtypes in tumor progression, metastasis, and prognosis still need to be systematically elucidated, especially the key functional subgroups closely related to poor prognosis, which may constitute potential prognostic biomarkers and therapeutic targets. Based on single-cell transcriptome gene expression profile clustering analysis, this study classified CAFs into four subtypes with distinct molecular characteristics: TINAGL1... + CAFs, WNT5A + CAFs, CLEC3B + CAFs and CXCL8 + CAFs ( Figure 2 B). Each CAF subtype exhibited significantly differentiated marker gene expression characteristics ( Figure 2 C), further identifying the top 10 differentially expressed characteristic genes in each subgroup ( Figure 2 D). Among them, TINAGL1 + The distribution characteristics of CAFs are particularly prominent: this subtype has a significantly higher proportion in tumor tissues and liver metastases than in normal tissues. Figure 2 E), suggesting that it may be a subtype of pro-tumor CAFs closely related to CRC progression and liver metastasis, and has the potential to serve as a biomarker for poor CRC prognosis.

[0090] Further analysis revealed that TINAGL1 in TINAGL1 + Significantly high expression was observed within the CAFs subset. Figure 2 F), verifying the specificity and reliability of this subtype classification at the molecular level. To investigate TINAGL1... + The association between cancer cells (CAFs) and the tumor immune microenvironment was investigated. Using the median TINAGL1 expression level as the cutoff, samples were divided into a TINAGL1-high expression group (TINAGL1-high) and a TINAGL1-low expression group (TINAGL1-low), and the differences in infiltration of different cell types between the two groups were compared. Results showed that compared to the TINAGL1-low group, the proportion of T cells was significantly lower in the TINAGL1-high group, while the proportions of myeloid cells and NK cells were relatively higher. Figure 2 G), prompt TINAGL1+ CAFs may be involved in the formation of an immunosuppressive microenvironment.

[0091] To further validate the expression characteristics of TINAGL1 in CAFs at the tissue level, this study collected tissue samples from 11 CRC patients and 1 liver metastasis patient, prepared TMA, and detected it using mIF. Alphasmooth muscle actin (α-SMA) was used as a marker for CAFs. The results showed that TINAGL1 was expressed in α-SMA+ CAFs in CRC tissue, liver metastasis tissue, and normal tissue. Figure 2 H). Compared with normal tissues, the expression level of TINAGL1 was significantly increased in primary CRC tissues and liver metastases, further confirming the presence of TINAGL1. + Enrichment characteristics of CAFs in CRC tissues. In summary, high expression of fibroblasts is closely associated with poor prognosis in CRC patients, and among the heterogeneous subsets of CAFs, TINAGL1... + CAFs were significantly enriched in tumor tissues and liver metastases, and may promote CRC progression by shaping an immunosuppressive microenvironment. These results provide a basis for further investigation into TINAGL1. + This provides important evidence on the functional mechanism of CAFs and their role in CRC liver metastasis.

[0092] 3. TINAGL1 + CAFs are an early functional subgroup driving the progress of CRC. During cell differentiation and development, different gene expression programs determine different cell fates and functional states. To explore the developmental relationships and dynamic evolution of CAF subtypes, this study further conducted a pseudo-time series analysis of CAFs. The results showed that the developmental trajectory of CAFs exhibits three functional state branches (S1, S2, and S3). Figure 3 A) suggests that it exhibits a continuous rather than static differentiation process in the TME. It is worth noting that TINAGL1 + CAFs and CXCL8 + CAFs are mainly enriched in the initial stage of pseudo-chronological trajectory development, representing an early stage of CAF differentiation and evolution. As the trajectory progresses, early subpopulations gradually evolve into CAFs with different functional states, forming terminal subtypes with clear functional differences. Figure 3 B, C). The above results suggest that TINAGL1 + CAFs and CXCL8 +CAFs may represent an early CAF state activated under TME stimulation, serving as a common origin for the subsequent differentiation of various functional CAF subpopulations. This finding not only reveals the developmental pathway of CAF heterogeneity from a developmental dynamics perspective, but also relates to TINAGL1. + CAFs are positioned as early key nodes in the TME-driven pro-cancer cycle. Combined with TINAGL1 + The high enrichment of CAFs in primary tumor lesions and liver metastases, along with their close association with the immunosuppressive microenvironment (decreased T cells and NK cells, increased myeloid cells), leads us to speculate that TINAGL1, as the initiating subset of CAFs, plays a crucial role. + CAFs may promote CRC progression and metastasis by shaping an immunosuppressive microenvironment.

