Method for identifying stress immune cells and application

Abstract

The invention belongs to the technical field of biomedicine, and particularly relates to a method for identifying stress immune cells and application. The invention provides a method for identifying stress immune cells. According to the method, the stress immune cells are effectively identified and distinguished by extracting a single cell transcriptome of the immune cells and detecting the expression level of genes in a sample; the invention also provides a substance for identifying the stress response immune cells. The substance comprises a reagent for detecting the expression quantity of 11 genes; the 11 genes comprise an HSPA1A gene, an HSPA1B gene, an HSPA6 gene, an HSP90AA1 gene, an HSPE1 gene, an HSPD1 gene, a DNAJB1 gene, a DNAJA4 gene, an HSPH1 gene, a CACYBP gene and a ZFAND2A gene. The invention also provides application of the substance in preparation of a kit for prognosis evaluation of liver cancer patients. The identification method can be more reliably applied to clinical practice, the prognosis of the liver cancer patient can be more accurately predicted, and a scientific basis is provided for medical decision making.

Description

Technical Field

[0001] The present invention belongs to the field of biomedical technologies, and particularly relates to a method for identifying stress immune cells and applications thereof. Background Art

[0002] Tumors, which are still regarded as difficult medical problems to overcome, have a very clear treatment goal, that is, after diagnosis, through scientific and effective treatment means, the survival period of patients is extended as much as possible. In the relevant research fields, both the activity of tumor cells and immune cells in the tumor microenvironment are involved in the occurrence and development of tumors. Therefore, tumor immunology has attracted attention. Chu et al. (Chu, Yanshuo et al. “Pan-cancer T cell atlas links a cellular stress response state to immunotherapy resistance.” Nature medicine vol. 29, 6 (2023): 1550-1562.) defined a group of T cells in the stress response state (stress response T cell, T STR ) in the tumor microenvironment by constructing a pan-cancer single-cell T cell atlas and combining spatial multi-omics technologies for verification. These T cells are characterized by high expression of heat shock genes, and it is speculated that they may play a key role in the process of immunotherapy resistance. Further clinical cohort analysis found that T STRCells were significantly enriched in patient samples unresponsive to immune checkpoint inhibitors (ICB), suggesting their potential value as resistance markers; Tang et al. (Tang F, Li J, Qi L, et al. A pan-cancer single-cell panorama of human natural killer cells. Cell. 2023;186(19):4235-4251.e20.) used a pan-cancer single-cell transcriptome analysis method to identify DNAJB1+CD56dimCD16hi NK cells highly expressing stress response-related proteins, and named them tumor-associated NK cells (TaNK cells). The study observed that such cells presented an exhausted phenotype, specifically manifested as decreased killing ability and increased expression of inhibitory receptors, and their enrichment was closely related to the low sensitivity to immune checkpoint inhibitor (ICB) treatment. Yang et al. (Yang Y, Chen X, Pan J, et al. Pan-cancer single-cell dissection reveals phenotypically distinct B cell subtypes. Cell. 2024;187(17):4790-4811.e22.) integrated single-cell data of B cells from 19 cancers, combined with BCR sequence analysis and spatial validation methods to identify a B cell stress response subset c06_Bm_stress in tumor-infiltrating B cells. This subset of cells was similar to previously reported stress response T cells and tumor-associated dysfunctional NK cells, and was associated with poor survival. The research ideas of the above researchers were all based on pan-cancer single-cell sequencing data to systematically analyze the functional status and heterogeneity of a single cell type. However, there are still many unclear issues: What is the unified identification standard for stress immune cells? And how to identify them? How many subtypes of stress response immune cells are there in tumors? Can stress response immune cells be used to reclassify tumors? If so, how to guide clinical treatment based on this and how to develop reagents (kits) for tumor treatment. Therefore, there is an urgent need for a scientific and accurate identification method to identify stress response immune cells in the tumor microenvironment. Summary of the Invention

[0003] The purpose of the present invention is to provide a method for identifying stress response immune cells, enabling scientific and accurate identification of stress response immune cells in the tumor microenvironment.

[0004] The first purpose of the present invention is to provide a method for identifying stress response immune cells, the method comprising the following steps:

[0005] S1: Extract the single-cell transcriptome of immune cells from the biological sample to be tested;

[0006] S2: Detect the expression levels of the following genes in the sample: HSPA1A, HSPA1B, HSPA6, HSP90AA1, HSPE1, HSPD1, DNAJB1, DNAJA4, HSPH1, CACYBP, and ZFAND2A;

[0007] S3: Determine whether stress-responsive immune cells exist based on the expression levels of the genes in step S2.

[0008] Preferably, the immune cells in step S1 are T cells, NK cells, macrophages, and plasmacytoid dendritic cell subsets.

