A method for detecting mesenchymal stem cells of different tissue sources, passages and culture conditions based on single cell sequencing

Abstract

The application provides a method for detecting mesenchymal stem cells (MSCs) of different tissue sources, generations and culture conditions based on single-cell sequencing, and belongs to the technical field of stem cells. The single-cell transcriptional atlas of MSCs is detected by single-cell sequencing technology, the functional gene expression level is obtained through bioinformatics analysis, and the enrichment items are obtained through GO and KEGG analysis. In addition, the functional evaluation model of MSCs is created based on the OCLR algorithm to obtain the functional score. The heterogeneity of MSCs in different groups is obtained by combining the three indexes. The method provided by the application can detect MSCs from multiple different classification perspectives, and has the characteristics of comprehensive detection, high efficiency, high accuracy of results, strong readability and the like.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of stem cells, and particularly relates to a method for detecting mesenchymal stem cells of different tissue sources, passages and culture conditions based on single-cell sequencing. BACKGROUND

[0002] Mesenchymal stem cells (MSCs) are a kind of multipotent stem cells with self-renewal and multi-directional differentiation potential, which can differentiate into bone cells, adipocytes and chondrocytes, and can be derived from various tissues such as adult (fat, peripheral blood, bone marrow, synovial membrane, dental pulp, etc.) and neonatal tissues (amniotic membrane, umbilical cord, chorionic membrane, decidua, amniotic fluid, etc.). Mesenchymal stem cells (MSCs) can regulate the inflammatory environment, support the development and maintenance of neurons, and promote angiogenesis and wound healing, etc., which mainly produce effects through the following mechanisms: 1) involving paracrine activity of secreted proteins / peptides and hormones; 2) transferring mitochondria through tunnel nanotubes or microvesicles; 3) transferring extracellular vesicles (EVs) or microvesicles containing ribonucleic acid (RNA) and other molecules. These studies have greatly promoted the application of cell therapy and developed cell-derived therapies for regenerative medicine. At the same time, the development trend of stem cell clinical translation application is very fast, and the number of clinical experiment registrations related to cell therapy shows exponential growth. As an important part of cell therapy, stem cells play an important role in organ repair and tissue regeneration and have broad application prospects. However, there are few studies on the heterogeneity between MSCs of different sources, passages and culture conditions, which limits the development of basic research and clinical application of MSCs.

[0003] The previous microarray hybridization technology and next-generation sequencing (NGS) technology known as RNA sequencing (RNA-seq) have been widely applied. Cell-based RNA-seq provides a large amount of information, which promotes biomedical discovery and innovation, but this technology usually performs overall evaluation on samples containing thousands to millions of cells, which hinders direct evaluation of cells as the basic unit of biology. Single-cell RNA sequencing (scRNA-seq) can describe RNA molecules in a single cell at a high resolution on a genomic scale, providing researchers and clinicians with a powerful method as a technology and tool needed for research.

[0004] The current method for detecting MSCs mainly detects the contents of the "three categories": multi-directional differentiation potential, represented by osteogenic, adipogenic and chondrogenic induction differentiation function; immune regulation function, represented by total lymphocyte proliferation inhibition, suppression of Th1 / Th17 cell subsets and promotion of Treg cell subsets, suppression of inflammatory factor tumor necrosis factor-α (TNF-d) release and IDO and other immune regulatory molecules; and tissue regeneration function, represented by anti-apoptosis, pro-angiogenesis, secretion of various active factors (such as HGF and IL6) that promote tissue regeneration, and promotion of M2 macrophage polarization. Various instruments including RT-qPCR, ELISA, flow cytometry, etc. are used. The existing MSCs detection methods and technologies have the following shortcomings: 1. In the test of osteogenic, adipogenic and chondrogenic differentiation, oil red O staining, alizarin red staining or alizarin blue staining is required after 14 days (osteogenic and adipogenic) or 21 days (chondrogenic) of induction differentiation, which has a large time span, and the totipotency of MSCs also includes the differentiation potential of fibroblasts, neurons and other types. The current detection scheme does not detect other differentiation potential, and at the same time, this scheme cannot accurately evaluate the differentiation potential of MSCs, i.e. cannot evaluate the differentiation potential heterogeneity of MSCs from different tissues. In addition, RT-qPCR is used to detect TNFR1, 1DO, VEGF and ELISA to detect CD73, CD90, CD105, HGF, IL6 and other factors, which involve the immune regulation potential and tissue regeneration potential of MSCs, but RT-qPCR can only detect a small number of factors and cannot comprehensively detect the heterogeneity of MSCs from different tissues. And before detection, lymphocyte extraction, cell co-culture and other tests are required, and a large amount of time is required for a large number of experimenters. SUMMARY

[0005] Therefore, the purpose of the present application is to provide a method for detecting mesenchymal stem cells from different tissues, generations and culture conditions based on single cell sequencing, which can detect MSCs from multiple different classification perspectives, has the characteristics of comprehensive detection, high efficiency, high accuracy of results and strong readability, etc.

[0006] The present application provides a method for detecting the biological function of mesenchymal stem cells based on single cell sequencing, comprising the following steps:

[0007] Single cell sequencing is performed on the single cell suspension of mesenchymal stem cells to obtain mesenchymal stem cell transcriptome data;

[0008] After quality control processing, the mesenchymal stem cell transcriptome data is subjected to principal component analysis dimension reduction to obtain dimension reduction data;

[0009] grouping the dimensionality reduction data, determining marker genes in different groups, performing GO and KEGG enrichment analysis based on the marker genes in the different groups, obtaining enrichment entries;

[0010] comparing the average expression levels of the biological function related genes of each mesenchymal stem cell in different groups;

[0011] constructing a function evaluation model of mesenchymal stem cells based on human stem cells and embryo transcriptome data using a linear regression algorithm, inputting each single cell transcriptome data in different groups into the function evaluation model, and calculating the biological function scores of each mesenchymal stem cell;

[0012] comprehensive analysis of the average expression levels of the function related genes, the enrichment entries and the various function scores in each group to determine the biological function of the mesenchymal stem cells.

[0013] Preferably, the grouping method is based on clustering analysis results, tissue source, passage and culture conditions.

[0014] Preferably, the clustering analysis result is based on the execution of the "FindClusters" function with a resolution of 0.2, and the first 20 principal components are used to define cell identity.

[0015] Preferably, the tissue source includes amniotic membrane source, umbilical cord source, chorion source and adipose tissue source.

[0016] Preferably, the culture conditions include conventional oxygen culture conditions and low oxygen concentration culture conditions.

[0017] The oxygen concentration in the conventional oxygen culture condition is 21%.

[0018] The oxygen concentration in the low oxygen concentration culture condition is 1%.

[0019] Preferably, the quality control method is to exclude cells with the lowest 5% and the highest 5% of detected gene numbers and more than 15% of mitochondrial genes.

[0020] Preferably, the marker genes are genes with |avg_log2FC|>0.25 and p_val<0.05 and min.pct>0.1.

[0021] Preferably, the biological function related genes include at least two of the following: genes related to evaluating tissue repair potential, genes related to evaluating MSCs differentiation potential, genes related to evaluating proliferation potential, genes related to evaluating exosome transport capacity and genes related to evaluating cell stemness.