[0093] To further explore the functional differences among the CAF subgroups, this study conducted KEGG enrichment analysis on the above four CAF subgroups. Figure 3 D). The results showed that TINAGL1 + CAFs are significantly enriched in signaling pathways such as MAPK, Notch, and Ras. These pathways regulate cell proliferation and differentiation and participate in tumor immune escape, suggesting that TINAGL1 is involved. + CAFs may enhance malignant phenotypes by activating pro-cancer signaling pathways; in contrast, CXCL8 + CAF is mainly enriched in inflammation-related signaling pathways such as NF-κB, IL-17 and TNF, suggesting that it may remodel the TME by mediating inflammatory responses and inflammation-related signal transduction.

[0094] 4. CAF knockdown of TINAGL1 inhibits HCT116 cell proliferation, migration, invasion, and epithelial-mesenchymal transition. Based on the aforementioned bioinformatics analysis, TINAGL1 was suggested. + CAFs may promote CRC progression by shaping an immunosuppressive microenvironment. To further clarify the pro-cancer function of TINAGL1 in CAFs, this study conducted in vitro cell experiments. Using a third-generation primary CAF cell line isolated and cultured from CRLM tissue previously established in our laboratory, we established a stable CAF cell line with low TINAGL1 expression by knocking down TINAGL1 expression via lentivirus-mediated shRNA, and evaluated its effects on HCT116 cell proliferation, migration, invasion, and EMT. CAF cells were infected with a lentivirus carrying TINAGL1-targeting shRNA, and after selection with puromycin to obtain stable knockdown cell lines, the knockdown efficiency was verified by RT-qPCR. The results showed that compared with the sh-NC group, the TINAGL1 mRNA expression level in the sh-TINAGL1-1 group was significantly decreased (…). Figure 4A) indicates the successful construction of a stable knockdown cell line. sh-TINAGL1-1, with the most significant knockdown effect, was selected for subsequent functional experiments. CCK-8 assay results showed that, compared to the sh-NC group, HCT116 cells treated with CAFs-conditioned medium in the sh-TINAGL1 group exhibited significantly reduced proliferation after 48 h of culture. This indicates that knockdown of TINAGL1 in CAFs can significantly inhibit the proliferation of HCT116 cells. Figure 4 B). Transwell assays were further used to examine the effect of TINAGL1 knockdown in CAFs on the migration and invasion abilities of HCT116 cells. The results showed that compared with the sh-NC group co-cultured with CAFs, the number of HCT116 cells co-cultured with CAFs in the sh-TINAGL1 group was significantly reduced, with both the number of migrating cells (169±14.84) ​​and the number of invading cells (99±12.01) being significantly lower than those in the sh-NC group (migration: 272±23.59; invasion: 174±6.03). Figure 4 The results showed that knocking down TINAGL1 in CAFs significantly inhibited the migration and invasion of HCT116 cells. To further investigate whether the above phenotypic changes were related to the EMT process, this study used Western blot analysis to analyze the expression changes of EMT-related markers. The results showed that compared with co-cultured CAFs in the sh-NC group, the epithelial marker E-cadherin was significantly upregulated at the protein level in HCT116 cells co-cultured with CAFs in the sh-TINAGL1 group, while the expression of N-cadherin and Vimentin was significantly decreased at the protein level. Figure 4 F), suggesting that CAFs knocking down TINAGL1 expression can effectively inhibit the EMT process in CRC cells.

[0095] In summary, knocking down TINAGL1 expression in CAFs significantly inhibits the proliferation, migration, and invasion of HCT116, and suppresses its epithelial-mesenchymal transition process by regulating the expression of EMT-related molecules, suggesting that TINAGL1... + CAFs play an important role in promoting cancer progression in CRC.

