CN115198017B - Application of macrophage subgroup based on high-expression PLIN2 in diagnosis and treatment of ovarian cancer ascites - Google Patents

Abstract

The invention belongs to the technical field of treatment of gynecological diseases, and discloses application of macrophage subgroup with high expression of PLIN2 to diagnosis and treatment of ovarian cancer ascites. By treating patients with ovarian cancer with CD45 in primary tumor, metastatic focus and ascites ⁺ Single cell RNA sequencing of immune cells revealed a new macrophage subset enriched in ascites and positively correlated with ovarian cancer progression, named ascites-associated macrophages (AAM). AAM is M2 phenotype, with promotion of tumor pathways and high expression of adiponectin PLIN 2. The knockdown of macrophage PLIN2 reduces the permeability of endothelial cells in vitro and the migration of tumor cells, and inhibits the generation of ascites in vivo and the metastasis of cancer. AAM is similar to tumor-associated macrophages in primary and metastatic lesions, but the cell adhesion pathway is down-regulated, with enhancement of CD4 ⁺ T lymphocytes interact to present an immunosuppressive phenotype, and provide a potential immunotherapy target for treating ovarian cancer.

Description

Application of macrophage subgroup based on high expression PLIN2 in diagnosis and treatment of ovarian cancer ascites

Technical Field

The invention relates to the technical field of treatment of gynecological diseases, in particular to application of macrophage subgroup with high expression of PLIN2 to diagnosis and treatment of ovarian cancer ascites.

Background

Ovarian Cancer (OC) is the most lethal of the gynaecological malignancies, and high-grade serous ovarian cancer (HGSOC) is the most common and most aggressive subtype. Patients with HGSOC are often diagnosed as advanced, with 1/3 of them having ascites at the time of diagnosis. Malignant ascites not only affects quality of life, but also leads to high mortality, thus becoming an unfavorable prognostic indicator of OC. Numerous studies have attempted to reveal the mechanisms by which ascites develops, the most common culprit being increased capillary permeability and impaired peritoneal lymphatic drainage. However, the pathogenesis of malignant ascites is much more complex than that disclosed so far. In addition, ascites is considered to be an indispensable mediator of transperitoneal metastasis, and even in the early stages of the disease, tumor cells are shed and spread, eventually leading to relapse and poor clinical outcome. However, it is still unknown how ascites mediates intraperitoneal dissemination, making current treatments ineffective.

Immunotherapy targeting the Tumor Microenvironment (TME) has shown better results in various cancer types. However, due to the complexity and heterogeneity of TME, only a few patients respond to this treatment. TME in OC malignant ascites, which consists of cellular and acellular components, is completely different from that in solid tumors. Macrophages are one of the major immune cell types of TME and account for 40-95% of ascites TME. Although tumor-associated macrophages (TAMs) can promote tumor growth, metastasis, immunosuppression, angiogenesis and drug resistance, the heterogeneity and function of macrophages in malignant ascites has not been fully characterized.

Single cell RNA sequencing (scRNA-seq) has become an effective method to dissect the heterogeneity of complex systems such as TME by revealing novel immune cell subsets and cell-cell interactions. Immune microenvironment profiles of many cancer types have been deciphered at the single cell level, including hepatocellular carcinoma, breast cancer, lung adenocarcinoma, and even pan-cancer. For example, in OC, scRNA-seq identified 16 distinct cell populations, associated with a unique one of the populations associated with high-grade, low-grade, benign and breast cancer relapsing patients. Cell-cell interactions in OC TME reveal the underlying mechanisms by which cell interactions model different tumor immunophenotypes. Recently, scRNA-seq studies on OC ascites describe the variation of malignant cells and the composition of immune cells in ascites. However, the phenotypic diversity and pathogenic function of macrophages in ascites remains to be further validated.

Disclosure of Invention

In order to overcome the above problems of the prior art, the present invention firstly provides the application of the ascites-associated macrophage subset mac _ c4_ PLIN 2.

A second object of the present invention is to provide the use of the PLIN2 gene or its expression product in the preparation of functional products for ovarian cancer.

The purpose of the invention is realized by the following technical scheme:

application of ascites-related macrophage subgroup mac _ c4_ PLIN2 in developing and screening ovarian cancer functional products.

The invention adopts scRNA-seq to analyze immune microenvironment composition of three different parts (primary focus, metastatic focus and ascites) from OC patients, highlights a new subgroup mac _ c4_ PLIN2 (AAM) of macrophage in ascites, the AAM is characterized by high expression of fat differentiation related protein (PLIN 2), and reduces endothelial cell permeability and in-vitro tumor cell migration by knocking down PLIN2 in the macrophage. This work revealed PLIN2 in ascites ⁺ A new subset of macrophages that promote vascular permeability and OC metastasis and provide potential immunotherapeutic targets for the treatment of OC.

Preferably, in the above application, the ascites-associated macrophage subset mac _ c4_ PLIN2 has M2 phenotype and is positively correlated with the generation and metastasis of ascites due to ovarian cancer.

The invention also provides application of the functional product with the inhibition effect on the PLIN2 gene or the expression product thereof in preparing products for preventing or/and treating ovarian cancer.

Alternatively, the invention provides the use of the PLIN2 gene or its expression product in developing and screening functional products for ovarian cancer, which have an inhibitory effect on the PLIN2 gene or its expression product.

Preferably, in the above application, the functional product is used for reducing the generation and/or metastasis of ascites due to ovarian cancer.

More preferably, the functional product includes: one or more of a PLIN2 nucleic acid inhibitor, a PLIN2 protein inhibitor, an immune-related cell deficient or silenced in a PLIN2 gene, a differentiated cell thereof, or a recombinant construct of a gene.