[0022] Preferably, the genes related to the potential of tissue repair include the following genes: MMP2, MMP9, CXCR2, CXCR4, CXCR5, CXCR6 and CDH2.

[0023] The genes related to the potential of MSCs differentiation include the following genes: ACTA2, VEGF, FBLN5, DCN, UBE2C, CTGF and POSTN.

[0024] The genes related to the potential of proliferation include the following genes: PLK1, CCNE1, MKI67, CCNB1, E2F1, FOXM1, MCM3, MCM4 and TOP2A.

[0025] The genes related to the potential of extracellular vesicle transport include the LIN7A gene.

[0026] The genes related to the potential of cell stemness include the following genes: SOX2, LIFR and MEST.

[0027] Preferably, the method further comprises performing pseudotemporal analysis to evaluate the differentiation of mesenchymal stem cells.

[0028] Preferably, the method of pseudotemporal analysis comprises a Monocle 3-based process, combining pseudotemporal graphs and selecting the starting and ending positions of development, then identifying genes related to MSCs development, summarizing the genes related to MSCs development into multiple gene modules, observing the expression changes of the above module genes with time, and obtaining the changes of time with cell differentiation.

[0029] The application provides a method for detecting biological functions of mesenchymal stem cells based on single-cell sequencing, comprising the following steps: performing single-cell sequencing on a single-cell suspension of mesenchymal stem cells to obtain mesenchymal stem cell transcriptome data; performing principal component analysis dimension reduction on the mesenchymal stem cell transcriptome data after quality control processing to obtain dimension reduction data; grouping the dimension reduction data to determine marker genes in mesenchymal stem cell subgroups in different groups, performing GO and KEGG enrichment analysis based on the marker genes in different groups to obtain enrichment entries; comparing the average expression levels of biological function-related genes in different groups; constructing a function evaluation model of mesenchymal stem cells based on human stem cells and embryonic transcriptome data using a linear regression algorithm to input each single-cell transcriptome data in different groups into the function evaluation model to calculate the scores of various biological functions of each mesenchymal stem cell; and comprehensively judging the biological functions of mesenchymal stem cells based on the average expression levels of function-related genes, enrichment entries and various function scores in different groups. The application detects the single-cell transcriptome atlas of MSCs through single-cell sequencing technology, analyzes through bioinformatics, and creates a function evaluation model of MSCs through OCLR algorithm, and then detects the heterogeneity of MSCs from different tissues. The single-cell sequencing technology explores the gene expression profile at single-cell resolution, which is greatly improved compared with the resolution of the original cell population, and can accurately detect the state of each cell in the MSC cell population and judge the potential of each cell. At the same time, the application obtains MSCs in different groups based on the expression levels of biological function genes, GO and KEGG enrichment analysis and OCLR algorithm evaluation, and observes that different MSCs have different gene expression levels, different GO and KEGG enrichment entries, and different function scores obtained by OCLR algorithm. Compared with the general scheme which only detects a single function of a certain classification of MSCs, the scheme is more comprehensive in detecting MSCs, and the detection result is more accurate.

[0030] The method provided by the application further limits the grouping method, which comprises clustering analysis results, tissue sources, generations and culture conditions. Compared with the general scheme which only detects a single function of a certain classification of MSCs, the scheme is more comprehensive in detecting MSCs, and characterizes the tissue repair potential, differentiation potential, proliferation potential, exosome secretion potential and cell stemness of MSCs from different tissue sources, different generations and different culture conditions at the level of a large number of genes. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1The analysis results of P5 generation MSCs grouped by different tissue sources, wherein A: subpopulation distribution of MSCs; B: average expression heat map of genes of different functions in each subpopulation; C: top 10 pathways of GO BP enrichment of specific genes of each subpopulation; D: distribution of MSCs grouped by sources; E: average expression heat map of genes of different functions of MSCs of different sources; F: top 10 pathways of GO BP enrichment of specific genes of MSCs of different sources;

[0032] Figure 2 The scores and marker genes of MSCs of different sources, wherein A-E: scores of tissue repair potential (A), differentiation potential (B), proliferation potential (C), exosome secretion potential (D) and cell stemness (E) of MSCs of different sources; F: top 3 marker genes of MSCs of different sources;

[0033] Figure 3 The functional change results of UCMSCs with the increase of generation;

[0034] Figure 4 The pseudo-time sequence analysis results of MSCs of different generations, wherein A: cell distribution of pseudo-time sequence analysis re-clustering, and the serial number in the figure represents the order of development process; B: time sequence change of MSCs of different generations with the development process;

[0035] Figure 5 The qPCR detection results of function-related genes of MSCs of different sources, wherein A-E: expression levels of tissue repair-related genes (A), differentiation-related genes (B), proliferation-related genes (C), exosome secretion-related genes (D) and cell stemness-related genes (E) of MSCs of different sources, and T cells are control cells;

[0036] Figure 6 The three-line differentiation experiment results of MSCs of different sources, wherein the control group is UCMSCs cultured without induction;

[0037] Figure 7 The CCK-8 proliferation experiment results of MSCs of different sources, and T cells are control cells;

[0038] Figure 8 The electron microscope scanning graph and concentration detection graph of exosomes of UCMSCs, wherein A is the electron microscope scanning graph of exosomes, and B is the concentration detection of exosomes. DETAILED DESCRIPTION

[0039] The application provides a method for detecting biological functions of mesenchymal stem cells based on single cell sequencing, comprising the following steps:

[0040] Single cell sequencing is performed on a single cell suspension of mesenchymal stem cells to obtain transcriptome data of the mesenchymal stem cells;

[0041] The mesenchymal stem cell transcriptome data were subjected to quality control processing and then dimensionality reduction by principal component analysis to obtain dimensionality-reduced data.

[0042] The dimensionality-reduced data were grouped to identify marker genes in different subpopulations of mesenchymal stem cells. Based on the marker genes in these different groups, GO and KEGG enrichment analyses were performed to obtain enriched entries.

[0043] Compare the average expression levels of genes related to biological function in each mesenchymal stem cell in different groups;

[0044] A functional assessment model for mesenchymal stem cells was constructed using a class of linear regression algorithms based on human stem cell and embryonic transcriptome data. The single-cell transcriptome data from different groups were input into the functional assessment model to calculate the scores of various biological functions for each mesenchymal stem cell.

[0045] The biological function of mesenchymal stem cells was determined by combining the average expression levels of function-related genes, enrichment items, and various functional scores in each group.

[0046] This invention performs single-cell sequencing on a single-cell suspension of mesenchymal stem cells to obtain mesenchymal stem cell transcriptome data.

[0047] This invention does not impose any particular limitation on the preparation method of the mesenchymal stem cell single-cell suspension; any single-cell suspension preparation method well-known in the art can be used. In the embodiments of this invention, the concentration of the mesenchymal stem cell single-cell suspension is preferably 1–5 × 10⁻⁶. 5 More preferably 3.5×10 5 .

[0048] In this invention, the single-cell sequencing method is preferably based on a microfluidic chip. Single-cell sequencing kit and Singleton Matrix TM The platform performs single-cell sequencing and quality control. This invention detects MSCs based on single-cell transcriptome sequencing, offering high throughput and speed, and simultaneously detecting the expression of a large number of genes, thus uncovering the heterogeneity of MSCs from different tissue sources at the whole transcriptome level. Simultaneously, it detects MSC function at the single-cell level, offering higher accuracy than conventional methods, identifying impurity cells in the MSC population, and the OCLR algorithm identifies multiple potentials of MSCs from the perspective of a large number of gene expression profiles, making it more reliable than existing methods.