[0096] 5. High expression of TINAGL1 is associated with poor prognosis in CRC patients. To systematically evaluate the clinical significance of TINAGL1 in CRC, the online tool TIMER was used to analyze the expression specificity of TINAGL1 across various cancer types. The results showed that TINAGL1 was upregulated in multiple malignant tumor tissues, with particularly significant increases in COAD and READ. Figure 5A, B). Subsequently, we collected primary CRC tissue samples from 11 CRC patients and 1 CRLM tissue sample from our hospital, and selected 4 normal tissue samples as controls. IHC analysis was performed to determine the overall expression level of TINAGL1 protein in CRC tissues. ImageJ software was used to convert the IHC results into semi-quantitative data. The analysis results showed that, compared with normal tissues, the expression level of TINAGL1 protein was significantly increased in tumor tissues (A, B). Figure 5 C, D). Since only one CRLM tissue sample was analyzed, it was not included in the statistical analysis. However, IHC staining results showed that TINAGL1 protein expression was increased in CRLM compared to normal tissue and primary CRC tissue. Further validation of the IHC results using the TCGA (TCGA-COAD and TCGA-READ) datasets confirmed that TINAGL1 protein expression was significantly higher in CRC tissue than in normal tissue. No relevant data on CRLM tissue are currently available in the TCGA database. Figure 5 E). Based on this, we further explored the relationship between TINAGL1 expression levels and the prognosis of CRC patients. In the independent public cohorts GSE14333 and GSE17536, the optimal cutoff value for TINAGL1 expression was determined using the "survminer" R package, and the Log-Rank test was used to compare DFS and OS between the high and low expression groups. Figure 5 As shown in G, in the GSE14333 cohort, patients with high TINAGL1 expression had significantly shorter DFS than those with low expression (median DFS: approximately 120 months vs. 150 months, P < 0.01). In the GSE17536 cohort ( Figure 5 F), high TINAGL1 expression was significantly associated with shorter OS, with a median OS of approximately 140 months in the low expression group and approximately 120 months in the high expression group (P<0.05). This finding was further validated in the TCGA-COAD and TCGA-READ datasets in the TCGA database, where high TINAGL1 expression was also significantly associated with shorter OS in CRC patients. Figure 5 H). The results of multi-cohort and multi-level analyses consistently showed that TINAGL1 was significantly highly expressed in CRC tissues, and its high expression level was closely related to poor patient prognosis, suggesting that TINAGL1 may serve as an important molecular marker for CRC prognostic assessment.

[0097] 6. CAFs in CRC are correlated with myeloid cells. Complex and important cell communication networks exist among different cell subpopulations in the tumor microenvironment (TME), which are crucial for tumorigenesis and development. Numerous studies have shown that myeloid cells play a key role in CRC progression, and that functionally similar cell populations may exist within the TME, collectively participating in tumorigenesis and development. To further explore TINAGL1...+ The potential mechanisms by which CAFs regulate the immune microenvironment of CRC were investigated. BayesPrism algorithm was used to perform deconvolution analysis on bulk transcriptome data to estimate the proportion of each cell subpopulation in bulk RNA-seq data and to analyze their specific expression profiles. Results showed that myeloid cells were correlated with CAF subpopulations, particularly with TINAGL1. + CAFs showed the most significant correlation ( Figure 6 A). The scatter plot results showed that the proportion of TINAGL1⁺CAF cells was positively correlated with the proportion of myeloid cells. Figure 6 B). An intercellular communication network was constructed; Circos diagrams showed the strength and number of interactions between CAFs and myeloid cells, including TINAGL1. + CAFs communicate most frequently and with high intensity with myeloid cells, suggesting that they may play a key role in the cell-to-cell interaction network of the TME. Figure 6 C). The above results indicate that TINAGL1 + There is a positive correlation and active cell communication between CAF subsets and myeloid cells. Given that myeloid cells have been widely shown to have pro-tumor effects, this suggests that TINAGL1... + The synergistic interaction between CAFs and myeloid cells may jointly shape the pro-cancer microenvironment of CRC. Subsequently, we will analyze myeloid cell subsets and screen for those related to TINAGL1. + CAFs are a cell subpopulation with potential synergistic effects.

