Colorectal cancer biomarkers and their applications in diagnosis, prevention, treatment and prognosis

Abstract

The present invention discloses colorectal cancer biomarkers and their applications in diagnosis, prevention, treatment, and prognosis. The present invention also discloses distinct single-cell characteristics of blood and intestinal ILCs subsets under healthy conditions and in the setting of colorectal cancer. The healthy intestine is composed of ILC1s, ILC3s, and ILC3s / NK cells, but lacks ILC2s. Additional tumor-specific ILC1s and ILC2s subsets were identified in CRC patients. The present invention also discloses that SLAMF1 is selectively expressed on tumor-specific ILCs, and higher levels of SLAMF1+ ILCs are observed in the blood of CRC patients. The survival rate of rectal cancer patients with high SLAMF1 expression was significantly higher than that of those with low SLAMF1 expression, indicating that SLAMF1 is an anti-tumor biomarker in CRC. The present invention also discloses the use of unsupervised hierarchical clustering methods to investigate the heterogeneity of auxiliary ILCs in blood, normal mucosa, and intestinal tumors during steady state and CRC. These biomarkers can effectively diagnose, prevent, treat, assess prognosis, and predict survival for CRC.

Description

Technical Field

[0001] The present invention belongs to the field of medical technology, and relates to colorectal cancer biomarkers and their applications, and unsupervised cluster analysis methods of biomarkers. It reveals the relationship between tumor-specific innate lymphocytes and colorectal cancer progression and immunity through single-cell transcriptional maps. Background Art

[0002] T cell-based immunotherapy has been highly successful in the clinical treatment of malignancies, but its efficacy is limited to a small subset of patients (Baumeister et al., 2016; Chen and Mellman, 2017; Okazaki et al., 2013; Okazaki and Honjo, 2007; Schumacher and Schreiber, 2015; Sharma and Allison, 2015). T cell-based immunotherapy is often combined with treatment targeting other immune components to increase the cure rate in patients receiving immunotherapy. Innate lymphoid cells (ILCs) are tissue-resident, antigen-independent lymphocytes that regulate immunity against pathogens and commensals to maintain tissue homeostasis (Spits et al., 2013; Vivier et al., 2018). ILCs can form heterogeneous cell populations and are currently divided into five major categories (natural killer (NK) cells, helper ILC1s, ILC2s, ILC3s, and lymphoid tissue inducer cells) based on the cytokines produced and transcription factors expressed by these cell populations (Vivier et al., 2018). ILCs participate in the body's immune functions, including pathogen response, inflammation, tissue development, remodeling, repair, and homeostasis.

[0003] Given the quantity and nature of cytokines produced by ILCs, ILC subsets may participate in cancer immunity but may also be associated with tumor-associated inflammation. NK cells are known to play a role in cancer through tumor suppressor properties and can effectively control metastasis (Lopez-Soto et al., 2017). The role of helper ILCs in tumorigenesis and cancer immunity is less clear and may depend on the tumor microenvironment. ILC1s can produce large amounts of proinflammatory cytokines such as IFN-γ and TNF-α, thereby promoting tumorigenesis (Chiossone et al., 2018). However, in certain specialized tumor microenvironments, IFN-α can also limit tumor growth (Castro et al., 2018; Zaidi, 2019). ILC2s have been shown to be largely detrimental in various tumor settings; for example, ILCs are abundant in the peripheral blood of patients with gastric cancer (Bie et al., 2014) and acute promyelocytic leukemia (Trabanelli et al., 2017). In acute promyelocytic leukemia (Trabanelli et al., 2017) and in human bladder and mouse prostate tumors, ILC2-derived IL-13 stimulates the immunosuppressive activity of myeloid-derived suppressor cells (Chevalier et al., 2017). However, ILC2-derived IL-5 may contribute to the suppression of primary and metastatic lung tumors in mouse models (Saranchova et al., 2016), and ILC2s can activate tissue-specific tumor immunity in pancreatic cancer (Moral et al., 2020). ILC3s have been reported to have tumor suppressor properties in, for example, the B16 melanoma mouse model (Eisenring et al., 2010; Nussbaum et al., 2017) and in patients with non-small cell lung cancer (NSCLC) (Carrega et al., 2015). In contrast, ILC3s-derived IL-17 and IL-22 may contribute to the development of colorectal cancer (Chanet et al., 2014; Kirchberger et al., 2013). Therefore, it is urgent to investigate the presence and mechanisms of action of ILCs subsets in various cancer indications.

[0004] Despite significant improvements in cancer treatment strategies, colorectal cancer (CRC) remains the third most common cancer in both men and women and the second most common cause of cancer mortality worldwide (Bray et al., 2018). Dysregulated responses of ILCs have been implicated in the development of intestinal cancers. In many human pathological conditions, ILC2s are found to be low in abundance (Fuchs et al., 2013; Simoni et al., 2017), as are abnormally low levels of ILC3s (Ikeda et al., 2020; Simoni et al., 2017). These ILC3s are typically densely distributed in the colon at steady state (Fuchs et al., 2013; Ikeda et al., 2020; Simoni et al., 2017), whereas ILC1s are abundant in the intestines of CRC patients (Carrega et al., 2020; Fuchs et al., 2013; Ikeda et al., 2020; Simoni et al., 2017). Furthermore, a decreased ILC3s / ILC1s ratio is associated with the severity of CRC (Simoni et al., 2017). The landscape of ILCs subpopulations in the human intestine remains largely undefined in terms of the composition, diversity, and functional status of these cells in the steady-state and tumor contexts. Summary of the Invention

[0005] Innate lymphoid cells (ILCs) are tissue-resident lymphocytes that, unlike conventional T lymphocytes, lack antigen-specific receptors. ILCs include natural killer (NK) cells, ILC1s, ILC2s, ILC3s, and lymphoid tissue inducer (LTi) subsets. Tumor ILCs are present in a variety of cancers, but their roles in cancer immunity and immunotherapy remain far less well-defined than those of other lymphocytes, such as T cells and NK cells.

[0006] The present invention uses an unsupervised hierarchical clustering method to study the heterogeneity of ILCs subpopulations in blood, normal mucosa and intestinal tumors under steady-state and CRC conditions. The healthy intestine is composed of ILC1s, ILC3s and ILC3s / NK, but no ILC2s. The present invention proposes for the first time that ILCs from CRC patients contain two additional tumor-specific ILCs subpopulations TILCs: a tumor-specific ILC1s-like subpopulation (similar to TILC1s) and an ILC2s (TILC2s) subpopulation. SLAMF1 (signaling lymphocyte activation molecule family member 1, CD150) is selectively expressed on TILCs, and the frequency of ILCs expressing SLAMF1 is higher in the blood of CRC patients. The survival rate of CRC patients with high SLAMF1 expression is significantly higher than that of patients with low SLAMF1 expression, indicating that SLAMF1 is an anti-tumor biomarker for CRC.

[0007] The present invention proposes the use of any one or a combination of SLAMF1, HPGD, TLE4, PRDM1, AQP3, cytokine IL33, tumor-specific ILC1-like subpopulation, and ILC2 subpopulation as biomarkers for colorectal cancer. The application uses any one or a combination of SLAMF1, HPGD, TLE4, PRDM1, AQP3, cytokine IL33, tumor-specific ILC1-like subpopulation, and ILC2 subpopulation as targets for preparing diagnostic reagents for the occurrence and / or metastasis of colorectal cancer, or for preparing drugs for treating colorectal cancer, or for preparing reagents for predicting colorectal cancer survival time, or for preparing reagents for evaluating the prognosis of colorectal cancer.

[0008] In the present invention, tumor-specific ILC1-like and ILC2 subpopulations can also be combined with ILC3 subpopulations as biomarkers for the diagnosis, treatment, survival prediction, and prognosis assessment of colorectal cancer.

[0009] The present invention also provides a method for diagnosing colorectal cancer in a subject, or a method for preventing or treating colorectal cancer in a subject in need thereof, or a method for predicting or prognostically evaluating the survival time of a patient with colorectal cancer, the method comprising: determining the level of any one or a combination of several of SLAMF1, HPGD, TLE4, PRDM1, AQP3, cytokine IL33, tumor-specific ILC1-like subpopulation, and ILC2 subpopulation in a sample obtained from the patient or subject, wherein the level is used to diagnose whether the subject has colorectal cancer, or to predict or prognostically evaluate the survival time of the patient; or,

[0010] The method comprises: using any one or a combination of several of the SLAMF1, HPGD, TLE4, PRDM1, AQP3, cytokine IL33, tumor-specific ILC1-like subpopulation, and ILC2 subpopulation as targets to prevent or treat colorectal cancer in the patient.

[0011] In the application or method of the present invention, antibodies are prepared by immunizing animals with any one or a combination of SLAMF1, HPGD, TLE4, PRDM1, AQP3, and the cytokine IL33; and / or immune cells, proteins and / or small molecules are prepared by targeting any one or a combination of SLAMF1, HPGD, TLE4, PRDM1, AQP3, the cytokine IL33, tumor-specific ILC1-like subpopulations, and ILC2 subpopulations.

[0012] In the application or method of the present invention, the sample for measuring the biomarker is a sample of tumor tissue or blood obtained from the patient, or a sample of intestinal tissue or blood from the subject.

[0013] In the use or method of the present invention, the level of any one or more of SLAMF1, HPGD, TLE4, PRDM1, AQP3, and cytokine IL33 is determined at the protein level; or, the level of any one or more of SLAMF1, HPGD, TLE4, PRDM1, AQP3, and cytokine IL33 is determined at the nucleic acid level.

[0014] In the uses or methods of the present invention, when the level of any one or more of SLAMF1, HPGD, TLE4, PRDM1, AQP3, and cytokine IL33 is determined at the protein level, the level of any one or more of SLAMF1, HPGD, TLE4, PRDM1, AQP3, and cytokine IL33 is determined by immunohistochemistry; or

[0015] When the level of any one or more of SLAMF1, HPGD, TLE4, PRDM1, AQP3, and cytokine IL33 is determined at the nucleic acid level, the level of any one or more of SLAMF1, HPGD, TLE4, PRDM1, AQP3, and cytokine IL33 is determined by quantifying the mRNA encoding any one or more of SLAMF1, HPGD, TLE4, PRDM1, AQP3, and cytokine IL33.

[0016] The use or method of the present invention comprises determining the level of any one or more of SLAMF1+ILC, HPGD+ILC, TLE4+ILC, PRDM1+ILC and AQP3+ILC in the sample.

[0017] In the use or method of the present invention, a group of binding partners is used to determine the level of SLAMF1+ cells in a sample obtained from the patient, wherein the binding partners are specific for the following cell surface markers: CD45, CD127, CD117, CRTH2, CD5, TIGIT and SLAMF1; or,

[0018] The level of any one or more of SLAMF1+ILC, HPGD+ILC, TLE4+ILC, PRDM1+ILC, AQP3+ILC, and IL33+ILC is determined by flow cytometry; or,

[0019] The level of any one or more of the SLAMF1+ILC, HPGD+ILC, TLE4+ILC, PRDM1+ILC, AQP3+ILC, and IL33+ILC is determined by single-cell RNA sequencing.

[0020] In the use or method of the present invention, the higher the level of SLAMF1, the higher the probability that the patient will have a long survival time.