[0009] Preferably, the determination criterion in step S3 is that the expression levels of at least 3 of the genes in step S2 exceed a preset threshold; the preset threshold is that the expression levels of at least 3 of the genes exceed 2 times the expression levels of the corresponding genes in the healthy control sample. The healthy control sample refers to the liver tissue or peripheral blood immune cells of a healthy person.

[0010] The second object of the present invention is to provide an application of the above method in constructing a prognostic evaluation model for liver cancer patients.

[0011] The third object of the present invention is to provide a substance for identifying stress-responsive immune cells, and the substance includes reagents for detecting the expression levels of 11 genes; the 11 genes include HSPA1A, HSPA1B, HSPA6, HSP90AA1, HSPE1, HSPD1, DNAJB1, DNAJA4, HSPH1, CACYBP, and ZFAND2A genes.

[0012] The fourth object of the present invention is to provide an application of the above substance in preparing a prognostic evaluation kit for liver cancer patients.

[0013] The fifth object of the present invention is to provide a tumor patient typing method based on the above substance, including the following steps:

[0014] (a) Identify the stress-responsive immune cells in the tumor sample by the above method;

[0015] (b) Combine the expression levels of the above substance with the malignant cell marker gene FN1 and the ligand-receptor pair genes C3, C3AR1, C5, C5AR1, TNFSF14, LTBR, HBEGF, EGFR, SEMA4A, and PLXNB2 to construct a "stress-responsive immune cell-malignant cell interaction" gene set;

[0016] (c) Calculate the gene set score by consensus clustering method, and divide the patients into high-score group and low-score group according to the score, where the high-score group indicates poor prognosis and immunotherapy resistance.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0018] 1) The present invention provides a method for identifying stress-responsive immune cells, which can significantly improve the efficiency of identifying and distinguishing various stress-responsive immune cells, and thus contribute to the in-depth study of their action mechanisms in the tumor microenvironment, providing more accurate cell targets for subsequent immunotherapy research.

[0019] 2) The present invention also provides a hepatocellular carcinoma reclassification gene set constructed based on stress-responsive immune cells and malignant cells with their interactions, realizing the accurate classification of hepatocellular carcinoma patients, thereby predicting the patient survival period, guiding clinical medication, and monitoring the treatment effect from the immunological perspective. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0021] Figure 1 To preprocess all single-cell sequencing data through the Seurat software package, a schematic diagram of the integrated and annotated liver tissue single-cell atlas is obtained. Among them, (A) UMAP shows all cell types from healthy (Health) individuals (n = 8), chronic hepatitis B (Hepatitis) (n = 6), tumor (Tumor) (n = 8), and adjacent normal (AdjacentNormal) (n = 5) patients. (B) The bubble chart shows the average expression levels of characteristic genes of each cell population. The color depth represents the average expression level, and the size of the dots represents the gene expression percentage. (C) UMAP shows malignant and non-malignant cell populations. (D) The average proportion of each cell type in the integrated samples. (E) The average proportion of each cell type in different samples.

[0022] Figure 2It is a schematic diagram for observing the clustering annotation of T cell subsets and the expression of characteristic genes. Among them, (A) t-SNE shows the naming of T cell subsets. (B) The bubble chart shows the average expression levels of characteristic genes of each T cell subset. The color depth represents the average expression level, and the size of the dots represents the gene expression percentage. (C) The expression patterns of CD4 / 8+ T cell subsets in selected genes. (D) The line chart shows the changes in the cell numbers of healthy individuals, hepatitis patients, and liver cancer patients, and the grouping is based on the dynamic changes in cell numbers. (E) The bar chart shows the composition of T cell clusters at the single-sample level, and the bar chart is colored according to different subcluster types. (F) The cell fractions of hepatitis-specific T cell populations are colored by rank and fluctuate with the progression of inflammation. (G) GSVA is used to analyze the pathway changes in the groups with decreasing and increasing cell numbers. (H) The correlation of marker gene expression in tumor-specific T cell subsets. (I) The circular diagram of GO enrichment analysis of the DN-c12 subset. (J) The specific transcription factors of each T cell subset. (K) The t-SNE map of RXRA and its regulated target genes.

[0023] Figure 3 It is as follows: By clustering and classifying other immune cells, stress-responsive NK cells, stress-responsive macrophages, and stress-responsive plasmacytoid dendritic cells are screened and discovered, and the schematic diagram of the expression of their respective characteristic genes is observed. Among them, (A) t-SNE shows the naming and grouping of NK cell subsets. (B) The expression levels of FCGR3A and NCAM1 in different groups. (C) The top genes with the highest expression levels in each NK cell subset. (D) The heat map shows the different expressions of cell markers used for developmental identification in NK cell subsets. The color represents the gene expression level after score normalization. (E) The heat map shows the expression of the corresponding genes used to define the functional score. (F) The heat map shows the expression of activation and inhibitory receptors. The color represents the gene expression level after score normalization. (G) t-SNE shows the naming of macrophage subsets. (H) The top genes with the highest expression levels in each macrophage subset. (I) The M1 and M2 scores of macrophage subsets. (J) t-SNE shows the naming of pDC cell subsets. (K) The top genes with the highest expression levels in each pDC cell subset. (L) The heat map shows the expression of cell markers for developmental migration. (M) The heat map shows the expression of function-related genes. The color represents the gene expression level after score normalization.

