KR20230144759A - Gene-defined cancer-origin cell, and a classification method thereof - Google Patents

Abstract

The present invention is a group consisting of NRXN1, It relates to a method of isolating brain cancer origin cells, comprising the step of confirming the expression level of one or more selected proteins or genes encoding them. The present inventors have found that in brain cancer, especially glioblastoma, for which it is difficult to find an effective treatment and the prognosis and survival rate are very poor, the above factors not only function as reliable diagnostic markers, but also suppress the tumor by regulating their expression, thereby suppressing the intractable disease, glioblastoma. It can be useful as a fundamental therapeutic composition that significantly improves the survival rate of patients with blastoma.

Description

Gene-defined cancer-origin cell, and a classification method thereof}

The present invention relates to cancer origin cells defined by genes, and methods for classifying them.

Cancer is one of the most common causes of death worldwide. Approximately 10 million new cases occur each year, accounting for approximately 12% of all deaths, making it the third most common cause of death. Among various types of cancer, brain cancer in particular occurs regardless of age and has a higher incidence in children than other cancers. Brain cancer is divided into primary brain tumors that develop in brain tissue and the meninges surrounding the brain, and secondary brain tumors that metastasize from cancer that originates in the skull or other parts of the body. Symptoms include motor paralysis, sensory paralysis, language impairment, and visual impairment. , local symptoms such as balance disorders, and symptoms of intracranial hypertension. The brain cancer includes various types of cancer such as glioblastoma multiforme, malignant glioma, lymphadenoma, germ cell tumor, and metastatic tumor, unlike cancer that occurs in other tissues in which one or two types of cancer are tissue-specific. , Among them, glioma is a tumor that accounts for 60% of primary brain tumors. It has a high incidence and is difficult to treat, so it is a malignant tumor for which there is no specific treatment other than radiation therapy. Among them, glioblastoma (GBM), which is classified as the most malignant, has a very high resistance to radiation and anticancer drug treatment compared to other cancers, and once diagnosed, the survival period is only 1 year, so each patient's Proper diagnosis and understanding of the origin and course are important. In addition, in the case of brain tumors, not only is it difficult for drugs for treatment to be delivered to the target brain area due to the presence of the brain blood barrier, but also the development of treatments is difficult due to the relative lack of understanding of brain neurobiology. The reality is that it is not active. Moreover, compared to other brain tumors, glioblastoma shows an aggressive variant, which can lead to fatal results within a few weeks if not treated quickly. In addition to surgical treatment, glioblastoma is treated with radiation therapy and chemical drug treatment, but there is no perfect treatment due to causes such as the occurrence of resistant mutations and recurrence by tumor stem cells.

Therefore, as a study for early diagnosis and understanding of the origin of glioblastoma and development of new treatments based thereon, it was confirmed that cells isolated from the subventricular zone were glioblastoma origin cells (Oc1 and Oc2), and the above Glioblastoma cells of origin were defined by gene expression patterns. The glioblastoma origin cells of the present invention and the method for classifying them from the subventricular zone are expected to be greatly used in the prevention and treatment of glioblastoma by understanding the origin of glioblastoma development and blocking the progression from cancer origin cells to cancer cells.

The present inventors have made extensive research efforts to discover the origin of brain cancer, especially glioblastoma, for which it is difficult to find an effective treatment and its prognosis and survival rate are very poor. As a result, the present invention was completed by isolating and defining glioblastoma origin cells (Oc1 and Oc2) through gene expression analysis in cells isolated from the subventricular zone.

Therefore, the object of the present invention is to produce one or more proteins selected from the group consisting of LUZP2, RORA, SLC1A2, TNR, The aim is to provide a method of isolating brain cancer origin cells, which includes the step of confirming the expression level of the gene.

Another object of the present invention is to measure the expression level of one or more proteins selected from the group consisting of LUZP2, RORA, SLC1A2, TNR, XKR4, CERCAM, FTL, LAMA2, MBP, PLP1, PLXDC2, and TF, or genes encoding them The object is to provide a composition for diagnosing glioblastoma containing the following agent as an active ingredient.

However, the technical problem to be achieved by the present invention is not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the description below.

DETAILED DESCRIPTION OF THE INVENTION Various embodiments described herein are described below with reference to the drawings. In the following description, various specific details, such as specific forms, compositions, and processes, are set forth in order to provide a thorough understanding of the invention. However, certain embodiments may be practiced without one or more of these specific details or in conjunction with other known methods and forms. In other instances, well-known processes and manufacturing techniques are not described in specific detail so as not to unnecessarily obscure the invention. Reference throughout this specification to “one embodiment” or “an embodiment” means that a particular feature, form, composition or characteristic described in connection with the embodiment is included in one or more embodiments of the invention. Accordingly, the phrases “in one embodiment” or “an embodiment” expressed in various places throughout this specification do not necessarily refer to the same embodiment of the invention. Additionally, particular features, shapes, compositions, or properties may be combined in any suitable way in one or more embodiments.

Unless there is a special definition in the specification, all scientific and technical terms used in the specification have the same meaning as commonly understood by a person skilled in the art in the technical field to which the present invention pertains.

Throughout the specification, when a part is said to “include” a certain element, this means that it may further include other elements rather than excluding other elements, unless specifically stated to the contrary.

According to one aspect of the present invention, NRXN1 (Neurexin 1), 1 Member 2), MEG3 (Maternally Expressed 3), TNR (Tenascin R), NRCAM (Neuronal Cell Adhesion Molecule), PTPRZ1 (Protein Tyrosine Phosphatase Receptor Type Z1), LUZP2 (Leucine Zipper Protein 2), RORA (RAR Related Orphan) Receptor A), PLP1 (Proteolipid Protein 1), TF (Transferrin), MBP (Myelin Basic Protein), CERCAM (Cerebral Endothelial Cell Adhesion Molecule), TMEM144 (Transmembrane Protein 144), PLXDC2 (Plexin Domain Containing 2), FTL (Ferritin) Light Chain), and LAMA2 (Laminin Subunit Alpha 2), confirming the expression level of one or more proteins selected from the group consisting of, or genes encoding the same.

The present inventors have made extensive research efforts to discover the origin of brain cancer, especially glioblastoma, for which it is difficult to find an effective treatment and its prognosis and survival rate are very poor. As a result, the present invention was completed by isolating and defining glioblastoma origin cells (Oc1 and Oc2) through gene expression analysis in cells isolated from the subventricular zone.

As used herein, the term “subventricular zone (SVZ)” refers to a region located on the outer wall of each lateral ventricle of the vertebrate brain. This region is present in both fetal and adult brains, and in embryonic life the subventricular zone represents a secondary proliferative zone containing neural progenitor cells that divide during neurogenesis to generate neurons.

As used herein, the term “brain tumor” refers to a solid (liquid) neoplasm formed within the skull, a tumor within the brain or spinal canal, and is a general term for tumors occurring in the brain matter and meninges. This is a tumor that grows in brain cells. It includes all neoplasms that develop within the skull, originating from blood vessels, nerves, and meninges. Symptoms vary depending on where they occur, but commonly include headaches, vomiting, abnormal movements of the limbs (convulsions, paralysis), and visual impairment.

According to a specific embodiment of the present invention, there is a separation method in which the brain cancer origin cells are cells that develop into glioblastoma (GBM).

As used herein, the term “glioblastoma (GBM)” refers to a type of glioma, the most common and severe form of tumor that occurs primarily in the brain. Symptoms and signs that first occur are not disease-specific and may be stroke-like, such as headaches, personality changes, or nausea. Symptoms worsen rapidly and may lead to unconsciousness.

According to a specific embodiment of the present invention, there is a separation method further comprising: confirming an increase in the copy number of chromosome 7 in cells isolated from the subventricular zone.

According to another aspect of the present invention, the present invention provides a group consisting of NRXN1, Provided is a composition for predicting the development of brain cancer, comprising as an active ingredient an agent for measuring the expression level of one or more proteins selected from, or genes encoding the same.

As used herein, the term “prediction” refers to evaluating whether a subject has developed a specific disease or disease to a severe degree or is at risk of developing a severe disease in the future based on a marker that has a significant correlation with the severity.