[0021] In the use or method of the present invention, the diagnosis comprises the following steps: i) determining the level of SLAMF1 or IL33, a cytokine that activates ILC2, in a sample obtained from the subject; ii) comparing the level determined in step i) with a predetermined reference value; iii) when the level determined in step i) is higher than the predetermined reference value, the individual being measured is a CRC patient, or when the level determined in step i) is close to the predetermined reference value, the individual being measured is a healthy individual; or,

[0022] The diagnosis comprises the following steps: i) determining the levels of SLAMF1, HPGD, TLE4, PRDM1, and AQP3 in an intestinal sample obtained from the subject; ii) comparing the levels determined in step i) with predetermined reference values; iii) when the levels of SLAMF1, HPGD, TLE4, and PRDM1 on the surface of intestinal ILC cells determined in step i) are characteristically upregulated, and AQP3 is downregulated, the individual is determined to be a CRC patient; or,

[0023] The prediction or prognostic assessment comprises the following steps: i) determining the level of SLAMF1 or IL33, a cytokine that activates ILC2, in a sample obtained from the patient; ii) comparing the level determined in step i) with a predetermined reference value; iii) when the level determined in step i) is higher than the predetermined reference value, the patient has a good prognosis, or when the level determined in step i) is lower than the predetermined reference value, the patient has a poor prognosis; or,

[0024] The prevention or treatment includes the following steps: preventing or treating colorectal cancer in a subject in need by targeting colorectal cancer tumor-specific biomarkers, signaling molecules, and cell subpopulations, including: administering a pharmaceutical composition to the subject, the pharmaceutical composition comprising a pharmaceutically acceptable carrier, an effective amount of a regulator of a colorectal cancer tumor-specific biomarker, or a regulator of a signaling molecule, or a regulator of a cell subpopulation, and optionally an additional therapeutic agent, thereby preventing or treating colorectal cancer; wherein the colorectal cancer tumor-specific biomarker is selected from SLAMF1, HPGD, TLE4, and PRDM1, the signaling molecule is AQP3, and the cell subpopulation is a specific ILC1-like and ILC2 subpopulation.

[0025] In the applications or methods of the present invention, the regulators include promoters, inhibitors, improvers, etc., for example, selected from small molecule chemical agents, antisense oligonucleotides, small interfering RNA (siRNA), short hairpin RNA (shRNA), antibodies, and biologically active fragments or homologs of the antibodies.

[0026] The present invention also provides an anti-tumor biomarker SLAMF1 and its application in diagnosing and treating colorectal cancer (CRC) diseases.

[0027] The present invention also proposes the use of SLAMF1 as a biomarker in the diagnosis and treatment of colorectal cancer (CRC) and the prediction of colorectal cancer (CRC) survival rate.

[0028] The present invention also proposes the use of a detection reagent for the biomarker SLAMF1 in the preparation of drugs for diagnosing and treating colorectal cancer (CRC) and predicting the survival rate of colorectal cancer (CRC).

[0029] In the present invention, high levels of SLAMF1 are associated with higher survival rates in CRC patients and can be used as an anti-tumor biomarker for CRC, wherein the high levels are obtained based on TCGA data analysis.

[0030] The present invention also proposes the use of tumor-specific ILC1s-like subpopulations and ILC2s subpopulations as biomarkers in the diagnosis and treatment of colorectal cancer (CRC).

[0031] The present invention also proposes the use of tumor-specific ILC1s-like subpopulations and ILC2s subpopulations as biomarkers in predicting colorectal cancer (CRC) survival rate.

[0032] Among them, the "tumor-specific ILC1s-like subpopulation" and "tumor-specific ILC2s subpopulation" are a new group of cell populations.

[0033] The present invention also proposes the use of tumor-specific genes PTGDR2 and / or GATA3 as biomarkers in the diagnosis and treatment of colorectal cancer (CRC).

[0034] The present invention also proposes the use of tumor-specific genes PTGDR2 and / or GATA3 as biomarkers in predicting colorectal cancer (CRC) survival rate.

[0035] The present invention also proposes the use of the tumor-specific gene (or gene combination) AQP3 as a biomarker in the diagnosis and treatment of colorectal cancer (CRC).

[0036] The present invention also proposes the use of the tumor-specific gene (or gene combination) AQP3 as a biomarker in predicting the survival rate of colorectal cancer (CRC).

[0037] The present invention also proposes the use of a combination of ILC1s, ILC2s and ILC3s subpopulations in the blood as biomarkers in the diagnosis and treatment of colorectal cancer (CRC).

[0038] The present invention also proposes the use of a combination of ILC1s, ILC2s and ILC3s subpopulations in the blood as biomarkers in predicting colorectal cancer (CRC) survival rate.

[0039] The present invention also proposes the use of SLAMF1, HPGD, TLE4 and PRDM1 gene combinations in blood as biomarkers in the diagnosis and treatment of colorectal cancer (CRC).

[0040] The present invention also proposes the use of SLAMF1, HPGD, TLE4 and PRDM1 gene combinations in blood as biomarkers in predicting CRC survival rate.

[0041] In the present invention, any one or a combination of SLAMF1, HPGD, TLE4, PRDM1, AQP3, cytokine IL33, tumor-specific ILC1-like subpopulation, and ILC2 subpopulation, including, for example, SLAMF1 alone, is used as a biomarker or target in the diagnosis or treatment of colorectal cancer (CRC), in the prediction of colorectal cancer (CRC) survival rate, or in the prognosis evaluation of colorectal cancer. By detecting SLAMF1 on the surface of intestinal ILC cells, if SLAMF1 is characteristically upregulated, healthy individuals and CRC patients can be distinguished.

[0042] In the present invention, any one or combination of SLAMF1, HPGD, TLE4, PRDM1, AQP3, cytokine IL33, tumor-specific ILC1-like subpopulation, and ILC2 subpopulation, including, for example, the use of cytokine IL33 alone as a biomarker or target in the diagnosis or treatment of colorectal cancer (CRC), in predicting colorectal cancer (CRC) survival rate, or in the prognosis evaluation of colorectal cancer. The cytokine IL33 refers to the cytokine IL33 that activates ILC2, and high expression of the cytokine IL33 is associated with a longer survival time in CRC patients, indicating that TILC2 may indicate a good prognosis in CRC patients.

[0043] In the present invention, any one or more combinations of SLAMF1, HPGD, TLE4, PRDM1, AQP3, cytokine IL33, tumor-specific ILC1-like subpopulations, and ILC2 subpopulations, including, for example, a combination of SLAMF1, HPGD, TLE4, PRDM1, and AQP3, are used as biomarkers or targets in the diagnosis or treatment of colorectal cancer (CRC), in the prediction of colorectal cancer (CRC) survival rate, or in the prognosis of colorectal cancer. The characteristic upregulation of SLAMF1, HPGD, TLE4, and PRDM1, and the downregulation of AQP3 in intestinal TILCs, can distinguish healthy individuals from CRC patients.

[0044] In the present invention, the tumor-specific ILC1-like subpopulation is TIGIT+TILC1-like cells.

[0045] In one embodiment, the present invention provides a method for predicting the survival time of a colorectal cancer patient, comprising determining the level of SLAMF1 in a sample obtained from the patient, wherein the level is correlated with the patient's survival time.

[0046] As used herein, the term "colorectal cancer" includes the generally accepted medical definition, which defines colorectal cancer as a medical condition characterized by cancer of the intestinal cells below the small intestine, i.e., cancer of the large intestine (colon), which includes the cecum, ascending colon, transverse colon, descending colon, sigmoid colon, and rectum. Additionally, as used herein, the term "colorectal cancer" also includes medical conditions characterized by cancer of the cells of the duodenum and small intestine.

[0047] The method of the present invention is particularly suitable for predicting the overall survival (OS), progression-free survival (PFS) and / or the disease-free survival (DFS) of cancer patients.

[0048] As used herein, the term "short survival time" means that a patient's survival time will be shorter than the median (or mean) observed among patients in the general population. When a patient's survival time is short, it means that the patient will have a "poor prognosis." Conversely, the term "long survival time" means that a patient's survival time will be longer than the median (or mean) observed among patients with cancer in general. When a patient has a long survival time, it means that the patient will have a "good prognosis."

[0049] The samples involved in the present invention include blood, intestinal mucosa and colorectal cancer tissue samples obtained from patients in any storage method. The SLAMF1 level detected includes the nucleic acid and protein levels of SLAMF1 detected by any means (agents that can selectively bind to SLAMF1).

[0050] The methods of the present invention include determining the level of SLAMF1+ ILCs in a sample. In particular, for the described embodiments, the sample can be a tumor tissue sample or a blood sample. According to current definitions, ILCs refer to innate lymphoid cells, including natural killer (NK) cells, helper ILC1s, ILC2s, ILC3s, and lymphoid tissue induction cells. As described herein, the term "SLAMF1+ ILCs" refers to ILCs that express SLAMF1 (i.e., SLAMF1 protein or SLAMF1 nucleic acid, such as mRNA). The level of SLAMF1+ ILCs is measured by ILCs (e.g., mean fluorescence intensity (MFI)) as a measure of the expression intensity of a marker (e.g., protein and / or mRNA), or by the amount of ILCs expressing SLAMF1 (e.g., protein and / or mRNA) in a sample (e.g., frequency (e.g., percentage) of SLAMF1+ ILCs and density of SLAMF1+ cells). Quantification of SLAMF1+ ILCs generally involves the presence or absence of specific cell surface markers. It includes flow cytometry, mass spectrometry (CyTOF) and qPCR / RT-PCR detection of SLAMF1 levels after ILCs are sorted.

[0051] The cell surface marker combination used in the present invention includes CD45, CD127, CD117, CRTH2, CD5, TIGIT, and SLAMF1. The present invention includes any form of antibody coupling agent to detect these surface proteins. These antibodies include monoclonal or polyclonal antibodies from different species. Coupling agents include lanthanide metal markers for mass spectrometry and fluorescent conjugates for conventional flow cytometry.

[0052] The term "SLAMF1" as used herein refers to signaling lymphocyte activation molecule 1, also known as CDw150, IPO-3, SLAM family member 1, and CD150. The following is an exemplary amino acid sequence of SEQ ID NO: 1:

[0053] SEQ ID NO:1>sp|Q13291|SLAF1_HUMAN Signaling lymphocytic activation molecule OS=Homo sapiens OX=9606GN=SLAMF1 PE=1SV=1

[0054] MDPKGLLSLTFVLFLSLAFGASYGTGGRMMNCPKILRQLGSKVLLPLTYERINKSMNKSIHIVVTMAKSLENSVENKIVSLDPSEAGPPRYLGDRYKFYLENLTLGIRESRKEDEGWYLMTLEKNVSVQRFCLQLRLYEQVSTPEIKVLNKTQENGTCTLILGCTVEK GDHVAYSWSEKAGTHPLNPANSSHLLSLTLGPQHADNIYICTVSNPISNNSQTFSPWPGCRTDPSETKPWAVYAGLLGGVIMILIMVVILQLRRRGKTNHYQTTVEKKSLTIYAQVQKPGPLQKKLDSFPAQDPCTTIYVAATEPVPESVQETNSITVYASVTLPES.

[0055] The beneficial effects of the present invention include: SLAMF1 is selectively expressed in tumor-specific ILCs (TILCs), and higher levels of SLAMF1+ ILCs are observed in the blood of CRC patients. The survival rate of rectal cancer patients in the high SLAMF1 group is significantly higher than that in the low SLAMF1 group, indicating that SLAMF1 is an anti-tumor biomarker for CRC. This biomarker can effectively diagnose, prevent, treat, assess prognosis, and predict survival rates for CRC. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 : scRNAseq analysis showed that ILC1s, ILC3s, and ILC3s / NKs were observed in normal mucosa, but no ILC2s were observed.

[0057] A. UMAP images of 16,145 ILCs from four normal mucosa samples. Cell colors are defined according to the specific dataset.

[0058] B. UMAPs colored according to donor origin.

[0059] C. Unsupervised hierarchical clustering of each donor into six clusters based on the mean expression levels of variably expressed genes within the cells. Each sample is colored according to its relationship to a specific subset.

[0060] D. Heatmap showing 542 genes that differentiated the three ILC groups using the Wilcoxon rank-sum test (nmC0-3: 256 genes, nmC4: 97 genes, nmC5: 189 genes). Cells are represented in columns, genes in rows, and genes are sorted by adjusted p-value < 0.05. Gene expression intensity is indicated by color based on a z-score distribution ranging from -2.5 (purple) to 2.5 (yellow). Boxes indicate transcriptome information for specific ILC subpopulations.

[0061] E. Principal component analysis (PCA) was performed to cluster the three ILCs of each sample according to the average expression levels of variably expressed genes.

[0062] F. Distribution of genes that account for 20% of the total information of each PC (E) in each ILCs subset.