[0024] Figure 4It is: a developmental trajectory analysis diagram of each stress-responsive cell subset. Among them, (A) Pseudotime differentiation trajectory of T cell subsets, with arrows indicating the pseudotime differentiation process. (B) Cell density diagram of T cell subsets along the time axis. (C) Pseudotime differentiation trajectory of NK cell subsets, with arrows indicating the pseudotime differentiation process. (D) Cell density diagram of NK cell subsets along the time axis. (E) Pseudotime differentiation trajectory of macrophage subsets, with arrows indicating the pseudotime differentiation process. (F) Cell density diagram of macrophage subsets along the time axis. (G) Pseudotime differentiation trajectory of pDC cell subsets, with arrows indicating the pseudotime differentiation process. (H) Cell density diagram of pDC cell subsets along the time axis. (I) Expression changes of genes along the pseudotime differentiation trajectory. (J) Bubble chart showing the average expression levels of 11 stress-responsive signature genes in each immune cell subset (T cells; NK cells; macrophages; pDC cells). (K) Relationship between MALAT1 and stress-responsive signature gene expression levels in pan-cancer calculated by Timer.

[0025] Figure 5 It is: an analysis diagram of pseudotime differentiation genes and enrichment analysis results of stress-responsive immune cells; among them, (A-D) Heatmaps show the dynamic changes of gene expression along pseudotime. Among them, A represents the dependent gene changes of the T cell stress-responsive subtype in the branch, B represents the dependent gene changes of the NK cell stress-responsive subtype in the branch, C represents the dependent gene changes of the macrophage stress-responsive subtype in the branch, and D represents the dependent gene changes of the plasmacytoid dendritic cell stress-responsive subtype in the branch. E represents the GO and KEGG enrichment analysis of pseudotime genes.

[0026] Figure 6 It is: a schematic diagram of the correlation analysis between the stress-responsive signature gene set and multiple physical and chemical environment response pathways in each stress-responsive subtype. Among them, hypoxia, acidic pH, stress, lipid overload, oxidative stress, and inflammation are studied in sequence from top to bottom.

[0027] Figure 7 It is: a schematic diagram of the expression of stress-responsive signature genes in stress-responsive subtypes under different physical and chemical conditions; among them, (A) Heatmap shows the correlation between stress-responsive signature genes and physical and chemical environment response pathways in stress-responsive macrophages. (B) Changes in the relative mRNA expression levels of each gene at different time points in the hypoxia model. (C) Changes in the relative mRNA expression levels of each gene at different time points in the oxidative stress model. (D) Changes in the relative mRNA expression levels of each gene when treated with different concentrations of H2O2 for 8 h. Error bars represent standard deviation. (ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001)

[0028] Figure 8It is: the enrichment score analysis diagram of gene sets in the HALLMARK and KEGG cohorts analyzed by irGSEA in different immune cell subsets; among them, A represents T cells; B represents NK cells; C represents macrophages; D represents plasmacytoid dendritic cells.

[0029] Figure 9 It is: a schematic diagram of the interaction analysis of each cell component in the tumor microenvironment; among them, (A) t-SNE shows the grouping and naming of malignant cell subsets. (B) The infiltration degree of malignant cell subsets is evaluated by ssGSEA. (C) The bubble chart visualizes the ligand-receptor pairs that significantly play roles as signal senders and signal receivers of stress-responsive immune cells in the HCC tumor microenvironment. The color of the dots represents the average gene expression level, red indicates higher expression, and blue indicates lower expression. The size of the dots represents the significance of the p-value. (D) The violin chart visualizes the expression of ligands / receptors in cell subsets, and different subsets are distinguished by different colors. (E) The protein interaction between DEGs of malignant_2 and its specifically expressed receptors or ligands (such as C3, C5, EGFR, etc.). (DEG selection criteria: logFC > ±1, p < 0.05).

[0030] Figure 10 It is: a schematic diagram of signal transduction related to stress-responsive immune cells; among them, (A-I) the hierarchical chart visualizes the COMPLEMENT, LIGHT, EGF, SEMA4, LIGHT, CD86, CD137, IL10 signal networks and ligand-receptor signal networks. (A-E) between malignant cells and stress-responsive immune cells, (F-I) between immune cells. In the figure, Source is the cell subset that emits signals. On the left side of the figure, Target is the stress-responsive immune cell subsets (T12; NK12; Macro_6; pDC_3) and all subsets of malignant cells. On the right side of the figure, Target is the subsets other than those in the left figure. The thicker the line, the stronger the signal.