As used herein, the term “predictive composition” refers to NRXN1, , an integrated mixture or device containing an agent for measuring the expression level of one or more proteins selected from the group consisting of MBP, CERCAM, TMEM144, PLXDC2, FTL, and LAMA2, or the genes encoding them. This means that it can also be expressed as a “prediction kit.”

According to a specific embodiment of the present invention, from the group consisting of NRXN1, The agent for measuring the expression level of one or more selected proteins may be a composition that is an aptamer, antibody, or antigen-binding fragment thereof that specifically binds to the protein.

According to the present invention, the protein of the present invention can be detected according to an immunoassay method using an antigen-antibody reaction and used to analyze glioblastoma in an individual. This immunoassay can be performed according to various immunoassay or immunostaining protocols developed conventionally.

For example, when the method of the present invention is performed according to a radioimmunoassay method, antibodies labeled with radioisotopes (e.g., C14, I125, P32, and S35) may be used. In the present invention, one or more proteins selected from the group consisting of NRXN1, Antibodies that are recognized as adversaries are polyclonal or monoclonal antibodies, and more specifically, monoclonal antibodies.

The antibody of the present invention can be produced by methods commonly practiced in the art, for example, a fusion method, a recombinant DNA method, or a phage antibody library method.

Glioblastoma can be predicted by analyzing the intensity of the final signal by the above-described immunoassay process. That is, one or more selected from the group consisting of NRXN1, If the signal for the protein is weaker or stronger than that of a normal sample, the risk of glioblastoma is judged to be increased.

As used herein, the term “antigen binding fragment” refers to a portion of a polypeptide that can bind an antigen in the overall structure of an immunoglobulin, such as F(ab')2, Fab', Fab, Fv, and Including, but not limited to scFv.

As used herein, the term “specifically binding” has the same meaning as “specifically recognizing,” meaning that an antigen and an antibody (or a fragment thereof) specifically interact through an immunological reaction. It means to do.

The present invention may use an aptamer that specifically binds to the marker protein of the present invention instead of an antibody. As used herein, the term “aptamer” refers to a single-stranded nucleic acid (RNA or DNA) molecule or peptide molecule that binds to a specific target substance with high affinity and specificity.

According to a specific embodiment of the invention, selected from the group consisting of NRXN1, An agent for measuring the expression level of a gene encoding one or more proteins may be a composition that is a primer or probe that specifically binds to the gene.

As used herein, the term "nucleic acid molecule" is meant to comprehensively include DNA (gDNA and cDNA) and RNA molecules, and nucleotides, which are the basic structural units in nucleic acid molecules, include not only natural nucleotides but also those with modified sugars or base sites. Also includes analogues.

As used herein, the term "primer" refers to conditions that induce the synthesis of a primer extension product complementary to a nucleic acid chain (template), that is, the presence of nucleotides and a polymerizing agent such as DNA polymerase, and synthesis under conditions of appropriate temperature and pH. refers to an oligonucleotide that acts as the starting point. Specifically, the primer is a single strand of deoxyribonucleotide. Primers used in the present invention may include naturally occurring dNMP (i.e., dAMP, dGMP, dCMP, and dTMP), modified nucleotides, or non-natural nucleotides, and may also include ribonucleotides.

The primer of the present invention may be an extension primer that anneals to the target nucleic acid to form a sequence complementary to the target nucleic acid by template-dependent nucleic acid polymerase, which extends to the position where the immobilized probe is annealed, thereby forming the probe. Occupies the annealed area.

Extension primers used in the present invention target nucleic acids, such as NRXN1, It includes a hybridized nucleotide sequence complementary to a specific base sequence of one or more genes selected from the group consisting of. The term “complementary” means that a primer or probe is sufficiently complementary to selectively hybridize to a target nucleic acid sequence under predetermined annealing or hybridization conditions, including substantially complementary and perfectly complementary. ), and specifically means completely complementary cases. As used herein, the term “substantially complementary sequence” refers not only to completely identical sequences, but also to sequences that are partially mismatched with the sequence being compared, to the extent that they can anneal to a specific sequence and serve as a primer.

Primers should be sufficiently long to prime the synthesis of the extension product in the presence of a polymerizing agent. The suitable length of the primer depends on a number of factors, such as temperature, pH, and the source of the primer, but is typically 15-30 nucleotides. Short primer molecules generally require lower temperatures to form sufficiently stable hybrid complexes with the template. The design of such primers can be easily performed by a person skilled in the art by referring to the target nucleotide sequence, for example, by using a primer design program (eg, PRIMER 3 program).

As used herein, the term “probe” refers to a linear oligomer having a natural or modified monomer or linkage containing deoxyribonucleotides and ribonucleotides that can hybridize to a specific nucleotide sequence. Specifically, the probes are single-stranded, more specifically deoxyribonucleotides, for maximum efficiency in hybridization. As a probe used in the present invention, a sequence that is completely complementary to the specific base sequence of the marker gene of the present invention may be used, but a substantially complementary sequence may be used to the extent that it does not interfere with specific hybridization. This may also be used. In general, because the stability of the duplex formed by hybridization tends to be determined by the match of the terminal sequences, it is preferable to use a probe complementary to the 3'-end or 5'-end of the target sequence. do.

According to a specific embodiment of the invention, the composition may be one in which the brain cancer is glioblastoma (GBM).

According to another aspect of the present invention, the present invention consists of NRXN1, Provided is a kit for predicting the development of brain cancer, which includes a composition for predicting the development of brain cancer containing as an active ingredient an agent for measuring the expression level of one or more proteins selected from the group, or genes encoding the same.

According to a specific embodiment of the invention, the kit may be used for cells isolated from the subventricular zone (SVZ).

According to a specific embodiment of the invention, the cells isolated from the subventricular zone may be cells with an increased copy number of chromosome 7, which may be a kit.

According to a specific embodiment of the invention, the kit may be an RT-PCR kit, a DNA chip kit, an ELISA kit, a protein chip kit, a rapid kit, or a multiple reaction monitoring (MRM) kit.

According to another aspect of the present invention, NRXN1, , measuring the expression level of one or more proteins selected from the group consisting of TF, MBP, CERCAM, TMEM144, PLXDC2, FTL, and LAMA2, or genes encoding the same. .

According to a specific embodiment of the invention, the prediction method may further include the step of confirming an increase in the copy number of chromosome 7 in cells isolated from the subventricular zone.

According to a specific embodiment of the invention, when the expression of one or more proteins selected from the group consisting of NRXN1, It may be a prediction method that predicts that the individual has a high possibility of developing brain cancer.

According to a specific embodiment of the invention, when the expression of one or more proteins selected from the group consisting of LUZP2, RORA, SLC1A2, TNR, and XKR4, or the genes encoding them, is increased, the likelihood of the subject developing brain cancer is high. It may be a predictive or predictive method.

According to a specific embodiment of the invention, when the expression of one or more proteins selected from the group consisting of PLP1, TF, MBP, CERCAM, TMEM144, PLXDC2, FTL, and LAMA2, or the genes encoding them, is reduced, the It may be a predictive method that predicts that the likelihood of developing brain cancer is high.

According to a specific embodiment of the invention, when the expression of one or more proteins selected from the group consisting of CERCAM, FTL, LAMA2, MBP, PLP1, PLXDC2, and TF, or the genes encoding them, is reduced, the subject develops brain cancer. It may be a prediction method that predicts what is likely.

According to a specific embodiment of the invention, the brain cancer may be glioblastoma (GBM), a prediction method.

According to another aspect of the present invention, the present invention includes the steps of (a) treating brain cancer origin cells isolated by the method of claim 1 with a candidate substance for preventing brain cancer; and,

(b) determining that the candidate material has a brain cancer prevention effect when the brain cancer origin cells are killed.

The term "candidate" used when referring to the screening method of the present invention is added to an unknown sample for the purpose of testing the brain cancer prevention effect, and is used to determine the brain cancer prevention effect of NRXN1, , one or more proteins selected from the group consisting of PLP1, TF, MBP, CERCAM, TMEM144, PLXDC2, FTL, and LAMA2, or used in screening to test whether they affect the activity or expression level of the gene encoding them. means substance. The test substances include, but are not limited to, compounds, nucleotides, peptides, and natural extracts. In biological samples treated with test substances, selected from the group consisting of NRXN1, The step of measuring the expression level or activity of one or more proteins, or the genes encoding them, can be performed by various expression level and activity measurement methods known in the art.