[0063] G. Table showing the top 10 differentially expressed genes among ILCs groups, including the top 10 genes encoding transcription factors, secreted proteins, and cell membrane markers. Gene symbols and annotations were retrieved from public databases. Genes are ranked by p-value.

[0064] H. Based on the gene characteristics of tonsil ILCs, module scores were performed on the gene expression groups of the three groups of ILCs at the single-cell level.

[0065] Figure 2 : scRNA-seq analysis showed the presence of tumor-specific ILC1s-like and ILC2s subsets in CRC tissues.

[0066] A. UMAP images of 15,101 tumor-specific ILCs (TILCs) from CRC tissues of four patients. Cells are displayed in different colors according to the defined cell populations.

[0067] B. Different donors are shown in different colors in the UMAP image.

[0068] C. Unsupervised hierarchical clustering of four clusters per donor based on the mean of variable gene expression levels.

[0069] D. Heat map analysis of 982 genes (463 in TILCs C0, 100 in TILCs C1, 314 in TILCs C2, and 105 in TILCs C3) was performed, and the Wilcoxon rank sum test was used to distinguish the four TILC subpopulations in tumor tissue. Cells are represented in columns, and genes are represented in rows, arranged by adjusted p-value (<0.05). Gene expression is shown in color based on the z-score distribution, ranging from -2.5 (purple) to 2.5 (yellow). Boxes identify the transcriptomes of specific TILC subpopulations.

[0070] E. Principal component analysis (PCA) of the four TILCs subpopulations in each sample based on the mean expression levels of variably expressed genes.

[0071] F. The genes that contribute to 20% of the total to each PC in (E) are displayed, and their distribution in each subgroup is distinguished by different colors.

[0072] G. List showing the top 10 expressed genes ranked by p-value from the total list of genes with significant differences between TILCs groups, as well as the top 10 expressed genes encoding transcription factors, secreted proteins, and cell membrane markers. Gene symbols and annotations were retrieved from public databases.

[0073] H. At the single-cell level, modular scoring of four TILCs subsets was performed based on the tonsillar ILCs gene signatures.

[0074] Figure 3 : Characteristics of tumor tissue-specific ILCs subpopulations.

[0075] A. UMAP images of ILCs from 41,603 samples of normal blood, CRC blood, and CRC tissue. Cells are colored according to the defined subpopulations.

[0076] B. Unsupervised hierarchical clustering of normal blood, CRC blood, and CRC tissue ILCs from each donor based on the mean expression levels of variably expressed genes.

[0077] C. Heatmap showing 899 genes (44 in normal blood, 70 in CRC blood, and 775 in CRC tissue) that distinguished the three tissues using the Wilcoxon rank sum test. Cells are represented in columns, and genes are represented in rows, arranged by adjusted p-value (<0.05). Gene expression is displayed in color, with a z-score distribution ranging from -2.5 (purple) to 2.5 (yellow). Boxes identify transcriptomic signatures of specific ILC subpopulations.

[0078] D. Unsupervised hierarchical clustering of normal blood ILCs (nbILCs), CRC blood ILCs (cbILCs), and CRC tissue ILCs (TILCs) of each donor was performed at the subset level based on the mean expression levels of the variably expressed genes.

[0079] E. Venn diagram showing the gene intersection of the four ILC1s subsets from normal blood, CRC blood, normal mucosa, and CRC tissue.

[0080] F. Venn diagram showing the gene intersection of the three ILC2s subsets in normal blood, CRC blood, and CRC tissue.

[0081] Figure 4 : ILCs-specific gene signatures in CRC.

[0082] A. UMAP images of 31,246 ILCs from normal mucosa and CRC tissues. Cells are colored according to defined cell populations.

[0083] B. Unsupervised hierarchical clustering of ILCs from normal mucosa and CRC tissues of each donor based on the mean expression levels of variably expressed genes.

[0084] C. Heatmap showing 331 genes (266 in CRC tumors and 51 in normal mucosa) that differentiated between the two organs using the Wilcoxon rank sum test. Cells are represented in columns, and genes are represented in rows, arranged by adjusted p-value (<0.05). Gene expression is shown in color based on the z-score distribution, ranging from -2.5 (purple) to 2.5 (yellow). Transcriptome information for specific ILC subpopulations is shown in boxes.

[0085] D. UMAP plots of 26,502 ILCs in normal and CRC blood. Cells are colored according to the defined dataset.

[0086] E. Unsupervised hierarchical clustering of normal blood and CRC blood ILCs from each donor based on the mean expression levels of variably expressed genes.

[0087] F. Heatmap showing 254 genes (233 in CRC blood and 23 in normal blood) that differentiated between the two organs using the Wilcoxon rank sum test. Cells are represented in columns and genes are represented in rows, arranged by adjusted p-value (<0.05). Gene expression is represented by color based on the z-score distribution, ranging from -2.5 (purple) to 2.5 (yellow). Boxes identify transcriptome profiles of specific ILC subpopulations.

[0088] G. Venn diagram showing the gene intersection between normal blood and CRC blood, and normal mucosa and CRC tissue.

[0089] H. Feature plot showing the relative expression level of SLAMF1 in each ILCs in normal blood, CRC blood, normal mucosa, and tumor tissues.

[0090] Figure 5 : SLAMF1 is a biomarker for CRC.

[0091] A. Representative flow cytometry analysis showing surface TIGIT expression in the TILC1s-like subpopulation.

[0092] B. Flow cytometric analysis of ILCs subsets in normal mucosa (n=8-16) and CRC tissues (n=7-16). Data show the proportions of the indicated ILCs subsets in total ILCs.

[0093] C. Kaplan-Meier curves of overall survival for CRC patients stratified by IL-33 expression in the TCGA dataset. The Cox proportional hazards model was used to determine the optimal cutoff points for patient stratification, and the log-rank test was used to calculate p-values. IL-33-high group (n = 192); IL-33-low group (n = 91).

[0094] D. Flow cytometric analysis of ILCs subsets in blood from healthy blood donors (n=18) and CRC patients (n=16).

[0095] E. Representative FACS images show the expression of SLAMF1 on the surface of ILCs.

[0096] F. Flow cytometry analysis of SLAMF1 expression in ILCs from normal mucosa (n=7) and CRC tissues (n=7). The data show the proportion of SLAMF1+ ILCs to total ILCs.

[0097] G. Flow cytometric analysis of SLAMF1 expression in total blood ILCs from healthy blood donors (n=14) and CRC patients (n=5). The data show the proportion of SLAMF1+ cells in total ILCs.

[0098] H. Kaplan-Meier curves of CRC patients stratified by SLAMF1 expression levels in the TCGA analysis. A Cox proportional hazards model was used to determine the optimal cutoff for patient stratification, and p-values ​​were calculated using the log-rank test. The SLAMF1-high group (n = 73) and the SLAMF1-low group (n = 20) were included.

[0099] (B) and (D) Kruskal-Wallis test with Dunn's multiple comparison test, and p values ​​were adjusted using the Benjamin-Hochberg method. (F) and (G) Mann-Whitney test with nonparametric t test.

[0100] *p-value<0.05; **p-value<0.01; ***p-value<0.001; ****p-value<0.0001.

[0101] Figure 6 :The experimental design of this study

[0102] A. Single-cell transcriptome analysis workflow for ILCs isolated from blood, adjacent normal mucosa, and CRC tissues from healthy donors or patients with colorectal cancer (CRC).

[0103] B. Gating strategy for human ILCs defined as lineage marker negative (Lin-) and CD127 positive (CD127+). SSC, side scatter; FSC: forward scatter.

[0104] C. Percentage of ILCs among CD45+ lymphocytes in tissues (left). Statistics regarding ILC proportions and sample size (right). Data are presented as mean ± SD, with p values ​​calculated using the Mann-Whitney test and nonparametric t-test. ns, not significant. ****p value < 0.0001.

[0105] D. Graph summarizing the scRNAseq data from this study. This includes the total cell number for normal blood (n=4), CRC blood (n=3), adjacent normal mucosa (n=4), and CRC tissue (n=4), the number of genes detected per tissue, the number of genes detected per cell in each tissue, the number of cells captured per sample, the total number of genes detected per sample, and the average number of genes detected per cell in each sample. # = number.

[0106] Figure 7 : Distribution of normal mucosal ILCs subsets based on known characteristics or genes.

[0107] A. Feature plot of the relative expression of IL7R, GATA3, NCR3, EOMES, TBX21, PTGDR2, KIT, RORC, NCR1, NCR2, and KLRF1 in nmILCs, separated by colored dashed lines, e.g. Figure 7 As shown in A.

[0108] B. Six nmILCs populations were scored at the single-cell level using the signature gene lists of human jejunal (ileal) ILC1s, ILC3s, NK cells, and spleen ILC2s (Yudanin et al., 2019).

[0109] C. Feature plot of the relative expression of CD5 in each nmILCs cell, separated by colored dashed lines as indicated.

[0110] Figure 8 : Cell markers that assist in defining ILCs subpopulations in tumor tissue.

[0111] FeaturePlot shows the relative expression levels of IL7R, GATA3, NCR3, EOMES, TBX21, PTGDR2, KIT, RORC, NCR1, NCR2, and KLRF1 in TILCs. Colored dotted lines are used to distinguish each ILCs subpopulation, as shown in the figure.

[0112] Figure 9 : Characteristics of ILC3s subsets in normal mucosa and CRC tissues.

[0113] A. UMAP image of 15,701 normal mucosal ILC3s. Cells are colored according to the cell population to which they belong.

[0114] B. Different colors in UMAP indicate different donors.

[0115] C. Unsupervised hierarchical clustering of the four nmILC3s clusters from each donor based on the mean expression levels of differentially expressed genes.

[0116] D. A Wilcoxon rank-sum test was used to identify 316 genes that distinguished nmILC3s (0: 74 genes, 1: 85 genes, 2: 48 genes, 3: 109 genes) and displayed as a heatmap. Cells are represented in columns, and genes are represented in rows, arranged by adjusted p-value (< 0.05). Gene expression is represented by color based on a z-score distribution ranging from -2.5 (purple) to 2.5 (yellow). Boxes identify transcriptomic signature genes of specific ILCs subpopulations.

[0117] E. Table showing the top 10 expressed genes ranked by p-value from the total list of genes significantly differentially expressed between ILCs groups, as well as the top 10 expressed genes encoding transcription factors, secreted proteins, and cell membrane markers. Gene symbols and annotations were retrieved from public databases.

[0118] F. UMAP shows 9260 ILC3s cells in tumor tissue. Different colors represent cells belonging to different cell populations.

[0119] G.UMAP shows individuals from different sources and represents them with different colors.

[0120] H. Unsupervised hierarchical clustering of ILC3s from CRC tissues of each donor based on the mean expression levels of variably expressed genes.

[0121] I. Heatmap showing 420 genes (0,98 genes; 1,106 genes; 2,49 genes; 3,167 genes) that distinguish ILC3s from tumor tissues using the Wilcoxon rank sum test. Cells are represented in columns and genes are represented in rows, arranged by adjusted p-value (<0.05). Gene expression is represented by color based on the z-score distribution, ranging from -2.5 (purple) to 2.5 (yellow). Boxes identify transcriptome information for specific ILCs subpopulations.

[0122] J. The top 10 expressed genes ranked by p-value from the total list of genes with significant differences between ILCs groups are shown. The top 10 expressed genes encoding transcription factors, secreted proteins, and cell membrane markers are also listed. Gene symbols and annotations were retrieved from public databases.

[0123] Figure 10 : scRNAseq defines ILC1s, ILC2s, and ILC3s in the blood of healthy blood donors.

[0124] A. UMAP image of 19,603 healthy blood ILCs. Cells are colored according to the cell population to which they belong.

[0125] B. Different colors in UMAP indicate different donors.

[0126] C. Unsupervised hierarchical clustering of the four nbILCs from each donor based on the mean expression levels of differentially expressed genes.