[0031] Figure 11 It is: a schematic diagram of the multiplex immunofluorescence verification of stress immune cells. Among them, (A-C) the spatial distribution of stress-responsive immune subsets. The scatter plot shows the number of CD3+HSPA1B+ cells (T STR ), CD56+HSPA1B+ cells (NK STR ), and CD68+HSPA1B+ cells (Macro STR ). In the tumor tissue sections of hepatocellular carcinoma patients, stress-responsive T cells (T STR ), stress-responsive NK cells (NK STR ), and stress-responsive macrophages (Macro STRRepresentative examples (arrows). The scale bar in the figure represents 20 μm. (D-F) Macro STR Spatial correlation between CD68+HSPA1B+ cells and FN1+ cells. The StrataQuest software was used to quantify the positional relationship between CD68+HSPA1B+ cells and FN+ cells, and the number of CD68+HSPA1B+ cells at different distances from FN+ cells was quantified: 0-20 μm, 20-50 μm. Sample 01 (A, D): Tumor tissue section of in situ hepatocellular carcinoma. Sample 02 (B, E): Tumor tissue section of hepatocellular carcinoma with capsular invasion. Sample 03 (C, F): Tumor tissue section of hepatocellular carcinoma with capsular and vascular invasion.

[0032] Figure 12 Schematic diagram for performing individual survival analysis of genes in the "Stress response immune cell - malignant cell interaction" gene set.

[0033] Figure 13 Schematic diagram for performing survival analysis and consensus clustering-based TIDE analysis using TCGA LIHC data, where (A) Survival analysis results of high and low groups (p < 0.05). (B) Unsupervised clustering analysis with K = 2 identified 2 Subtypes (Subtype 1 / 2). (C) Box plot visualizing the ssGSEA score distribution of gene sets in different groups of HCC, where the horizontal line represents the median, and the edges of the box represent the upper and lower quartiles. The significance of the difference between the two groups was evaluated by kruskal.test (p < 0.05). (D) TIDE immunotherapy response prediction score. The scores between the two risk groups were compared by Wilcoxon test (ns, not significant, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001). (E) Heat map of clinical information.

[0034] Figure 14 Schematic diagram of the relationship between some genes in the gene set and the stage and grade of hepatocellular carcinoma. (A) Relationship between the expression levels of HSPA1A, HSPA1B, ZFAND2A, LTBR, SEMA4 and the combination and the clinical grade of HCC patients. (B) Relationship between the expression level of HSPE1 and the clinical stage of HCC patients. Evaluated by Kruskal-Wallis rank test.

[0035] Figure 15Schematic diagram for gene set scoring of spatial transcriptome data. Among them, (A) Annotation of HCC neoadjuvant treatment response group and non-response group. Different spots represent different cell types. (B) Cell type annotation, mainly divided into Tumor, CAFs, and Immune regions. Each point is colored with different cell types. (C) Visualize the proportion of each cell type from different sample sources. (D) Expression of gene set in the treatment response group and non-response group. (E) Expression of gene set in each sample. (F-G) High and low grouping of immune cells and malignant cells from different sample sources based on some stress response characteristic genes and malignant cell characteristic genes.

[0036] Figure 16 For the processing of spatial transcriptome data and the expression of gene sets of stress response immune cell interactions. (A) ElbowPlot of principal component analysis of spatial transcriptome data. (B) Cell distribution before batch correction by Harmony. (C) Cell distribution after batch correction by Harmony. (D) UMAP plot of cell clusters after dimensionality reduction and clustering. (E) Bubble plot visualizing the average expression levels of marker genes in each cell cluster. The darker the color, the higher the average expression level, and the size of the point represents the percentage of gene expression. (F) Violin plot visualizing the expression distribution of genes in spatial transcriptome data, shown by different sample sources.

[0037] Figure 17 For the discovery of co-infiltration of stress response macrophages and Malignant_2 in spatial transcriptome HCC4R samples using MIA analysis. (A) ElbowPlot of principal component analysis of spatial transcriptome HCC4R data. (B) UMAP plot of cell clusters after dimensionality reduction and clustering. (C) Spatial feature map after dimensionality reduction and clustering in tissue sections. (D) Heat map showing the distribution of each cell component in the HCC tumor microenvironment on spatial transcriptome tissue sections. Detailed implementation manners

[0038] To further illustrate the present invention, the technical solutions provided by the present invention will be described in detail below with reference to the accompanying drawings and embodiments, but they should not be construed as limiting the protection scope of the present invention.

[0039] The production processes, experimental methods, or detection methods involved in the embodiments of the present invention are all conventional methods in the prior art unless otherwise specified, and their names and / or abbreviations are all conventional names in the field and are very clear and definite in the relevant application fields. Those skilled in the art can understand the conventional process steps according to the names and apply the corresponding equipment, and implement them under conventional conditions or conditions recommended by the manufacturer.