According to another aspect of the present invention, the present invention includes the steps of (a) treating brain cancer origin cells isolated by the method of claim 1 with a candidate substance for preventing brain cancer;

(b) selected from the group consisting of NRXN1, Reconfirming the expression level of one or more proteins or genes encoding them; and,

(c) determining that the candidate material has a brain cancer prevention effect when the expression level of the protein or gene is significantly different compared to before treatment with the candidate material; a screening method for a candidate material for preventing brain cancer, including to provide.

The features and advantages of the present invention are summarized as follows:

(a) The present invention is NRXN1, Provided is a method of isolating brain cancer origin cells, comprising the step of confirming the expression level of one or more proteins selected from the group consisting of, or the genes encoding them.

(b) The present inventors have found that in brain cancer, especially glioblastoma, for which it is difficult to find an effective treatment and the prognosis and survival rate are very poor, the above factors not only function as highly reliable diagnostic markers, but also suppress the tumor by regulating their expression, resulting in incurable disease. It can be useful as a fundamental therapeutic composition that significantly improves the survival rate of patients with glioblastoma disease.

Figure 1 relates to a human tissue-based identification study to find glioblastoma progenitor cells. Next-generation sequencing of the transcriptome can be used to predict patterns of structural changes at the chromosome level, such as chromosome gains and losses from human tissues. A stem cell niche, a subventricular sample of the human brain, was included in this study. In addition to bulk tissue and cellular level (single cell and mononuclear) RNA sequencing, we included complete sequencing data (SRP145073) for chromosomal pattern identification. 1b is for chromosome copy number patterns calculated from bulk tissue RNA sequencing samples, with sample percentages summarized in a pie chart as patterns for chromosomes 7 and 10. 1c concerns mononuclear RNA array-based cell identification, where two types of cells showed distinct gene expression patterns and were labeled as cell 1 and cell 2. After confirming the cell-level chromosomal patterns of chromosome 7 q-arm gain and chromosome 10 loss, cells showing only chromosome 7 q-arm gain were designated as GBM origin-cell 1 (Oc1), chromosome 7 q. The cell showing -arm gain and chromosome 10 loss was defined as origin-cell 2 of GBM. 1d shows the tumor-progenome profiling strategy with alteration patterns of chromosomes 7 and 10, copy number patterns and cellular signatures of the two origin cells as vulnerable cells before transformation into glioblastoma tumor cells. It was assumed to be a marker tool to find.
Figure 2A shows that two types of origins were discovered from integrated human tissue-derived data (total cells n = 59,417). Origin cell 1 (brown) and origin cell 2 (yellow) are adjacent to the main cluster of tumor cells consisting of neural progenitor cell group (light blue), neuron-like cell group (gray), cell cycle activated cell group (blue), and mesenchymal cell group (red). There is. The present invention focuses on the main cluster of tumors. 2b shows tumor-free subventricular zone samples (left, tumor-free subventricular zone samples from control disease, tumor-free subventricular zone samples from glioblastoma) or tumor-related samples (right, tumor-glioblastoma or infiltration of glioblastoma tumor tissue). 2c focuses on mononuclear patient GN1, and two cell types, Origin-cell 1 and 2, were found in the subventricular zone and in the tumor. 2D examines chromosomal gain events of 1 copy gain (orange), 2 copy gain (red), or 1 copy loss (blue) in tumor-free subventricular samples. 2e, matched tumor samples show that the Origin-cell 1 cluster is more likely to be in a chromosomally neutral state than Origin-cell 2 and other cells. 2f, cells from patient GN1 show the cumulative copy number of chromosome 7q arm from Origin-cell 1 to Origin-cell 2. 2g Comparing the two types of chromosomal alterations at the level of individual cells revealed the ordering of chromosomal copy number changes. Left, chromosome 7q is followed by an increase in chromosome 7p. Right, gain of chromosome 7q is followed by loss of chromosome 10.
Figure 3 shows the connectome of ligand receptor signaling for each cell type. 3a, top, a tumor-free subventricular sample of glioblastoma without tumor infiltration. Bottom, glioblastoma tumor tissue, with different modes defined by mode-specific receptors and ligand genes. The stem cell mode is defined by three receptors (CD44, ITGA4, NRCAM) and four ligands (BCAN, CD14, LPL, VIM). 3b, connectome edgeweight by gene Origin-cell 1 (brown), Origin-cell 2 (yellow), and cycling cell (blue) were selected for subventricular specimens and tumor specimens. At the top, a cycle plot of interconnected genes in a subventricular sample is shown, and at the bottom, a cycle plot of interconnected genes in a glioblastoma tumor sample is shown. 3c examines two types of glioblastoma tumors in terms of progenitor cells, circulating cells, and cancer stem cells. Glioblastoma tumor samples are subclassified according to the mutational status of the TERT promoter in the tumor tissue. Matched subventricular samples are labeled according to their respective TERT promoter mutation status, and cell presence is determined using four sets of genes (cell of origin 1, cell of origin 2, cycle cell, and cancer stem cell 24) extracted from single-cell level RNA array data. Predictions were made by applying them to expression values from bulk tissue RNA array data sets (TERT promoter available data samples were used for this analysis). Left, TERT promoter wild-type glioblastoma tumor and patient matched subventricular sample. TERT promoter mutant glioblastoma tumor and patient matched subventricular sample. student t-test, <0.001**, <0.01**, non-significant results are not shown.
Figure 4 relates to a mouse cell model of CRISPR/Cas9-based gene editing technology 2. In 4a, subventricular cells of mice with gene mutations in Trp53 and Pten were isolated by induction of the EGFRvii gene. Mutant cells in the subventricular region transform into mouse glioblastoma tumors approximately 11 to 16 weeks after mutation introduction. Three types of cells were labeled and compared. 4b used cell-specific markers to identify subpopulations of cells from the three cell types. Other cell markers were negative for these three types of cells. 4c is for tumorigenesis in tumor cells, no tumorigenesis was found in other cells. That is, they were cells isolated from the lower part of the contralateral ventricle without genetic modification or original cells isolated from the lower part of the CRISPR/Cas9-edited ventricle. 4d for individual cell integration maps of the three cell types, left, showing the three cell types mixed, and right, showing the four cell states identified in the three cell types. 4e is for quantitative comparison of four cell states in three mouse cell types, and in 4f, gene module scores were calculated by single-cell RNA sequencing of three mouse cell types. Gene module scores were used to predict the relative cell number of a given cell state (arbitrary relative units). Additionally, the mouse cell score was calculated by converting the cell 1 specific gene of human origin into the mouse homolog gene. Mouse cell scoring uses cell 2 specific genes of human origin. 4g Chromosomal changes were calculated from single-cell RNA arrays, and the most significant changes were noted in cell state maps. It was confirmed that the front and rear parts of chromosome 7 had different chromosomal change patterns in three mouse cell models.
Figure 5 shows a subventricular sample collected from an adult human brain for identification of glioblastoma cells. Figure 5A is an overview of the human study part, which used a new and previously published data set (SRP145073)2 for identification of glioblastoma progenitor cells. Figure 5B numbers single cell-level RNA sequencing of patient specimens from 1 to 14, but not bulk tissue-level RNA sequencing patient specimens, and patient case 4 was excluded due to specimen preparation quality. Figure 5C is a simple graphical summary of preparing subventricular samples from adult brains during surgery. 5D to 5F show magnetic resonance medical images of a patient and sampling locations in the subventricular region. Figure 5D represents a single nuclear RNA sequencing patient case with glioblastoma (label starting with GN), and all subventricular specimens were found to be uncontaminated by tumor by pathological evaluation. That is, mononuclear RNA is sequenced from patients with control disease (labels starting with CN). No tumor cell contamination was found in the samples. Figure 5f relates to single-cell RNA sequencing of a glioblastoma patient, and single-cell RNA sequencing failed in two ventricular zone specimens (GC11 and GC13). Tumor mixed subventricular samples were also included in the study as a control group (GC12).
Figure 6 shows cell of origin evidence for the ventricular region obtained from bulk tissue RNA array and bulk tissue whole exome array data. Figure 6a is a comparison of gene expression in the subventricular group, in which the tumor-free subventricular region of left glioblastoma was compared with control disease. Right, set enrichment analysis results of genes highly upregulated in the tumor-free lower ventricles than in control disease. Figure 6B shows gene set enrichment analysis comparing subventricular and glioblastoma tumors, resulting in genes upregulated in the subventricular region of glioblastoma. Subventricular areas of control disease were excluded. Figures 6C-6E show the chromosomal patterns, variant allele frequencies and mutation times of tumor-free ventricular regions of adult human brain in the whole exome sequencing dataset of SRP145073. Figure 6c shows the chromosomal pattern of copy number gain on the q-arm side of chromosome 7 in patient P245. Figure 6D shows that the chromosomal pattern was relatively distinct in the subventricular specimens identified as unidirectional shared mutations from the subventricular region to the tumor (P520, P276, P26, P499). Figure 6e confirms that chromosome 7 q-arm does not clearly increase in patient samples (P37, P396, P881, P246). Figure 6f relates the chromosomal pattern of copy numbers across all chromosomes, calculated from bulk tissue RNA sequencing using a previously published algorithm. All copy number results are adjusted and normalized across the entire sample. A reference group was used for copy number calculations with control disease-free subventricular zone (n = 9). Figure 6g shows chromosome pattern verification in whole sequencing data and bulk tissue-level RNA sequencing data. Since these two methods show limitations, we prepared single-cell level analysis.