[0127] D. Heatmap showing 356 genes (nbC0, 90; nbC1, 191; nbC2, 75 genes) that differentiate ILCs from normal blood tissue using the Wilcoxon rank sum test. Cells are represented in columns, and genes are represented in rows, arranged by adjusted p-value (<0.05). Gene expression is represented by color based on the z-score distribution, ranging from -2.5 (purple) to 2.5 (yellow). Boxes identify transcriptome profiles of specific ILC subpopulations.

[0128] E. Principal component analysis of the three ILCs clusters for each sample based on the average expression levels of variably expressed genes.

[0129] F. Distribution of genes that account for 20% of the total information of each PC (E) in each ILCs subset.

[0130] G. Table showing the top 10 expressed genes ranked by p-value from the total list of genes with significant differences between ILCs groups, as well as the top 10 expressed genes encoding transcription factors, secreted proteins, and cell membrane markers. Gene symbols and annotations were retrieved from public databases.

[0131] H. At the single-cell level, modular scoring was performed for the three ILCs subsets based on the tonsil ILCs gene signatures.

[0132] I. Feature plot showing the relative expression levels of IL7R, GATA3, NCR3, EOMES, TBX21, PTGDR2, KIT, RORC, NCR1, NCR2, and KLRF1 in nbILCs, separated by colored dashed lines, as shown.

[0133] Figure 11 : scRNAseq defines ILC1s, ILC2s, and ILC3s in the blood of CRC patients.

[0134] A. UMAP images of ILCs from the blood of 6,899 CRC patients. Cells are colored according to the cell population to which they belong.

[0135] B. Different colors in UMAP indicate different donors.

[0136] C. Unsupervised hierarchical clustering of the three cbILCs from each donor based on the mean expression levels of differentially expressed genes.

[0137] D. Heatmap showing 359 genes (cbC0, 143; cbC1, 81; cbC2, 135 genes) that differentiate ILCs from normal blood tissue using the Wilcoxon rank sum test. Cells are represented in columns, and genes are represented in rows, arranged by adjusted p-value (<0.05). Gene expression is represented by color based on the z-score distribution, ranging from -2.5 (purple) to 2.5 (yellow). Boxes identify transcriptome profiles of specific ILC subpopulations.

[0138] E. Principal component analysis of the three ILCs clusters for each sample based on the average expression levels of variably expressed genes.

[0139] F. Distribution of genes that account for 20% of the total information of each PC (E) in each ILCs subset.

[0140] G. Table showing the top 10 expressed genes ranked by p-value from the total list of genes with significant differences between ILCs groups, as well as the top 10 expressed genes encoding transcription factors, secreted proteins, and cell membrane markers. Gene symbols and annotations were retrieved from public databases.

[0141] H. At the single-cell level, modular scoring was performed for the three ILCs subsets based on the tonsil ILCs gene signatures.

[0142] I. Feature plot showing the relative expression levels of IL7R, GATA3, NCR3, EOMES, TBX21, PTGDR2, KIT, RORC, NCR1, NCR2, and KLRF1 in cbILCs, separated by colored dashed lines, as shown.

[0143] J. RNA velocity of ILCs in the blood of CRC patients was projected onto the UMAP graph using Gaussian smoothing on a regular grid.

[0144] Figure 12 : Expression patterns of AQP3, HPGD, TLE4 and PRDM1 among different tissues.

[0145] FeaturePlot shows the relative expression levels of AQP3, HPGD, TLE4, and PRDM1 in normal blood, CRC patient blood, normal mucosa, and tumor tissues. DETAILED DESCRIPTION

[0146] The present invention will be further described below in conjunction with the examples and accompanying drawings. These examples should be understood to be merely illustrative of the present invention and not to limit the scope of the present invention. Without departing from the spirit and scope of the present invention, variations and advantages that can be imagined by those skilled in the art are included in the present invention and are protected by the appended claims and their equivalents.

[0147] Example 1 The healthy intestine contains ILC1s, IL3Cs, and ILCs / NKs, but no ILC2s.

[0148] 1. Materials and Methods

[0149] (1.1) Clinical sample collection

[0150] Healthy blood samples for scRNAseq were obtained from individuals undergoing routine colonoscopy who were generally healthy and had no other relevant medical history, such as inflammatory bowel disease (IBD) or CRC.

[0151] (1.2) Isolation of human lymphocytes

[0152] Fresh intestinal tissue was prepared immediately after surgery. Fat tissue and visible blood vessels were removed. The samples were weighed and washed with PBS and then cut into small pieces. Normal tissue was incubated in 10 mL of freshly prepared endothelial lymphocyte solution (PBS containing 5 mM EDTA, 15 mM HEPES, 10% FBS, and 1 mM DTT) at 37°C for 1 hour with shaking at 200 rpm. CRC tissue was washed with 10 mL of freshly prepared PBS containing 65 mM DTT at 37°C for 15 minutes with shaking. After incubation, the tissue pieces were rinsed twice with PBS and enzymatically digested at 37°C for 1 hour. The digestion solution mainly consisted of RPMI 1640 containing 0.38 μg / mL collagenase VIII, 0.1 mg / mL DNase I, 100 U / ml penicillin, 100 mg / mL streptomycin, and 10% FBS. The digested tissue was then shaken vigorously by hand for 5 minutes and then mechanically dissociated using a 21-gauge syringe. The resulting cell suspension was filtered through a 100 μm pore size cell strainer into a new 50 mL conical tube. The volume was brought up to 30 mL with PBS. The cells were then centrifuged at 1800 rpm for 5 minutes. The supernatant was removed and the cells were resuspended in RMPI 1640 containing 10% FBS. Density gradient centrifugation was performed using Ficoll and the cells were resuspended in PBS after centrifugation.

[0153] Peripheral blood mononuclear cells (PBMCs) are obtained from human blood samples by centrifugation on a Ficoll density gradient. Briefly, blood is mixed in equal parts with 2% FBS in PBS and gently layered on a Ficoll density gradient. Centrifugation is performed at 1000 x g for 25 minutes with the brake at zero. The cells in the middle layer are aspirated and resuspended in 2% FBS in PBS for use.

[0154] (1.3) Removal of contaminating cells

[0155] The R package SingleR (Aran et al., 2019) was used with default parameters, with the Human Primary Cell Atlas Data as the reference dataset. The cell type information in "label.main" was used to annotate each cell with its cell type. Because the dataset does not include a human ILCs dataset, and given the similarities between ILCs and cells or NK cells, the present invention retains cells labeled as NK and T cell types. Small clusters of fewer than 200 cells were removed to eliminate contigs. Certain individual-specific cell populations with strong specific NK cell characteristics were considered true NK cell contamination and removed from downstream analyses.

[0156] (1.4) ILCs sorting

[0157] Freshly prepared cells were resuspended in PBS and incubated with a cell death dye at 4°C for 10 minutes. Cells were resuspended in FACS buffer (PBS containing 2% FBS and 2mM EDTA) containing 10% mouse serum and 40% Brillilant Stain buffer. Blocked with Fc blocking antibodies for 10 minutes, and then incubated with antibodies against human CD45, CD127, CD117, CRTH2, and lineage antibodies (TCRγδ, TCRαβ, CD3, CD19, CD14, CD16, CD94, CD123, CD34, CD303, and FcεRI) at room temperature for 30 minutes. After incubation, cells were washed with FACS buffer and centrifuged. Collected cells were resuspended in FACS buffer and sorted for live ILCs using a BD FACSAria III instrument.

[0158] 2. Experimental results

[0159] The present invention studies paired samples of CRC tissue and adjacent mucosal tissue (used as a control) and compares blood from patients with blood from age-matched healthy donors ( Figure 6 A) Analyze the role of ILCs in CRC. Lin-CD127+ILCs are more abundant in normal mucosa and CRC tissues than in blood, which is consistent with the known tissue retention characteristics of ILCs (Gasteiger et al., 2015) ( Figure 6 BC). The content of ILCs in CRC tissues was significantly lower than that in normal mucosa, but the proportion of ILCs in blood samples of the normal group and CRC group was similar ( Figure 6 BC).

[0160] Example 2 The present invention performed single-cell transcriptome sequencing (scRNAseq) on 58,000 ILCs in blood samples of CRC patients, healthy blood, normal mucosa and CRC tissue samples ( Figure 6 D).

[0161] 1. Materials and Methods

[0162] (1.1) Single-cell RNA sequencing

[0163] Purified ILCs were resuspended in PBS containing 0.04% BSA and kept on ice. Cells were counted, and the cell density was adjusted to the recommended concentration for the 10X Genomics v3 kit. The libraries were then sequenced using the Illumina platform (NovaSeq 6000) from Crystalline Genetics, with a sequencing depth of approximately 90,000 reads per cell.

[0164] (1.2) Raw sequence alignment, quality control and normalization

[0165] FastQC software v0.11.9 was used

[0166] (https: / / www.bioinformatics.babraham.ac.uk / projects / fastqc / ) The raw sequencing reads were quality controlled. The sequencing data in the bcl file were converted to FASTQ format using the bcl2fastq2 conversion software v2.20 (https: / / support.illumina.com / downloads / bcl2fastq-conversion-software-v2-20.html). The sequencing data were processed, aligned, and summarized using the standard process and default parameters of the Cell Ranger Single Cell Software Suite v.2.2 software. In short, the present invention uses the Cell Ranger Count standard process to align the FASTQ sequence with the GRch38 genome. Low-quality sequencing reads were filtered out based on the base judgment score, and a cell barcode and unique molecular identifier were assigned to each read. The single-cell sequencing data were normalized using the default parameters of the Cell Ranger aggr standard process to obtain data of the same sequencing depth. The matrix containing gene features and cell features was used for subsequent analysis.

[0167] During quality control of the Seurat analysis, the raw UMI count matrix was filtered to remove genes with fewer than three cells, cells with fewer than 200 genes, cells with more than 4,000 genes, and cells with a high proportion of mitochondrial genes (more than 8%). The resulting matrix was then normalized by a global scaling method, transformed using a scaling factor (10,000 by default), and logarithmically transformed using the "LogNormalize" function in Seurat for downstream analysis.

[0168] 2. Experimental results

[0169] The heterogeneity of ILCs in normal mucosa of CRC patients was evaluated together with 16,145 Lin-CD127+ ILCs from adjacent colon tissue of colorectal tumors ( Figure 6 D). In the unified manifold approximation and projection (UMAP) analysis, the two-dimensional projection of cells showed that they were separated into six different cell populations: normal mucosal cell population (nmC) 0 to nmC5 ( Figure 1A). The two populations, nmC4 and nmC5, contain cells from all donors, indicating that these two ILCs populations do not have donor-specific transcriptome profiles ( Figure 1 BC). In contrast, most cells from nmC0 to nmC3 were specific to a single donor ( Figure 1 BC).

[0170] Example 3 Hierarchical Clustering

[0171] 1. Materials and Methods

[0172] (1.1) Unsupervised hierarchical clustering

[0173] Gene expression values ​​were calculated for individual cells within each cell population. Only genes previously selected as variable features were used. Unsupervised cluster maps were generated using the Heatmap.plus package. Euclidean distances were calculated for genes in all cell populations. For normal mucosa, only cell populations with clusters 0-4 were used for analysis.

[0174] (1.2) Dimensionality reduction and clustering

[0175] The top 2000 genes were screened using the "FindVariableGenes" function of Seurat (Stuart et al., 2019) and used for principal component analysis (PCA). For ILCs in normal mucosa, the present invention retained the top 40 principal components. For normal blood, CRC blood, and tumor tissue, the present invention retained the top 20 PCs. Clusters were identified using the "FindClusters" function, which is based on the nearest neighbor modular optimization implemented in Seurat and visualized using the uniform manifold approximation and projection (UMAP) algorithm. For comparisons between different individuals and different tissues, the "merge" function was used to merge individual Seurat objects. For visualization of donor and tissue data, when using the "DimPlot" function in Seurat for plotting, the tissue and donor information was used to present them using "group.by".