[0040] There are no special restrictions on the sources of various instruments, equipment, raw materials or reagents used in the embodiments of the present invention. They are all conventional products that can be obtained through regular commercial channels, and can also be prepared according to the conventional methods well-known to those skilled in the art.

[0041] Genes and their IDs in the NCBI database are shown as follows;

[0042] HSPA1A Gene ID: 3303; HSPA1B Gene ID: 3304; HSPA6 Gene ID: 3310; HSP90AA1 Gene ID: 3320; HSPE1 Gene ID: 3336; HSPD1 Gene ID: 3329; DNAJB1 Gene ID: 3337; DNAJA4 Gene ID: 55466; HSPH1 Gene ID: 10808; CACYBP Gene ID: 27101; ZFAND2A Gene ID: 90637.

[0043] Obtaining data of Example 1 and screening stress response cells

[0044] Data processing and screening of immune genes related to liver cancer prognosis: The present invention collected and integrated published single-cell sequencing data from healthy human liver, viral hepatitis, hepatocellular carcinoma infected with HBV and adjacent tissues. The data sources are from the GEO database: GSE156625 (hepatocellular carcinoma and adjacent tissues), GSE158723 (healthy human liver), GSE186343 (viral hepatitis). According to typical cell lineage markers and significantly differentially expressed genes significantly up-regulated in each subcluster, a total of 13 cell types were determined, as shown in Figure 1 shown; Through a comprehensive and systematic analysis of immune cell subsets, it was found that specific subsets in T cells, NK cells, macrophages and plasmacytoid dendritic cells showed high expression of heat shock genes.

[0045] For T cells, clustering annotation and expression observation of characteristic genes of T cell subsets were performed. It can be seen that through clustering analysis, all cells were further divided into 19 cell subclusters, and then all T cell subclusters were annotated according to classical markers combined with the expression of differentially expressed genes and characteristic gene sets, as shown in Figure 2 A - G.

[0046] From Figure 2 H, I, it can be found that: The stress response T cell T STR specifically present in the tumor environment, which highly expresses heat shock protein family genes (such as HSPA1A, HSP90A1B, etc.) and lacks the expression of function-related genes, may be in a dysfunctional state induced by the physical and chemical stress of the tumor microenvironment;

[0047] Evaluating upstream transcription factors of T cell subsets: Through SCENIC analysis, the transcription factor RXRA is specific to T STR cell subsets, as Figure 2 shown in J;

[0048] As Figure 2 shown in K, both the activity of the transcription factor RXRA and its regulated target genes ZFAND2A and DNAJA4 are significantly expressed in T STR cell subsets, indicating that these three genes can also be regarded as potential markers of T STR cell subsets.

[0049] For NK cells, all cells were further divided into 13 cell subclusters by clustering analysis and named and grouped according to the differential genes of each subcluster as Figure 3 shown in A and B. Through analysis, it was found that both NK11_EGR1 and NK12_HSPA1A upregulated stress-related genes, but NK12_HSPA1A mainly highly expressed heat shock genes and also highly expressed killer cell immunoglobulin-like receptors KIR2DL1 and KIR2DL3, as Figure 3 shown in C and D; as Figure 3 shown in E and F, a stress response subtype in NK cells was obtained, defined as stress response NK cells (NK STR ).

[0050] For macrophages, all cells were further divided into 14 cell subclusters by over-clustering analysis and named according to the differential genes of each subcluster, as Figure 3 shown in G and H. The polarization characteristic score showed that Macro6_HSPA1B had relatively close scores in M1 and M2 polarization scores, showing certain mixed characteristics and possibly being in an intermediate state between M1 and M2, as Figure 3 shown in I, and it showed high expression of heat shock genes. Therefore, a stress response subtype in macrophages was obtained, named stress response macrophages (Macro STR ).

[0051] For plasmacytoid dendritic cells, all cells were further divided into 4 cell subclusters by clustering analysis and named according to the differential genes of each subcluster, as Figure 3 shown in J and K. From Figure 3 L and M, it can be seen that pDC3_HSPA1A mainly highly expressed heat shock genes and significantly upregulated chemokines and MHC-II molecules. Therefore, it was classified into the stress response subtype in plasmacytoid dendritic cells, defined as stress response plasmacytoid dendritic cells (pDC STR ).

[0052] The present invention further performs pseudotime analysis on T cells, NK cells, macrophages, and plasmacytoid dendritic cells in the tumor microenvironment, and the research finds that these stress response subtypes are all in the terminal stage of differentiation in their respective cell populations, showing unique transcriptional characteristics, such as Figure 4 shown in A-H. Among all the pseudotime-dependent genes such as Figure 4 shown in I, only some gradually increase in expression level during the differentiation of stress response subtypes, which are HSP90AB1, HSPA8, HSPB1, DNAJA1, DNAJA4, and MALAT1. As shown in 4K, the relationship between MALAT1 and the expression level of stress response characteristic genes.