Figure 7 relates to specific gene sets and patient-specific proportions of cells of origin in all patient samples (related to Figures 2A-2B). Figure 7A shows a network plot of the gene signature of cell of origin 1. Enrichment analysis of the gene set was done based on the Kyoto Encyclopedia of Genes and Genomes (KEGG), and the top 100 genes defined cell of origin 1. Genes were extracted from an integrated single nuclear RNA array dataset of patient GN1 (tumor-free subventricular and glioblastoma). Figure 7b shows the network plot of cell of origin 2, where the top 100 genes of cell of origin 2 were extracted from the integrated single nuclear RNA sequencing patient sample GN1. Figure 7C extracts genes defining cell types and depicts normalized expression in a heatmap. Yellow indicates high expression and purple indicates low expression. All single-cell-level RNA sequencing data sets were integrated for analysis, and ribosomal protein genes or mitochondrial genes were included throughout the process and these genes were removed prior to the gene set extraction process. Figure 7D shows cell type percentages summarized by individual patient, with mononuclear RNA array sample GN1 comprised of a glioblastoma tumor and a tumor-free subventricular region of glioblastoma. Cells of origin 1 and 2 are indicated in brown and yellow, respectively. The cell percentage graph shows that oligodendrocytes (light brown) are predominantly expressed in the single nuclear RNA array dataset (GN-dataset). In the single-cell RNA sequencing dataset (GC-dataset), human tissue samples are predominantly expressed by neural progenitor cells (light blue) rather than oligodendrocytes (light brown). Among these samples, the tumor-mixed subventricular zone (GC12 and GC13) of glioblastoma (GC14) has a lower proportion of oligodendrocytes (light brown) than the tumor-free subventricular zone (GC12 and GC13) based on two single-cell RNA sequencing analysis. Tumor-free ventricular zone samples from control disease (CN8, CN9, and CN10) are predominantly expressed by oligodendrocytes (light brown) and microglia (green). Two types of cell of origin were identified in all samples.
Figure 8: Chromosome patterns on cell type maps of tumor-free subventricular zone and glioblastoma tumor tissue (related to Figures 2E-2G). The chromosomal patterns of chromosome 7 p-arm, 7 q-arm and chromosome 10 are summarized by patient and sample type, with the seven patient samples listed from top to bottom, with sample types divided from left to right. Subventricular zone samples (light blue) and tumor samples (light red) are shown, and seven single nuclear RNA sequencing patient cases are shown. Individual cell locations are derived from Figure 2A. The panel provides maps of cell location by tissue type (subventricular zone or glioblastoma tumor), cell type (cell of origin 1, cell of origin 2, neural progenitor-like cells, or mesenchymal cells), and chromosome pattern (chromosome 7 p-arm, chromosome 7 q-arm or chromosome 10). Overall, the tumor samples showed a chromosomal pattern of copy number gains in the 7p-arm and 7q-arm (except in glioblastoma of GN5), and loss of chromosome 10 was found in tumor samples except in GN2 and GN5. Two types of cells of origin are found in tumor-free subventricular zone samples. Additionally, cells affected by chromosome 7 q-arm copy number gain are more abundant in the two cell groups of origin in tumor-free subventricular zone samples than are chromosome 10 loss events in the samples. That is, subventricular zone samples excluding the cell of origin of GN7: chromosome pattern 7 p-arm, 7 q-arm and chromosome 10 cannot be compared with the few cells affected by the chromosome pattern. To summarize the above description of the tumor-free subventricular zone and glioblastoma tumor samples, two types of cells of origin are at the center of the chromosomal pattern.
Figure 9 shows that the genetic signature of the cell of origin was found in all samples of RNA sequencing at the single cell level, with cell of origin 1-rooted pseudotime showing cell of origin 2 at an early stage. Figure 9A shows a pseudo-time trajectory plot using cell of origin 1 as the starting point or root. Left, represents integrated trajectory plots for tumor-free subventricular zone samples from single nuclear RNA sequencing (GN1, GN2, GN3, GN5, GN6, and GN7) and single cell RNA sequencing data (GC14), with purple indicating the root is the initial event. If the initial differentiated cells in the root, yellow indicates further differentiated cells in the root only if the root is an initial event. Right, trajectory plot is based on all single cell level tumor samples (GN1, GN2, GN3, GN5, CN6, CN7, GC11, GC12, GC13 and GC14). Figures 9B and 9C show that the subventricular zone cells and tumor cells were subsampled by each patient, artificially selecting cell of origin 1 as the root, and found that the cell of origin 2 group appeared as relatively early cells in the trajectory analysis. Figure 9B relates to a trajectory plot of a single-nucleus RNA sequencing dataset, which, unlike the single-cell RNA sequencing dataset below, shows neuron-like clusters (dark gray) downstream of cell of origin 1 in patients GN2, GN3, and GN5. was discovered. Figure 9C shows a trajectory plot of the single cell RNA sequencing data set, confirming that patient GC11 had no trace lines crossing the cell clusters 1 or 2 of origin.
Figure 10: Impact on glioblastoma patient survival based on cell-of-origin-related gene signatures, enrichment in samples, and cell identification scores from bulk tissue RNA sequencing. Figure 10A Estimated identity of cells of origin 1 and 2 (cell identity score) from bulk tissue-level RNA sequencing data for three types of samples: tumor-free subventricular region of control disease (n = 9), tumor-free Subventricular glioblastoma areas (n = 40) and glioblastoma tumor samples (n = 126) are shown. Left, cell of origin 1 was assumed to be a set of genes derived from cell of origin 1 in the tumor-free subventricular zone of glioblastoma patient GN1. Right, cell of origin 2 was assumed to be the gene set of cell of origin 2 in the tumor-free subventricular zone of glioblastoma patient GN1. Figures 10B and 10C show patient survival after glioblastoma surgery by cell identity score of the two cells of origin. Figure 10B shows patient survival by cell of origin score in subventricular zone samples and Figure 10C shows patient survival by cell of origin score in tumor tissue. In Figure 10D, the cell identification scores of different gene sets are summarized in a heatmap. Public data sets or published gene sets were used to find signature-assigned genes. The heatmap is divided into groups (Brain cell type group26, Glioblastoma MGH group27, Columbia group28, and origin-cell type in this study), and subgroup signatures were clustered by Euclidean distance. Figures 10e and 10f show that the cross-correlation between signatures was calculated with the dot product of cell identification scores, with red indicating higher correlation between signatures and blue meaning lower correlation between signatures. Figure 10E shows the internal product heatmap of tumor-free subventricular zone samples of glioblastoma (n = 40), and Figure 10F shows the internal product heatmap of all tumor samples (n = 126).
Figure 11 shows that cells of origin were identified in the subventricular zone of the CRISPR/Cas9-based mutagenic mouse brain, and that the isolated cells of origin were confirmed to be non-tumorigenic in a brain allograft model. Figure 11A is a summary of the previously published method for generating CRISPR/Cas9-based genome-edited mice with tumor suppressor mutations and EGFRvII amplification, followed by electroporation of plasmids to induce tdTomato-positive cells (red) in the subventricular zone of the mouse brain. Create. After 4-14 weeks, mouse brain sections show cells migrating from the subventricular to subcortical areas. Tumors with necrosis, labeled as tumor cells or mouse glioblastoma, could be detected 14 to 23 weeks after electroporation. Figure 11B shows representative time series images of a coronal section of a mouse brain aligned. Top, confocal image (red: tdTomato-positive cells); bottom, H&E image. Scale bar: 1 mm. A conceptual time window during tumorigenesis can be used for the separation of cells of origin and tumor cells (tumor cells that have migrated from the subventricular zone). Figure 11c refers to the conceptual location where the cells used in the experiment are illustrated. Origin-cells are tdTomato-positive cells into which a plasmid has been inserted, and tumor cells are tdTomato-positive cells into which a plasmid has been inserted. Figure 11D shows representative microscopic images of cells captured in spherical culture conditions. Scale bar: 200 μm. BF, bright field; tdTomato filter: Red indicates tdTomato positive. Figure 11e shows an anatomy-based illustration of three types of cells. Figure 11f shows mouse brain injection experiments using cells isolated from different locations, subventricular regions, or tumors as an indicator of tumor formation potential. These cells were injected into the brains of mice, and while mice injected with tumor cells had poor survival, mice injected with cells of origin were unaffected (see Figure 4c). Representative images of IVIS, magnetic resonance imaging (9.4T), H&E, and confocal images using tdTomato (red) are presented. Scale bar: 1 mm.
Figure 12 shows the intermediate cell state from the control state to the tumor state for CRISPR/Cas9-based origin-cells in terms of transcriptome changes and chromosomal modification patterns at each stage. Figure 12A Heatmap specifies cell states in four cell states (control, early, late, and tumor state) from integrated single-cell RNA sequencing data from three different cells (control cells, origin cells, and tumor cells). Shows genes. Figure 12b compares the gene expression of Cspg4, an oligodendrocyte precursor cell marker, by state and cell type. Figure 12c shows the module scores of the four states in the integrated data set for the three types of cells, where the gene sets were converted from human to mouse and scores for each cell state were calculated (related to Figure 4f). Figure 12D summarizes the chromosomal pattern of the mouse subventricular zone cell model. The control condition of control cells was used as the reference cell group (related to Figure 4g). Figure 12e shows a pseudo-time trajectory plot for integrated data from three cell types, with the trajectories rooted in the control state among the four states.
Figure 13 shows that CRISPR/Cas9-based origin cells show superior ability in neurosphere formation (stem) and invade a longer distance (migration ability) than control cells and tumor cells. Figure 13A shows sphere formation assay to estimate stem cell phenotype. Cells from different mouse brains are separated on the graph. Scale bar: 200 μm. Figure 13B: Cell invasion assay to estimate cell velocity over 72 hours. Cells from different mouse brains are isolated on the graph. Scale bar: 500 μm. Figure 13c compares the differentiation ability between origin cells and tumor cells in the mouse subventricular zone model. After culturing in medium containing 10% fetal bovine serum for 2 weeks, GFAP was expressed in origin cells and tumor cells. The cells of origin expressed the neuronal marker TUBB3. Scale bar: 200 μm.
Figure 14 shows a graphical summary of cells of origin 1 and 2, two types of adult brain stem cells distinguished by single cell and single nuclear RNA sequencing (subventricular zone n = 11, glioblastoma n = 10). We found consistency in the chromosomal pattern, connectome pattern, and CRISPR/Cas9 cell model of origin.