More preferably, the functional product includes any one of:

(i) Small interfering RNA, dsRNA, shRNA, micro RNA and antisense nucleic acid which take PLIN2 or PLIN2 transcripts as target sequences and can inhibit the expression of PLIN2 gene expression products or gene transcription;

(ii) (ii) capable of expressing or forming the small interfering RNA, dsRNA, shRNA, microRNA, antisense nucleic acid construct of (i);

(iii) Constructs containing PLIN2, or the PLIN2 complement, and capable of forming interfering molecules that inhibit the expression of the PLIN2 gene expression product or gene transcription upon transfer into the body;

(iv) An immune-related cell, differentiated cell or construct thereof following suppression or knock-out of the PLIN2 gene sequence.

Compared with the prior art, the invention has the beneficial effects that:

the present invention utilizes the most advanced scRNA-seq to delineate the immune landscape of multiple foci of OC and to discover a new subpopulation of macrophages that are enriched in the ascites and associated with disease progression. This new subgroup is displaying the M2 phenotype and upregulates cancer-associated pathways, with high PLIN2 expression. Functionally, PLIN2 deficiency in macrophages impedes ascites production by protecting the vascular barrier and inhibits cancer progression by inhibiting tumor cell migration. Finally, we tracked the origin and cell-cell interaction of this subset and demonstrated that it might be shed from primary and metastatic lesions by modulating most cell adhesion molecules, while potentiating CD4 ⁺ The affinity of T lymphocytes thus leads to an immunosuppressive effect. This work revealed an active role for ascites TME in promoting transperitoneal metastasis and provided a new target for the treatment of ascites in OC patients.

Drawings

FIG. 1 is a multifocal immune microenvironment map of ovarian cancer patients; wherein, FIG. 1A is a study protocol design demonstrating the collection and processing of tissue samples for scRNA-seq for 6 non-malignant patients and 7 OC patients; FIG. 1B is specific information sampled for each patient; FIG. 1C is a PCA analysis showing transcriptome differences, NO, normal ovaries across multiple foci; OC, primary focus; ML, transfer cooker; AW, normal peritoneal irrigation; AC, malignant ascites; ellipses of different colors represent transcriptome-like tissues; FIG. 1D is a Unified Manifold Approximation and Projection (UMAP) diagram showing the major immune cell types; FIG. 1E is a UMAP panel showing marker gene expression levels for the immune cell types defined in FIG. 1D; FIG. 1F stacked bar graphs show the proportion of major immune cell types for different lesions; FIG. 1G is a group bar graph showing the percentage of cell types in different lesions;

FIG. 2 shows myeloid cell subsets and tissue preferences of OC; wherein FIG. 2A is a UMAP plot showing 15 subpopulations of myeloid lineage cells, different subpopulations being represented by different colors; FIG. 2B is a UMAP panel showing marker gene expression for identified cell subsets; FIG. 2C stacked bar graphs show the proportion of subpopulations of myeloid cells from different foci; FIG. 2D is a graph of the first 3 differentially expressed genes in each myeloid cell subpopulation, bubble size representing the percentage of the number of cells expressing the gene in the subpopulation, and the color of the circles representing the average expression level of the gene in the subpopulation (from green to red); the left heat map of fig. 2E shows tissue preference of macrophage subpopulations estimated by Ro/E values, 0.2-straw Ro/E <0.8, +; 0.8-straw Ro/e <1, +; ro/e >1, + ++; FIG. 2F is a bar graph showing the ratio of Ro/e values for AC and AW in macrophage subpopulations, with the dashed line indicating a ratio equal to 1;

FIG. 3 is a profile of two populations of macrophage subpopulations; wherein, figure 3A is a heatmap showing gene set expression scores for classical polarized phenotypes M1 and M2; FIG. 3B is a violin diagram showing the expression distribution of the classical polarized phenotype M1 and M2 associated genes; (P <0.01, { P < 0.0001); FIG. 3C is a scatter plot of the differentially expressed genes mac _ C4_ PLIN2 and mac _ C6_ ISG15, the x-axis and y-axis representing the average amount of expression of a gene in mac _ C4_ PLIN2 and mac _ C6_ ISG15, respectively, with each dot representing a gene, with color annotations indicating that the gene passed our p-value and logFC threshold (black, no significance); FIG. 3D bar graph shows different up-regulation paths for mac _ c4_ PLIN2 and mac _ c6_ ISG15 in GSVA analysis; FIG. 3E shows TCGA OC data overall survival Kaplan Meier survival curves, layered according to high and low expression of mac _ c4_ PLIN2 and mac _ c6_ ISG15 signature genes; (P <0.01, P < 0.0001);

FIG. 4 is an AAM high expression PLIN2 analysis; wherein, the stacked violin diagram of fig. 4A shows the first 10 marker genes of mac _ c4_ PLIN 2; FIG. 4B is a Venn diagram showing common differential genes expressed in ascites macrophages in comparison to the other 4 lesions; FIG. 4C violin plots showing the distribution of PLIN2 expression among 5 focal macrophages; FIG. 4D is the proportion of mac _ c4_ PLIN2 in different lesions of the same patient; FIG. 4E is a t-SNE plot of 8 ascites sample cells in a validation dataset, each color representing one cell type; FIG. 4F is a projection of the t-SNE plot showing PLIN2 expression levels, and the violin plot on the right shows the expression levels of PLIN2 in different cell types; FIGS. 4G and 4H are multi-color immunofluorescent staining to verify mac _ c4_ PLIN2 in NO and OC on a scale bar representing 20 microns, wherein FIG. 4H is the quantitative experimental plot of FIG. 4G; FIGS. 4I and 4J are flow cytometry analyses verifying mac _ c4_ PLIN2 in AW and AC, where FIG. 4J is the quantitative experimental plot of FIG. 4I; FIG. 4K is a Kaplan-Meier plot showing stratification of OS and PFS by PLIN2 expression in two different groups of ovarian cancer patients. OS, overall survival, PFS, progression-free survival (means ± SEM in fig. 4H and J plot data); (P <0.05, P < 0.001, P < 0.0001);