[0176] (1.3) Differential expression analysis

[0177] The present invention uses the "FindAllMarkers" function in Seurat to identify differentially expressed genes between each cluster in the sample. Using the non-parametric Wilcoxon rank sum test and based on the Bonferroni correction, the p-values ​​for comparison and the adjusted p-values ​​for all genes in the dataset are obtained. The present invention calculates the logarithmic fold change (logFC) of the expression values ​​and obtains the p-values ​​of all variable genes in each cluster using the following parameters: min.pct = 0.05, min.diff.pct = 0.1, logfc.threshold = 0.25. The logarithmic transformation and scaled expression values ​​of the genes are used to generate heat maps.

[0178] (1.4) Principal component analysis

[0179] Principal component analysis was performed based on the mean expression values ​​of variable genes within each cluster. The weights of the top 20 genes contributing to PCs 1 and 2, or PCs 1 and 3, were plotted. For normal mucosal ILCs, principal component analysis was performed after removing ILC subpopulations from clusters 0-4 that were not the primary individual of origin. The top 20 genes contributing most to PCs 1 and 2, or PCs 1 and 3, were graphically displayed.

[0180] (1.5) Scoring samples using ILCs gene signatures

[0181] The gene signatures of ILCs and NK cells from the tonsils were defined by Bjorklund et al. (Bjorklund et al., 2016). The gene signatures of ILC1s and ILC3s from the jejunum (ileum) and ILC2s from the spleen were derived from findings by Yudanin et al. (Yudanin et al., 2019). Seurat's "AddModuleScore" tool was used to assign a module score to each ILC. Briefly, the mean gene expression value for each cell was calculated, and the expression value of a control gene across all cells was subtracted. The control gene was randomly selected from the total gene set. The gene list for scoring ILC3s in tumor tissue was the differentially expressed genes in ILC3s from adjacent tissues. Violin plots were used to present the gene module scores for each subpopulation.

[0182] 2. Experimental results

[0183] Using hierarchical clustering ( Figure 1 C) and gene marker heatmap ( Figure 1 D), principal component analysis (PCA) Figure 1 EF), analysis of the 10 most highly expressed genes ( Figure 1 G) and module score analysis ( Figure 1H), we compared the gene signatures of nmC0 to nmC5 with the transcriptome signatures of previously described human ILCs subsets (Bjorklund et al., 2016).

[0184] Among the top 10 highly expressed genes, nmC0 to nmC3 shared common transcriptome features with ILC3s: REL, KIT, CXCL8, IL4l1, and IL1R1. Among them, REL can encode NF-κB family proto-oncogene signals through the IL22 promoter site in ILC3s (Victor et al., 2017) ( Figure 1 FG). nmC4 is characterized by the expression of NKG7, which encodes the cytolytic granule membrane protein (Medley et al., 1996), CD94, and KLRD1, which is a driver gene in T and NK cells, along with GNLY, GZMK, XCL2, and CCL4, which are the most highly expressed genes, a common overall feature of tonsillar NK cells and ILC3s. Therefore, nmC4 was identified as an ILC3s / NK subset. nmC5 resembles ILC1s: high expression levels of T cell markers (CD3D, CD3G, and CD3E), as previously described (Bjorklund et al., 2016; Ercolano et al., 2020; Robinette et al., 2015), transcription factors that control ILCs development (IKZF3, BCL11B, PRDM1, and ID3), and NK / ILC1s cell function cytokines (GZMM, IFNG, IL32, CCL4, and CCL5) ( Figure 1 These cell annotations were further confirmed by the selective expression of markers known to be ILCs, such as IL7R, GATA3, NCR3, EOMES, TBX21, KIT, RORC, NCR1, NCR2, and KLRF1 ( Figure 7 A). The discrepancy between nmC5 and previously reported healthy intestinal ILC1s (Yudanin et al., 2019) may be due to the fact that the gating strategy used in this paper did not exclude CD5+ cells ( Figure 7 BC). Thus, the normal intestinal mucosa, as defined by the Lin-CD127+ scRNAseq profile, contains ILC1s, ILC3s, and ILCs / NKs, but no ILC2s, consistent with the lack of PTGDR2 gene expression ( Figure 7 A).

[0185] Example 4 Tumor ILC1s-like and ILC2s subpopulations exist in CRC patients.

[0186] The present invention investigated the composition and diversity of 15,101 ILCs (hereinafter referred to as intestinal TILCs) from CRC tumor patients. UMAP analysis identified four distinct cell populations: TILCs C0 to C3 ( Figure 2 A). In contrast to what is observed in normal mucosa, no large batch effect was observed and each cell population was present in all samples ( Figure 2 BC). According to the method applied to common mucosal cell populations ( Figure 1 ), assigning TILCs C0 to ILC3s, consistent with their overexpression of KIT, CXCL8, NFIL3, and IL4l1, as in nmC0-3. TILCs C1 resemble ILC1s and, like nmC5, show differential expression of genes encoding T cell molecule 8 (CD3D, CD3G)-secreted effectors (CCL4, IFNG) and ILCs-associated transcription factors (IKZF3, PRDM1, and BCL11B). Two other subpopulations, TILCs C2 and TILCs C3, are absent in normal mucosa. TILCs C2 cells correspond to another ILC1s subset (hereafter referred to as the TILC1s-like subset) characterized by enrichment of genes encoding inhibitory and co-stimulatory markers (TIGIT, CTLA4, TNFRSF18, and TNFRSF4). TILCs C3 cells (defined as ILC2s, hereinafter referred to as TILC2s) highly express genes encoding transcription factors required for ILC2s development (GATA3, RORA and ZBTB16) and ILC2s-responsive cytokine receptor genes (IL1RL1 and IL17RB) ( Figure 2 DH). The selective expression of known ILCs markers, such as IL7R, GATA3, NCR3, EOMES, TBX21, KIT, RORC, NCR1, NCR2, and KLRF1, supports these assignments ( Figure 8 ). In particular, PTGDR2 and higher levels of GATA3 expression were found in ILC2s ( Figure 8 Thus, like normal mucosal ILCs, intestinal TILCs form a heterogeneous subpopulation comprising four distinct subsets: TILCs C0 (similar to ILC3s), TILCs C1 (similar to ILC1s), TILCs C2 (a novel subpopulation similar to ILC1s), and TILCs C3 (similar to ILC2s).

[0187] Example 5 Batch effect correction of normal mucosal and tumor tissue ILC3s cell populations.

[0188] ILC3s in tumor tissue and ILC3s and ILC3s / NK cells in normal mucosa were further subdivided for downstream clustering. The "IntegrateData" function of the Seurat standard workflow was used to correct for batch effects based on the detected anchor points.

[0189] Tumor tissue ILC3s heterogeneity appears to be less than that of normal mucosa. Therefore, the present invention focuses on nmC0-3, nmC4, and TILCs C0. After applying batch effect correction, the same analytical process as above was used to compare the heterogeneity of ILC3s between normal mucosal ILCs and intestinal TILCs ( Figure 9 Four distinct subpopulations were found in ILC3s from both tissues ( Figure 9 AI), including a subpopulation that may be immature SELL-expressing, and a subpopulation enriched in HLA-encoded transcripts also present in human tonsils ( Figure 9 E and J). Each subpopulation of normal mucosal ILC3s has a corresponding cell population in tumor tissue ( Figure 9 K). Given the overlap in heterogeneity between normal mucosal and intestinal TILCs in ILC3s, we can conclude that CRC does not affect the heterogeneity of ILC3s subsets. Therefore, intestinal TILCs are distinguished from nmILCs by the presence of a subpopulation of TILC2s and a second subpopulation similar to TILC1s.

[0190] Example 6 Heterogeneity of blood ILCs is stable in CRC.

[0191] The present invention investigates the differences between ILCs in the blood of healthy individuals and patients with CRC to identify potential biomarkers for this disease. UMAP analysis of 19,603 ILCs from healthy donors revealed three distinct cell populations (hereinafter referred to as nbC0, nbC1, and nbC2) ( Figure 10 AC). Based on the upregulation of CD3D, CD3E, CD3G, NK / ILC1s cell effector proteins (CCL5, GZMK, GZMM, and GZMA), and ILCs transcription factors (BCL11B, PRDM1, and IKZF3), nbC0 was considered to correspond to ILC1s ( Figure 10 DH). nbC1 was identified as ILC3s, characterized by the expression of ILC3s transcription factors (MAFF, RUNX3) and co-stimulatory markers (TNFRSF4, TNFRSF18). nbC2 showed upregulation of ILC2s characteristic (GATA3, RORA) genes and genes encoding regulatory receptors (KLRB1, KLRG1) ( Figure 10DH). These assignments were supported by the selective expression of markers of known ILCs, such as IL7R, GATA3, NCR3, EOMES, TBX21, PTGDR2, KIT, RORC, NCR1, and KLRF1 ( Figure 10 I).

[0192] Example 7 scRNAseq defines ILC1s, ILC2s, and ILC3s in the blood of CRC patients.

[0193] 1. Materials and Methods

[0194] (1.1) RNA velocity evaluation

[0195] We analyzed the dynamics of gene expression in single-cell sequencing data by analyzing the presence of spliced ​​and unspliced ​​RNA transcripts in the bam files to assess RNA velocity. RNA velocity values ​​for each gene and embedded RNA velocity parameters were calculated using the R package veloycyto.R (La Manno et al., 2018) (https: / / github.com / velocyto-team / velocyto.R). RNA velocities were projected onto the UMAP plot using Gaussian smoothing on a regular grid.

[0196] 2. Experimental results

[0197] This example further identified three subpopulations (hereinafter referred to as cbC0, cbC1 and cbC2) based on the UMAP profiles of 6,899 blood ILCs from CRC donors ( Figure 11 AC). Driver genes, top ten genes, and module score features highlight the similarities between cbILCs and nbILCs ( Figure 11 DH). cbC0, like nbC1, had an ILC3s profile, enriched for MAFF, RUNX3, TNFRSF18, NCR1, and KIT. cbC1 was identified as an ILC2s because, like nbC2, it expressed high levels of GATA3, RORA, KLRB1, KLRG1, and PTGDR2. cbILC2s, like nbC0, were also enriched for genes that are hallmarks of ILC1s: CD3D, CD3G, CD3E, CCL5, GZMK, GZMM, GZMA, BCL11B, PRDM1, IKZF3, and TBX21 ( Figure 11 DH). Selective expression of IL7R, GATA3, NCR3, EOMES, TBX21, PTGDR2, KIT, RORC, NCR1, and KLRF1 also supports these assignments ( Figure 11I). However, despite the similarity between the cbILCs subpopulation and the nbILCs subpopulation, RNA velocity analysis predicted that ILC1s may transform into ILC3s only in the context of CRC tumor blood ( Figure 11 J) In conclusion, blood ILCs from healthy donors and CRC patients formed heterogeneous subpopulations containing ILC1s, ILC2s, and ILC3s.

[0198] Example 8 Identification of a new subpopulation of CRC-specific TILC1s

[0199] Tumor tissue ILCs contain two subpopulations that are not present in normal mucosal ILCs, and their transcriptome characteristics are similar to those of ILC2s and ILC1s ( Figure 1 A and H, Figure 2 A and H). The present invention investigated the correlation between these two tumor tissue-specific cell populations and ILCs subsets from healthy blood and CRC patient blood by grouping 41,603 ILCs into a single global analysis. This analysis revealed organ-specific signatures between ILCs, with overlap between the two blood samples (healthy blood and CRC patient blood, as mentioned above) ( Figure 3 A), while TILCs clustered separately. nbILCs and cbILCs had high similarity in gene signatures, which is consistent with the finding that TILCs ( Figure 3 The present invention further analyzed the relationship between defined ILCs subpopulations from CRC tissue, normal blood, and CRC blood samples. Although this subpopulation has the core ILC1s transcriptional signature, the TILC1s-like subpopulation appears to be separated from other cell populations (especially TILC1s). Figure 3 D). Similarly, another specific subset of TILCs, TILC2s, aggregates with other TILCs and ILC2s in the blood. In the blood, each nbILC aggregates with its corresponding cbILC subset ( Figure 3 D). The present invention creates a Venn diagram ( Figure 3 E and F) Comparison of their whole transcriptome signatures to assess whether tumor-specific ILCs share more genes with their counterparts in normal or CRC blood. TILC1s-like subpopulations and cbILC1s (57 genes) share more genes than nbILC1s (34 genes). Figure 3 E). TILC2s shared a similar number of genes with cbILC2s and nbILC2s, 39 and 34, respectively ( Figure 3 F).