[0049] After obtaining the mesenchymal stem cell transcriptome data, the present invention performs principal component analysis to reduce the dimensionality of the mesenchymal stem cell transcriptome data after quality control processing, and obtains dimensionality-reduced data.

[0050] In the present application, the quality control processing method preferably excludes cells with the lowest 5% and highest 5% of detected genes and more than 15% of mitochondrial genes. The quality control processing is preferably completed using the Seurat R package (Version 4.0.3, https: / / satijalab.org / seurat). The quality control processing preferably excludes cells with the lowest 5% and highest 5% of detected genes and more than 15% of mitochondrial genes after the initial Cell Ranger indicator evaluation.

[0051] In the present application, in order to avoid batch effects between samples and experiments before principal component analysis dimension reduction, the top 2000 highly variable genes are used for canonical correlation analysis (CCA) implemented in Seurat to align samples. The aligned canonical correlation coefficients are then used for integration analysis by the IntegrateData() function, and the integrated data is used for downstream dimension reduction and clustering analysis. Seurat is an R language software package specially used for single-cell sequencing data analysis, which provides cell quality control, clustering and other analysis processes and gives visualization results. The principal component analysis is preferably completed using the "RunPCA" function. Given the specificity of G2M and S phase genes, the cell cycle score of each cell is evaluated by the "CellCycleScoring" function to determine the cycle in which the cell is located, and then the "SCTransform" function is used based on the regularized negative binomial regression method (SCT) to remove the influence of cell cycle effects on the analysis. The dimension reduction is preferably performed using the "RunUMAP" function. The uniform manifold approximation and projection (UMAP) algorithm in the Seurat software package has the characteristics of fast running speed and small memory occupation, and can reduce the feature dimension to the greatest extent while preserving the characteristics of the original data, thereby faster completing the heterogeneity detection of MSCs.

[0052] After obtaining the dimension-reduced data, the present application groups the dimension-reduced data, determines marker genes in different groups, performs GO and KEGG enrichment analysis based on the marker genes in the different groups, and obtains enrichment entries.

[0053] In the present application, the method of grouping is preferably performed according to the clustering analysis result, tissue source, passage, and culture condition. The clustering analysis result is performed by the "FindClusters" function with a resolution of 0.2, and the first 20 principal components are used to define cell identity. The tissue source preferably includes amniotic membrane source, umbilical cord source, chorion source, and adipose tissue source. The culture condition preferably includes conventional oxygen culture condition and low-concentration oxygen culture condition. The oxygen concentration in the conventional oxygen culture condition is 21%. The oxygen concentration in the low-concentration oxygen culture condition is 1%.

[0054] In the present application, the method of determining marker genes in different groups is preferably to determine marker genes for each cluster using the Wilcoxon rank-sum test of the "FindAllMarkers" function. The marker genes are preferably genes with |avg_log2FC|>0.25 and p_val<0.05 and min.pct>0.1.

[0055] In the present application, the method of GO and KEGG enrichment analysis is preferably performed using Metascape (Version 3.5, http: / / metascape.org / ) and visualized using the R package ggplot2 (Version 3.3.5, https: / / github.com / tidyverse / ggplot2). The significance-Log 10 (P) The top 10 enrichment entries represent the biological functions of the MSC cell population. If multiple biological function entries are enriched, the one with the highest number of genes and the highest significance is selected as the biological function.

[0056] After grouping, the present application compares the average expression levels of biological function-related genes of each mesenchymal stem cell in different groups.

[0057] In the present application, the biological function related genes preferably include at least one of the following: evaluation of tissue repair potential related genes, evaluation of MSCs differentiation potential related genes, evaluation of proliferation potential related genes, and evaluation of extracellular vesicle transport capacity related genes, and evaluation of cell stemness related genes; the evaluation of tissue repair potential related genes preferably include the following genes: MMP2, MMP9, CXCR2, CXCR4, CXCR5, CXCR6 and CDH2; the evaluation of MSCs differentiation potential related genes preferably include the following genes: ACTA2, VEGF, FBLN5, DCN, UBE2C, CTGF and POSTN; the evaluation of proliferation potential related genes preferably include the following genes: PLK1, CCNE1, MKI67, CCNB1, E2F1, FOXM1, MCM3, MCM4 and TOP2A; the evaluation of extracellular vesicle transport capacity related genes preferably include the LIN7A gene; and the evaluation of cell stemness related genes include the following genes: SOX2, LIFR and MEST. The average of the expression amount of a function related gene of mesenchymal stem cells in each group is a comparison index, and compared with other groups, the potential of the biological function corresponding to the gene in the group is obtained.

[0058] After grouping, the present application uses a linear regression algorithm to construct a function evaluation model of mesenchymal stem cells based on human stem cell and embryo transcriptome data, and inputs each single cell transcriptome data in different groups into the function evaluation model to calculate the biological function scores of each mesenchymal stem cell.

[0059] In the present application, the human stem cell and embryo transcriptome data are preferably obtained from the synapse database. The human stem cell and embryo transcriptome data are used as a training set to create a function evaluation model of MSCs by using a linear regression algorithm (OCLR, One Class Linear Regression). The higher the function score is, the stronger the function of MSCs is, and the function heterogeneity of MSCs is obtained.

[0060] The average expression amount of the function related genes, the enrichment items and the various function scores of each group are comprehensively considered to determine the biological functions of mesenchymal stem cells.

[0061] In the present application, preferably further comprising performing pseudo-time analysis to evaluate the differentiation of mesenchymal stem cells. The method of pseudo-time analysis includes a Monocle 3-based process, combining pseudo-time graphs and selecting the starting position and ending position of development, then identifying genes related to MSCs development, summarizing the genes related to MSCs development into multiple gene modules, observing the expression changes of the above module genes with time, and obtaining the changes of time with cell differentiation. Monocle3 provides a single-cell pseudo-time analysis scheme. Compared with conventional pseudo-time analysis, it introduces a graph-based trajectory inference process based on the Partition-based graph abstraction (PAGA) algorithm. The trajectory inference form based on the dimensionality reduction graph makes the pseudo-time analysis process more in line with natural laws, has efficient operation, beautiful visualization and novel differential gene algorithm, and can effectively eliminate noise branches, thereby accurately identifying the cell stemness heterogeneity of MSCs from different tissues; OCLR algorithm is an innovative linear regression machine learning algorithm, which has been used for functional evaluation and feature screening of tumor cells, and the test results prove that the algorithm result is accurate.

[0062] The method for detecting mesenchymal stem cells from different tissues, passages and culture conditions based on single-cell sequencing provided by the present application will be described in detail below in combination with examples, but they should not be understood as limiting the scope of protection of the present application.

[0063] Abbreviations and English meanings are shown in Table 1.