FIG. 5 is an analysis of PLIN2+ AAM in promoting cross-body-cavity dissemination and ascites production; wherein, fig. 5A is a graph demonstrating the knockdown efficiency of PLIN2 at THP1 at protein level; FIG. 5B is a representative graph of a migration experiment in which SKOV3 was co-cultured indirectly with macrophages transiently transfected with PLIN 2-specific siRNA or control siRNA, with a scale bar representing 50 μm and a right histogram of quantitative statistics; FIG. 5C is a permeability assay using TRITC-dextran-spiked fluorescence detection of transient transfection of PLIN 2-specific siRNA or control siRNA by macrophages in a direct co-culture system; FIG. 5D is a qPCR method to verify the expression of HUVEC permeability-associated genes in the co-culture system; FIG. 5E is a workflow of animal experimental design; FIG. 5F is a representative photograph of ascites volume collected from mice inoculated 14 days after mixing HM-1 with macrophage stable knock-out PLIN2, with quantitative statistics of ascites on the right bar chart; FIG. 5G is a representative photograph of mesenteric metastases, the dashed circles indicate metastases, and the right histogram is a quantitative statistic of tumor weight (. Times.P.ltoreq.0.01,. Times.P.ltoreq.0.0001);

FIG. 6 is a diagram of AAM provenance analysis; wherein, FIG. 6A shows that ascites macrophages match macrophages in other tissues by expression similarity analysisA scaled chord plot, the right histogram showing the results of the same analysis performed on the same patient; figure 6B heat map shows GSVA analysis of macrophage subpopulation adhesion-associated pathways; FIG. 6C is a histogram overlay showing CAMs GSVA scores in mac _ C4_ PLIN2 and mac _ C6_ ISG 15; figure 6D is a GSEA graph showing the down-regulated pathway of ascites macrophages; FIG. 6E is a representative photograph and statistical plot of an adhesion experiment; FIG. 6F heat map shows adhesion gene expression in macrophage subpopulations; FIG. 6G is a heat map showing the expression of the corresponding ligand and receptor in E in a subset of T cells; FIG. 6H histograms show CD4+ T cell numbers adhering to macrophages; FIG. 6I is a diagram showing mac _ c4_ PLIN2 and CD4 ⁺ T and CD8 ⁺ T interaction. The size of the dots represents the P value, the color represents the interaction score (. Times.P.ltoreq.0.01,. Times.P.ltoreq.0.001,. Times.P.ltoreq.0.0001);

FIG. 7 is a set of single-cell transcriptome data for multiple lesions from ovarian cancer patients, wherein FIGS. 7A and 7B are graphs of Nandinger roses showing the number of cells from different lesions (A) and samples (B); FIG. 7C is the Upset chart showing the number of 5 lesion-specific high expression genes; FIGS. 7D and 7E are Seurat analysis of CD45+ immune cells from 19 specimens, each color representing lesion (D) and patient (E), respectively, and the right side the effect of Harmony batch removal on the left; FIG. 7F shows the expression of epithelial and fibroblast gene signatures that have been excluded in downstream analysis;

FIG. 8 is a tissue enrichment of a subpopulation of myeloid cells, wherein FIG. 8A is UMAP showing myeloid cells labeled by patient ID; FIGS. 8B-8F are UMAP plots showing different focal myeloid cell subsets;

FIG. 9 shows subpopulation characteristics mac _ c4_ PLIN2 and mac _ c6_ ISG15, where FIG. 9A is an assessment of macrophage activity and polarization, AW, peritoneal rinse from M1 and M2 phenotypes in different microenvironments; AC, ascites, NO normal ovaries; OC, primary tumor; ML, transfer cooker; FIG. 9B is a map of mac _ c4_ PLIN2 and mac _ c6_ ISG 15M 1 and M2 gene expression; FIG. 9C is a violin diagram of the genes significantly different for mac _ C4_ PLIN2 and mac _ C6_ ISG 15; FIG. 9D is mac _ c4_ PLIN2 and mac _ c6_ ISG15 differentially expressed gene GO analysis; (P is less than or equal to 0.0001);

FIG. 10 is the expression of macrophage PLIN2 in different foci, wherein FIGS. 10A-10D are differentially expressed genes of macrophages from normal ovary, primary foci, metastases and normal peritoneal irrigation in ascites, with the dotted line indicating the significance threshold (logFC >1; -log10 (adjusted p value) > 0); FIG. 10E set of bar graphs showing macrophage PLIN2 expression in different foci of the same patient; FIG. 10FUMAP panel shows marker gene expression for the identified cell types, (expression levels are indicated in blue to red); FIG. 10G is a collection of cells from ascites and peritoneal washes, using the CD45+ CD68+ PLIN2+ flow analysis circle strategy to identify PLIN2+ AAMs; (P <0.05, P < 0.001, P < 0.0001);

FIG. 11 is an analysis of the effect of PLIN2 in mouse macrophages, wherein FIG. 11A is a qPCR validation of the reduction of PLIN2 expression in THP-1 macrophages transiently transfected with PLIN2 specific siRNA (sipIN 2-01, sipIN2-02) or control siRNA (NC); FIG. 11B is the relative mRNA expression levels of HUVEC permeability-associated genes in an indirect co-culture system; figure 11C is validation of knock-down efficiency of Raw264.7 by qPCR; FIG. 11D shows the determination of permeability by TRITC-dextran tracing fluorescence in a direct co-culture system; FIG. 11E is a representation of a migration experiment in which ID8 was indirectly co-cultured with macrophages transfected with plin 2-specific siRNA or control siRNA, with scales representing 50 μm; FIG. 11F shows the proliferation rate of shPLIN2 macrophages relative to the control group measured by CCK 8; FIG. 11G is the relative change in M1 and M2 labeling following PLIN2 knockdown in Raw 264.7; (P <0.05, P < 0.001, P < 0.0001);