[0053] Using the BEAM method to analyze the specified nodes, revealing the dependent gene changes of stress response subtypes in the branches of T cells, NK cells, and macrophages, and at the same time showing the change trend of pseudotime-dependent genes of plasmacytoid dendritic cells along the pseudotime axis, such as Figure 5 shown in A-D. A total of 59 pseudotime-dependent intersection genes were counted, and GO and KEGG enrichment analyses were performed on them. As shown in the results of Figure 5 E, these genes are mainly involved in protein folding, ribosome structure and function. Some immune response-related biological processes and signaling pathways are enriched, such as antigen processing and presentation, MAPK signaling pathway, suggesting that they participate in immune regulation in the tumor microenvironment.

[0054] Based on the above genes, combined with the Top10 genes of each stress response subtype and the specific transcription factor RXRA and its target genes of T STR previously discovered, 11 genes that are generally highly expressed in different stress response subtypes were finally screened out, and a stress response characteristic gene set was constructed, including HSPA1A, HSPA1B, HSPA6, HSP90AA1, HSPE1, HSPD1, DNAJB1, DNAJA4, HSPH1, CACYBP, ZFAND2A, as marker genes for stress response immune cells( Figure 4 J).

[0055] Example 2 Verification of the induction effect of the tumor microenvironment on stress response subtypes

[0056] 1. Verify the correlation between the stress response characteristic gene set and physical and chemical indexes through in vitro experiments

[0057] The present invention analyzes whether physical and chemical factors affect the formation of stress response immune cells by regulating the expression of heat shock genes by evaluating the correlation between the stress response characteristic gene set and physical and chemical indexes (hypoxia, acidic pH, lipid overload, oxidative stress, etc.).

[0058] Result analysis: As shown in Figure 6As shown, there are varying degrees of associations between different stress response subtypes and physical and chemical factors. Among them, the correlation between stress response macrophages and these physical and chemical factors is particularly prominent.

[0059] Re-analyze the correlation of stress response characteristics in stress response macrophages combined with qRT-PCR experiments under different physical and chemical conditions.

[0060] Result analysis: As Figure 7 shown in A, focusing on stress response macrophages, its stress response characteristic genes are significantly positively correlated with multiple physical and chemical characteristics, namely hypoxia, stress, oxidative stress, lipid, and inflammatory response. And it has been reported that these physical and chemical factors such as inflammation, alcohol stress, and lipid effectively regulate the expression of heat shock genes.

[0061] 2. Effects of other physical and chemical factors on stress response characteristic genes in macrophages

[0062] Based on the above correlation analysis, the present invention screened out 4 key genes (HSPD1, HSPH1, DNAJB1, HSPA1B) and designed in vitro hypoxia and oxidative stress models.

[0063] In the mouse macrophage hypoxia model induced by treatment with 100.00 μM cobalt dichloride, it was observed that as Figure 7 shown in B, the mRNA expression levels of HSPH1 and HSPA1B were up-regulated in a time-dependent manner in the hypoxia model. As Figure 7 shown in C, in the oxidative stress cell model induced by hydrogen peroxide, it was observed that when exposed to 1000.00 μM hydrogen peroxide, the mRNA expression levels of HSPH1, HSPA1B, and DNAJB1 genes reached the highest at 8 h. As Figure 7 shown in D, at different concentrations and after 8 h of treatment, the mRNA expression level of the gene changed most significantly for HSPA1B, and the gene mRNA expression increased with the increase in concentration.

[0064] From the above results, it can be seen that HSPA1B is the most sensitive molecular marker in macrophages in response to hypoxia and oxidative stress.

[0065] Example 3 Explore the functional heterogeneity of each stress response subcluster in immune cells

[0066] By using the irGSEA software package, analyze the enrichment of HALLMARK and KEGG gene sets in different cell subsets to comprehensively understand the biological processes and molecular mechanisms in gene expression data.

[0067] Result analysis: As Figure 8As shown, the stress-responsive immune cell subtypes show varying degrees of inhibition in immune function. There are certain similarities in protein secretion, metabolic activity, and energy generation, and a widespread inhibitory trend is also shown. The downregulation of the pathways related to development and proliferation further supports that these cells are in a terminally differentiated state.

[0068] Example 4 Communication Network between Tumor Origin and Stress-Responsive Immune Cells

[0069] Cell function not only depends on its own internal biochemical reactions and molecular mechanisms but also is closely related to surrounding cells.

[0070] In the present invention, CellChat was used to analyze the potential communication network between tumor origin and stress-responsive immune cells, and the pathways of the interaction between malignant cells and stress-responsive subtypes were screened.

[0071] The results showed that, as Figure 9 shown in C-E, the stress-responsive immune cells mainly interacted with the Malignant_2 subset with a higher infiltration fraction. As Figure 9 shown in A, B, Malignant_2 is the subset with the highest infiltration level among all malignant cell subsets in the TCGA LIHC samples.