Hereinafter, the present invention will be described in more detail through examples. These examples are only for illustrating the present invention in more detail, and it will be apparent to those skilled in the art that the scope of the present invention is not limited by these examples according to the gist of the present invention. .

Example

Experiment method

Ethical approval

This study was approved by the Institutional Review Board of Severance Hospital for transcriptomic analysis of human samples (IRB 4-2012-0212, 4-2021-1319). The procedures used in this study adhered to the tenets of the 1964 Declaration of Helsinki.

human patient tissue

A total of 196 human tissue samples were included in this study. These samples were obtained as residual samples after neurosurgical diagnosis of brain tumor. These samples consist of 175 bulk tissue-level RNA sequencing data and 21 single-cell level RNA sequencing data (16 mononuclear RNA sequencing data, 5 single-cell RNA sequencing data). Bulk RNA sequencing data consists of three sample groups. Tumor-free SVZ of glioblastoma (n = 40), glioblastoma tumor tissue (n = 126), and tumor-free SVZ of control disease (n = 9).

Single-cell-level RNA sequencing cohort of human samples

Patients in the single-cell level database were labeled according to method and preparation sequence. Patient samples processed for single nuclear RNA sequencing were designated GN1, GN2, GN3, GN5, GN6, GN7, CN8, CN9, and CN10. GN patient samples are glioblastoma patient-derived samples, and CN patient samples are tumor-free SVZ samples from non-GBM patients. The GN4 patient sample was lost during preparation and was excluded. Single cell RNA sequencing samples are indicated as GC11, GC12, GC13, and GC14. All these samples were prepared from GBM patients. Among these samples, two SVZs from two different patients failed the quality control step, including a single tumor-free SVZ and a mixed SVZ sample with one tumor.

Single-nuclei isolation of human samples

Single nuclei were isolated from fresh frozen tissue samples according to previous reports with some modifications. Briefly, tissue samples were triturated with a scalpel in lysis buffer [0.2% Triton Homogenized tissue samples were filtered using a 40-μm strainer, then DAPI (0.5 μg/ml; Sigma) was added to stain the nuclei, and the nuclei were strained free using a 20-μm pluriStrainer Mini for analysis by FACS (BD FACSAria® systems, BD). The nuclei were filtered again. Single nuclei from the DAPI-positive population were sorted for subsequent snRNA sequencing.

Single-cell isolation of human samples

Single cell dissociation was performed on SVZ and GBM tumor samples. Each living tissue was cryopreserved immediately after surgery, and single cell dissociation was performed approximately 1 month later. In summary, on the day of surgery, a cryopreservation solution mixed with 90% FBS and 10% DMSO was added to the raw tissue, chopped into small pieces, and stored in liquid nitrogen for about a month. After approximately 1 month, each cryopreserved sample was centrifuged at 300 g and washed twice with serum-free DMEM to obtain a pellet. The pellet was minced as finely as possible in medium containing 1 M PIPES solution and treated with Papain (Worthington Biochemical) and 0.4 U/mL DNase I solution (Worthington Biochemical) at 37°C for 13 min. After centrifugation at 310 g for 10 minutes, 100 μl of 7 mg/ml Ovomucoid (Sigma-Aldrich) and 0.4 U/mL DNase I were added to the washing medium to inactivate the papain enzyme. Afterwards, myelin sheath and debris were removed using Percoll gradation with 22% Percoll solution (MP Biomedicals), and the supernatant was removed by centrifugation at 594 g to obtain the final cell pellet. If the cells in the pellet layer contained red blood cells (RBCs), the RBC removal process was performed according to the manufacturer's specifications using 10x RBC lysis buffer (Stemcell Technologies). Due to high aggregation, single cells cannot be obtained in some samples. In this case, Magnetic Activated Cell Sorting (MACS) sorting (Miltenyi Biotec, Germany) was performed according to the manufacturer's protocol, supplemented with 10% BSA, 1% FBS and Rho kinase (ROCK) inhibitor in 1xPBS to prevent aggregation and single cell death. The separated cells were diluted by adding.