[0098] 7. MMP9 + Macs are related to CRC progress. Given TINAGL1 + CAFs may promote immune infiltration in CRC, thereby driving tumor progression. This study analyzed myeloid cell subsets with tumor immune infiltration function associated with tumor progression and metastasis. First, unsupervised cluster analysis was performed on myeloid cells, classifying 9292 myeloid cell subsets into major cell subpopulations such as macrophages, classical dendritic cells, and mast cells. Figure 7 A). Considering the important role of macrophages in tumor immune regulation, this study focuses on macrophages and classifies them into S100A9. + Macs, MMP9 + Macs, HSPA1A + Macs, FORL2 + Macs, CLEC10A + Macs and SPP1 + Macs 66 subtypes ( Figure 7B). Significant differences exist in the expression of marker genes among the subgroups, and UMAP visualization results clearly reveal the distribution of subgroup-specific gene expression. Figure 7 (C, D) The heatmap further presents the characteristic gene expression profiles of each subtype ( Figure 7 E). Figure 7 F displays, MMP9 + Macs were significantly enriched in tumor tissues and liver metastases, but their proportion was low in normal tissues, indicating that this subtype is significantly associated with tumor progression. Gene expression analysis also confirmed that the MMP9 gene is specifically highly expressed in MMP9⁺ Macs. Figure 7 G). To investigate MMP9 + Functional characteristics of Macs were analyzed, and gene set enrichment analysis (GSEA) was performed on their transcriptome. The results showed that the expression of genes involved in metastasis and metabolism (such as interferon-α response, EMT, and oxidative phosphorylation) was significantly upregulated (NES>0), while typical inflammatory pathways such as TNFα / NF-κB and inflammatory responses were significantly downregulated (NES<0). Figure 7 H). This transcriptional signature is consistent with the typical M2 macrophage phenotype, indicating that MMP9 + Macs exhibited enhanced metabolic and tissue remodeling capabilities, coupled with reduced inflammatory activity, suggesting that this cell population may promote tissue repair by remodeling the immunosuppressive microenvironment, thereby exacerbating CRC progression and metastasis. Further analysis of MMP9 gene expression characteristics in multiple cancer types was conducted using the online tool TIMER. Results showed that MMP9 was expressed at moderate to high levels in tumor tissues across various tumor types, with particularly significant increases in COAD and READ. Figure 7 I, J). Given that macrophages may play a pro-tumorigenic role in tumors, MMP9 was further analyzed in multiple independent cohorts. + The correlation between Macs expression levels and patient prognosis. In the GSE14333 cohort, CRC patients with high MMP9 expression had significantly lower DFS than those with low expression. Figure 7 K). In the GSE39582 cohort, the overall survival (OS) of CRC patients in the high MMP9 expression group was significantly lower than that in the low expression group (K). Figure 7 K). Although high MMP9 expression in the TCGA-COAD cohort was not statistically significantly associated with shorter overall survival (OS) in CRC patients, its survival trend was consistent with the results of the GEO cohort analysis, which may be related to the heterogeneity of the clinical samples in the TCGA cohort. Figure 7 L). The above results indicate that MMP9 + High expression of Macs may serve as a potential molecular marker for poor prognosis in CRC patients.

[0099] 8. TINAGL1 in CRC+ CAFs and MMP9 + Macs have potential interactions To clarify the interaction between CAF subsets and macrophage subsets, the BayesPrism algorithm was used to perform deconvolution analysis on batch transcriptome data to investigate the cell-specific expression patterns of CAFs and macrophage subsets. The results showed that TINAGL1... + CAFs and MMP9 + There is a significant positive correlation among Macs. Figure 8 A). The scatter plot further validates TINAGL1. + CAF expression levels and MMP9 + Macs expression levels were significantly positively correlated ( P <0.01)( Figure 8 B). Subsequently, this study conducted an in-depth analysis of TINAGL1 based on the intercellular communication network constructed from single-cell data. + CAFs and MMP9 + Functional interactions between two macrophage subpopulations. The Circos diagram illustrates the strength and number of interactions between CAFs and macrophage subpopulations, including TINAGL1. + CAFs and MMP9 + Macs have the most frequent and most powerful communication capabilities. Figure 8 C). The above results indicate that TINAGL1 + CAFs and MMP9 + Macs showed a positive correlation with cell percentage and exhibited active intercellular communication, suggesting that TINAGL1... + CAFs and MMP9 + Macs may synergistically play a pro-tumor role in the TME. To assess the significance of these two cell subsets in clinical prognosis, we used TINAGL1... + CAFs and MMP9 + The expression levels of Macs divided patients into four groups, namely TINAGL1 + CAFs and MMP9 + Macs high expression group, TINAGL1 + CAFs high expression group and MMP9 + Macs low expression group, TINAGL1 + Low expression of CAFs and MMP9 + Macs high expression group, TINAGL1 + Low expression of CAFs and MMP9 +In the low Macs expression group, the differences in survival rates among the four groups of patients were compared. The results showed that CRC patients with high expression of both Macs and Macs had the worst prognosis. Figure 8 D), which is closely related to poor prognosis.

