Method for detecting non-target cells in human neural stem cells

Abstract

The invention discloses a method for detecting non-target cells in human neural stem cells, which is realized by detecting one or more reagents in the following gene markers: HBA1, HBB, GYPA, PECAM-1, VWF, CD105, CD34, IBA1, CD11b, PTPRC and ADGRE1, the non-target cells comprise red blood cells, endothelial cells and microglial cells, and particularly the red blood cells are detected by detecting the expression level of the combination of HBA1, HBB and GYPA. The expression level of the combination of PECAM-1, VWF, CD105 and CD34 is used for detecting endothelial cells, the expression level of the combination of IBA1, CD11b, PTPRC and ADGRE1 is used for detecting microglial cells, and ROC AUC of non-target cells in human neural stem cells is detected to be 0.814. The method for detecting the non-target cells in the human neural stem cells is provided by combining single cell sequencing and gene detection, so that time, manpower and material cost are saved.

Description

Technical Field

[0001] The present invention belongs to the field of cell biotechnology, and particularly relates to a method for detecting non-target cells in human neural stem cells. Background Art

[0002] Neural stem cells (NSCs) are a type of cells with self-renewal ability and multi-directional differentiation potential. They play important roles in the development, regeneration, and repair of the nervous system. They can differentiate into various types of nerve cells, including neurons, astrocytes, and oligodendrocytes. However, existing extraction or induction techniques cannot completely remove non-target cells, which will exist. Non-target cells generally refer to other types of cells that are not the main target but coexist during the preparation of cell therapy products or stem cell products. These cells may come from different cell lineages, may be incompletely differentiated cells, or are unwanted cells. In the production and quality control of cell therapy products, the management of non-target cells is an important consideration. According to the characteristics of cell therapy products, if non-target cells have no adverse effects on the safety and effectiveness of the products, their composition and proportion are mainly studied in quality research, and the batch-to-batch consistency is controlled if necessary. However, for impurity components that may affect the safety of the products, they need to be removed in the process, detected in quality research, and qualitatively and quantitatively controlled.

[0003] Researchers are aware of the existence of non-target cells and also understand the significance of determining the types of non-target cells and removing non-target cells. However, the screening mechanism is complex (most current techniques use single-cell sequencing), the detection and analysis take too long, the data analysis is time-consuming and laborious. There are suitable methods for detection such as qPCR, flow cytometry, immunofluorescence, etc., but there is no direction on what cell type to use as non-target cells for detection, and there is no targeted non-target cell as an application standard to adapt to, thus forming a unified industry understanding.

[0004] Single-cell sequencing is a technique for high-throughput sequencing analysis at the single-cell level of the genome, transcriptome, and epigenome. This technique can reveal the heterogeneity in cell populations, even if these cells are from the same tissue or sample. The application of single-cell sequencing technology is very extensive and can be used in the fields of developmental biology, oncology, immunology, neuroscience, etc. It can help understand the diversity of cell types, discover new cell subpopulations, reveal the processes of cell development and differentiation, and study the mechanisms of disease occurrence. At the same time, single-cell sequencing faces problems such as technical complexity, data analysis challenges, and high costs. With the continuous development of technology, these problems are gradually being solved, and single-cell sequencing technology is gradually becoming an important tool for life science research. Summary of the Invention

[0005] To overcome the above-mentioned deficiencies of the existing technologies, the purpose of the present invention is to provide a method for detecting non-target cells in human neural stem cells. Raw data is obtained by performing single-cell sequencing analysis using the Illumina or BGI high-throughput sequencing platform, and bioinformatics analysis is carried out through Seek Tools, which mainly includes: data quality control, construction of expression matrix, cell clustering, differential gene analysis, and enrichment analysis; the data of bioinformatics analysis is further screened and evaluated, including cell types, signaling pathways, mechanisms of action, and marker markers, and the common cell types and the proportion of non-target cells are screened out, and confirmed by flow cytometry and qPCR analysis to identify the types of non-target cells in human neural stem cells, providing a basis for subsequent purification of target cells, discarding non-target cells, and understanding the changes of non-target cells during the culture process of human neural stem cells, so as to achieve the purpose of controlling impurities.

