WO2010131641A1 - Method for determination of condition of cell - Google Patents

Abstract

Disclosed is a technique for identifying the condition of a cell rapidly, conveniently and accurately. Specifically disclosed is a method for determining the condition of a cell, in which a sugar chain present on the surface of a cell to be determined or a sugar chain-containing sample that contains the sugar chain is measured under the conditions where the sugar chain or the sugar chain-containing sample remains bound to multiple types of lectins in a liquid phase, and the state of the cell is determined based on the quantity of the cell to be determined or the sugar chain-containing sample that is bound to each of the multiple types of lectins obtained by the measurement.

Description

How to determine cell status

The present invention relates to the determination of the state of cells.

Regenerative medicine refers to medical treatment that recovers and treats defects and insufficiency of organs and tissues due to diseases and disorders by transplanting cultured cells and tissues obtained from them, and is actually skin, bone / cartilage, blood vessels, cornea In such fields, clinical applications are being carried out.

Stem cell research is deeply involved in understanding the life of individual development and differentiation, and at the same time leads to application to regenerative medicine using cells differentiated from stem cells. As a result, it is expected that regenerative medicine can be performed even in fields that were not possible in the past.

Stem cell evaluation methods include cell morphology, gene expression analysis, and surface antigen evaluation.

Patent Document 1 describes a method of diagnosing cell differentiation by observing stem cell morphology.

Patent Document 2 describes a method for monitoring cell differentiation from the methylation pattern of cytosine in DNA collected from cells.

Undifferentiated human embryonic stem (hES) cells are (1) in addition to the observation of morphological characteristics, (2) positive alkaline phosphatase activity, (3) stage-specific emblemonic antigen (SSEA)- 3 and SSEA-4 and TRA-1-60 recognizing keratan sulfate proteoglycan, TRA-1-81 positive cell surface marker, (4) expression detection of undifferentiated marker genes OCT3 / 4, SOX2, NANOG, and (5) Evaluation is performed by confirming the formation of teratomas (teratomas) by transplantation into immunodeficient mice.

It is known that various proteins and lipids exist on the cell surface and various antigens are expressed. Furthermore, many of them are sugar chain-modified, and it is known that the cell surface is covered with various sugar chains. Since the sugar chain on the surface varies depending on the type of cell, the sugar chain is also called the face of a cell. For example, K. A. Wearne et al. Have shown that various changes in sugar chain structure occur before and after differentiation of human ES cells using fluorescently labeled lectins (Non-patent Document 1).

In Patent Document 3, it is constituted by a lectin group including at least one lectin whose sugar binding site is modified at least, or selected by panning a cell or a pseudo-cell body having a predetermined sugar chain. A sugar chain analysis method using a lectin library composed of a lectin group including at least one lectin has been proposed.

Patent Document 4 proposes a method of selecting a lectin having a predetermined analysis ability from a plurality of types of lectins and a lectin selected by the method.

Patent Document 5 proposes a method of analyzing an interaction between a protein and a sugar chain that interacts with a sugar chain by applying excitation light without washing the substrate.

In Patent Document 6, a substrate made of a light-guiding material in which a plurality of types of sugar chain-binding proteins are arranged and fixed, and light is introduced into the side end face of the substrate to generate an evanescent wave on the surface of the substrate, thereby fluorescent labeling is performed. There has been proposed a sugar chain or complex carbohydrate analyzing device having means for exciting and fluorescence intensity measuring means for measuring the intensity of fluorescence generated by the means for each arrangement position of the sugar chain binding protein.

In Patent Document 7, an array substrate for observing structural information of molecules such as proteins, sugar chains, and lipids existing on the cell surface without destroying the cells, and a method for achieving a cell surface profile using the array substrate Has been proposed.

In applying stem cells to regenerative medicine, quality control technology is very important. In order to perform quality control, a technique for discriminating the state of stem cells including undifferentiation, differentiation, and the direction of differentiation is required. Therefore, establishment of a cell evaluation technique and a cell standardization method is awaited.

For example, it has been reported that in human ES cells, there is a difference in differentiation tendency depending on established cell lines (Non-patent Document 2).

In ES-like cells such as iPS cells, the evaluation criteria are not based on ES cells at the present time, and there are similar problems.

WO2006 / 057444 JP 2004-194655 A Patent 3658394 WO2004 / 018513 WO2005 / 064333 JP2007-3357 JP2008-187932

K. A. Wearne et al., Glycobiology 2006, 16, 981-990 K. Osafune et al., Nature Biotechnology 2008, 26, 313-315

In order to proliferate while maintaining the state of the cell and to properly differentiate the cell, it is necessary to be able to evaluate information on the state and change of the cell quickly, simply and accurately.

Therefore, what the present invention intends to solve is to provide a technique for quickly, simply and accurately discriminating the state of cells.