FIG. 12 is a characterization of T cell subsets in ovarian cancer, wherein FIG. 12A is the origin of myeloid cell subsets enriched in ascites; figure 12B is validation of stable knock-down of cell knock-down efficiency at protein level; fig. 12C is a sourta analysis of T cells from ovarian cancer tumor microenvironment, with UMAP color encoded by subpopulations; figure 12D is a heatmap of the first 20 up-regulated genes of all subpopulations in T cells; figure 12E is a heatmap of functional gene expression in T cell subpopulations; figure 12F dot plot shows CD4+ T cell purity efficiency by magnetic bead sorting.

Detailed Description

The technical solutions of the present invention will be described clearly and completely with reference to the following embodiments of the present invention, and it should be understood that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.

The experimental methods used in the examples of the present invention are all conventional methods unless otherwise specified; the materials, reagents and the like used are, unless otherwise specified, commercially available reagents and materials.

Procedure of experiment

1. Sampling specimen

The present invention recruited 15 patients diagnosed with HGSOC and 10 patients diagnosed with gynecological tumors. 19 samples, including primary foci, non-neoplastic ovarian tissue, metastases, ascites, and peritoneal washes were collected from 10 of the patients and characterized with scRNA-seq, with 3 patients having multiple matched foci. Samples from the remaining 15 patients were subjected to flow and immunofluorescence analysis.

All samples were taken from the fifth hospital affiliated with Zhongshan university and all patients in this study signed written informed consent for sample collection and data analysis.

2. Experimental methods

2.1 cell lines and cell cultures

HUVECs were obtained from ATCC, mouse EC line (C166) and mouse macrophage line (RAW 264.7) and human macrophage line (THP-1) were donated by professor university of middle school biology academy of sciences \37021dong. The mouse ovarian cancer cell line OV2944-HM-1 (HM-1) was given by Nelson Teng from Stanford university. Cells were identified by short tandem repeat analysis and confirmed to be mycoplasma free prior to use. HUVECs were cultured in ECM (ScienCell) containing 5% FBS,1% EC growth supplement (ECGS), 100U/mL penicillin and 100 μ g/mL streptomycin. THP-1 cells and HM-1 cells were cultured in RPMI 1640 medium (Invitrogen) containing 10% FBS,100U/mL penicillin and 100. Mu.g/mL streptomycin. C166 and RAW264.7 cells were cultured in DMEM (Invitrogen) containing 10% FBS,100U/mL penicillin and 100. Mu.g/mL streptomycin. Culturing all cells5% of CO cultured at 37 deg.C ₂ The humidifying incubator of (1).

2.2 tissue dissociation

After surgical resection, the surgical specimens were transported from the operating room to the laboratory in ice boxes within 10 minutes. Tissue samples were cleaned with 1 x phosphate buffered saline (Invitrogen) and cut into small pieces. The tissue was digested enzymatically in a tissue homogenizer with a heater (MiltenyiBiotec) for about 1 hour. The tissue mass was gently dissociated on a 70 μm cell filter (BD) in 1 × PBS using a 20mL syringe plunger until a homogeneous cell suspension was obtained. The suspension cells were then centrifuged at 450g,4 ℃ for 5 minutes. Cell deposits were collected for subsequent operations. The peritoneal irrigation fluid and malignant ascites specimens were transported in the same manner as described above and centrifuged directly at 450g,4 ℃ for 5 minutes. The supernatant was removed to collect the cell pellet for subsequent manipulation.

2.3 CD45 ⁺ Immune cell isolation

Single cell suspensions from tumor tissues and fluids were incubated with Fc receptor blocking reagents (Biolegend). The cells were then stained with PE/Cyanine7 anti-human CD45 antibody (Biolegend) for 30 min at 4 ℃ in the dark. Isotype antibody staining was used to control background. Dead cells were excluded by the annexin V-FITC apoptosis detection kit (KeyGEN). Targeting live CD45 Using BD FACScaria FUSION ⁺ The cells were sorted.

2.4 Single cell library preparation and sequencing

After Fluorescence Activated Cell Sorting (FACS), the concentration of CD45 single cell suspension was adjusted to 300-350 cells/μ L. Cells were loaded between 7,000 and 10,000 cells/Chip location using the chromosome Next GEM Single Cell 3' library, gel Bead & i7 Multiplex Kit and Chip G Kit (10X Genomics, kit v 3) according to the instructions provided by the reagent supplier. All subsequent steps are performed according to standard protocols. The purified library was analyzed by illumina Novaseq with 150bp paired end reads.

2.5 Single cell RNA-seq data analysis

The single cell transcriptome data for each sample was aligned and quantified using cellanger software for GRCh38 human reference genomes. The Scrublet python packet was then used to kick out the double cells in an attempt to identify the true single cell from each droplet. To eliminate the batch effect, we followed the method previously described in the Harmony R package. Data are subject to strict quality control according to three indexes: (1) total UMI counts per cell; (2) the detected gene count for each cell; (3) ratio of mitochondrial Gene counts. If the number of genes detected is below 600 per cell and above 4000 per cell, cells are filtered out and, in addition, low quality cells are removed from the data set, i.e., the percentage of mitochondrial gene counts is above 25%. A total of 52948 cells were retained for downstream analysis. The standardized single cell analysis pipeline performed in the Seurat package was then followed. First, we identified 2000 highly variable genes for Principal Component Analysis (PCA) and extracted 20 PCs to identify the major immune cell types using the Luwen clustering algorithm (unified flow approximation and projection (UMAP). The process was repeated in myeloid cells and T cells to break the data down into smaller subpopulations and further explore the salient features.Wilcoxon Rank Sum test the FindMarkers function by Seurat was used for differential expression analysis and to extract statistically significant markers for the subpopulations according to an adjusted p-value threshold of 0.05.