[0006] To achieve the above purpose, the present invention adopts the following technical solutions:

[0007] The present invention provides a detection product for non-target cells in human neural stem cells, which includes reagents for detecting one or more of the following gene markers: HBA1, HBB, GYPA, PECAM-1, VWF, CD105, CD34, IBA1, CD11b, PTPRC, ADGRE1;

[0008] The non-target cells in the human neural stem cells include one or more of red blood cells, endothelial cells, and microglial cells.

[0009] Preferably, the detection product includes reagents for detecting the combination of HBA1, HBB, and GYPA gene markers, and the expression levels of the combination of HBA1, HBB, and GYPA are detected by it for detecting red blood cells;

[0010] And / or the detection product includes reagents for detecting the combination of PECAM-1, VWF, CD105, and CD34 gene markers, and the expression levels of the combination of PECAM-1, VWF, CD105, and CD34 are detected by it for detecting endothelial cells;

[0011] And / or the detection product includes reagents for detecting the combination of IBA1, CD11b, PTPRC, and ADGRE1 gene markers, and the expression levels of the combination of IBA1, CD11b, PTPRC, and ADGRE1 are detected by it for detecting microglial cells.

[0012] Preferably, the ROC AUC of the reagent for detecting non-target cells in human neural stem cells is 0.814.

[0013] The present invention also provides a method for detecting non-target cells in human neural stem cells, including:

[0014] Detecting the expression level of one or more of the following gene markers by any of the detection products described above: HBA1, HBB, GYPA, PECAM-1, VWF, CD105, CD34, IBA1, CD11b, PTPRC, ADGRE1; performing qPCR relative quantitative analysis based on the expression level data obtained in step (1), then performing differential analysis △CT, and then plotting an ROC curve including sensitivity and specificity and the corresponding area under the curve (AUC) and performing ROC curve analysis, and distinguishing neural stem cells from non-neural stem cells through ROC AUC, thereby judging non-target cells in human neural stem cells.

[0015] Preferably, the non-target cells in the human neural stem cells include red blood cells, endothelial cells and microglial cells;

[0016] Detecting the human neural stem cells to be tested by a reagent for detecting one or more of the following gene markers: HBA1, HBB, GYPA, PECAM-1, VWF, CD105, CD34, IBA1, CD11b, PTPRC, ADGRE1, wherein the combination of HBA1, HBB and GYPA is used to detect red blood cells; the combination of PECAM-1, VWF, CD105 and CD34 is used to detect endothelial cells; the combination of IBA1, CD11b, PTPRC and ADGRE1 is used to detect microglial cells.

[0017] Preferably, when the ROC AUC is 0.814, human neural stem cells and non-target cells can be distinguished.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The present invention performs bioinformatics analysis on human neural stem cells of different passages and different sources by single-cell sequencing, and performs differential gene analysis on the cells to obtain a non-target cell population, and performs relative quantitative analysis by flow cytometry and qPCR to identify that the non-target cells include red blood cells, endothelial cells and microglial cells. A method for detecting non-target cells in target cells is proposed by using the combination technology of single-cell sequencing + gene detection, which saves time, manpower and material costs for the subsequent method of culturing high-quality neural stem cells. Description of the Drawings

[0019] Figure 1 It is a flow chart of completing single-cell lysis and labeling within 2 hours for the fresh single-cell suspension in the example.

[0020] Figure 2 For the original sequencing data in the example through Seek Flow chart of bioinformatics analysis using Tools.

[0021] Figure 3 Results of differential expression gene enrichment analysis of samples in the examples.

[0022] Figure 4 Cell subpopulation clustering in the examples.

[0023] Figure 5 Results of cell type bar chart analysis in the examples.

[0024] Figure 6 Results of the analysis of the percentage of target cell population and non-target cell population in the examples.

[0025] Figure 7 Results of the confirmation analysis of non-target cells by flow cytometry in the examples.

[0026] Figure 8 Graph Pad Prism analysis chart of qPCR relative quantification results in the examples.

[0027] Figure 9 Agarose gel electrophoresis of qPCR products in the examples (P9 and negative).

[0028] Figure 10 Agarose gel electrophoresis of qPCR products in the examples (P00 and positive).