[0072] A total of four groups of signaling pathways were identified: COMPLEMENT, LIGHT, EGF, and SEMA4. From Figure 10 shown in A, B, it can be seen that the Malignant_2 subset mainly communicates with stress-responsive macrophages through C3-C3AR1 and HC-C5AR1 on the COMPLEMENT signaling pathway. At the same time, the Malignant_2 subset also acts as a signal receiver to receive signals from stress-responsive immune cells. For example, stress-responsive T cells act on Malignant_2 through LIGHT (TNFSF14-LTBR); stress-responsive macrophages act on Malignant_2 through EGF (HBEGF-EGFR); stress-responsive pDC cells act on Malignant_2 through SEMA4 (SEMA4A-PLXNB2), as Figure 10 shown in C-E.

[0073] Example 5 Interaction between Stress-Responsive Immune Cells and Other Immune Cells

[0074] Stress-responsive T cells act on stress-responsive macrophages through LIGHT (TNFSF14-TNFSRF14), as Figure 10 shown in F. Stress-responsive pDC cells act on effector T cell T7 and regulatory T cell T13 through CD137 (TNFSF9-TNFRSF9) asFigure 10 As shown in G. Stress-responsive macrophages act on regulatory T cell T13 through CD86 (CD86-CTLA4), and regulatory T cell T13 acts on stress-responsive macrophages such as Figure 10 as shown in H and I.

[0075] Example 6 Verification of stress-responsive immune cells in hepatocellular carcinoma

[0076] The present invention collected three tissue samples with different pathological progressions, namely 01 in-situ tumor, 02 tumor with capsule invasion, and 03 tumor with capsule and vascular invasion. Stress-responsive immune cells in the tumor microenvironment were detected by multiplex immunofluorescence technology. Since the number of plasmacytoid dendritic cells was small, they were not included in the detection indexes of this study.

[0077] The present invention focused on detecting the distributions of T cells (CD3+), NK cells (CD56+), and macrophages (CD68+), and simultaneously observed the expression of HSPA1B, as Figure 11 shown. There were significant infiltrations of stress-responsive immune cells in all three groups of samples.

[0078] Example 7 Verification of the interaction between stress-responsive immune cells and malignant cells in hepatocellular carcinoma

[0079] The present invention constructed a "stress-responsive immune cell-malignant cell interaction" gene set, including stress-responsive characteristic genes (HSPA1A, HSPA1B, HSPA6, HSP90AA1, HSPE1, HSPD1, DNAJB1, DNAJA4, HSPH1, CACYBP, ZFAND2A), the marker gene FN1 of Malignant_2 group, and key specific receptor-ligand pair genes (C3, C3AR1, C5, C5AR1, TNFSF14, LTBR, HBEGF, EGFR, SEMA4A, PLXNB2). Based on this gene set, survival analysis was performed using the TCGA cohort of hepatocellular carcinoma. As Figure 12 The results showed that in the single-gene survival analysis, except for EGFR, TNFSF14, C3, and C5, the survival trends of most genes were poor. Taking the total gene set as a unit, the overall survival rate of patients in the high-score group was low, while that of patients in the low-score group was high, as Figure 13 shown in A.

[0080] To study the predictive effect of the gene set on the efficacy of immunotherapy and auxiliary clinical classification, consensus clustering was used to analyze the samples, and "Subtype 1" with a higher gene set score and "Subtype 2" with a lower gene set score were obtained, as Figure 13As shown in B and C. The present invention further uses the Tumor Immune Dysfunction and Exclusion (TIDE) algorithm to predict the response to immunotherapy based on gene set expression, such as Figure 13 As shown in the analysis results of D and E, higher TIDE scores and Dysfunction scores were observed in the high-scoring "Subtype 2", and lower Exclusion scores, while there was no significant difference in the Microsatellite Instability (MSI) scores, indicating that HCC patients with high gene set expression may have a certain resistance to immunotherapy.

[0081] The present invention further analyzes the relationship between these genes and clinical grades and stages, and uses the "Kruskal-Wallis" test to compare the differences between different groups, and corrects the p-value with Bonferroni, and finds that the expression of some genes can be used to evaluate the disease progression of patients. Such as Figure 14 As shown in A, there was a significant correlation between the expression of HSPA1A and clinical Grade 1 and Grade 3, and between Grade 2 and Grade 3 (p < 0.05). The gene set composed of HSPA1A, HSPA1B, ZFAND2A, LTBR, and SEMA4A had a significant correlation with each clinical grade (p < 0.05), and the expression level increased with the increase of clinical grade. Such as Figure 14 As shown in B, there was also a significant correlation between the expression of HSPE1 and clinical Stage 1 and Stage 3 (p < 0.05), and the expression level increased with the increase of clinical grade.