[0064] Table 1

[0065]

[0066]

[0067] Example 1

[0068] After sampling from the source, it is sent to Source Cell Biotechnology Co., Ltd. within 12 hours for the following treatment:

[0069] Amniotic membrane: amniotic membrane tissue is separated from the placenta, mechanically chopped into 1x1cm 2 blocks, an equal volume of 0.4U / mL Collagenase NB6 (Norrdmark) is used for 37°C shaking digestion for 1h, the solution is filtered through a 70μm cell screen after digestion, and centrifuged at 1000xg for 5min, the supernatant is removed, and the cells are inoculated in a 75T culture flask;

[0070] Umbilical cord: After the umbilical cord tissue was washed with DPBS (BBI) to remove blood, it was cut into small pieces of 1.5 cm, the blood vessels and outer membrane were removed, Wharton's jelly tissue was collected, an equal volume of 0.2% collagenase type I was added for digestion for 40 min, after dilution with DMEM, it was filtered through a 70 μm cell strainer, centrifuged (400 g, 10 min) and the cells were inoculated in a 75T culture flask;

[0071] Chorion: Chorion tissue was separated from the placenta and washed with DPBS (BBI) to remove blood. It was mechanically chopped into small pieces, an equal volume of 0.2% collagenase type II (Sigma-Aldrich) was added for digestion at 37°C, 200 rpm for 2 h, after filtration through a 70 μm cell strainer, it was centrifuged (1000 g, 5 min) and the cells were inoculated in a 75T culture flask;

[0072] Fat: Fat tissue was oscillation digested (37°C, 200 rpm, 1 h) with an equal volume of 0.075% collagenase type I (Sigma-Aldrich). After dilution with MEM-α (Gibco), it was filtered through a 100 μm cell strainer and centrifuged (400 x g, 10 min), and the cells were inoculated in a 75T culture flask. The four different tissue-derived mesenchymal stem cells were cultured and expanded in T4 culture medium containing 5% serum substitute at 37°C, 5% CO2.

[0073] Single-cell transcriptome data analysis and functional detection were performed on P5 generation MSCs derived from normoxic cultured fat, amniotic membrane, chorion and umbilical cord tissues. The specific methods are as follows:

[0074] (1) Preparation of MSC single cell suspension: under a microscope, when the cultured cells appeared 90%-95% confluence at the bottom of the culture dish. Remove the culture medium, add Dulbecco's Phosphate Buffered Saline (DPBS) for continuous rinsing for 2 times; add an appropriate amount of trypsin (0.25% trypsin-ethylenediamine tetraacetic acid), digest at 37°C for 2 min; add an appropriate amount of T4 complete culture medium to terminate digestion, transfer the cell suspension into a 15 ml centrifuge tube, centrifuge at 400 g for 3 min, remove the supernatant; resuspend with 1 ml DPBS (without calcium and magnesium), detect the cell concentration; dilute with DPBS (without calcium and magnesium) to a cell concentration of 3.5 x 10 5 cell / ml to prepare a single cell suspension.

[0075] (2) Single cell sequencing to obtain MSC transcriptome data: after the sample was prepared into a single cell suspension, a single cell sequencing kit based on a microfluidic chip (Nanjing, China, Xinggenyuan Biotechnology Co., Ltd.) and a single cell sequencing instrument (Illumina NovaSeq 6000) were used to obtain the transcriptome data of MSCs. ​The automated single-cell sequencing library construction system (Nanjing, China, Xinggen Biotechnologies) captures single cells, which are The chip is placed into the instrument Singleron According to the Poisson distribution principle, cells fall into specially customized chip microwells under the action of gravity, and only one cell falls into each microwell, completing cell separation. Add millions of Barcoding Beads carrying unique cell labels to the chip microwells. Since the microwells are slightly larger than the Barcoding Beads, only one Barcoding Bead will fall into each microwell. After adding cell lysis solution to lyse the cells, Barcoding Beads with unique cell labels and molecular labels (Unique Molecular Identifiers, UMIs) capture mRNA by binding to the poly(A) tail on the mRNA, and label the cells and mRNA. Singleron Automatically collect Barcoding Beads in the chip, reverse transcribe mRNA captured by Barcoding Beads into cDNA and amplify. After cDNA goes through steps such as fragmentation and linker ligation, a sequencing library suitable for the illumina sequencing platform is constructed. Subsequently, the Singleron Matrix TM platform performs single-cell transcriptome sequencing and quality control on the sample. Using BD-RhapsodyTM to analyze and process sequencing data (.fastq files), first remove low-quality read pairs (R1 & R2). Analyze the screened R1 sequences to identify cell label sequences, unique molecular recognition labels, and poly-dT tail sequences. R2 reads are mapped to Genome Reference Consortium Human Build37 (GRCh37) (ftp: / / ftp.ncbi.nih.gov / genomes / Homo_sapiens / ), using Star (Version: 2.5.2b) and annotated according to GENCODE (Version 28, https: / / www.gencode.org / ). Further correction is performed using a distribution-based error correction algorithm, and all suspected cells are distinguished from background noise using second derivative analysis, thereby obtaining a single-cell transcriptome expression matrix.

[0076] (3) Single-cell transcriptome data quality control of MSCs: Seurat R package (Version 4.0.3, https: / / satijalab.org / seurat) was used for further analysis of scRNA-seq data. After the initial Cell Ranger metrics evaluation, cells with top 5% of detected genes and top 5% of mitochondrial gene proportion over 15% were excluded from downstream analysis, and the remaining cells were used for downstream bioinformatics analysis. Normalized UMI values were obtained by the “NormalizeData” function. The “ScaleData” function was used to scale and center expression levels in the dataset for dimensionality reduction.

[0077] (4) PCA dimensionality reduction and clustering analysis: To avoid batch effects between samples and experiments, the top 2000 highly variable genes were used for canonical correlation analysis (CCA) implemented in Seurat to align samples. The aligned canonical correlation coefficients were then used for integration analysis by the “IntegrateData” function, and the integrated data were used for downstream dimensionality reduction and clustering analysis. Subsequently, principal component analysis was performed by the “RunPCA” function. Since cell cycle has a large impact on the analysis of MSCs, we assessed the cell cycle score of each cell based on the specificity of G2M and S phase genes by the “CellCycleScoring” function to determine the cycle in which the cell was located, and then removed the influence of cell cycle effects on the analysis based on the regularized negative binomial regression method (SCT) using the “SCTransform” function. Dimensionality reduction was performed by the “RunUMAP” function, and visualization was performed. Cell clustering was performed by the “FindClusters” function with a resolution of 0.2, and the top 20 principal components were used to define cell identity. The Wilcoxon rank-sum test using the “FindAllMarkers” function was used to determine the marker genes of each cluster, and only those |avg_log2FC| > 0.25 and p_val < 0.05 and min.pct > 0.1 genes were considered as marker genes.

[0078] (5) Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis: Gene Ontology_Biological_Processes (GO_BP) and KEGG analysis of marker genes were performed by Metascape (Version 3.5, http: / / metascape.org / ) and visualized using R package ggPlot2- (Version 3.3.5, https: / / github.com / tidyverse / ggplot2). The top 10 enriched items with the highest -Log10(P) of GO or KEGG enrichment significance were selected, which represented the biological functions of MSCs cell populations. If multiple biological function items were enriched, the one with the highest number of genes and the highest significance was selected as the biological function.