[0029] Figure 11 Results of the analysis of the percentage of target cells and non-target cells after culturing cells of different passages and sources in the examples.

[0030] Figure 12 ROC curve and area under the curve (AUC) for detecting the prediction sensitivity and specificity of non-target cells in neural stem cells in the examples. Detailed implementation mode

[0031] The technical solution of the present invention will be further described below in conjunction with the examples.

[0032] Example 1 Analysis of cell population and percentage

[0033] 1. Test principle

[0034] Based on the principle of microfluidics technology, single-cell separation and capture are achieved through water-in-oil, and nucleic acid-modified Barcoded Beads are used to molecularly label the RNA from different cell sources respectively, and a high-throughput single-cell transcriptome library compatible with Illumina and MGI sequencers is constructed.

[0035] 2. Test process

[0036] The test samples are different passages of human neural stem cells that have passed the enterprise quality inspection, including 3 batches of whole brain P00, 3 batches of cortex-derived P00, 3 batches of midbrain-derived P00, 3 batches of hippocampus-derived P00, 3 batches of spinal cord-derived P00, and the P9-generation cells corresponding to the cortex, midbrain, and spinal cord.

[0037] 3. Sample type

[0038] There should be no large particle precipitates in the single-cell suspension. If there are large particle precipitates, the cell suspension needs to be filtered through a 40μm cell filter, and the cell diameter is required to be 5 - 40μm.

[0039] 4. Sample quality

[0040] Cell quantity: The minimum number of input cells per single chip should be no less than 1000;

[0041] Cell viability: The best analysis effect is achieved when the viable cell rate > 90% (counted by a cell counter);

[0042] Cell aggregation rate < 5%, and the nucleated cell rate > 70%.

[0043] 5. Sample storage

[0044] Fresh single-cell suspension, placed on ice, and the single-cell lysis and labeling experiments should be completed within 2 hours. The process is as Figure 1 shown.

[0045] After obtaining the original sequencing data (Sequenced Reads), perform bioinformatics analysis through Tools, mainly including: data quality control, single-cell data expression matrix, single-sample analysis, multi-sample analysis (cell clustering analysis, cell type identification, Marker gene analysis, functional annotation and analysis), multi-sample comparison analysis (cell type comparison analysis between different groups, differential analysis between different groups, functional annotation and analysis), etc. The process is as Figure 2 shown, specifically as follows:

[0046] First, perform data quality control on the downloaded data, mainly including sequencing data format description, removal of low-quality reads, removal of adapters, calculation of sequencing error rate, statistics of Q20, Q30, GC content, etc. After filtering the low-quality sequencing data, further evaluate the quality of the single-cell library, align the sequencing data to the reference genome for gene expression quantification, generate a preliminary cell-gene expression matrix, and then further filter the cells. This is because during the library construction process, a gel bead may have no cells or multiple cells bound to it, or due to cell apoptosis resulting in an increased proportion of mitochondrial genes. Therefore, these abnormal and low-quality cells need to be filtered, and the filtered cell-gene expression matrix is used for subsequent analysis.

[0047] After filtering out low-quality cells, the CCA method was used to merge the samples and remove batch effects, and the Seurat (Butler, Hoffman et al. 2018) software was used for clustering and result visualization. Seurat is an R package for analyzing single-cell transcriptome data, providing various functions such as t-SNE dimensionality reduction analysis, clustering analysis, and marker gene identification. After obtaining cell subset information by completing cell clustering, cell annotation was performed on the subsets, and subsequent data mining was carried out based on cell types. By calculating the differential genes between a subset and the remaining subsets, the genes specifically highly expressed in the subset could be known, and these genes were used for further manual annotation of the subset. Using the Find All Markers command in Seurat (Log2FC >= 0.25, min.pct = 0.1), the differentially expressed genes in each cluster were analyzed, and these characteristic marker genes were displayed in various forms, and genes were selected after sorting in descending order of avg_log2FC.