[0100] 9. TINAGL1 + CAFs and MMP9 + Key signaling pathways between Macs This study further explores TINAGL1 + CAFs and MMP9 + The role and potential molecular mechanism of Macs synergistically promoting CRC progression. First, a communication network model between CAF and macrophage subpopulations was constructed. Based on the analysis of CAF outward communication patterns, the CellChat algorithm was used to identify three signal transmission patterns (Pattern 1-3). The results showed that WNT5A + CAFs and TINAGL1 + CAFs are highly similar in outward communication characteristics, both being enriched in signaling pathways related to immune regulation and matrix remodeling, such as MIF, FN1, and COLLAGEN, and synergistically participating in the remodeling process of the TME. Figure 9 A). The above results suggest that WNT5A + CAFs and TINAGL1 + CAFs may play a key role in promoting the progress of CRC.

[0101] Macrophage migration inhibitory factor (MIF) has been reported to be associated with immune escape in the TME. Analysis of MIF-related ligand-receptor interactions between CAFs and macrophages revealed that CAFs, especially TINAGL1, play a crucial role in this process. + CAFs and WNT5A + CAFs can secrete MIF and mediate signal transduction by binding to macrophage surface receptor CD74 and co-receptor CD44. Figure 9 (B) These results suggest that the CAF-derived MIF-CD74 / CD44 ligand-receptor signaling axis may be involved in regulating the functional state of macrophages and promoting the formation of an immunosuppressive microenvironment in CRC TME. In the collagen signaling pathway, we observed TINAGL1... + CAFs highly express various collagen molecules, including COL1A1, COL1A2, and COL6A2. These collagens can interact with CD44, integrin receptors, and SDC4 on the surface of macrophages. Figure 9B, C). The results indicate that collagen secreted by CAFs may participate in extracellular matrix remodeling and the formation of a fibroblastic microenvironment by binding to macrophage-associated receptors, thereby regulating tumor growth. Figure 9 C). Further communication probability analysis showed that there were significant differences in the communication strength between different CAFs and macrophage subsets in the MIF signaling pathway ( Figure 9 D, E). Among them, TINAGL1 + CAFs act as signal transmitters, in conjunction with MMP9 + The high probability of communication between Macs suggests a possible functional synergy between them in CRC TME. Furthermore, TINAGL1... + CAFs can secrete FN1 and interact with specific receptors on myeloid cells, potentially promoting the activation of tumor signaling. Figure 9 F). The above results reveal TINAGL1 + CAFs and MMP9 + The synergistic regulatory mechanism among Macs mediated by multiple signaling pathways provides theoretical support for a deeper understanding of the CRCTME communication network and its role in tumor progression.

[0102] 10. TINAGL1 + CAFs and MMP9 + Macs exhibit spatial proximity distribution in CRC. To validate the TINAGL1 predicted by CellChat analysis + CAFs and MMP9 + Macs exhibit intercellular interactions, and we further explored TINAGL1 from a tissue spatial perspective. + CAFs and MMP9 + Spatial proximity characteristics of macrophages (MACs) in the CRC microenvironment. This study used primary CRC tumor tissue, normal tissue, and CRLM tissue from 11 patients at Tangshan People's Hospital to create TMAs, and mIF staining was used for cell localization analysis. α-SMA, a protein molecule encoded by the ACTA2 gene and secreted by myofibroblasts, is one of the preferred biomarkers for identifying CAFs in tumors. α-SMA (red fluorescence) and TINAGL1 (yellow fluorescence) were used as co-labeling markers for TINAGL1. + CAFs. MMP9 was co-labeled using CD68 (green fluorescence) and MMP9 (purple fluorescence). + Macs (Macro- ... +Macs gathered at TINAGL1 + In the region surrounding CAFs, the two cell types exhibited a significant co-localization pattern, suggesting possible direct or indirect cell-cell interactions. Figure 10 A, C); In CRLM tissue, α-SMA + TINAGL1 + CAFs are distributed in the interstitium of metastatic lesions, CD68 + MMP9 + Macs are closely associated with α-SMA + TINAGL1 + CAFs aggregated, and under fluorescence microscopy, two types of cells were observed to be closely adjacent, suggesting an interaction between them in the metastatic lesions. Figure 10 B). In normal tissues, α-SMA + TINAGL1 + CAFs showed weak fluorescence signals, with only a small amount of CD68 surrounding them. + MMP9 + Macs suggest that the spatial proximity of the two tissues is significantly reduced in normal tissues. The above tissue spatial co-localization results are highly consistent with the previous CellChat analysis based on scRNA-seq data, namely TINAGL1. + CAFs and MMP9 + The highest communication frequency and strong signal transmission strength between Macs provide strong evidence for their functional interaction in CRC TME from an organizational space perspective.