The software version collected data using Cell range software v3.1.0 and analyzed using rv.3.6.3. And the following packages and versions for analysis in R: seurat v3.2.3, harmony v1.0.0, scrublet v0.2.3, GSVA v1.34.0, limma v3.42.3, cluster profiler v3.14.3. Regulations were made using the following software packages and versions in R: RColorBrewer v1.1-2, pheatmap v1.0.12, ggplot v3.3.5, ggpubr v0.4.0, ggsignif v0.6.3, complexHeatmap v2.2.0 and paletteer v1.4.0.

2.6 analysis of cell-cell interaction

We analyzed the interaction of mac _ c4_ PLIN2 and T cells by using a publicly available ligand receptor pair CellPhoneDB. The interaction score is calculated by calculating the average amount of ligand expressed in a cell type and receptor expressed in the corresponding cell type. Statistical analysis of significance was based on the permutation test (1000 times). We performed a biologically significant screen of the interaction results based on the interaction score and P-value.

2.7 immunofluorescent staining

Ovarian cancer tissues and non-malignant ovarian sections were obtained from the pathologist and each tissue section was dewaxed, rehydrated, immersed in an antigen retrieval solution, and boiled at 100 ℃ for 10 minutes. Each tissue section was then infiltrated with 0.3% Triton X-100 (MEIJUN) for 15 minutes and blocked with 10% goat serum (ZSBG-BIO) for 1 hour. The tissue sections were then incubated overnight at 4 ℃ with primary antibody: ADFP (Abcam), CD68 (Biolegend), was then incubated with the appropriate secondary antibody for 1 hour at room temperature and DAPI stained in the dark. Images were acquired using a confocal microscope (zeiss 880) and a 10X objective.

2.8 flow cytometry

Flow cytometry analyses were carried out using samples from 15 other patients, first of all cell pellets were obtained according to the centrifugation step of the liquid samples already described above. Nonspecific antibody binding was prevented by incubation with Fc block (Biolegend). The cells were then incubated with the following antibody mixtures: PE/Cy7-CD45 (Biolegend, 1, 1000), APC/Cy7-CD68 (Biolegend, 1. After staining of the cell membrane, the cells were fixed and permeabilized (FOXP 3 Fix/Perm kit, biolegend) and stained with Alexa Fluor 647-PLIN2 antibody. Fluorescence was measured by cytoflexLX (beckmann coulter) and data were analyzed using Flowjo v 10.4.0.

2.9 Western blot

The protein concentration of the cell lysates was determined by BCA (Beyotime) and separated on 10% SDS-PAGE and transferred to PVDF membrane. The strips were incubated overnight at 4 ℃ with primary anti-ADFP (Abcam, 1, 1000) and then with secondary antibody GAPDH (Proteintech, 1 10000) at room temperature for 1 hour. GAPDH was used as an internal control.

2.10 RNA isolation and quantitative real-time fluorescent quantitative PCR

Total RNA was extracted using Total RNA kit I (omega). First strand cDNA was synthesized from 2. Mu.g total RNA (Vazyme). Amplification and detection were tested using ABI QuantStudio 7 Flex. The quantitative PCR conditions were as follows: 95 ℃ for 10 seconds, then 40 cycles, 95 ℃ for 10 seconds, 60 ℃ for 10 seconds, and finally 95 ℃ for 15 seconds, 60 ℃ for 30 seconds, 95 ℃ for 15 seconds, the relative expression of specific mRNA was analyzed by the comparative cycle threshold (Ct) method.

2.11 Small RNA interference

Raw 246.7 was seeded onto 6-well plates and transfected when 40-50% confluency was achieved. Cells were transfected with PLIN2 siRNA using Lipofectamine 3000 and then assayed 48 hours after transfection, according to the manufacturer's instructions.

2.12 Permeability determination

Mixing C166 (1X 10) ⁵ Individual cells) were seeded in a 24-well Transwell upper chamber and cultured for 24 hours until the confluence point was reached. Raw 246.7 with PLIN2 knockdown was seeded on endothelial cell monolayers for co-culture for 24 hours. After 24 hours of incubation, two washes were performed using 100 μ LPBS. Then 500. Mu.L of LPBS was added to the lower layer and 200. Mu.L of PBS containing TRITC-dextran (2 mg/mLSigma-Aldrich) was added to the upper layer. The entire cell was incubated at 37 ℃ for 5 hours. The PBS in the lower compartment was collected and analyzed in a fluorescence spectrophotometer (Synergy HTX, BIOTEK) with excitation and emission wavelengths of 530nm and 540nm, respectively.

2.13 transfection and establishment of Stable cell lines

PLIN2 stable knockdown in RAW264.7 cells was established by infecting the cells with lentiviral vectors purchased from GeneCopoeia. RAW264.7 cells were infected with lentivirus using 0.8. Mu.g/mL polystyrene (MOI 25). Stable cell lines were obtained by screening for about 2 weeks using 5. Mu.g/mL puromycin (Biofrox) and observing the virus infection efficiency by an inverted fluorescence microscope. The efficiency of RNA interference was assessed by qRT-PCR and western blot.

2.14 cell proliferation assay

1×10 ³ Individual cells were seeded in 96-well plates. Then, 10 μ LCCK-8 (KeyGEN) solution was added to each well and incubated for 3 hours to assess cell proliferation. From day 1 to day 4, measurements were made with a fluorescence spectrophotometer (Synergy HTX, BIOTEK) at OD 450Absorbance of each well.