[0079] (6) Functional heterogeneity of MSCs: Based on the expression levels of functionally relevant genes in cells, the average expression levels of function genes were calculated and standardized. The higher the average expression level of function genes, the stronger the function of MSCs population. For each MSCs subpopulation obtained by clustering analysis, the functional heterogeneity of different MSCs subpopulations was evaluated by comparing each MSCs subpopulation with other subpopulations. For MSCs in different states, the functional heterogeneity of MSCs in different states was also evaluated by comparing different groups of MSCs in different states. The genes MMP2, MMP9, CXCR2, CXCR4, CXCR5, CXCR6, and CDH2 were used to evaluate the tissue repair potential of MSCs; the genes ACTA2, VEGF, FBLN5, DCN, UBE2C, CTGF, and POSTN were used to evaluate the differentiation potential of MSCs; the genes PLK1, CCNE1, MKI67, CCNB1, E2F1, FOXM1, MCM3, MCM4, and TOP2A were used to evaluate the proliferation potential of MSCs; the gene LIN7A was used to evaluate the exosome transport capacity of MSCs; and the genes SOX2, LIFR, and MEST were used to evaluate the stemness of MSCs.

[0080] To more accurately evaluate the function of MSCs, human stem cell and embryonic data were obtained from the synapse database, and a one-class linear regression algorithm (OCLR, One Class Linear Regression) was used to create a function evaluation model for MSCs based on bulk gene expression. The data after sequencing were incorporated into the model to calculate the function scores of each MSC. Based on the separation of different MSC subgroups or different MSC states, the function scores of MSCs were evaluated. The higher the function score, the stronger the function of MSCs. The functional heterogeneity of MSCs was obtained. Based on the expression level of the above genes, GO and KEGG enrichment results, and OCLR algorithm evaluation, MSCs of different tissue sources, passages, and culture conditions were obtained. It was observed that different MSCs had different expression levels of genes, different GO and KEGG enrichment items, and different function scores obtained by OCLR algorithm.

[0081] (7) Monocle 3 constructs pseudo-time analysis: The pseudo-time analysis R package Monocle3 (Version 1.0.0, https: / / cole-trapnell-lab.github.io / monocle3 / ) is used to reconstruct the developmental trajectory of MSCs. Based on the Monocle 3 process, the starting position and the end position of the development were selected in combination with the time sequence diagram and using the "choose_graph_segments" function. Subsequently, the "graph_test" function was used to identify genes related to MSC development. The "find_gene_modules" function was used to summarize these genes into multiple gene modules, and the expression of these module genes was observed. The expression of MSCs related to stemness was observed to change with pseudo-time, and the change of stemness with MSCs was identified. For example, the expression of MSCs stemness genes decreased with the increase of MSCs time sequence.

[0082] (8) qPCR detection factor expression: 5 T cells were cultured as control cells, and 5 cases of umbilical cord, amniotic membrane, chorion and fat-derived P5 MSCs were cultured under the same normoxic conditions, which will be used for subsequent studies. Total RNA was extracted from mesenchymal stem cells according to the manufacturer's instructions. First, the cultured T cells and MSCs were loaded into 5 mL centrifuge tubes, and 1 mL Trizol extraction solution was added. The supernatant was removed to a new centrifuge tube, 1 mL chloroform was added, and it was shaken vigorously for 15 s. After 5 min, 12000 r / min, 4℃ centrifugation for 15 min, the phases were separated, and the supernatant was transferred to a new tube. An equal volume of isopropanol was added and mixed well. After 15 min, 4℃, 12000 r / min centrifugation for 10 min, the supernatant was discarded, and 1 mL of 75% ethanol was added for washing. After 4℃, 12000 r / min centrifugation for 5 min, the ethanol was discarded, and the total RNA was dissolved in DEPC water. 2 μL of RNA sample was taken in 398 μL of DEPC water and mixed well. The absorbance value of A 260 、A 280 of total RNA concentration and purity was determined by spectrophotometer. 2 μL of total RNA was electrophoresed on a 1.2% formaldehyde agarose gel at 120 V for 20 min, and the RNA integrity was detected.

[0083] According to the instructions of Goldenstar TM RT6 cDNA Synthesis Kit Ver.2), the cDNA reaction system was configured, 1 μl gDNA remover, 1 μl 10×gDNA remover Buffer, 100 / total RNA concentration (ng / μl) of total RNA extraction solution and RNase-free water were added to make up to 10 μl. After the configuration was completed, incubate in qPCR instrument at 42℃ for 2 minutes, and then incubate for 60 minutes. Then add the following reagents: 1 μl dNTP Mix, 1 μl Goldenstar TM Oligo(dT) 17 , 4 μl 5×Goldenstar TM Buffer, 1 μl DDT, 1 μl Goldenstar TM RT6, 2 μl RNase-free water. After the configuration was completed, incubate in qPCR instrument at 55℃ for 30 minutes, and then incubate at 85℃ for 5 minutes. After the end, the reverse transcription cDNA template was obtained.

[0084] The qPCR primer design is shown in Table 2. According to the qPCR kit ( Master qPCR Mix-SYBR(+UDG) according to the instruction, configure qPCR reaction system: 10 μl Master qPCR Mix-SYBR(+UDG), 0.8 μl 10Mm Primer F, 0.8 μl 10Mm Primer R, 1 μl reverse transcription cDNA template, 7.4 μl ddH2O. After configuration, set 20℃ for 2 minutes, 95℃ for 2 minutes in qPCR instrument, then cycle 40 times under the condition of 95℃ for 15 seconds and 60℃ for 30 seconds and collect fluorescence signal (Ct).2 -ΔΔCt For representing the relative expression amount of gene. ΔCt = target gene Ct-internal reference Ct, ΔΔCt = ΔCt-(T cell average Ct-T cell corresponding internal reference average Ct).

[0085] Table 2 qPCR primer sequence

[0086]

[0087]

[0088] (9) Mesenchymal stem cells (MSCs) three-lineage differentiation detection: adipogenic differentiation detection standard operation: prepare 24-well plate and centrifuge tube, count according to the standard operation procedure of cell counting, take 2.0×10 5 MSCs into the centrifuge tube, add MSC-T4 complete culture medium to 5 ml, blow and mix uniformly; take 0.5 ml of cell suspension into each identification hole, transfer the cells to 37℃, 5% CO2 incubator;

[0089] Adipogenic differentiation medium preparation: 10% AdipoGen supplement + 90% AdipoGen Differentiation Basal medium, mix well, store at 4°C; when the cell confluence reaches 70-80%, start induction, record as Day 0, discard the supernatant in the experimental wells, add 0.5 ml of adipogenic differentiation medium to the experimental wells, and transfer the 24-well plate to a 37°C, 5% CO2 incubator; replace the adipogenic differentiation medium in the experimental wells every 3 days, and replace the complete medium in the control wells; adipogenic differentiation is performed for 7-14 days, and obvious oil droplets can be observed under a microscope, i.e., oil red O staining identification is performed; discard the supernatant in the identification wells, add 0.5 ml of 4 wt% polyhydroxyalkanoate (PFA) to the identification wells, and fix for 20 min; after discarding the liquid, suck 0.5 ml of 60% (v / v) isopropanol into the identification wells for 1 time of rinsing; add 0.5 ml of oil red O staining solution to the identification wells, and stain for 15 min; after discarding the liquid, suck 0.5 ml of ultrapure water into the identification wells for 2 times of rinsing; add 0.5 ml of ultrapure water to the identification wells, and observe and take photos under a fluorescence microscope;