[0103] 11. TINAGL1 + CAFs regulate macrophages through the MIF-CD74 / CD44 ligand-receptor signaling axis. To parse TINAGL1 + CAFs and MMP9 + To investigate the molecular mechanisms of cell communication between macrophages (CAFs), this study constructed a CAF-macrophage co-culture system. Using a third-generation primary CAF cell line isolated and passaged from CRLM tissue in our laboratory, we established stably low-TINAGL1-expressing CAFs by knocking down TINAGL1 expression via lentivirus-mediated sh-RNA. THP-1 cells were induced into macrophages, and the two were co-cultured. Changes in macrophage-related protein expression were then detected. Western blot results showed that compared with the sh-NC control group, the protein expression levels of MMP9, CD74, CD44, and CXCR4 in macrophages in the sh-TINAGL1 co-culture system were significantly reduced. Figure 11The results suggest that downregulation of TINAGL1 in CAFs can significantly weaken its induction of pro-tumor phenotypes in macrophages. To further verify the key role of the MIF signaling axis in the above-mentioned regulatory process, this study set up an intervention experiment in the sh-TINAGL1 co-culture system, including the sh-TINAGL1 + MIF inhibitor group (ISO-1) and the sh-TINAGL1 + ISO-1 + recombinant MIF protein group (rMIF). Western blot results showed that, compared with the sh-TINAGL1 + ISO-1 group, exogenous supplementation of rMIF significantly restored the protein expression levels of MMP9, CD74, CD44 and CXCR4 in macrophages. Figure 12 This result further supports TINAGL1. + CAFs can regulate the expression of MMP9 and related molecules in macrophages through the CD74 / CD44 receptor complex on the surface of macrophages, revealing TINAGL1 + The key molecular basis of CAFs-macrophage interaction in CRC TME remodeling.

[0104] The above are merely preferred embodiments of the present invention and do not limit the patent scope of the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the patent protection scope of the present invention.

Claims

1. A molecular biomarker for prognostic assessment of colorectal cancer, characterized in that, The molecular markers include TINAGL1 and / or MMP9.

2. The molecular marker according to claim 1, characterized in that, The TINAGL1 is derived from a fibroblast subpopulation in the tumor microenvironment, and the MMP9 is derived from a macrophage subpopulation in the tumor microenvironment.

3. Application of reagents for detecting the expression levels of molecular markers TINAGL1 and / or MMP9 in samples in the preparation of products for assessing the prognosis of colorectal cancer and the risk of colorectal cancer liver metastasis.

4. The application according to claim 3, characterized in that, The assessment is based on the expression level or enrichment status of TINAGL1 and / or MMP9 in the biological samples of the test subjects to evaluate the risk of liver metastasis of colorectal cancer or the prognosis of colorectal cancer patients.

5. The application according to claim 4, characterized in that, When TINAGL1 and MMP9 are both highly expressed or highly enriched in a sample, it indicates the worst prognosis for colorectal cancer patients.

6. Application of TINAGL1 inhibitors in the preparation of drugs for the prevention or treatment of colorectal cancer and colorectal cancer liver metastases.

7. The application according to claim 6, characterized in that, The TINAGL1 inhibitors inhibit tumor cell proliferation, migration, invasion, or epithelial-mesenchymal transition by reducing the expression of TINAGL1 in tumor-associated fibroblasts.

8. The application according to claim 6, characterized in that, The TINAGL1 inhibitor blocks TINAGL1. + CAFs and MMP9 + Signal communication between macrophages.

9. The application according to claim 6, characterized in that, The TINAGL1 inhibitor is selected from shRNA or siRNA that targets the TINAGL1 gene.

Images