Single-cell-level RNA sequencing of the libraries

Sequencing was performed at the Yonsei University Genome Center or Macrogen using the 10x Chromium Controller (10x Genomics) according to the manufacturer's protocol based on 10x Genomics proprietary technology. All single nuclear and single cell RNA sequencing libraries were prepared using the Next GEM Single-cell 3' GEM, Library & Gel Bead Kit, V3.1 (10x Genomics) according to the manufacturer's protocol. The initial step consisted of performing an emulsion in which individual nuclei were separated into droplets with gel beads coated with a unique primer carrying a 10x cell barcode, a unique molecular identifier, and a poly(dT) sequence. A reverse transcription reaction was performed to generate barcoded full-length cDNA, and then the emulsion was disrupted using DynaBeads MyOne Silane Beads (10X genomics) as a retrieval agent and cDNA cleanup. Bulk cDNA was amplified using a SimpliAmp Thermal Cycler (98°C for 3 min, cycled 11-13x (depending on target cell recovery): 98°C for 15 s, 67°C for 20 s, 72°C for 1 min). (72°C for 1 minute, held at 4°C). The amplified cDNA product was washed using the SPRIselect Reagent Kit (Beckman Coulter). An indexed sequencing library was constructed using reagents from the Chromium Single-cell 3 v3.1 reagent kit as follows: fragmentation, end repair, and A-tailing; Select size with SPRIselect; adapter ligation; Post-ligation cleanup using SPRIselect; Sample index polymerase chain reaction and cleanup using SPRIselect beads. Library quantification and quality assessment were performed using the 4200Tapetation using HSD1000 screen tape (Agilent Genomics). Indexed libraries were sequenced on the Illumina HiSeq series using paired-end 26x98 bp.

Single-cell/nuclei RNAseq data analysis

Raw sequencing data were computed from the GRCh38 genome (cell coverage 6.0.1). Processed raw and filtered features were loaded into Seurat (version 4.0.6) and graphs were plotted using Seurat (version 4.0.6) and dittoSeq (version 1.6.0). Single nuclear RNA sequencing data samples were prepared as raw and filtered matrices of samples using the SoupX algorithm34 and the Seurat package35, and the dataset was subjected to SC transformation, PCA, UMAP (from 1 to 10), nearest neighbor graph construction ( dimensions from 1 to 15) and cluster determination (resolution 0.5). The pN-pK parameter sweeping algorithm was applied to remove doublets (1st to 10th principal components). Singlet assumption cells were calculated with a doublet ratio of 7.5%, artificial doublet number of 0.25, PC neighborhood size of 0.09, and principal component number of 1 to 10. Single cell RNA sequencing data samples were processed as filtered matrices and the SoupX algorithm was not applied. Data were normalized and resized and dimension reduction plots were calculated. Single nuclear RNA sequencing data and single cell RNA sequencing data were integrated into Seurat using the CCA algorithm.

Cell-type identification in the single-cell level data

The obtained Seurat individuals were scored as modules for the functional expression program of SCT analysis in single nuclear RNA sequencing data sets. For normal cell programs, oligodendrocyte, microglial, radial glial, or pericyte programs extracted from third-party data sets were used. For the GBM-related gene program, published gene lists from GBM tumor articles were used. After integrating all data sets (Figure 2A), cell type 1 of origin was enriched for neural progenitor signatures (BCAN, SEZ6L) and oligodendrocyte progenitor signatures (PTPRZ1, TNR, and PDGFRA). Cell of origin type 2 was enriched with an astrocyte signature (AQP4, SPARCL1, ATP1A2) and neural stem cell signature (ADGRV1, SLC1A2).

Magnetic resonance imaging

The SVZ biopsy site was visualized using pre- and postoperative axial or coronal magnetic resonance imaging (MRI). MRI was used for preoperative planning of the SVZ biopsy site. The planned SVZ removal area was identified by the surgeon and a thin layer was incised in the area of non-tumor involvement.

Pathologist review

An experienced pathologist (S.H.K.) distinguished tumor-free SVZ samples from tumor-mixed SVZ samples during disease diagnosis or during post-mortem pathological review for RNA sequencing analysis.

Bulk-tissue-level RNA sequencing

RNA sequencing followed the best protocol for hisat2 and stringtie37.

Survival analysis

Patient survival time was calculated from the first neurosurgical operation (unit: months) until the last clinical follow-up or from data collected from the death certificate (Severance Oncology Center). Subgroups were divided according to the GSVA scores of origin cell 1 and origin cell 2, with a threshold of 0.SVZ-based survival, which aims to investigate patient survival, is distinguished by finding specific cells of origin in the SVZ, and tumor-based Survival was determined whether patient survival was influenced by specific cell-of-origin characteristics of the GBM tissue.

Copy-number variation analysis

For human bulk tissue-level RNA sequencing, the chromosome pattern inference algorithm was based on a moving average of 500 genes. Tumor-free SVZ samples from the control group were used as the control reference tissue group. Copy number variation patterns of single cell/nucleus data sets were processed with infercnv using oligodendrocytes as control samples (version 1.10.1). Copy number changes in six states were calculated for UMAP illustration and cell-level sequence prediction. They are based on the hidden Markov model of the infercnv algorithm.

Deconvolution analysis

Gene set variation analysis was used with designated gene sets for cell type identification in bulk tissue-level RNA sequencing data sets. Normalization scores per signature were calculated using the GSVA package (version 1.42.0), and for the cancer stem cell gene set, the cluster 0 signature of GBM24 was used.

Pseudo-trajectory analysis

Pseudo-time trajectory plots were calculated with Monocle 3. For the human data set, origin cell 1 was selected as the starting point, and for the mouse cell data, the control state of the control cell was selected as the starting point.

Mouse origin-cells induced by somatic mutation

All mouse experiments were approved and performed in accordance with GLP principles and in accordance with the guidelines of the Institutional Animal Care and Use Committee of Yonsei University College of Medicine. Mice harboring LSL-tdtomato (C57BL/6 strain, The Jackson Laboratory) were bred with the LSL-EGFR viii47 allele (FVB strain) and newborns were administered plasmid (sgRNA for p53 and Pten genes) solution (2 μg/μl, containing 1% Fast Green). Briefly, newborn, 2- to 3-day-old pups (P2-P3) were anesthetized by hypothermia and then injected with the plasmid solution into the right ventricle. Successfully injected animals received five electrical pulses (100 V, 50 ms, 950 ms apart) using an ECM830 electroporator (BTX-Harvard Apparatus) and 1 mm tweezer electrodes (CUY650P1, Nepagene).

Mouse origin-cell isolation

Either tumor specimens or SVZ specimens were dissected from electrolyzed mice using a stereomicroscope (S9i, Leica). Subsequent cell isolation from the sample was brief. The cell isolation procedure consisted of performing mechanical dissociation with a scalpel in Dulbecco's modified Eagle's medium/nutrient mixture F-12 (DMEM/F-12; Corning), then straining the mixture through a 70-μm nylon mesh cell strainer. Passed through (BD Falcon). The cell suspension was then washed twice in DMEM/F-12 and cultured in DMEM/F-12 containing 20 ng/ml of B27 supplement (1X; Invitrogen), basic fibroblast growth factor (bFGF, Novoprotein), and epidermal growth factor (Novopein). Cultured in complete medium consisting of 12. All in vitro experiments were performed under previously mentioned culture conditions.

Sphere formation assay

Isolated single cells were grown in complete medium consisting of DMEM/F-12 containing B27 supplement (1 Dissociated single cells were cultured in 96-well plates. After culturing for 11 days under the same conditions, the number of 9-positive wells was calculated, and the ratio of 9-positive wells in each group was calculated and expressed as a percentage. Images of spherical positive wells were captured and analyzed using ToupView software (ToupTek Photonics).

3D Invasion assay

The invasiveness of mouse cells was determined using a 3D invasion assay. Single dissociated cells (3000 cells) were seeded into each well of a PrimeSurface 96M plate (Shimadzu). After 24 hours, each well was filled with Matrigel matrix consisting of Matrigel, collagen type I (Corning), and complete medium. After 30 min, complete medium (30 μL/well) was added to the gel matrix to prevent drying. Images were captured using Operetta CLS (Perkin Elmer), and the invasive area was quantified as occupied area at (72 h-0 h)/0 h.

Orthotopic engraftment

For the orthotopic allograft model, 6- to 8-week-old male athymic nude mice (Central Lab. Animal Inc.) were housed in microisolator cages under sterile conditions and monitored for at least 1 week prior to experiments to ensure appropriate health. Zoletil (30 mg/kg; Virbac Korea) and xylazine (10 mg/kg; Bayer Korea) solutions were administered intraperitoneally to anesthetize the mice. Dissociated cells (2 Х 105) were transplanted orthotopically as previously described (day 0). Bioluminescence image acquisition and analysis were performed using the In Vivo Imaging System (IVIS) imaging system and Living Image v4.2 software (Caliper Life Sciences). Fifteen minutes before signal acquisition, mice were injected intraperitoneally with 100 μL D-luciferin (30 mg/mL, dissolved in PBS; Promega) and administered under 2.5% isoflurane anesthesia. Mice that lost more than 15% of their maximum body weight were euthanized according to approved protocols.