[0082] Example 8 Determination of Spatial Colocalization and Drug Resistance

[0083] To further evaluate the predictive ability of the "Stress Response Immune Cell-Malignant Cell Interaction" gene set for immunotherapy, the spatial transcriptome data in the neoadjuvant therapy of cabozantinib combined with nivolumab were analyzed. The data included a total of 7 groups, among which HCC1R, HCC2R, HCC3R, and HCC4R were the immunotherapy response groups, and HCC5NR, HCC6NR, and HCC7NR were the immunotherapy non-response groups, as Figure 15 shown in A and B. The gene set scoring results are as Figure 15 shown in C and D, indicating that the overall score of the non-response group was higher than that of the response group. Figure 15 As shown in E, among all samples, HCC7NR had the highest score in the non-response group, and HCC4R had the highest score in the response group. Therefore, by visualizing the expression of each gene in the gene set, the present invention found that only some genes had high expression, which became the main contributors to the gene set score. These highly expressed genes can be divided into two categories: stress response immune cell-expressed genes (such as HSPA1A, HSP90AA1, HSPE1, HSPD1) and malignant cell-expressed genes (such as FN1, C3, C5, PLXNB2), such asFigure 16 as shown in Figure F.

[0084] The present invention further extracts immune cells and tumor cells from each sample, calculates the average expression of these genes in the corresponding cells, and defines it as the set score. The results are as Figure 15 shown in Figures F and G, indicating that the non-responsive group usually shows higher stress immunity and malignancy scores, and the co-infiltration of the two may be associated with treatment tolerance.

[0085] MIA analysis integrates single-cell and spatial transcriptome omics data and calculates the enrichment degree of cell type-specific gene sets in spatial positions. A cell distribution feature of coexistence of stress-responsive immune cells and malignant cells was found in the HCC4R sample. The results are as Figure 17 shown, and a high infiltration of Malignant_2 was observed in region 3 with a high infiltration of stress-responsive macrophages Macro_6. This result further verifies the interaction between stress-responsive macrophages and Malignant_2 malignant cells in the interaction analysis.

[0086] In addition, to clarify the interaction between stress-responsive macrophages and Malignant_2 malignant cells marked by FN1. The present invention calculates the number of CD68+HSPA1B+ cells (Macro STR ) at different distance gradients and finds that the number of stress-responsive macrophages gradually decreases with increasing distance from FN1+ cells, as Figure 11 shown in Figures D-F. The interaction between stress-responsive macrophages (Macro STR ) and FN1+ cells shows significant spatial dependence, and Macro STR mostly concentrates within 20 μm of FN1+ cells. These results indicate that Macro STR cells may have a positive interaction with FN1+ cells. Among them, the co-localization intensity between the two has no significant correlation with the degree of tumor invasion.

[0087] Although the above embodiments have described the present invention in detail, they are only a part of the embodiments of the present invention, not all embodiments. People can also obtain other embodiments without creative efforts based on this embodiment, and these embodiments all belong to the protection scope of the present invention.

Claims

1. A method for identifying stress response immune cells, characterized in that The method includes the following steps: S1: Extract the single-cell transcriptome of immune cells from a biological sample to be tested; S2: Detect the expression levels of the following genes in the sample: HSPA1A, HSPA1B, HSPA6, HSP90AA1, HSPE1, HSPD1, DNAJB1, DNAJA4, HSPH1, CACYBP, and ZFAND2A; S3: Determine whether stress-response immune cells exist based on the expression levels of the genes in step S2.

2. The method according to claim 1, wherein The immune cells in step S1 are T cells, NK cells, macrophages, and plasmacytoid dendritic cell subsets.

3. The method according to claim 1, wherein The determination criterion in step S3 is that the expression levels of at least 3 of the genes in step S2 exceed a preset threshold; the preset threshold is that the expression levels of at least 3 of the genes exceed 2 times the expression levels of the corresponding genes in a healthy control sample.

4. Use of the method according to claim 1 in constructing a prognostic evaluation model for liver cancer patients.

5. A substance for identifying stress-responsive immune cells, characterized in that, The substance includes reagents for detecting the expression levels of 11 genes; the 11 genes include HSPA1A, HSPA1B, HSPA6, HSP90AA1, HSPE1, HSPD1, DNAJB1, DNAJA4, HSPH1, CACYBP, and ZFAND2A genes.

6. Use of the substance according to claim 5 in preparing a prognostic evaluation kit for liver cancer patients.

7. A method for classifying tumor patients based on the substance described in claim 5, characterized in that, Includes the following steps: (a) Identify the stress-response immune cells in the tumor sample by the method according to claim 1; (b) The substance according to claim 5 combines the expression levels of the malignant cell marker gene FN1 and ligand-receptor pair genes C3, C3AR1, C5, C5AR1, TNFSF14, LTBR, HBEGF, EGFR, SEMA4A, and PLXNB2 to construct a "stress-response immune cell-malignant cell interaction" gene set; (c) Calculate the gene set score by the consensus clustering method, and divide the patients into a high-score group and a low-score group according to the score, where the high-score group indicates poor prognosis and immune therapy resistance.

Images