[0200] Example 9 ILCs-specific markers in CRC

[0201] The present invention seeks tumor-specific tissue characteristics of ILCs by clustering 31,246 ILCs from normal mucosa and tumor tissues. The two tissues share some common ILC subpopulations, but UMAP highlights the transitions between the two tissues, indicating differences at the transcriptional level ( Figure 4 A). Unsupervised hierarchical clustering also showed that tissue-of-origin features were stronger than ILCs subpopulation marker features ( Figure 4 BC). UMAP analysis of 26,502 ILCs cells showed that nbILCs and cbILCs had similar separation patterns ( Figure 4 D); Based on unsupervised hierarchical cluster analysis, the two subpopulations derived from blood samples were classified according to health status, revealing the differences in transcriptomes between them ( Figure 4 EF). The results showed that compared with the control group, one gene (AQP3) was upregulated in normal blood and normal mucosa, and four genes (SLAMF1, HPGD, TLE4 and PRDM1) were upregulated in CRC blood and intestinal TILCs ( Figure 4 G). The characteristic plots of these five target genes confirmed the characteristic upregulation of SLAMF1, HPGD, TLE4, and PRDM1, and downregulation of AQP3 in intestinal TILCs ( Figure 4 H and 11). SLAMF1 (signaling lymphocyte activation molecule family member 1 or CD150), encoding a protein involved in the activation of T cells, B cells, and NK cells (Gordiienko et al., 2019), is the main surface protein gene upregulated in tumors. This gene is expressed in cbILC2s, TILC2s, and TILC1s-like subpopulations, but is only weakly expressed in 12 healthy controls ( Figure 4 H), indicating that SLAMF1 expression on the cell surface of ILCs can distinguish healthy individuals from CRC patients.

[0202] Example 10 SLAMF1 is a biomarker for CRC

[0203] 1. Materials and Methods

[0204] (1.1) Flow cytometry detection of ILCs

[0205] As with the staining of sorted ILCs, freshly prepared cells were stained with a viability dye and blocked with an Fc antibody before surface antibody staining. Surface antibody staining was performed using an antibody cocktail (anti-lineage antibody cocktail, including CD45, CD127, CD117, CRTH2, CD5, TIGIT, and SLAMF1) for 30 minutes at room temperature. PBMCs derived from CRC patient blood, adjacent paracancerous tissue, and tumor samples were also stained, while PBMCs from healthy donors served as controls. Stained cells were incubated at 4°C and analyzed by flow cytometry on a BD Symphony instrument. Flow cytometry data were analyzed using FlowJo software. Statistical analysis was performed using the nonparametric unpaired t-test with the Mann-Whitney rank sum test or the Kruskal-Walis test with Dohn's multiple comparisons. P values ​​were corrected for multiple comparisons using the Benjamini-Hochberg method. *p-value < 0.05, **p-value < 0.01, ***p-value < 0.001, ****p-value < 0.0001.

[0206] (1.2) TCGA analysis

[0207] RNAseq data from primary tumors and clinical annotations were downloaded using the TCGAbiolinks package. Kaplan-Meier curves were constructed using the R package survminer. To stratify expression levels into two groups, the expression level that gave the lowest p-value was used as the cutoff. The optimal cutoff for patient stratification was determined using the Cox proportional hazards model, and the p-values ​​shown in the figures were calculated using the log-rank test.

[0208] 2. Experimental results

[0209] The present invention confirmed by flow cytometry that the expansion of ILC1s subpopulation in CRC patient tumor tissues was at the expense of the reduction of ILC3s subpopulation compared with normal adjacent cancer tissues ( Figure 5 AB).

[0210] The present invention also observed the presence of a novel TIGIT+TILC1s-like cell and TILC2s subpopulation in tumors, but not in normal tissues ( Figure 5 AB), and scRNAseq data ( Figure 1 Patient data from The Cancer Genome Atlas (TCGA) showed that high expression of IL33, a cytokine that activates ILC2s, in tumors was associated with longer survival time in CRC patients, suggesting that TILC2s may indicate a good prognosis for CRC patients ( Figure 5 C). In contrast to the findings for intestinal ILCs, the frequencies of each ILC subset relative to total ILCs were similar in the blood of CRC patients and healthy donors ( Figure 5 D).

[0211] The number of ILCs expressing the SLAMF1 surface molecule in tumors was greater than in adjacent tissues, which in fact expressed almost no SLAMF1. Figure 5 E and F). In contrast, blood ILCs from healthy donors expressed SLAMF1, and a high proportion of ILCs expressing SLAMF1 was also found in the blood of CRC patients ( Figure 5 G). By studying the clinical outcomes of cancer patients, the present invention further explored the potential role of SLAMF1 in the development and progression of CRC disease. The survival rate of rectal cancer patients with high SLAMF1 expression was much higher than that of patients with low SLAMF1 expression ( Figure 5 H), strongly suggesting that SLAMF1 is an anti-tumor biomarker for CRC.

[0212] discuss

[0213] Over the past decade, helper ILCs have emerged as key for pathogen protection, tissue remodeling, and maintenance of homeostasis (Vivier et al., 2018). However, the contribution of helper ILCs to cancer is poorly understood, as they may promote tumor-associated inflammation or, conversely, exhibit antitumor properties, depending on the tumor microenvironment.

[0214] In this study, we investigated the heterogeneity of helper ILCs in the human intestine by constructing single-cell transcriptional profiles of Lin-CD127+ cells from healthy individuals and patients with CRC. This unbiased characterization of helper ILCs differs from a recent study that analyzed the transcription factors of intestinal ILCs, in which these cells were flow cytometrically sorted for CD103, CD300LF, and CD196 cell surface markers before transcriptome profiling (Cella et al., 2019). This study further revealed that the healthy intestine contains ILC1s, ILC3s, and an ILC3s / NK population, but no ILC2s. Interestingly, ILC2s are almost completely absent from healthy human tissues, with the exception of lung and adipose tissue (Trabanelli et al., 2018), in contrast to previously reported mouse studies. We detected tumor-infiltrating TILC2s in CRC patients. Similar to existing literature, TILC2s have also been observed in the urine of patients with breast cancer (Salimi et al., 2018), gastric cancer (Salimi et al., 2018), pancreatic cancer (Moralet al., 2020), and bladder cancer (Chevalier et al., 2017). Previously reported data have shown that in one model, ILC2s infiltrate tumors through an IL-33-dependent pathway (Chevalier et al., 2017; Moral et al., 2020; Saranchova et al., 2016) and mediate tumor immune surveillance by promoting cytolytic CD8+ T cell responses. IL-33 is overexpressed in colorectal tumors (Cui et al., 2015), and high levels of IL-33 are frequently observed in low-grade adenocarcinomas and early colorectal tumors (Mertz et al., 2016). Patients with colorectal cancer who express high levels of IL-33 have a better survival rate than those who express low levels of IL-33 ( Figure 5 C), suggesting that TILC2s may predict a good prognosis in CRC. Therefore, further investigation of the anti-tumor immune role of TILC2s in CRC and other cancer indications is highly warranted.

[0215] The present study discovered an additional subset of helper-like ILC1s, designated TILC1s-like TIGIT+ (TILCs-likeTIGIT+), which is present only in tumors but not in the blood of CRC patients. The transcriptional profile of TILC1s-like TIGIT+ is more similar to the ILC1s gene signature than that of any other ILCs, but they are separated from TILC1s, indicating that they are clearly distinct from "conventional" intestinal TILC1s. ILC1s-like cells, termed "intermediate ILC1s" (inTILC1s), have also been described in mouse models of methacholine (MCA)-induced tumors and experimental RM-1 and B16F10 lung metastases (Gao et al., 2017). In humans, CD56-CD16+ ILC1s-like cells have been found in the peritoneum and pleural effusions of patients with solid tumors and cancer (Levi et al., 2015), and the cytotoxic function of these cells has been altered in the peripheral blood of donors with acute myeloid leukemia (Salome et al., 2019). Intratumoral intiLC1s may arise from NK cell differentiation driven by TGF-β signaling, a phenomenon known as ILC plasticity (Cortez et al., 2016; Gao et al., 2017). Studies in humanized mice have demonstrated that TGF-β signaling induces the conversion of ILC3s to ILC1s, and a transitional ILC3s-ILC1s subpopulation has been identified in the human intestine (Cella et al., 2019). No such phenomenon was observed in the intestinal ILCs dataset of the present invention, and none of the algorithms tested could determine the correlation between TILC1s-like TIGIT+ and another subset of intestinal ILCs that reflects possible differentiation (data not shown). Therefore, the mechanism by which TILC1s-like TIGIT+ appears in CRC tumors remains to be determined. The present invention observed that ILC1s in the blood of CRC patients were plastic to ILC3s, but not in healthy donors, suggesting that there may be soluble signals that drive ILC1ss-ILC3ss plasticity, such as sustained IL-23 levels (Koh et al., 2019). The biological relevance of this ILC1s-ILCs plasticity in the blood of CRC patients is unclear.

[0216] In TILC1s and ILC1s produce large amounts of TNF-α, which has been found to be ineffective in controlling carcinogens and may even promote tumor metastasis in mouse models (Gao et al., 2017). In humans, CD56+CD16- ILC1s-like cells express the pro-angiogenic factor VEGF, which may also promote tumor growth (Levi et al., 2015). In CRC patients, the proportion of ILC1s in tumor tissue is higher than in normal mucosal tissue, and with tumor progression, ILC1s increase at the expense of ILC3s (Ikeda et al., 2020). These results suggest that high levels of ILC1s may predict poor cancer prognosis. The specific biological functions of the TILC1s-like subpopulation relative to classical TILC1s in CRC tumors remain to be further explored, as TILC1s-like cells have high levels of PD-1 and TIGIT and may be unleashed by anti-PD-1 and anti-TIGIT immunotherapy.

[0217] The present study characterized three subsets (ILC3s, ILC3s / NK cells, and ILC1s) in the normal mucosa of all donors. Donor-specific effects were observed in the ILC3s subset, suggesting that the microbiome may be imprinted by ILC3s. Interestingly, CRC tumors had much lower levels of ILC3s and exhibited a significant loss of this donor-specificity. CRC is frequently associated with tumor immune dysregulation, which involves profound changes in microbiome composition (Feng et al., 2015; Liang et al., 2017; Nakatsu et al., 2015; Yazici et al., 2017; Yu et al., 2017). ILC3s are key regulators of intestinal barrier integrity and immune homeostasis. Therefore, promoting the recolonization and diversification of ILC3s in CRC patients may be therapeutically beneficial. Increasing microbial diversity may also increase LC3 heterogeneity.

[0218] The present invention also defines a subset of ILC3s / NK cells in the healthy intestinal mucosa, which are absent in CRC patient tumor tissue. These cells have the same transcriptional functions as ILC3s and NK cells. They differ from ILC3s primarily in the expression of NKG7, KLRD1 (CD94), GNLY, GZMK, XCL2, and CCL4. The biological role of this ILC3s / NK subset and its relationship with "classical" ILC3s remain to be studied.

[0219] SLAMF1 is the only cell surface marker expressed at higher levels in TILCs and blood ILCs from CRC patients. Compared with healthy donors, ILCs expressing SLAMF1 on their surface are more frequent in tumors and blood from CRC patients. SLAMF1 is a single-chain type I transmembrane receptor with two immunoreceptor tyrosine-based switch motifs (ITSMs) in its cytoplasmic tail (Gordiienko et al., 2019). SLAMF1 is an autoligand and a microbial receptor for measles virus. It also functions as a bacterial receptor involved in the elimination of Gram-negative bacteria (Gordiienko et al., 2019). SLAMF1 is expressed by nearly all hematopoietic cells except natural killer (NK) cells, particularly those with an activated phenotype, and is upregulated upon cell activation. In the blood, SLAMF1 is expressed stably on the surface of most ILCs, but this expression is not observed in ILCs from the normal intestinal mucosa. In contrast, SLAMF1 is expressed in TILCs from CRC patients, suggesting that TILCs in the tumor bed are more active than those in the adjacent tissue mucosa. However, the biological impact of SLAMF1 expression on the surface of helper ILCs on these cells remains to be investigated. High levels of SLAMF1 are associated with improved survival in CRC patients. Therefore, the results of the present invention indicate that SLAMF1 is an anti-tumor biomarker for CRC.