As a means to solve the above-mentioned problems, the inventors of the present application have repeatedly studied about a method for discriminating cells with lectins. As a result, the state of cells is discriminated by using binding measurement data of sugar chains of cells to a plurality of lectins. Invented a method to do this.

The invention described in claim 1
Obtained by measuring a sugar chain of a cell to be discriminated or a sugar chain-containing sample containing the sugar chain while bound to a plurality of types of lectins in a liquid phase, the plurality of types In this method, the state of a cell is discriminated using the amount of the cell to be discriminated or the amount of a sugar chain-containing sample each bound to lectin.

The invention according to claim 2
The method according to claim 1, wherein the cell is a stem cell or a progenitor cell.

The invention described in claim 3
The method according to claim 2, wherein the stem cells are somatic stem cells, embryonic stem cells, nuclear transfer stem cells, or induced pluripotent stem cells.

The invention according to claim 4
The method according to claim 3, wherein the somatic stem cell is a hematopoietic stem cell or a mesenchymal stem cell.

The invention according to claim 5
The method for determining a state of a cell according to any one of claims 2 to 4, wherein the cell is derived from a mammal or a human.

The invention described in claim 6
6. The method for discriminating a cell state according to any one of claims 1 to 5, wherein the state of the cell to be discriminated includes an undifferentiated state of the cell and a state of cell differentiation.

The invention described in claim 7
The method for discriminating a cell state according to any one of claims 1 to 5, wherein the cell state to be discriminated is a cell differentiation tendency or a resistance to cell differentiation.

The invention described in claim 8
6. The method for discriminating a cell state according to any one of claims 1 to 5, wherein the cell state to be discriminated is a tendency of cell canceration or cell canceration.

The invention according to claim 9
The method for discriminating the state of a cell according to any one of claims 1 to 8, wherein the sugar-containing molecule is any one or a plurality of glycoproteins, glycopeptides, glycolipids, and proteoglycans. It is.

The invention according to claim 10 is:
10. The state of a cell according to any one of claims 1 to 9, wherein the plurality of types of lectins are fixed to any one of a lectin array, an ELISA plate, a magnetic bead, and a latex bead. It is a method to do.

The invention according to claim 11
The method for determining a state of a cell according to any one of claims 1 to 10, wherein the amount of the cell or sugar chain-containing sample is measured by fluorescence measurement or luminescence measurement.

The invention according to claim 12
12. The method according to claim 11, wherein the fluorescence measurement is an evanescent excitation method or a confocal fluorescence measurement method.

The invention according to claim 13
12. The method for determining a state of a cell according to claim 11, wherein the fluorescence measurement is flow cytometry.

The invention according to claim 14
The cell state discrimination method is performed in combination with one or both of gene expression analysis and binding measurement with a specific marker antibody of a cell in a target state. This is a method for determining the state of cells.

The invention according to claim 15 is:
The method for discriminating a cell state according to any one of claims 1 to 13, wherein the cell state discrimination method is performed by linear discriminant analysis or nonlinear discriminant analysis.

The invention according to claim 16
The method according to claim 15, wherein the nonlinear discriminant analysis is discriminant analysis based on Mahalanobis distance.

The invention described in claim 17
14. The method for determining a cell state according to any one of claims 1 to 13, wherein the cell state determination method is a supervised machine learning method or a semi-supervised machine learning method.

The invention according to claim 18
The machine learning method is performed by any method selected from a k-nearest neighbor method, a naive Bayes classifier, a decision tree, a neural network, a support vector machine, a bagging method, a boosting method, and a random forest method. Item 18. A method for determining the state of a cell according to Item 17.

The invention according to claim 19
The cell state determination method according to any one of claims 1 to 13, wherein the cell state determination method is performed by any one method selected from principal component analysis, cluster analysis, and self-organizing map. It is a method to do.

The invention according to claim 20 provides
The state of the cells to be identified is undifferentiated and differentiated human embryonic stem cells or human induced pluripotent stem cells, and the lectins used for the identification are canine endlectin I, kidney bean lectin (L), kidney bean lectin (E), spindle tree The lectin is one or more selected from the group consisting of molecules each having a sugar-binding specificity similar to any one of these lectins or any of these lectins. 20. A method for discriminating the state of a cell according to any one of items 1 to 19.

The invention according to claim 21
The state of the cells to be distinguished is undifferentiation and bone differentiation of human mesenchymal stem cells. 20. One or a plurality of molecules selected from the group consisting of molecules each having a sugar-binding specificity similar to any one or any of these lectins. This is a method for discriminating the state of a cell.

By using the present invention, the state of a cell can be determined quickly, simply and accurately.