2.15 cell migration assay

Raw 246.7/THP-1 with PLIN2 knockdown was seeded in the lower chamber of a Transwell plate and ID8/SKOV3 cells (1X 10) ⁴ ) The cells were seeded in a Transwell upper chamber with a pore size of 8 μm (Corning) and cultured for 48 hours. Cells that migrated to the lower surface were stained with crystal violet (Beyotime) and quantified by counting 3 randomly selected fields at x 200 magnification.

2.16 Extraction of CD4+ T cells

And (3) separating the mouse spleen CD4+ T cells by adopting a magnetic bead negative selection method. MojoCoort is used according to the manufacturer's instructions ^TM The mouse CD 4T cell isolation kit (Biolegend) enriches and purifies target cells. CD4+ T cells were 90% pure when cultured in RPMI 1640 medium containing CD3 (5. Mu.g/mL Biolegged), CD28 (1. Mu.g/mL Biolegged), and IL2 (10 ng/mL PeproTech).

2.17 adhesion test

The 96-well plates were coated with fibronectin (ScienCell, 10 μ g/mL,80 μ L/well) at 37 degrees for 2 hours, then washed 2 times with PBS. Use 1% BSA (Beyotime) at 37 degrees in CO ₂ The incubator was blocked for 1 hour and washed 2 times with PBS. PLIN2 stably knockdown cell lines and control cells were counted to 5X10 ⁵ one/mL and added to the well (100. Mu.L/well). CO at 37 degree ₂ After incubation in the incubator for 45 minutes, the cells were washed 2 times with PBS. The number of cells adhering to the plate was counted by observing the fluorescence of GFP by an inverted fluorescence microscope.

2.18 animal experiments

Animals were housed in SPF-grade animal houses in the biomedical imaging stress laboratory in guangdong province. B6C3F1 mice were used as HM-1 animal models, aged 6 to 8 weeks, generated as crosses between male C3H mice and female C57BL/6 mice. Mice were divided into 2 groups according to experimental design. PLIN2 knockdown and control macrophages and HM1 cells (resuspended 10 each in 100. Mu.L PBS, respectively ⁶ Individual cells) are mixed and injected into the peritoneal cavity. Mice were weighed every 2 days and observed for behavior, and ascites were collected after 2 weeks.

2.19 statistical analysis

For statistics related to scRNA-seq data analysis, P values for the function "FindAllMarkers" in Seurat and when comparing proprietary scores, we corrected with Bonferroni using unpaired Wilcoxon rank-sum test. To compare the proportion of subpopulations in different lesions, we used an independent t-test. Survival analysis was performed by Kaplan-Meier curve. Other data are expressed as mean + SEM, with P values significant when P < 0.05.

Analysis of Experimental results

1. Immune microenvironment of multiple foci in HGSOC under single cell resolution

To understand the ascites-specific immune microenvironment in OCs, the present invention performed single cell RNA sequencing analysis in multiple lesions and corresponding controls (see fig. 1A). A total of 13 malignant and 6 non-malignant samples were obtained from 7 primary treatment high-grade serous ovarian cancer (HGSOC) patients and 6 uterine fibroids and cervical high-grade squamous intraepithelial neoplasia patients (non-malignant controls) (see fig. 1B). It is noted that at least two different lesions are derived from the same individual to further illustrate the heterogeneity of TME. Fresh tissue was rapidly made into single cell suspensions and CD45 was collected by Fluorescence Activated Cell Sorting (FACS) ⁺ An immune cell. After high quality filtration and twin removal, we obtained 52948 cells from primary tumor (OC), metastatic Lesion (ML), ascites (AC), non-malignant ovary (NO) and non-malignant peritoneal lavage (AW) (see fig. 7A and 7B).

First, the present invention performed a pseudo bulk Principal Component Analysis (PCA) to assess the similarity between different lesions (see fig. 1C). It distinguishes solid tissues from body fluids and both have their unique set of highly expressed genes (fig. 7C). The present invention uses the Harmony algorithm to minimize the batch effect of transcriptomics data from these 19 samples (fig. 7D and 7E). To define the cellular composition of these tissues, we performed unsupervised map-based clustering (see fig. 1D) and analysis of differentially expressed genes, and annotated these major cell populations by mean expression of known gene sets, including myeloid lineage cells (CD 68, CD1E, LAMP3, ILR 3A), T cells (CD 2, CD3D, CD3E, CD 3G), natural Killer (NK) cells (NKG 7) and B cells (MS 4A1, CD79A, MZB1, IGHA 1) (see fig. 1E). Epithelial and fibroblasts were removed from the downstream analysis (fig. 7F).

The present invention also assesses differences in cell type distribution in different samples. Generally, myeloid lineage cells are more enriched in body fluids than in solid tissues. It is noted that the composition of immune cell types is more heterogeneous in malignant tissue (OC, ML, AC) than in non-malignant tissue (NO, AW) (fig. 1F and fig. 1G). These data depict different immune component compositions in different lesion environments.

2. Identification of novel macrophage subgroup enriched in malignant ascites

Because the myeloid population is the highest in the fluid, the present invention places emphasis on the myeloid cells. First, we extracted myeloid lineage cells from different immune cells, totaling 19336 cells, and divided them into three major lineages (dendritic cells, macrophages and monocytes) by known marker genes (see fig. 2A and 2B). Consistent with previous reports, macrophages were the most prominent component of all tissues we analyzed (see fig. 2C).