Image analysis of mouse brain

Mouse brains were obtained, fixed in 10% formalin, cryoprotected overnight in 30% buffered sucrose, and stored in cryoblocks (Optimal Cutting Temperature Compound (OCT Compound)) stored at -80°C. Cryostat cut sections (10 μm thick) were obtained at coronal section points 1, −1, −2, and −3 mm distal to bregma. Mounting medium with DAPI (H-1200-10, Vector Laboratories) was used for nuclear staining, and images were obtained using a Zeiss LSM700 confocal microscope (Zeiss). For H&E staining, 4 μm cryo-stat-cut sections were used.

Mouse cell preparation for single-cell RNA sequencing

Cells of mouse origin and tumor cells isolated from the Trp53/Pten/hEGFRvIII mutant mouse model were incubated with B27 supplement (1X; Invitrogen), 20 ng/ml basic fibroblast growth factor (bFGF, Novoprotein), and 20 ng/ml epidermal growth factor (EGF). , Novoprotein) and 1% penicillin-streptomycin (15140122, Thermo Fisher). Once cells of mouse origin or tumor cells formed spheres, the spheres were collected and dissociated with Accutase solution (A6954, Sigma-Aldrich) for 1 min. Dissociated single cells were washed three times with 0.04% BSA solution in PBS. Cell viability was monitored by trypan blue staining, and surviving cells were used for subsequent single-cell RNA sequencing.

Experiment result

Chromosome patterns in the subventricular zone tissues

The present invention used chromosome patterns as indicators for tracking glioblastoma (GBM)-originating cells (gain of chromosome 7 and loss of chromosome 10). Patient samples were collected for bulk tissue-level RNA sequencing and single-cell-level RNA sequencing (Figure 5a), along with the publicly available full extract sequencing data set (SRP145073)2. Single-cell-level RNA sequencing includes single-cell RNA sequencing and mononuclear RNA sequencing (Figure 5b).

The most important aspect of this study is the tumor-free status of the SVZ of the brain (yellow box, Figure 1a). SVZ tissue samples were defined as tumor-free if tumor admixture status was not confirmed by the pathologist after sampling of SVZ tissue ( Fig. 5C ). In this study, all SVZ samples except one tumor-mixed SVZ sample in single-cell RNA array analysis were tumor-free ( Fig. 1A ). The sampling location of the tumor-free SVZ of GBM differed depending on the tumor location ( Fig. 5D ). Tumor-free SVZ controls are obtained from non-GBM patients (Figure 5e). Sampling of the SVZ was planned with preoperative medical imaging, but pathological examination revealed two cases with tumor-mixed samples for single-cell RNA array analysis (for downstream analysis, one sample failed and only one tumor-mixed sample was included; Figure 5f). Therefore, we refer to the SVZ of the brain as otherwise specified tumor-free samples.

A pilot study using bulk tissue-level RNA sequencing showed differences in the SVZ of GBM from the control SVZ through differences in gene expression and upregulation of neurotransmitter-related signaling ( Fig. 6A ). The SVZ of GBM was again compared to GBM tissue and cell migration-related signatures were upregulated in the SVZ of GBM.

Based on the differences between the SVZ of GBM and other tissues with bulk RNA sequences, we hypothesized that the SVZ of GBM may contain evidence for GBM-progenitor cells. Because GBM has a pattern of chromosome 7 q-arm gain and chromosome 10 loss, we reviewed the entire data set focusing on the chromosomal patterns of these two chromosomes. Interestingly, we found a pattern of chromosome 7q gain in the SVZ of GBM (Figure 6C). The chromosome 7 pattern was also evident in other samples from the SVZ of GBM ( Fig. 6D ). Additionally, because chromosome 7 was not prominent in SVZ samples (Figure 6e), we attempted to validate the chromosome pattern in a larger bulk RNA array data set and compared copy numbers (Figures 1b and 6f).

The defining chromosomal pattern of GBM, chromosome 7 gain or chromosome 10 loss, was found in 98% of tumor tissues (Figure 1B). The SVZ of GBM showed two CNV changes in 65% of samples, whereas the SVZ of controls showed a chromosomal pattern in 33% of samples. Having discovered the limitations of bulk tissue-based chromosome pattern analysis (Figure 6g), we investigated RNA sequencing at the single cell level.

As a result of single cell level analysis, two GBM-origin cells were detected in the SVZ.

Single-cell-level RNA sequencing results revealed two types of cells with GBM-related chromosomal patterns in the SVZ of one patient's GBM sample: a chromosome 7 q-arm gain and a chromosome 10 loss pattern ( Fig. 1C ). The common chromosomal pattern of these two cell types was chromosome 7 q-arm amplification. One cell type showed loss of chromosome 10 in half of the population, while the other cell type showed no loss of chromosome 10 in the population (Figure 1c). Because these cells showed the theoretical chromosome pattern of the GBM-origin cell, we called the cell showing only chromosome 7 q-arm acquisition as GBM origin-cell 1 (Oc1), a cell that only showed chromosome 7 q-arm acquisition. And the cell showing loss of chromosome 10 was defined as origin-cell 2 of GBM. We also hypothesized that the cytotype signature of these two cells could be applied to an expanded number of SVZ samples ( Fig. 1D ).

In the expanded cohort, we found two clusters of cells of origin (Figure 2A) with defined genetic signatures obtained from the pilot mononuclear RNA array SVZ sample above (Patient GN1). This cell-of-origin signature is summarized by genes representing cell-of-origin 1 for PTPRZ1 (Figure 7A) and cell-of-origin 2 for AQP4 (Figure 7B). Briefly, we labeled each single-cell level data set with the GN-1 SVZ extracted gene set and integrated the individual data sets to find a harmonized data set (Figures 2A and 7C). These two cell types of origin were found in tumor-free SVZ samples and GBM tumor tissue (Figures 2B and 7D).

Sequential accumulation of chromosome patterns

Although two types of cell-of-origin signatures were found throughout the samples (Figure 2b), we sought to examine the temporal order of chromosomal patterns in GBM. Patient GN1 samples from the SVZ and tumor show overlapping areas in the dimension reduction plot excluding neuron-like cells (Figure 2c). In SVZ samples from patient GN1, we found that the two types of cells of origin were dominated by chromosome 7q-arm gain events and chromosome 10 loss events rather than neuron-like cell clusters ( Fig. 2D ). In tumor cells of GN1, most chromosomal change patterns were found in cell clusters associated with two types of cells of origin (Figure 2e). Since gain of chromosome 7 and loss of chromosome 10 were found in the protocells of the SVZ and the tumor, we combined these two data sets in patient GN1, and seven q-arm accumulation patterns were found in cells of the protocell junction cluster (Figure 2f). In the same integrated data set, we found that chromosome 7q gain events precede chromosome 7p gain or chromosome 10 loss events at the single cell level ( Fig. 2G ). We can find that both types of origin cells display chromosomal patterns linking SVZ samples and tumor samples. In other words, more p-arms and q-arms are acquired in the origin cells of the SVZ than in other cells of the SVZ. The increase rate of chromosome 7 p-arm and q-arm is lower in the original cells of GBM tumors than in other cells of GBM. Chromosome 10 loss events also show a connection pattern from the SVZ to the tumor (Figure 8).

Interconnected origin-cells

Since the cell cluster of origin is connected to the positive cell clusters (neural precursors, mesenchymal, and circulating cells) of the chromosomal pattern, a pseudotrajectory plot was calculated. Because cell of origin 1 was expected to precede other cell types, we set the cell as the starting point: integrated datasets from the tumor-free SVZ showed that cell of origin 1 could lead to neuron-like cells, circulating cells, or mesenchymal cells. showed. The integrated tumor map showed that cell of origin 1 could lead to all cell types (Figure 9A). Patient-specific data show that the cells of origin are connected to tumor cell clusters and have connections with neuron-like cells, neural progenitor cells, mesenchymal cells, or circulating cells (Figure 9B). In single-cell RNA sequencing datasets, the connectivity of protocells to other tumor-related cells, such as neural progenitor cells, circulating cells, or mesenchymal cells, is not altered, even though the number of neuron-like cells is low across samples (Figure 9C).