The figure showing an example of multiple types of lectin currently fixed to the lectin microarray used for the method of this invention. The figure showing an example of the cluster analysis result in the method of this invention. The figure showing an example of the principal component analysis result in the method of the present invention. The figure by which the discrimination | determination result of the undifferentiated cell group and the differentiated cell group was represented by the method of this invention. The figure of the other example by which the discrimination | determination result of the undifferentiated cell group and the differentiated cell group was represented by the method of this invention. The figure showing the measurement result of the undifferentiated gene marker of the human ES cell and EB used by this invention. The figure showing an example of the principal component analysis result in the other method of this invention. The figure by which the discrimination | determination result of the undifferentiated mesenchymal stem cell group and the bone differentiated cell group was represented by the other method of this invention. The figure showing an example of the cluster analysis result in other methods of the present invention. The figure showing an example of the principal component analysis result in still another method of the present invention. The figure of the further another example by which the discrimination | determination result of the undifferentiated cell group and the differentiated cell group was represented by the method of this invention. The figure of the example by which the discrimination result of the cell group from which a differentiation tendency differs by the method of this invention was represented. The figure of the other example by which the discrimination result of the cell group from which a differentiation tendency differs by the method of this invention was represented. The figure showing an example of the cluster analysis result which discriminate | determined the kind of cell by the method of this invention. The figure showing an example of the result of having discriminate | determined the cell used by this invention by gene expression analysis data. The figure showing an example of the cluster analysis result which discriminate | determined the origin tissue of the cell by the method of this invention. The figure showing the discrimination | determination result (signal obtained from the antibody microarray) by the method of this invention in case a sample is a cell extract. The figure showing the discrimination (cluster analysis) result by the method of this invention in case a sample is a cell extract. The figure showing the discrimination | determination (principal component analysis) result by the method of this invention in case a sample is a cell extract. The figure showing the examination result of the precision of the discriminant by this invention when a sample is a cell extract. The figure showing an example of the cluster analysis result by the method of this invention in case a sample is the state after the undifferentiation of a human ES cell, and differentiation. The figure showing an example of the cluster analysis result by the method of this invention in case a sample is a human mesenchymal stem cell and a fetal cancer cell. The figure showing an example of a cluster analysis result when a sample is a human mesenchymal stem cell and a fetal cancer cell, and the purpose is specified and discriminated by the method of the present invention. The figure showing another example of the cluster analysis result when a sample is a human mesenchymal stem cell and a fetal cancer cell, and the purpose is specified and discriminated by the method of the present invention.

The method for discriminating the state of a cell according to the present invention comprises a sugar chain possessed by a cell to be discriminated or a sugar chain-containing sample containing the sugar chain (for example, a sugar chain possessed by a cell to be discriminated) Cell extract, which is a sugar chain-containing sample containing glycine, and obtained by measuring in a liquid phase while being bound to a plurality of types of lectins. The state of the cells is discriminated using the amount of the formed cells or the amount of the sugar chain-containing sample.

Since cells have a wide variety of sugar chains, multiple types of lectins, which are sugar chain binding molecules, are used at the same time, so that they can be identified by binding to each of the multiple types of lectins. A glycan-containing sample containing the glycan of the cell to be discriminated using the amount of cells or bound to each of the plurality of types of lectins (for example, from the cell to be discriminated) The cell state is discriminated using the amount of the cell extract).

It is known that the binding between sugar chains and sugar chain-containing molecules and lectins is weak compared to biomolecular interactions such as antigen-antibody reactions, and as a result, it is measured by techniques such as fluorescence measurement using conventional ELISA plates. In the washing step and / or drying step that is sometimes performed, the binding between the sugar chain or the sugar chain-containing molecule and the lectin is dissociated, so that the detection sensitivity is lowered and the reproducibility is lowered, and the measurement accuracy is lowered. In addition, the measurement accuracy decreases due to the generation of a noise signal due to non-specific adsorption or the like in the drying process.

Therefore, for the measurement of the binding between the sugar chain of the cell to be discriminated or the sugar chain-containing sample containing the sugar chain (for example, a cell extract from the cell to be discriminated) and the lectin. It is desirable that the sugar chain or the sugar chain-containing sample reacts with the lectin and the measurement can be performed in the liquid phase without washing as much as possible.

Examples of methods that can be measured in the liquid phase without washing include confocal fluorescence measurement and fluorescence measurement by evanescent excitation.

Conventional methods could only obtain qualitative information such as whether or not they were combined, but by using a method that can be measured in the liquid phase as described above, quantitative data that can withstand statistical analysis can be obtained with good reproducibility. Will be able to.

It is reasonable to consider that there are almost no differences in the sugar chain structure of cells in different states and that there are multiple. Therefore, it is considered that the accuracy of the discrimination is improved by using a plurality of types of lectins for discriminating the state of the cells at the same time instead of one type.