[0090] Osteogenic differentiation detection: prepare 24-well plates and centrifuge tubes, count the cells according to the cell counting standard operating procedure, take 2.0 x 10 5 to the centrifuge tubes, add MSC-T4 complete medium to 5 ml, and mix well by blowing; suck 0.5 ml of the cell suspension into each identification well; transfer the 24-well plate to a 37°C, 5% CO2 incubator;

[0091] Osteogenic differentiation medium preparation: 10% OsteoGen supplement + 90% OsteoGen Differentiation Basal medium, mix well, store at 4°C; when the cell confluence reaches 60-70%, start induction, record as Day 0, discard the supernatant in the experimental wells, add 0.5 ml of osteogenic differentiation medium to the experimental wells, and transfer the 24-well plate to a 37°C, 5% CO2 incubator; replace the osteogenic differentiation medium in the experimental wells every 3 days, and replace the complete medium in the control wells; osteogenic differentiation is performed for 14-21 days, and calcium nodules can be observed under a microscope, i.e., 0.2% alizarin red staining identification is performed; discard the supernatant in the identification wells, add 0.5 ml of 4 wt% PFA to the identification wells, and fix for 20 min; after discarding the liquid, suck ultrapure water 0.5 ml to rinse the identification wells 1 time; add 0.5 ml of 0.2% alizarin red solution to the identification wells, and stain for 30 min; after discarding the liquid, suck ultrapure water 0.5 ml to rinse the identification wells 2 times; add 0.5 ml of ultrapure water to the identification wells, and observe and take photos under a fluorescence microscope.

[0092] Chondrogenic differentiation assay: Prepare 24-well plates and centrifuge tubes. After counting cells according to standard operating procedures, collect 2.0 × 10⁻⁶ stem cells. 5 Add 5 ml of MSC-T4 complete medium to centrifuge tubes and mix thoroughly by pipetting. Add 0.5 ml of stem cell suspension to each identification well. Add 0.5 ml of D-PBS to the wells surrounding the identification wells. Transfer the 24-well plate to a 37°C, 5% CO2 incubator. Prepare the chondrogenic differentiation medium: 10% Chondrosis supplement + 90% Chondrosis Differentiation Basal medium, mix thoroughly, and store at 4°C. Induction begins when cell confluence reaches 60-70%, designated as Day 0. Discard the supernatant from the wells and add 0.5 ml of chondrogenic differentiation medium to each well. Transfer the 24-well plate to a 37°C, 5% CO2 incubator. Replace the chondrogenic differentiation medium in the wells every 3 days, and replace the control wells with complete medium. At Day 14-21, spherical cell clusters are visible under a microscope; perform Alcian blue staining for identification. Discard the supernatant from the identification wells and add 0.5 ml of D-PBS. Wash twice with D-PBS, add 0.5 ml of 4 wt% PFA to the identification wells, and fix for 20 min. After discarding the solution, rinse the identification wells once with 0.5 ml of ultrapure water. Add 0.5 ml of alicin blue acidification solution and soak for 5 min. Add 0.5 ml of alicin blue staining solution to the identification wells and stain for 30 min. After discarding the solution, rinse the identification wells twice with 0.5 ml of ultrapure water. Add 0.5 ml of ultrapure water to the identification wells and observe and photograph under a fluorescence microscope.

[0093] (9) CCK-8 cell proliferation assay: Prepare 100 μl of cell suspension in a 96-well plate. Pre-culture the plate in an incubator for 24 hours (at 37°C and 5% CO2). Add 10 μl of different concentrations of the test substance to the plate and incubate for 24 hours. Add 10 μl of CCK-8 solution to each well and incubate for 1-4 hours. Measure the absorbance at 450 nm using an ELISA reader.

[0094] (10) Exosome content detection: Based on the exosome extraction kit, collect the supernatant of MSC culture, centrifuge at 3000xg for 15 minutes to remove cell debris. Transfer the supernatant to a new test tube. Add an appropriate volume (1 ml per 5 ml sample) of ExoQuick-TC to the clarified biological fluid. Mix well by inverting the test tube and incubate on ice at 4°C. No rotation is required during incubation. Centrifuge the ExoQuick-TC / biological fluid mixture at 3,000xg for 10 minutes. Remove the supernatant, remove the residual ExoQuick solution, resuspend in 200 μl of buffer B, measure and record the protein concentration in the sample. Add 200 μl of buffer A to the resuspended extracellular vesicles, remove the purification column, loosen the screw cap, break the bottom seal, and place the column in a collection tube. Centrifuge at 1,000xg for 30 seconds to remove the storage buffer, discard the fluid after centrifugation, and then place the column back in the collection tube. When washing the column, remove the cap, apply 500 μl of buffer B to the resin, centrifuge at 1000xg for 30 seconds, discard the liquid that passes through, and repeat the washing 1 time. Insert the bottom seal into the bottom of the column with the bottom cap. Apply 100 μl of buffer B to the resin. Add the resuspended exosomes to the resin, place the screw cap on the top of the cylinder, and agitate at room temperature for 5 minutes. Loosen the screw cap, remove the bottom seal, and immediately transfer to a 2 ml EP (eppendorf) tube, centrifuge at 1000xg for 30 seconds to obtain purified MSC exosomes.

[0095] Take out 10 μL of exosomes; 10 μL of sample is dropped onto a copper mesh and precipitated for 1 min, and the supernatant is absorbed with filter paper; 10 μL of uranyl acetate is dropped onto a copper mesh and precipitated for 1 min, and the supernatant is absorbed with filter paper; dry for a few minutes at room temperature; perform electron microscope detection imaging at 100 kv; and obtain transmission electron microscope imaging results. Dilute the exosomes with PBS and directly use them for NTA detection of particle concentration and particle size range.

[0096] (11) Real-time quantitative PCR method for detecting telomerase activity: using Hela cell line as positive control and MRC-5 cell line as negative control, total RNA of P5 generation UCMSC, AMMSC, CMMSC, ADMSC and Hela cells and MRC-5 cells cultured in normal oxygen were extracted according to the instructions of KEGG Telomerase Activity Real-time Quantitative PCR Kit (human), and the total mRNA of the tissue was used as a template to configure a 20 μL reverse transcription cDNA reaction system: 2 μL 10xRT Buffer, 2 μL 10xRTRandom Primers, 1 μL MultiScribe Reverse Transcriptase, 0.8 μL 25xdNTP Mix, 10 μL extracted total RNA of cells, 4.2 μL RNase-Free Water. After mixing and centrifuging, incubate in PCR instrument at 25°C for 10 minutes, 37°C for 120 minutes, 85°C for 5 minutes, and then keep at 4°C. Then use telomerase primers (TS 5'-AATCCTCGAGGAGAGTT-3', SEQ ID NO: 57; ACX 5'-GCGCGG(CTTATT)3CTAACC-3', SEQ ID NO: 58) to configure a 25 μL qPCR reaction system according to the following dose: 0.5 μL TERT Forward Primer, 0.5 μL TERT Reverse Primer, 12.5 μL Maxima SYBR Green Qpcr Master Mix (2x), 100 ng cDNA template (volume according to actual concentration), RNase-Free Water to 25 μL. Activate UDG enzyme in qPCR instrument at 50°C for 30 minutes, pre-denature at 94°C for 3 minutes, then cycle 35 times at 94°C for 30 seconds, 60°C for 32 seconds, 72°C for 60 seconds, 72°C for 5 minutes, and then collect data for analysis.