[0048] By performing functional enrichment analysis on the marker genes of each cluster, it was possible to know which biological functions or pathways the cluster might be involved in, further assisting in determining cell types. The cluster Profiler software was used to perform GO functional enrichment analysis, KEGG pathway enrichment analysis, and Reactome enrichment analysis on the marker genes of each cluster. Enrichment analysis is based on the principle of hypergeometric distribution, and both differential genes and background genes need to be considered simultaneously. The p-value calculation formula for this hypothesis test is as Figure 3 shown: where N is the background gene set, which is the number of proteins with GO / KEGG / Reactome annotations among all genes; n is the number of differentially expressed proteins in N; M is the number of proteins in N annotated as a specific GO term (or a specific KEGG / Reactome); m is the number of differentially expressed proteins in n annotated as a specific GO term.

[0049] The results of differential expression gene enrichment analysis of these samples were obtained from the above process, the common cell types were screened, and confirmed by flow cytometry and qPCR analysis methods to determine the cell types and their detection methods for the detection of non-target cells of human neural stem cells.

[0050] After processes such as data integration, dimensionality reduction, and clustering, the cell subset clustering situation as shown in Figure 4 was obtained:

[0051] Cell types: According to the test results, they can generally be divided into nerve cells and non-neural stem cells. Non-neural stem cells are mainly non-target cell populations, which can be divided into red blood cells, endothelial cells, and microglial cells; nerve cells include neural stem cells, transitional nerve cells, and differentiated and mature nerve cells.

[0052] Confirmation of non-target cell types: The cell populations of CLUSTER31, CLUSTER32, and CLUSTER34 circled by red solid lines are significantly separated from the rest of the cell populations as a whole, as Figure 4 shown. The characteristic genes of each subgroup were found through the Find Marker function to confirm the subgroup attributes and Gene Markers. See Table 1.

[0053] Gene Marker:

[0054] ① Red blood cells: HBA1, HBB, GYPA, GYPC, ALAS2, as shown in Table 1:

[0055] Table 1

[0056]

[0057]

[0058] ② Endothelial cells: PECAM-1, VWF, CD105, CD34, as shown in Table 2:

[0059] Table 2

[0060] PECAM-1-F GAACGGAAGGCTCCCTTGAT PECAM-1-R GGCAGCTTAGCCTGAGGAAT VWF-F CCGATGCAGCCTTTTCGGA VWF-R TCCCCAAGATACACGGAGAGG CD105-F TGCACTTGGCCTACAATTCCA CD105-R AGCTGCCCACTCAAGGATCT CD34-F CTACAACACCTAGTACCCTTGGA CD34-R GGTGAACACTGTGCTGATTACA GAPDH-F CCACTCCTCCACCTTTGAC GAPDH-R ACCCTGTTGCTGTAGCCA

[0061] ③ Microglial cells: IBA1, CD11b, PTPRC, ADGRE1, as shown in Table 3:

[0062] Table 3

[0063] IBA1-F TGTCTCCCCACCTCTACCAG IBA1-R TGGAGGGCAGATCCTCATCA CD11b-F ACTTGCAGTGAGAACACGTATG CD11b-R TCATCCGCCGAAAGTCATGTG PTPRC-F TGCAGCTAGCAAGTGGTTTG PTPRC-R ATCTGGAAGTCAGCCGTGTC ADGRE1-F CCAGTGTTAATGCCGAAGTCT ADGRE1-R GTGAACAGGTAAGCCATGACA GAPDH-F CCACTCCTCCACCTTTGAC GAPDH-R ACCCTGTTGCTGTAGCCA

[0064] As Figure 5 and 6 shown, further histogram analysis of cell types was carried out, and percentage analysis of the target cell population and non-target cell population was carried out. In the initial extraction stage, the proportion of non-target cells was less than 3%. As the cell culture process was purified and removed, the proportion of non-target cells was less than 0.2% or none. Specific operation of the culture process: Cells were suspended in a low-attachment culture flask, and the culture period was 2 weeks. The culture conditions were 35°C and 5% CO2. The medium was changed regularly during the period. After the period, digestion treatment was carried out. After centrifugation, the supernatant was discarded and the cell precipitate was resuspended to obtain a cell suspension. Cell counting was performed, and cells were inoculated at a concentration of 0.15×10^6 / mL. The cycle operation was repeated in turn until the P9 generation of cells.