[0220] ILCs can control various aspects of immunotherapy and have become a tissue-specific regulatory target for cancer immunity. Because ILCs and T cells coexist in human cancers and share common stimulation and inhibition pathways, immunotherapy strategies targeting anti-cancer ILCs may be as important as those targeting T cells. The results of the present invention show that there are TILCs subpopulations in CRC, so ILCs are part of the tumor microenvironment. It can be speculated that they can regulate the immunity of the tumor bed or have a direct effect on tumor cells. Whether there are more tumor-specific subpopulations of ILCs in cancers of different tissue origins and different activation states, and whether they have a promoting or inhibitory effect on cancer, remains to be further studied.

[0221] Gene annotation

[0222] Transcription factor-related genes were identified based on four transcription factor-related databases: JASPAR (Khan et al., 2018)

[0223] (http: / / jaspar.genereg.net / ),DBD (Wilson et al., 2008)

[0224] (http: / / www.transcriptionfactor.org / ), AnimalTFDB (Hu et al.,

[0225] (http: / / bioinfo.life.hust.edu.cn / AnimalTFDB / ), and TF2DNA (Pujato et al.

[0226] ( http: / / www.fiserlab.org / tf2dna_db / ). Cell membrane surface and secretory protein genes were mapped according to The Human Protein Atlas (Uhlen et al., 2015) ( https: / / www.proteinatlas.org / humanproteome / tissue / secretome ) to make comments.

[0227] References:

[0228] Aran, D., Looney, AP, Liu, L., Wu, E., Fong, V., Hsu, A., Chak, S., Naikawadi, RP, Wolters, PJ, Abate, AR, et al. (2019). Reference-based analysis of lung single-cell sequencing reveals a transitional profibrotic macrophage. Nat Immunol 20, 163-172.

[0229] Bumeister, SH, Freeman, GJ, Dranoff, G., and Sharpe, AH (2016). Coinhibitory Pathways in Immunotherapy for Cancer. Annu Rev Immunol 34, 539-573.

[0230] Bie, Q., Zhang, P., Su, Z., Zheng, D., Ying,

[0231] Bjorklund, A.K., Forkel, M., Picelli, S., Konya, V., Theorell, J., Friberg, D., Sandberg, R., and Mjosberg, J. (2016).The heterogeneity of human CD127(+) innatelymphoid cells revealed by single-cell RNA sequencing. Nat Immunol 17,451-460.

[0232] Bray, F., Ferlay, J., Soerjomataram, I., Siegel, R.L., Torre, L.A., and Jemal, A. (2018).

[0233] Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin 68,394-424.

[0234] Carrega, P., Loiacono, F., Di Carlo, E., Scaramuccia, A., Mora, M., Conte, R., Benelli, R., Spaggiari, G. M., Cantoni, C., Campana, S., et al. (2015). NCR(+)ILC3sconcentrate in human lung cancer and associate with intratumoral lymphoidstructures. Nat Commun 6,8280.

[0235] Carrega, P., Orecchia, P., Quatrini, L., Tumino, N., Vene, R., Benelli, R., Poggi, A., Scabini, S., Mingari, MC, Moretta, L., and Vacca, P. (2020). Characterisation of innate lymphoid cell subsets infiltrating colorectal carcinoma. Gut.

[0236] Castro,F.,Cardoso,A.P.,Goncalves,R.M.,Serre,K.,and Oliveira,M.J.(2018).Interferon-Gamma at the Crossroads ofTumor Immune Surveillance orEvasion.Front Immunol 9,847.

[0237] Cella,M.,Gamini,R.,Secca,C.,Collins,P.L.,Zhao,S.,Peng,V.,Robinette,M.L.,Schettini,J.,Zaitsev,K.,Gordon,W.,et al.(2019).Subsets ofILC3s-ILC1s-like cells generate a diversity spectrum ofinnate lymphoid cells in humanmucosal tissues.Nat Immunol 20,980-991.

[0238] Chan,I.H.,Jain,R.,Tessmer,M.S.,Gorman,D.,Mangadu,R.,Sathe,M.,Vives,F.,Moon,C.,Penaflor,E.,Turner,S.,et al.(2014).Interleukin-23 is sufficient toinduce rapid de novo gut tumorigenesis,independent ofcarcinogens,throughactivation ofinnate lymphoid cells.Mucosal Immunol 7,842-856.

[0239] Chen,D.S.,and Mellman,I.(2017).Elements ofcancer immunity and thecancer-immune set point.Nature 541,321-330.

[0240] Chevalier,M.F.,Trabanelli,S.,Racle,J.,Salome,B.,Cesson,V.,Gharbi,D.,Bohner,P.,Domingos-Pereira,S.,Dartiguenave,F.,Fritschi,A.S.,et al.(2017).ILC2s-modulated T cell-to-MDSC balance is associated with bladder cancerrecurrence.J Clin Invest 127,2916-2929.

[0241] Chiossone,L.,Dumas,P.Y.,Vienne,M.,and Vivier,E.(2018).Natural killercells and other innate lymphoid cells in cancer.Nat Rev Immunol 18,671-688.

[0242] Cortez,V.S.,Cervantes-Barragan,L.,Robinette,M.L.,Bando,J.K.,Wang,Y.,Geiger,T.L.,Gilfillan,S.,Fuchs,A.,Vivier,E.,Sun,J.C.,et al.(2016).Transforming Growth Factor-beta Signaling Guides the DifferentiationofInnate Lymphoid Cells in Salivary Glands.Immunity44,1127-1139.

[0243] Cui,G.,Qi,H.,Gundersen,M.D.,Yang,H.,Christiansen,I.,Sorbye,S.W.,Goll,R.,and Florholmen,J.(2015).Dynamics ofthe IL-33 / ST2 network in theprogression of human colorectal adenoma to sporadic colorectal cancer.CancerImmunol Immunother 64,181-190.Eisenring,M.,vom Berg,J.,Kristiansen,G.,Saller,E.,and Becher,B.(2010).IL-12 initiates tumor rejection via lymphoid tissue-inducer cells bearing the natural cytotoxicity receptor NKp46.Nat Immunol 11,1030-1038.

[0244] Ercolano,G.,Wyss,T.,Salome,B.,Romero,P.,Trabanelli,S.,and Jandus,C.(2020).

[0245] Distinct and shared gene expression for human innate versus adaptivehelper lymphoid cells.J Leukoc Biol.

[0246] Feng,Q.,Liang,S.,Jia,H.,Stadlmayr,A.,Tang,L.,Lan,Z.,Zhang,D.,Xia,H.,Xu,X.,Jie,Z.,et al.(2015).Gut microbiome development along the colorectaladenoma-carcinoma sequence.Nat Commun 6,6528.

[0247] Fuchs,A.,Vermi,W.,Lee,J.S.,Lonardi,S.,Gilfillan,S.,Newberry,R.D.,Cella,M.,and Colonna,M.(2013).Intraepithelial type 1 innate lymphoid cellsare a unique subset ofIL-12-and IL-15-responsive IFN-gamma-producingcells.Immunity 38,769-781.

[0248] Gao,Y.,Souza-Fonseca-Guimaraes,F.,Bald,T.,Ng,S.S.,Young,A.,Ngiow,S.F.,Rautela,J.,Straube,J.,Waddell,N.,Blake,S.J.,et al.(2017).Tumorimmunoevasion by the conversion ofeffector NK cells into type 1 innatelymphoid cells.Nat Immunol 18,1004-1015.

[0249] Gasteiger,G.,Fan,X.,Dikiy,S.,Lee,S.Y.,and Rudensky,A.Y.(2015).Tissueresidency of innate lymphoid cells in lymphoid and nonlymphoid organs.Science350,981-985.

[0250] Gordiienko,I.,Shlapatska,L.,Kovalevska,L.,and Sidorenko,S.P.(2019).SLAMF1 / CD150in hematologic malignancies:Silent marker or active player?ClinImmunol 204,14-22.

[0251] Hu,H.,Miao,Y.R.,Jia,L.H.,Yu,Q.Y.,Zhang,Q.,and Guo,A.Y.(2019).AnimalTFDB 3.0:a comprehensive resource for annotation and predictionofanimal transcription factors.

[0252] Nucleic Acids Res 47,D33-D38.

[0253] Ikeda,A.,Ogino,T.,Kayama,H.,Okuzaki,D.,Nishimura,J.,Fujino,S.,Miyoshi,N.,Takahashi,H.,Uemura,M.,Matsuda,C.,et al.(2020).Human NKp44+group 3innate lymphoid cells associate with tumor-associated tertiary lymphoidstructures in colorectal cancer.Cancer Immunol Res.

[0254] Khan,A.,Fornes,O.,Stigliani,A.,Gheorghe,M.,Castro-Mondragon,J.A.,vander Lee,R.,Bessy,A.,Cheneby,J.,Kulkarni,S.R.,Tan,G.,et al.(2018).JASPAR 2018:update of the open-access database oftranscription factor binding profilesand its web framework.Nucleic Acids Res 46,D260-D266.

[0255] Kirchberger,S.,Royston,D.J.,Boulard,O.,Thornton,E.,Franchini,F.,Szabady,R.L.,Harrison,O.,and Powrie,F.(2013).Innate lymphoid cells sustaincolon cancer through production ofinterleukin-22 in a mouse model.J Exp Med210,917-931.

[0256] Koh,J.,Kim,H.Y.,Lee,Y.,Park,I.K.,Kang,C.H.,Kim,Y.T.,Kim,J.E.,Choi,M.,Lee,W.W.,Jeon,Y.K.,and Chung,D.H.(2019).IL23-Producing Human Lung CancerCells Promote Tumor Growth via Conversion ofInnate Lymphoid Cell 1(ILC1s)intoILC3s.Clin Cancer Res 25,4026-4037.

[0257] La Manno,G.,Soldatov,R.,Zeisel,A.,Braun,E.,Hochgerner,H.,Petukhov,V.,Lidschreiber,K.,Kastriti,M.E.,Lonnerberg,P.,Furlan,A.,et al.(2018).RNAvelocity of single cells.Nature 560,494-498.

[0258] Levi,I.,Amsalem,H.,Nissan,A.,Darash-Yahana,M.,Peretz,T.,Mandelboim,O.,and Rachmilewitz,J.(2015).Characterization oftumor infiltrating naturalkiller cell subset.

[0259] Oncotarget 6,13835-13843.

[0260] Liang, Q., Chiu, J., Chen, Y., Huang, Y., Higashimori, A., Fang, J., Brim, H., Ashktorab, H., Ng, SC, Ng, SSM, et al 23,2061–2070.

[0261] ControlofMetastasis by NK Cells.Cancer Cell 32,135-154.

[0262] Characterization ofGMP-17, a granule membrane proteinthat moves to the plasma membrane ofnatural killer cells following targetcell recognition.Proc Natl Acad Sci USA 93,685-689.

[0263] Mertz, KD, Mager, LF, Wasmer, MH, Thiesler, T., Koelzer, VH, Ruzzante, G., Joller, S., Murdoch, JR, Brummendorf, T., Genitsch, V., et al 5,e1062966.

[0264] Moral,J.A.,Leung,J.,Rojas,L.A.,Ruan,J.,Zhao,J.,Sethna,Z.,Ramnarain,A.,Gasmi,B.,Gururajan,M.,Redmond,D.,et al.(2020).ILC2s amplify PD-1 blockadeby activating tissuespecific cancer immunity.Nature 579,130-135.

[0265] Nakatsu,G.,Li,X.,Zhou,H.,Sheng,J.,Wong,S.H.,Wu,W.K.,Ng,S.C.,Tsoi,H.,Dong,Y.,Zhang,N.,et al.(2015).Gut mucosal microbiome across stagesofcolorectal carcinogenesis.Nat Commun 6,8727.