[0097] Results

[0098] It was found that these cells could be unbiasedly clustered into 7 different cell subpopulations after UMAP dimension reduction Figure 1 A), by average gene expression analysis Figure 1In B), it can be found that cluster 1, cluster 3 and cluster 5 subgroups have stronger tissue repair ability; cluster 2 subgroup has higher differentiation potential and proliferation potential; cluster 0 subgroup also has strong proliferation potential; cluster 4-6 subgroups all have higher exovesicle secretion capacity; cluster 0 and cluster 3 subgroups have higher stemness. OG BP enrichment We can see that cluster 0 and cluster 2 are enriched in the "mitotic cell cycle" and "cell division" pathways, which also indicates that they have higher proliferation potential; cluster 1, cluster 3 and cluster 5 are also enriched in the "collagen fibril organization" pathway, indicating that they have higher tissue repair capacity; cluster 3 and cluster 1 are enriched in the "wound healing" and "wound healing" pathways at the same time, and cluster 3 is specifically enriched in the "regulation of cell growth", combined with the high expression of cluster 3 stemness genes, which indicates that cluster 3 repairs damaged tissues by promoting cell growth; cluster 5 and cluster 1 are enriched in the "extracellular matrix organization" and "extracellular structure organization" and multiple extracellular matrix (Extracellular matrix) related pathways, and are specifically enriched in the "tissue morphogenesis" pathway, combined with the strong exovesicle secretion potential of cluster 5, which indicates that cluster 5 repairs tissues by regulating extracellular matrix through a paracrine pathway; cluster 4 is enriched in the "regulation of cell activation" pathway, and has the strongest exovesicle secretion capacity among all subgroups, indicating that cluster 4 promotes cell activation through a paracrine pathway; cluster 6 is specifically enriched in the "ATP metabolic process", which is related to the strong exovesicle secretion potential Figure 1 In C).

[0099] Mesenchymal stem cells are grouped and compared according to different sources Figure 1In the middle D), it was found that adipose-derived MSCs (ADMSCs) had strong tissue repair potential and differentiation potential, chorionic membrane-derived MSCs (CMMSCs) had strong proliferation potential and stemness, umbilical cord-derived MSCs (UCMSCs) had strong exosome secretion potential, and amniotic membrane-derived MSCs (AMMSCs) had slightly stronger exosome secretion potential than other functions Figure 1 In the middle E), UCMSCs were specifically enriched in the "cytoplasmic translation" pathway, which is an upstream pathway of exosome secretion; AMMSCs were enriched in the "ATP metabolic process" pathway, which is consistent with the above, related to the strong exosome secretion potential; CMMSCs were enriched in the "cell division" and "regulation of cell development" pathways, which is consistent with the high expression of proliferation potential and stemness genes in gene expression; ADMSCs were specifically enriched in the "ribonucleoprotein complex bio genesis" and "ribosome bio genesis" pathways, indicating that the ribosome basic function of ADMSCs is strong, which is consistent with its high tissue repair potential and differentiation potential Figure 1 In the middle F).

[0100] By analyzing the transcriptional profiles of MSCs from different tissues using OCLR algorithm, it was confirmed that the tissue repair potential, differentiation potential, proliferation potential, exosome secretion potential and cell stemness were heterogeneous (see Tables 3-7). Consistent with the results obtained above, ADMSCs had the highest tissue repair potential and cell differentiation potential Figure 2 In the middle A-B), CMMSCs had the highest cell proliferation potential and stemness Figure 2 In the middle C and E), UCMSCs had the highest exosome secretion potential Figure 2 In the middle D). At the same time, we obtained 71, 3, 85 and 2 marker genes for adipose, amniotic membrane, chorionic membrane and umbilical cord-derived MSCs through marker gene screening, which can be used as marker genes for identifying their cell sources Figure 2 In the middle F, Tables 3-7).

[0101] Table 3 Evaluation results of tissue repair potential

[0102]

[0103]

[0104] Table 4. Results of Differentiation Potential Assessment

[0105] Gene Coefficient Gene Coefficient SRF 0.016666937 RAB1B -0.007052014 BAD 0.016605127 HTR3A -0.007162265 PTK2B 0.015022533 GIT2 -0.007231085 GDPD2 0.014109388 TRA2B -0.00733905 SYNM 0.01310936 GTF2H3 -0.007352425 JAM3 0.012856526 STON1 -0.007458427 NRP2 0.011347748 RRAGB -0.007527965 KRAS 0.010841981 DHCR7 -0.007687071 MAPK3 0.009864419 HIGD2A -0.007987162 PLCB3 0.009847613 RBSN -0.008069572 CDC14A 0.009597653 LOXL1 -0.008190658 MARCHF9 0.009186887 PSMB9 -0.008536161 RAF1 0.009148642 CSTF3 -0.008619948 NRP1 0.008612333 IL11 -0.00864246 PRKCI 0.008596354 INTS1 -0.008801491 MAPK9 0.008403374 VDAC3 -0.008821739 PLCB1 0.008059177 ENOX1 -0.008878944 C1QTNF3 0.008058438 OGA -0.009275984 CHD2 0.008042322 RBM23 -0.009399024 RHOD 0.008006732 TMEM170A -0.009736804

[0106] Table 5. Results of cell proliferation potential assessment

[0107]

[0108]

[0109] Table 6. Assessment Results of External Vesicle Transport Potential

[0110] Gene Coefficient Gene Coefficient RAB11A 0.017002385 SYNGR2 -0.008299468 STK38 0.015007761 INTS8 -0.008370659 RAB27A 0.014608173 IFITM1 -0.00842380l RPS14 0.012886212 BCKDK -0.008446885 C4orf33 0.012744559 OSMR -0.008599538 LIN7A 0.012477828 DCBLD1 -0.008629689 OGFRL1 0.01223786 RCAN3 -0.008675687 ALKAL2 0.011254874 SELENBP 1 -0.008689698 AIFM1 0.011117278 WDR53 -0.008784271 RCC1 0.011000533 GPATCH11 -0.008804189 GRM2 0.010742862 ZNF544 -0.008864593 CES2 0.010613249 SDHAF4 -0.0090494 CEP104 0.010223936 TBCCD1 -0.009089044 DAZL 0.010078553 EP400 -0.009095914 UBXN2B 0.009201197 KLK13 -0.009202732 SERPINB8 0.009027859 TMEM254 -0.009276022 RAP1GAP 0.008918347 SENP1 -0.00952783 MEIS3 0.00885454 MTRES1 -0.009616417 ARHGAP19 0.008791629 TIE1 -0.010843839 RAB35 0.00861634 IGSF3 -0.011158928

[0111] Table 7 Results of Dry Potential Assessment

[0112]

[0113]

[0114] Single-cell transcriptome sequencing was performed on P0, P3, and P5 generation umbilical cord-derived mesenchymal stem cells cultured under normoxic conditions and P5 generation UCMSCs cultured under hypoxia. Hypoxia culture was conducted in a tri-gas incubator (Thermo Scientific, USA) with 5% CO2, 90% N2, and 5% O2 conditions. TM The UCMSCs were cultured under aerobic conditions, while the rest of the culture was the same as that of normoxic culture. The analysis showed that as the passage number of UCMSCs increased, the tissue repair potential increased, while the cell differentiation potential, cell proliferation potential, extravesicle secretion potential and stemness all decreased significantly.