[0065] Example 2 Flow Cytometry Confirmation Analysis

[0066] For the non-target cell types (red blood cells, endothelial cells, microglia) determined in Example 1, antibodies CD235a, CD34, CD45, and CD11b were used for content determination. Taking the P00-generation cells of neural stem cells obtained from the spinal cord as an example, flow cytometry was used to identify the content of non-target cells. The specific operation steps are as follows: After washing and mixing the freshly obtained cell suspension with PBS and centrifuging to discard the supernatant, the cell pellet was resuspended. Then, it was divided into a blank group, an isotype control group, and a sample group, with 100 μL in each group. Corresponding antibodies and isotype antibodies were added according to the group, with 5 μL of each antibody. Incubate at room temperature in the dark for 30 min, add the washing solution, centrifuge, and discard the supernatant, repeating 2 times. After resuspending the cell pellet, it was subjected to machine detection. As Figure 7 shown, for red blood cells: CD235a: 0.66%; for endothelial cells: CD34: 0.41%; CD45: 0.28%; for microglia: CD11b: 0.48%. The trend of the percentage analysis results of the non-target cell population is consistent with that in Example 1.

[0067] Example 3 qPCR Relative Quantification Analysis

[0068] According to the selected genes, three batches of P00 cells, three batches of P9 cells, positive cells (red blood cells, endothelial cells, microglia), and negative cells (hepatoma cells) were selected for method confirmation, and the qPCR detection method was used for detection. The specific operation steps are as follows: Neural stem cells obtained from tissue sources (spinal cord sources) were used as P00 (3 batches), passaged to the P9 generation (3 batches), positive cells (red blood cells, endothelial cells, microglia), and negative cells (hepatoma cells) were cultured. After centrifuging the cell suspension of each group to discard the supernatant, 1 mL of Trizol was added and mixed well, 200 μL of chloroform was added and shaken vigorously, then centrifuged. 400 μL of the supernatant was taken and an equal volume of isopropanol was added and allowed to stand, followed by centrifuging to discard the supernatant. The RNA was purified with 75% ethanol and diluted with enzyme-free water to measure the concentration. After reverse transcription as needed, an appropriate amount of cDNA was taken and used as a template to prepare a reaction system for qPCR, and the amplified product was subjected to agarose gel electrophoresis, as shown in Table 4 (taking the qPCR relative quantification results of red blood cells as an example).

[0069] Table 4

[0070]

[0071]

[0072]

[0073]

[0074] The relative quantitative results of red blood cell qPCR are as Figure 8 shown. Analyzed by Graph Pad Prism, during the culture process from P0 to P9, the expression levels of HBA1, HBB, and GYPA genes should show a significant downward trend, and the normalized gene expression difference △CT is much lower than that of the positive control. After culturing to the P9 generation, the expression of non-target cell gene markers is basically the same as that of negative cells.

[0075] The results of agarose gel electrophoresis are as Figure 9 and 10 shown. The amplified products are subjected to agarose gel electrophoresis. The electrophoresis instrument is set at 120V for 25 minutes. After electrophoresis, turn off the power supply, open the electrophoresis tank, take out the template, place the agarose gel on the gel imager to observe the results and take pictures. The purpose is to confirm the amplified products of qPCR by gel electrophoresis, which can determine the size and integrity of the product DNA and can corroborate the results of Example 1 and Example 2 are correct.

[0076] Example 4: Cultivation process for removing non-target cells

[0077] Cell culture is inseparable from the cultivation process. The cell bank establishment process includes the seed cell bank (P3), the master cell bank (P5), and the working cell bank (P8). The culture includes neural stem cells (P00) from sources such as the whole brain, cortex, midbrain, hippocampus, and spinal cord. According to the cultivation process, cycle operations are carried out in sequence until the P9 generation of cells. Quality identification of the final cells and analysis of the proportion of target cells and non-target cells are performed to determine that the cultivation process can achieve the expected purification and removal of non-target cells. The results are as Figure 11 shown.

[0078] Specific operation of the cultivation process: Cells are cultured in suspension in a low-attachment culture flask for a 2-week culture cycle. The culture conditions are: temperature 35°C, 5% CO2. During the cycle, the culture medium is changed regularly. After the due date, digestion treatment is carried out. After centrifugation, the supernatant is discarded and the cell pellet is resuspended to obtain a cell suspension. Cell counting is performed, and cells are inoculated at a concentration of 0.15×10^6 / mL. The cycle operation is repeated in sequence until the P9 generation of cells.