For interconnectivity, ligand-receptor linkage analysis was applied on single-cell level RNA sequencing data (Figure 3a). In stem cell mode, we found that origin cell 1 was connected to origin cell 2 in SVZ samples. In the tumor sample, the cycling cell was connected to the original cell 2 (Figure 3a). Since three types of cells were found in the centroid plot, we analyzed the interconnections between these cells at the gene level (Figure 3b). We found that BCAN-NRCAM connections were commonly found in circulation plots of SVZ and tumors. Particularly in SVZ samples, EGFR was upregulated across origin cell 1, origin cell 2, and circulation cells (Figure 3B).

Origin-cells and the clinical tumors

The research results were expanded from single cell-level RNA sequencing data to large-scale tissue-level RNA sequencing data. Although single cell level data can show high-resolution interconnections between cell types, population level data was not available in the above analyses. A cell identification score was calculated from the bulk tissue-level RNA array data (FIG. 10A) using the genetic signature of the above analysis. Two types of origin cells were found in the SVZ of controls and the SVZ of GBM, with no statistical differences in SVZ samples ( Fig. 10A ). The cell number of cell of origin 2 was predicted to decrease in tumor samples. Additionally, the survival of GBM patients was compared by cell identification score. There was no survival effect by origin cell score identification for tumor-free SVZ specimens (Figure 10B) or tumor specimens (Figure 10C). When we narrowed down the TERT promoter mutation status of GBM to the samples for which it was available, we found different patterns of the two cell types of origin (Figure 3C). Cell of origin 1 was common in TERT promoter wild-type GBM (Figure 3C). Cell of origin 2 appeared more commonly in TERT promoter mutant GBM. According to all data sets, original cell 2 showed a decreasing trend in GBM of both TERT promoter types (Figure 3C and Figure 10A). In contrast, these two types of GBM showed a trend toward increased intratumoral cell circulation and cancer stem cell signatures than SVZ samples. Additionally, the signatures of the two transmitting base cells were compared with other signatures (Figure 10d). In the SVZ of GBM, the two types of origin cells showed high correlation in the correlation matrix (Figure 10e). However, in the tumor, the sintering of protocell 2 was separated from the protocell 1 cluster (Figure 10f).

Mouse model of the GBM origin-cells

In light of human SVZ samples and GBM samples, protocells and protocell-derived tumor cells were generated with CRISPR/Cas9, and unmodified cells were used as controls (FIGS. 4A and 11A). This model was previously reported to form mouse GBM (Figure 11B). GBM-origin cells and tumor cells were obtained from the mouse subventricular region and tumor, respectively (Figure 11c). These cells are culturable and are verified by high tdTomato positivity, suggesting introduction of the CRISPR/Cas9 plasmid into the cells (Figure 11D). These three types of cells (Figure 11e) are verified to be capable of forming tumors when injected into the mouse brain (Figure 11f). Similar gene expression patterns of marker genes (Figure 4B) were not sufficient to explain different tumor formation in the mouse brain injection model (Figures 4C and 11F). In the integrated analysis, three types of cells could be classified into four types of cell states (Figure 4d). Briefly, we labeled these four cell states by comparing cell state composition (Figure 4e), gene signature scores of cells of human origin (Figure 4f), and chromosome pattern analysis (Figure 4g).

Details of single-cell RNA sequencing of mouse cells are summarized in Figures 12A-12E. In the three types of cell integration analysis, cell states and initial states that specify control genes are similarly overexpressed (Figure 12a). The oligodendrocyte progenitor cell marker Cspg4 is expressed in the same order as the two origin cell scores, which gradually decrease from high in the control state to tumor state (Figures 4f and 12b). The tumor-related cell cycle signature or mesenchymal signature showed an ascending pattern from the control state to the tumor state (Figure 12c). The chromosomal patterns and increase in abnormalities were calculated as the cell state changed from early and late to tumor state (Figure 12D). Finally, we chose the control state as the starting point and pseudotime analysis revealed tumor cells at the end of the trajectory.

An interesting phenotype of cells of origin was discovered in the neurosphere-based analysis (Figures 13A-13B). A high neurosphere formation phenotype was found in two different origin cells from different mouse brains (Figure 13A). Cell invasion assays showed that the protocell was the most distant cell by 72 hours (Figure 13b). When trying to differentiate between origin cells and tumor cells, it was confirmed that only origin cells formed TUBB3-positive cells (Figure 13c).

Briefly, we identified two types of human GBM-origin cells and the results were validated in mouse GBM-origin cells with gene expression patterns and chromosomal patterns. In particular, cell-of-origin 1 was identified for the first time in the adult brain, and TERT promoter wild-type GBM can arise from cell-of-origin 1 (Figure 14).

As the specific parts of the present invention have been described in detail above, it is clear to those skilled in the art that these specific techniques are merely preferred embodiments and do not limit the scope of the present invention. Accordingly, the substantial scope of the present invention will be defined by the appended claims and their equivalents.

Claims (20)

In cells isolated from the subventricular zone (SVZ), NRXN1, A method of isolating brain cancer origin cells comprising; confirming the expression level of one or more proteins selected from the group consisting of, or genes encoding the same.
According to claim 1,
Isolation method, wherein the brain cancer origin cells are cells that develop into glioblastoma (GBM).
According to claim 1,
A separation method further comprising: confirming an increase in the copy number of chromosome 7 in cells isolated from the subventricular zone.
One or more proteins selected from the group consisting of NRXN1, A composition for predicting the development of brain cancer, comprising as an active ingredient an agent that measures the expression level of a gene.
According to claim 4,
The expression level of one or more proteins selected from the group consisting of NRXN1, A composition wherein the agent to be measured is an aptamer, antibody, or antigen-binding fragment thereof that specifically binds to the protein.
According to claim 4,
A gene encoding one or more proteins selected from the group consisting of NRXN1, The composition for measuring the expression level is a primer or probe that specifically binds to the gene.
According to clause 4,
The composition, wherein the brain cancer is glioblastoma (GBM).
A kit for predicting the development of brain cancer, comprising the composition of claim 4.
According to claim 8,
The kit is used for cells isolated from the subventricular zone (SVZ).
According to clause 9,
The kit, wherein the cells isolated from the subventricular zone are cells with an increased copy number of chromosome 7.
According to clause 9,
The kit is an RT-PCR kit, DNA chip kit, ELISA kit, protein chip kit, rapid kit, or multiple reaction monitoring (MRM) kit.
In cells isolated from the subventricular zone (SVZ) of the subject of interest, NRXN1, , and measuring the expression level of one or more proteins selected from the group consisting of LAMA2, or the genes encoding them.
According to claim 12,
A prediction method further comprising: confirming an increase in the copy number of chromosome 7 in cells isolated from the subventricular zone.
According to claim 12,
When the expression of one or more proteins selected from the group consisting of NRXN1, A prediction method that predicts high.
According to claim 12,
A prediction method for predicting that the individual is likely to develop brain cancer when the expression of one or more proteins selected from the group consisting of LUZP2, RORA, SLC1A2, TNR, and XKR4, or the genes encoding them, is increased.
According to claim 12,
Predicting that the individual is likely to develop brain cancer when the expression of one or more proteins selected from the group consisting of PLP1, TF, MBP, CERCAM, TMEM144, PLXDC2, FTL, and LAMA2, or the genes encoding them, is reduced , prediction method.
According to claim 12,
A prediction that predicts that the individual is likely to develop brain cancer when the expression of one or more proteins selected from the group consisting of CERCAM, FTL, LAMA2, MBP, PLP1, PLXDC2, and TF, or the genes encoding them, is reduced. method.
According to clause 12,
A prediction method wherein the brain cancer is glioblastoma (GBM).
(a) treating brain cancer origin cells isolated by the method of claim 1 with a candidate substance for preventing brain cancer; and,
(b) determining that the candidate material has a brain cancer prevention effect when the brain cancer origin cells are killed.
(a) treating brain cancer origin cells isolated by the method of claim 1 with a candidate substance for preventing brain cancer;
(b) selected from the group consisting of NRXN1, Reconfirming the expression level of one or more proteins or genes encoding them; and,
(c) determining that the candidate substance has a brain cancer prevention effect when the expression level of the protein or gene is significantly different compared to before treatment with the candidate substance.