[0115] Monocle3 analysis showed that UCMSCs differentiated from P0 to P3, and after P5, finally became P5L UCMSCs. Figure 4 From the pseudotimeline plot, branch number 8 predicts that some cells will redevelop from generation P5 back to generations P3 and P0. This, combined with pseudotimeline analysis, suggests... Figure 4 In section B), it can be seen that the progression from branch 8 to P0 generation increases in time. Cell development also reaches its endpoint in branches 10 and 6. Although the P5 generation UCMSCs cultured under hypoxia are at the terminal position in P0-P3-P5-P5L, their time progression is still before branches 6, 8, and 10. This indicates that the stemness of MSCs decreases with increasing generation, and hypoxia culture preserves the stemness of MSCs, which is consistent with known results.

[0116] To verify the method, further culture of UCMSC, AMMSC, CMMSC and ADMSC each 5 cases, and prepared a number of biological repeats for subsequent experiments. By qPCR detection of MMP2, MMP9 and other 27 genes related to the function of MSC expression, and use T cells as control cells, found that its gene expression and the above scRNA-seq analysis results are the same, ADMSC tissue repair potential and differentiation potential related gene expression is the highest Figure 5 A, B), CMMSC proliferation potential and stemness related gene expression is the highest Figure 5 C, E), and UCMSC exosome secretion potential related gene expression is the highest Figure 5 D).

[0117] Using normal culture of UCMSC as control group, the three-line differentiation experiment of MSC from different sources, found that among them, ADMSC adipogenic, osteogenic and chondrogenic differentiation are the most, while in the other 3 kinds of tissue-derived MSC, UCMSC adipogenic differentiation ability is stronger, AMMSC chondrogenic differentiation ability is stronger, AMMSC and CMMSC osteogenic differentiation ability is higher Figure 6 ). This is consistent with the above results, found that the differentiation potential of ADMSC is higher, while the differentiation potential of AMMSC, CMMSC and UCMSC is smaller. CCK-8 experiment found that the proliferation ability of MSC from different sources is CMMSC > UCMSC > AMMSC > ADMSC > control T cells Figure 7 ), which is also consistent with the results above.

[0118] To evaluate the exosome secretion potential, extract the exosomes of MSC from different tissues, electron microscopy scanning found obvious exosome cup structure Figure 8 A), and the concentration of exosomes of the same volume of MSC from different tissues was determined Figure 8 B), found that the concentration of exosomes is UCMSC > AMMSC > ADMSC > CMMSC (Table 8).

[0119] To verify the stemness of MSC from different tissues, we measured the telomerase activity with MRC-5 and Hela cells as negative and positive controls, respectively, and found that the telomerase activity was CMMSC > AMMSC > UCMSC > ADMSC (Table 9). Telomerase activity to some extent represents the stemness of MSC, the higher the telomerase activity, the stronger the stemness. The results of these experiments are consistent with the results of scRNA-seq, which verifies the accuracy of the method proposed in this patent.

[0120] Table 8 Particle size and concentration of exosomes of MSC from different tissues

[0121]

[0122] Table 9 Telomerase activity qPCR detection results of MSCs from different tissue sources

[0123]

[0124] The above merely describes the preferred embodiments of the present application, and it should be noted that, for those skilled in the art, several improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements should also be considered as falling within the protection scope of the present application.

Claims

1. A method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing, characterized in that, Includes the following steps: Single-cell sequencing was performed on a single-cell suspension of mesenchymal stem cells to obtain mesenchymal stem cell transcriptome data; The mesenchymal stem cell transcriptome data were subjected to quality control processing and then dimensionality reduction by principal component analysis to obtain dimensionality-reduced data. The dimensionality-reduced data were grouped to identify marker genes in different groups. GO and KEGG enrichment analyses were then performed on the marker genes in these different groups to obtain enriched entries. Compare the average expression levels of genes related to biological function in each mesenchymal stem cell in different groups; A functional assessment model for mesenchymal stem cells was constructed using a class of linear regression algorithms based on human stem cell and embryonic transcriptome data. The single-cell transcriptome data from different groups were input into the functional assessment model to calculate the scores of various biological functions for each mesenchymal stem cell. The biological function of mesenchymal stem cells was determined by combining the average expression levels of function-related genes, enrichment items, and various functional scores in each group.

2. The method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing according to claim 1, characterized in that, The grouping method is based on cluster analysis results, tissue origin, generation, and culture conditions.

3. The method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing according to claim 2, characterized in that, The clustering analysis results are based on the "FindClusters" function performed at a resolution of 0.2, with the first 20 principal components used to define cell identity.

4. The method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing according to claim 2, characterized in that, The tissue source includes one or more of the following: amnion, umbilical cord, chorion, and adipose tissue.

5. The method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing according to claim 2, characterized in that, The culture conditions include conventional oxygen culture conditions and low-concentration oxygen culture conditions; The oxygen concentration in the conventional oxygen culture conditions was 21%. The oxygen concentration in the low-concentration oxygen culture strip is 1%.

6. The method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing according to claim 1, characterized in that, The quality control method involves statistically analyzing the proportion of mitochondrial gene expression in cells relative to the total cellular expression level, and excluding cells with the lowest 5% and highest gene counts, as well as cells with more than 15% mitochondrial genes.

7. The method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing according to claim 1, characterized in that, The marker genes are those with |avg_log2FC| > 0.25, p_val < 0.05, and min.pct > 0.

1.

8. The method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing according to claim 1, characterized in that, The biological function-related genes include at least one of the following: genes related to assessing tissue repair potential, genes related to assessing MSC differentiation potential, genes related to assessing proliferation potential, genes related to assessing extravesicle transport capacity, and genes related to assessing cell stemness.

9. The method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing according to claim 8, characterized in that, The genes related to assessing tissue repair potential include the following genes: MMP2, MMP9, CXCR2, CXCR4, CXCR5, CXCR6, and CDH2; The genes related to assessing the differentiation potential of MSCs include the following genes: ACTA2, VEGF, FBLN5, DCN, UBE2C, CTGF, and POSTN; The genes related to assessing proliferation potential include the following genes: PLK1, CCNE1, MKI67, CCNB1, E2F1, FOXM1, MCM3, MCM4 and TOP2A; The genes related to assessing the transport capacity of external vesicles include the LIN7A gene: The genes assessed for cell stemness-related characteristics include one or more of the following genes: SOX2, LIFR, and MEST.

10. The method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing according to any one of claims 1 to 9, characterized in that, It also includes conducting pseudo-time series analysis to assess the differentiation of mesenchymal stem cells.

11. The method for detecting the biological function of mesenchymal stem cells based on single-cell sequencing according to claim 10, characterized in that, The pseudo-time series analysis method includes a Monocle 3-based workflow, combining a pseudo-time series diagram and selecting the start and end positions of development, then identifying genes related to MSC development, summarizing these genes into multiple gene modules, observing the expression changes of these gene modules over time, and obtaining the temporal changes with cell differentiation.

Images