Furthermore, the myeloid lineage cell population can be further divided into multiple subpopulations based on differential genes. We first identified four different DC subsets, including three classical DC subsets and one plasmacytoid dendritic cell population. Further unsupervised clustering of macrophages/monocytes yielded 11 subpopulations, including 7 macrophage subpopulations and 4 monocyte subpopulations and named by their specific genetic signature (see fig. 2D). We also calculated the tissue distribution of the myeloid cell subsets in different tissues, indicating that different myeloid cell subsets have different tissue preferences (fig. 8A-8F). We used the ratio of observed cell number to expected cell number (Ro/E) for each subpopulation that has been reported to assess each foci-enriched macrophage subpopulation (see fig. 2E). mac _ c6_ ISG15 is enriched in all malignant tissues, including OC, ML and AC, while mac _ c3_ FN1 is enriched in non-malignant tissues (NO and AW). We found that a new macrophage subset mac _ c4_ PLIN2 was mainly enriched in ascites and therefore we named it as "ascites-associated macrophages" (see FIG. 2F).

3. Ascites-associated macrophage subgroup with M2 phenotype and tumor-promoting effect

Macrophage polarization states can be quite complex, with two extremes, classical activation (M1) and surrogate activation (M2). First, the present invention explores the overall polarization state of macrophages in different lesions, indicating that M2-polarized macrophages account for approximately 50-70% of all macrophages in solid tissues, and are present in slightly higher proportions in malignant tissues. However, more than 90% of macrophages in body fluids are biased towards M2 (see fig. 9A). We examined the polarization state of 7 macrophage subsets and found that most of them were mixed polarized (see fig. 3A). mac _ c4_ PLIN2 and the other 5 subsets shifted towards the M2 phenotype, while mac _ c6_ ISG15 expressed more M1-phenotype macrophage markers (see FIG. 3B, FIG. 9B). Given the enrichment of mac _ c4_ PLIN2 in ascites, it may have a function similar to tumor-associated macrophages to promote cancer progression. In comparing differentially expressed genes, we noted that mac _ C4_ PLIN2 highly expressed CXCL2, CXCL8, EREG and VCAN, which are widely reported to be associated with tumor progression and metastasis, while mac _ C6_ ISG15 highly expressed pro-inflammatory cytokines (CXCL 9, CXCL10, CXCL 11) and interferon-induced proteins may inhibit tumor progression (see FIG. 3C, FIGS. 9C and 9D). The present invention performed a genotypic variation analysis (GSVA) to assess the expression of the enriched pathway in two TAM subpopulations and revealed a high enrichment of pathways such as Epithelial Mesenchymal Transition (EMT), angiogenic and hypoxic pathways in mac _ c4_ PLIN2, along with a significant upregulation of inflammatory responses, complement system, interferon (IFN) - α and IFN- γ response-related pathways in mac _ c6_ ISG15 (see fig. 3D). Survival analysis showed that the prognosis for these two subpopulations was opposite (see figure 3E). These results indicate that the newly discovered mac _ c4_ PLIN2 subset is biased towards M2 polarization and likely promotes tumor progression.

4. Increased PLIN2 expression in AAM

To identify genes characteristic of AAM, the present invention performs a differential expression gene analysis between AAM and other macrophage subsets. Of the first ten signature genes of AAM, both SPP1 and PLIN2 genes were highly expressed in subpopulations enriched in ascites (see FIG. 4A). Since SPP1 has been reported to play a tumor promoting role in macrophages, we focused on PLIN2, which encodes a fat differentiation-associated protein and is involved in macrophage to foam cell conversion in plaque formation.

To test the specificity of PLIN2 expression in AAM, the present invention compared the highly expressed gene in ascites macrophages with the highly expressed gene of macrophages in other lesions, PLIN2 being the only specific gene representing AAM (see FIG. 4B, FIGS. 10A-10D). In addition, PLIN2 expression in malignant ascites was higher than any other lesions (see FIG. 4C). Then, we used data from patients with multiple matching lesions to verify enrichment of mac _ c4_ PLIN2 and high expression of PLIN2 in the ascites, and obtained the same results (see FIG. 4D, FIG. 10E). In addition, we also used a validation dataset (GSE 146026) to support our findings, which included 8 ascites samples from 6 patients. Based on the data from the article, we divided the cells in the ascites into 6 populations (fig. 10F) and found that PLIN2 is mainly expressed in macrophages between the immune cells, consistent with our results (see fig. 4E and 4F).

To confirm PLIN2 expression in clinical samples. We performed multicolor Immunofluorescence (IF) staining of tumor sections of OC patients and non-malignant controls and showed more PLIN2 macrophages in OC tissues (see fig. 4G and 4H). Meanwhile, we found that PLIN2 macrophages are more abundant in AC than AW (see FIGS. 4I and 4J, FIG. 10G). More importantly, high expression of PLIN2 could predict poor prognosis in OC patients, which was validated in two independent datasets (GSE 30161 and GSE 15622) (FIG. 4K). Taken together, the results indicate that PLIN2 is up-regulated in malignant macrophages, especially ascites-associated macrophages.

5、PLIN2 ⁺ AAM promotes peritoneal transmission and ascites production

To investigate the function of PLIN2, we first knocked down the expression level of PLIN2 in THP1 macrophages (see FIG. 5A, FIG. 11A). We then attempted to confirm that PLIN2 could promote transperitoneal transmission. We cultured human ovarian cancer cells in Transwell supernatants using macrophages with low PLIN2 expression. After 24 hours, cells that migrated to the bottom of the chamber were counted. The data show that there were much fewer cells migrating to the other side of the chamber in the PLIN2 knockdown group relative to the control group (see FIG. 5B).