[0266] Nussbaum,K.,Burkhard,S.H.,Ohs,I.,Mair,F.,Klose,C.S.N.,Arnold,S.J.,Diefenbach,A.,Tugues,S.,and Becher,B.(2017).Tissue microenvironment dictatesthe fate and tumorsuppressive function oftype 3 ILCs.J Exp Med 214,2331-2347.

[0267] Okazaki,T.,Chikuma,S.,Iwai,Y.,Fagarasan,S.,and Honjo,T.(2013).Arheostat for immune responses:the unique properties ofPD-1 and theiradvantages for clinical application.Nat Immunol 14,1212-1218.

[0268] Okazaki,T.,and Honjo,T.(2007).PD-1 and PD-1 ligands:from discovery toclinical application.Int Immunol 19,813-824.

[0269] Pujato,M.,Kieken,F.,Skiles,A.A.,Tapinos,N.,and Fiser,A.(2014).Prediction of DNA binding motifs from 3D models oftranscription factors;identifying TLX3 regulated genes.

[0270] Nucleic Acids Res 42,13500-13512.

[0271] Robinette,M.L.,Fuchs,A.,Cortez,V.S.,Lee,J.S.,Wang,Y.,Durum,S.K.,Gilfillan,S.,Colonna,M.,and Immunological Genome,C.(2015).Transcriptionalprograms define molecular characteristics ofinnate lymphoid cell classes andsubsets.Nat Immunol 16,306-317.

[0272] Salimi,M.,Wang,R.,Yao,X.,Li,X.,Wang,X.,Hu,Y.,Chang,X.,Fan,P.,Dong,T.,and Ogg,G.(2018).Activated innate lymphoid cell populations accumulate inhuman tumour tissues.BMC Cancer 18,341.

[0273] Salome,B.,Gomez-Cadena,A.,Loyon,R.,Suffiotti,M.,Salvestrini,V.,Wyss,T.,Vanoni,G.,Ruan,D.F.,Rossi,M.,Tozzo,A.,et al.(2019).CD56 as a marker ofanILC1s-like population with NK cell properties that is functionally impairedin AML.Blood Adv 3,3674-3687.

[0274] Saranchova,I.,Han,J.,Huang,H.,Fenninger,F.,Choi,K.B.,Munro,L.,Pfeifer,C.,Welch,I.,Wyatt,A.W.,Fazli,L.,et al.(2016).Discovery ofa MetastaticImmune Escape Mechanism Initiated by the Loss ofExpression ofthe TumourBiomarker Interleukin-33.SciRep 6,30555.

[0275] Schumacher,T.N.,and Schreiber,R.D.(2015).Neoantigens in cancerimmunotherapy.Science 348,69-74.

[0276] Sharma,P.,and Allison,J.P.(2015).Immune checkpoint targeting incancer therapy:toward combination strategies with curative potential.Cell161,205-214.

[0277] Simoni,Y.,Fehlings,M.,Kloverpris,H.N.,McGovern,N.,Koo,S.L.,Loh,C.Y.,Lim,S.,Kurioka,A.,Fergusson,J.R.,Tang,C.L.,et al.(2017).Human Innate LymphoidCell Subsets Possess Tissue-Type Based Heterogeneity in Phenotype andFrequency.Immunity 46,148-161.

[0278] Spits,H.,Artis,D.,Colonna,M.,Diefenbach,A.,Di Santo,J.P.,Eberl,G.,Koyasu,S.,Locksley,R.M.,McKenzie,A.N.,Mebius,R.E.,et al.(2013).Innatelymphoid cells--a proposal for uniform nomenclature.Nat Rev Immunol 13,145-149.

[0279] Stuart,T.,Butler,A.,Hoffman,P.,Hafemeister,C.,Papalexi,E.,Mauck,W.M.,3rd,Hao,Y.,Stoeckius,M.,Smibert,P.,and Satija,R.(2019).ComprehensiveIntegration ofSingle-Cell Data.Cell 177,1888-1902 e1821.

[0280] Trabanelli,S.,Chevalier,M.F.,Martinez-Usatorre,A.,Gomez-Cadena,A.,Salome,B.,Lecciso,M.,Salvestrini,V.,Verdeil,G.,Racle,J.,Papayannidis,C.,etal.(2017).Tumour-derived PGD2 and NKp30-B7H6 engagement drives animmunosuppressive ILC2s-MDSC axis.Nat Commun 8,593.

[0281] Trabanelli,S.,Gomez-Cadena,A.,Salome,B.,Michaud,K.,Mavilio,D.,Landis,BN,Jandus,P.,and Jandus,C.(2018).Human innate lymphoid cells(ILCs):Toward auniform immunephenotyping.Cytometry B Clin Cytom 94,392-399.

[0282] Uhlen,M.,Fagerberg,L.,Hallstrom,BM,Lindskog,C.,Oksvold,P.,Mardinoglu,A.,Sivertsson,A.,Kampf,C.,Sjostedt,E.,Asplund,A.,et al.(2015).Proteomics.Tissue-based map of the human proteome.Science 347,1260419.

[0283] Victor,AR,Nalin,AP,Dong,W.,McClory,S.,Wei,M.,Mao,C.,Kladney,RD,Youssef,Y.,Chan,WK,Briercheck,EL,et al.(2017).IL-18 Drives ILC3sProliferation and Promotes IL-22 Production viaNF-kappaB.J Immunol 199,2333-2342.

[0284] Vivier, E., Artis, D., Colonna, M., Diefenbach, A., Di Santo, J.P., Eberl, G., Koyasu, S., Locksley, R.M., McKenzie, A.N.J., Mebius, R.E., et al. (2018). Innate Lymphoid Cells: 10 Years On. Cell 174, 1054-1066.

[0285] Wilson,D.,Charoensawan,V.,Kummerfeld,S.K.,and Teichmann,S.A.(2008).DBD--taxonomically broad transcription factor predictions:new content andfunctionality.Nucleic Acids Res 36,D88-92.

[0286] Yazici,C.,Wolf,P.G.,Kim,H.,Cross,T.L.,Vermillion,K.,Carroll,T.,Augustus,G.J.,Mutlu,E.,Tussing-Humphreys,L.,Braunschweig,C.,et al.(2017).Race-dependent association ofsulfidogenic bacteria with colorectalcancer.Gut 66,1983-1994.

[0287] Yu,J.,Feng,Q.,Wong,S.H.,Zhang,D.,Liang,Q.Y.,Qin,Y.,Tang,L.,Zhao,H.,Stenvang,J.,Li,Y.,et al.(2017).Metagenomic analysis offaecal microbiome as atool towards targeted noninvasive biomarkers for colorectal cancer.Gut 66,70-78.

[0288] Yudanin,N.A.,Schmitz,F.,Flamar,A.L.,Thome,J.J.C.,Tait Wojno,E.,Moeller,J.B.,Schirmer,M.,Latorre,I.J.,Xavier,R.J.,Farber,D.L.,et al.(2019).Spatial and Temporal Mapping ofHuman Innate Lymphoid Cells Reveals ElementsofTissue Specificity.

[0289] Immunity 50,505-519 e504.

[0290] Zaidi,MR(2019).The Interferon-Gamma Paradox in Cancer.J InterferonCytokine Res 39,30-38. SEQUENCE LISTING <110> Shanghai Institute of Immunology, Shanghai Jiao Tong University School of Medicine, French National Institute of Health and Medical Research, French National Center for Scientific Research, Aix-Marseille University <120> Colorectal cancer biomarkers and their applications in diagnosis, prevention, treatment and prognosis <160> 1 <170> PatentIn version 3.3 <210> 1 <211> 335 <212> PRT <213> Artificial sequence <400> 1 Met Asp Pro Lys Gly Leu Leu Ser Leu Thr Phe Val Leu Phe Leu Ser 1 5 10 15 Leu Ala Phe Gly Ala Ser Tyr Gly Thr Gly Gly Arg Met Met Asn Cys 20 25 30 Pro Lys Ile Leu Arg Gln Leu Gly Ser Lys Val Leu Leu Pro Leu Thr 35 40 45 Tyr Glu Arg Ile Asn Lys Ser Met Asn Lys Ser Ile His Ile Val Val 50 55 60 Thr Met Ala Lys Ser Leu Glu Asn Ser Val Glu Asn Lys Ile Val Ser 65 70 75 80 Leu Asp Pro Ser Glu Ala Gly Pro Pro Arg Tyr Leu Gly Asp Arg Tyr 85 90 95 Lys Phe Tyr Leu Glu Asn Leu Thr Leu Gly Ile Arg Glu Ser Arg Lys 100 105 110 Glu Asp Glu Gly Trp Tyr Leu Met Thr Leu Glu Lys Asn Val Ser Val 115 120 125 Gln Arg Phe Cys Leu Gln Leu Arg Leu Tyr Glu Gln Val Ser Thr Pro 130 135 140 Glu Ile Lys Val Leu Asn Lys Thr Gln Glu Asn Gly Thr Cys Thr Leu 145 150 155 160 Ile Leu Gly Cys Thr Val Glu Lys Gly Asp His Val Ala Tyr Ser Trp 165 170 175 Ser Glu Lys Ala Gly Thr His Pro Leu Asn Pro Ala Asn Ser Ser His 180 185 190 Leu Leu Ser Leu Thr Leu Gly Pro Gln His Ala Asp Asn Ile Tyr Ile 195 200 205 Cys Thr Val Ser Asn Pro Ile Ser Asn Asn Ser Gln Thr Phe Ser Pro 210 215 220 Trp Pro Gly Cys Arg Thr Asp Pro Ser Glu Thr Lys Pro Trp Ala Val 225 230 235 240 Tyr Ala Gly Leu Leu Gly Gly Val Ile Met Ile Leu Ile Met Val Val 245 250 255 Ile Leu Gln Leu Arg Arg Arg Gly Lys Thr Asn His Tyr Gln Thr Thr 260 265 270 Val Glu Lys Lys Ser Leu Thr Ile Tyr Ala Gln Val Gln Lys Pro Gly 275 280 285 Pro Leu Gln Lys Lys Leu Asp Ser Phe Pro Ala Gln Asp Pro Cys Thr 290 295 300 Thr Ile Tyr Val Ala Ala Thr Glu Pro Val Pro Glu Ser Val Gln Glu 305 310 315 320 Thr Asn Ser Ile Thr Val Tyr Ala Ser Val Thr Leu Pro Glu Ser 325 330 335

Claims

1. The use of SLAMF1 in preparing a colorectal cancer biomarker reagent, characterized in that: The application uses SLAMF1 as a target to prepare a diagnostic reagent for the occurrence and / or metastasis of colorectal cancer, or to prepare a reagent for predicting the survival time of colorectal cancer, or to prepare a reagent for evaluating the prognosis of colorectal cancer; ILCs from colorectal cancer patients contain two additional tumor-specific ILCs subpopulations TILCs: a tumor-specific ILC1s-like subpopulation and an ILC2s subpopulation; SLAMF1 is selectively expressed on TILCs, and the level of SLAMF1 expression in the blood of colorectal cancer patients is higher.

2. The use according to claim 1, characterized in that The method comprises immunizing an animal with an immunogen of SLAMF1 to prepare antibodies; and / or preparing immune cells, proteins and / or small molecules with SLAMF1 as a target.

3. The use according to claim 1 or 2, characterized in that The sample for measuring the biomarker is a sample obtained from the patient's tumor tissue, intestinal tissue or blood.

4. The use according to claim 3, characterized in that The level of SLAMF1 is determined at the protein level; or, the level of SLAMF1 is determined at the nucleic acid level.

5. The use according to claim 4, characterized in that When the level of SLAMF1 is determined at the protein level, the level of SLAMF1 is determined by immunohistochemistry; When the level of SLAMF1 is determined at the nucleic acid level, the level of SLAMF1 is determined by quantifying the mRNA encoding SLAMF1.

6. The use according to claim 1, characterized in that The higher the level of SLAMF1, the higher the probability that the patient has a long survival time.

Images