[0079] Example 5: Detection of non-target cells in neural stem cells by integrating 14 gene markers of target cells and non-target cells

[0080] By integrating 14 gene markers (erythrocytes: HBA1, HBB, GYPA; endothelial cells: PECAM-1, VWF, CD105, CD34; microglia: IBA1, CD11b, PTPRC, ADGRE1; neural stem cells: Nestin, SOX2, Pax6), relative quantitative analysis by qPCR was performed to establish a prediction mixing model for non-target cells in neural stem cells. Specific operations: Prepare a total of 26 samples including erythrocytes, endothelial cells, microglia, immune cells, cancer cells, iPSCs, and neural stem cells (P00 generation, P9 generation). After relative quantitative analysis by qPCR, differential analysis of ΔCT was performed on the 26 data. After accumulating the differential data of the 14 gene markers, ROC curve analysis of the risk prediction mixing model was performed using the R plot.roc and auc functions of IBM SPSS software.

[0081] Evaluation of prediction accuracy: Sensitivity and specificity were determined using the ROC curve and the corresponding AUC. The prediction accuracy of the model was represented by the value of AUC. The AUC value was 0.814. As Figure 12 shown, the results indicated that the prediction mixing model for non-target cells could effectively distinguish non-neural stem cells from neural stem cells and determine the presence or absence of non-target cells in neural stem cells, and could be used as a high-performance strategy for neural stem cell purification.

[0082] The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit the protection scope of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the protection scope of the present invention.

Claims

1. A product for detecting non-target cells in human neural stem cells, characterized in that: It includes reagents for detecting one or more of the following gene markers: HBA1, HBB, GYPA, PECAM-1, VWF, CD105, CD34, IBA1, CD11b, PTPRC, ADGRE1; The non-target cells in the human neural stem cells include one or more of red blood cells, endothelial cells and microglial cells.

2. The detection product for non-target cells in human neural stem cells according to claim 1, characterized in that: The detection product includes reagents for detecting a combination of HBA1, HBB and GYPA gene markers.

3. The detection product for non-target cells in human neural stem cells according to claim 1, characterized in that: The detection product includes reagents for detecting a combination of PECAM-1, VWF, CD105 and CD34 gene markers.

4. The detection product for non-target cells in human neural stem cells according to claim 1, characterized in that: The detection product includes reagents for detecting a combination of IBA1, CD11b, PTPRC and ADGRE1 gene markers.

5. The method for detecting non-target cells in human neural stem cells according to claim 1, characterized in that: The ROC AUC of the reagent for detecting non-target cells in human neural stem cells is 0.

814.

6. A method for detecting non-target cells in human neural stem cells, characterized in that: include: The detection product according to any one of claims 1 to 5 is used to detect the expression level of one or more of the following gene markers: HBA1, HBB, GYPA, PECAM-1, VWF, CD105, CD34, IBA1, CD11b, PTPRC, ADGRE1; based on the expression level data obtained in step (1), qPCR relative quantitative analysis is performed, and then a differential analysis ΔCT is performed, and then a ROC curve including sensitivity and specificity and the corresponding area under the curve (AUC) is drawn, and neural stem cells and non-neural stem cells are distinguished by ROC AUC, thereby determining non-target cells in human neural stem cells.

7. The method for detecting non-target cells in human neural stem cells according to claim 6, characterized in that: The non-target cells of human neural stem cells include red blood cells, endothelial cells and microglia; The human neural stem cells to be tested are detected by reagents that detect one or more of the following gene markers: HBA1, HBB, GYPA, PECAM-1, VWF, CD105, CD34, IBA1, CD11b, PTPRC, and ADGRE1, wherein the combination of HBA1, HBB and GYPA is used to detect red blood cells; the combination of PECAM-1, VWF, CD105 and CD34 is used to detect endothelial cells; the combination of IBA1, CD11b, PTPRC and ADGRE1 is used to detect microglia.

8. The method for detecting non-target cells in human neural stem cells according to claim 6, characterized in that: When ROCAUC was 0.814, human neural stem cells and non-target cells could be distinguished.

Images