Since TAMs can promote vascular permeability, we speculate that PLIN2 AAM may have a similar effect. To confirm that AAM can promote ascites production, we also co-cultured directly and indirectly with a monolayer of endothelial cells HUVEC in the upper well of Transwell using the PLIN2 knockdown macrophages, and barrier function was tested in the upper chamber with TRITC fluorochrome-labeled dextran added, then collected in the lower chamber and quantified. Much less glucan passage was observed in the PLIN2 knockdown group in the direct co-culture system (see FIG. 5C). To validate this result at the molecular level, we evaluated mRNA levels of molecules associated with vascular integrity, such as the adhesion connexin VE-cadherin and the tight junction protein ZO-1. We noted a significant increase in the levels of these molecules in endothelial cells cultured directly with PLIN2 knockdown macrophages for 24 hours (see FIG. 5D). The above experiment was also repeated with mouse-derived cells (FIGS. 11B-11E). To further confirm the role of PLIN2 in vivo, we constructed a mouse OC model, injected intraperitoneally with HM1 cells (1 x10 < Lambda > 6/mouse), and then divided into two groups according to the schematic (see FIG. 5E). We observed a significant reduction in ascites production in the PLIN2 knockout animals (see FIG. 5F), with a reduction in mesenteric tumor burden (see FIG. 5G).

The present invention also investigated the potential effect of knockdown PLIN2 on macrophage proliferation and polarization. The data results show that while it had no effect on cell proliferation (see fig. 11F), M1 markers, such as IL6, NOS2 and CD80, were significantly enhanced in polarization with a tendency to decrease M2 marker levels, such as VEGFA and CD206 (fig. 11G). Taken together, all data suggest that highly expressed AAM of PLIN2 may play a role in promoting transperitoneal transmission and ascites production.

6. Accumulation of AAM in ascites due to low adhesive capacity

To clarify the origin of AAM, we compared it transcriptomic with cells of other tissues and found that AAM maps predominantly to cells from primary and metastatic foci (45.6%, 40.6% and 13.8% respectively toFrom OC, ML, AW; see fig. 6A). Data from two patients with multiple matching lesions also showed that AAM is largely similar to cells from metastatic sites (see fig. 6A). The mac _ c4_ PLIN2 subpopulation was more similar to cells from malignant solid tissue than other ascites-enriched myeloid lineage cell populations (FIG. 12A). In the process of inferring functional differences between AAM and other macrophage subsets, we noted significant down-regulation of cell adhesion molecule-associated pathways in AAM (see fig. 6B and C), suggesting that low levels of cell adhesion molecules of AAM may be a potential factor in promoting accumulation of ascites in AAM. This finding also yielded consistent results in the validation dataset (see fig. 6D). We verified the adhesion ability of the PLIN2 knockout macrophages. Adhesion experiments showed that macrophage adhesion to extracellular matrix fibronectin was significantly enhanced after PLIN2 knockdown (see FIG. 6E). ITGA5, ITGB8, ICAM5, ITGB1 were expressed in all CAM genes in the reverse direction, and were highly expressed in mac _ c4_ PLIN2 (see FIG. 6F). The results showed that, due to some low-expressed CAM, these AAMs accumulated in ascites, while it still functioned as specific CAM like ITGA5, ITGB1, ICAM5 and ITGB 8. To clarify their function, we matched their corresponding ligands/receptors in various T subsets (see fig. 6G, fig. 12C-12E), which were found to be predominantly linked to CD4+ T lymphocytes, especially CD4_ C3_ CXCL13 and CD4_ C4_ IKZF2 for T cells, respectively _fh And T _reg (FIG. 6G). In contrast, mac _ c6_ ISG15 enriched CAM mainly matched CD8+ T lymphocytes. To verify the affinity of PLIN2+ macrophages for CD4+ T cells, we extracted CD4+ T cells from mouse spleen and co-cultured them with PLIN2 knockout or knockout-free GFP-labeled macrophages (FIG. 12F). The adherence to CD4+ T cells was significantly reduced in the PLIN2 knockout group compared to the control group (see fig. 6H). Furthermore, the interaction between mac _ c4_ PLIN2 and CD4+ T lymphocytes showed a stronger immunosuppressive phenotype in malignant ascites, exemplified by the enhancement of CTLA4 and LGALS9 mediated responses (see FIG. 6I). Taken together, these results show that the mac _ c4_ PLIN2 subset is more prone to interact with CD4+ T lymphocytes, thereby promoting immunosuppression and tumorigenesis.

Claims (7)

1. Application of ascites-related macrophage subgroup mac _ c4_ PLIN2 in developing and screening ovarian cancer functional products, which is characterized in that the ascites-related macrophage subgroup mac _ c4_ PLIN2 is characterized in that the fat differentiation-related protein PLIN2 is highly expressed.

2. The use of claim 1, wherein the ascites-associated macrophage subset mac _ c4_ PLIN2 has an M2 phenotype and is positively correlated with the production and metastasis of ascites from ovarian cancer.

3. Application of a functional product with an inhibitory effect on PLIN2 gene or an expression product thereof in preparing a product for preventing or/and treating ovarian cancer.

Application of the PLIN2 gene or the expression product thereof in developing and screening ovarian cancer functional products, wherein the functional products have an inhibiting effect on the PLIN2 gene or the expression product thereof.

5. Use according to claim 3 or 4, wherein the functional product is for reducing the production and/or metastasis of ascites from ovarian cancer.

6. The use according to claim 5, wherein the functional product comprises: one or more of a PLIN2 nucleic acid inhibitor, a PLIN2 protein inhibitor, an immune-related cell deficient or silenced in a PLIN2 gene, a differentiated cell thereof, or a recombinant construct thereof.

7. Use according to claim 6, wherein the functional product comprises any of:

(i) dsRNA, shRNA and micro RNA which take PLIN2 or PLIN2 transcripts as target sequences and can inhibit the expression of PLIN2 gene expression products or gene transcription;

(ii) (ii) capable of expressing or forming the dsRNA, shRNA, microRNA construct of (i);

(iii) Constructs containing PLIN2, or the PLIN2 complement, and capable of forming interfering molecules that inhibit the expression of the PLIN2 gene expression product or gene transcription upon transfer into the body;

(iv) An immune-related cell, differentiated cell or construct thereof following suppression or knock-out of the PLIN2 gene sequence.

Images