The lectin used to detect sugar chains is not very different in the target cells or does not need a lectin that recognizes sugar chains that are not present in the target cells. Since it may cause noise, lectins used for measurement or discrimination are desirably selected as necessary.

For selection of lectins used for cell discrimination, it is desirable to measure the binding between the sugar chain of the cell to be discriminated and multiple types of lectins, and to perform screening to extract the lectins in which differences are found from the results. . The measurement method preferably uses a lectin array loaded with various types of lectins, and the glycine contained in the cells to be identified or the sugar chain-containing sample containing the sugar chain and the lectin Since it is a binding measurement, it is desirable to use a measurement method that allows measurement in a liquid phase. For example, a confocal fluorescence measurement method or a fluorescence measurement method using an evanescent excitation method can be used.

Lectins are defined as sugar-binding proteins, but anti-glycan antibodies that are products of immune reactions are not included in lectins.

Lectins can be obtained from various organisms such as plants, fungi, animals and microorganisms. Each lectin has specificity for sugars and sugar chains. There are lectins having affinity for monosaccharides such as mannose, galactose, N-acetylgalactosamine, fucose and sialic acid, and lectins having affinity for oligosaccharides containing the above monosaccharides. Any lectin can be used in the present invention.

When using a sugar chain-containing sample (for example, cell extract) containing the sugar chain of the cell to be discriminated, a sugar chain-containing sample that is a sugar chain-containing molecule extracted from the cell and a plurality of lectins For the binding measurement, lectin binding assay using ELISA plate, magnetic beads, latex beads, surface plasmon resonance method, lectin array, etc. can be used, and lectin array is preferable. Furthermore, it is desirable that the lectin array is based on a measurement method that allows measurement in a liquid phase state. For example, measurement can be performed using a confocal fluorescence measurement method or an evanescent field excitation method.

When using cells as samples, flow cytometry, ELISA plates that can be measured in liquid phase, lectin binding assays using magnetic beads, latex beads, surface plasmons, etc. A lectin array that can be measured by a resonance method or a liquid phase state can be used.

A lectin array is to measure the binding of a plurality of lectins to a carrier and react with sugar chains to be reacted or a sugar chain-containing sample. As the carrier, glass, plastic, metal or the like is used. For example, a cell to be discriminated or a sugar chain-containing sample containing a sugar chain possessed by the cell can be fluorescently labeled and reacted with a lectin immobilized on a carrier. At this time, the fluorescence intensity of the fluorescently labeled cells bound to the lectin and the fluorescence intensity of the fluorescently labeled sugar chain-containing sample can be measured. The fluorescence intensity depends on the amount of fluorescently labeled cells and the fluorescently labeled sugar chain. Increase or decrease in accordance with the amount of sample contained.

Measured fluorescence intensity can be digitized and handled. This numerical value is desirably processed according to the purpose. For example, in order to determine the state of a cell, it is desirable to normalize the value obtained from a sugar chain-containing sample (for example, a cell extract) or a binding measurement between a cell and a lectin. Further, it is desirable that a cutoff is performed on the value. Furthermore, it also includes deleting a lectin having no signal or very small signal that may become noise during statistical analysis processing before analysis. Furthermore, it is desirable that the value is corrected by a value measured with a standard substance such as a glycoprotein having a sugar chain corresponding to the specificity of each lectin.

As the lectin used for discriminating the state of a cell, a lectin necessary for discrimination can also be extracted using the result of measuring the binding between the sugar chain of the cell and a plurality of lectins. For example, using cells in multiple states to be distinguished as a sample, the cell extract or the results of binding measurements between cells and multiple types of lectins can be used to extract lectins that exhibit characteristic differences in cells in those states. can do. As a method for extracting lectins, a statistical method such as a significant difference test such as t-test or a principal component analysis can be used.

In the determination of the state of cells, not only the measurement of binding to lectins but also other methods can be combined. For example, gene expression analysis (DNA array or PCR), epigenetics analysis (DNA methylation analysis, histone chemical modification analysis), measurement of binding to specific marker antibodies in cells to be discriminated (immunostaining, flow site) Measurement, ELISA, antibody array) in combination with cell results.

The cell state discriminated by the above-described method of the present invention includes an undifferentiated state of the cell and a state after differentiation of the cell. Here, the undifferentiated state of the cells includes stem cells and progenitor cells. Stem cells refer to cells that are undifferentiated and have self-renewal and multipotency, somatic stem cells such as hematopoietic stem cells and mesenchymal stem cells, embryonic stem cells (ES cells), nuclear transplanted ES cells (ntES cells), Includes induced pluripotent stem cells (iPS cells).

In the above, differentiation means that undifferentiated cells such as stem cells and progenitor cells lose differentiation ability or pluripotency and acquire characteristics specific to the cell type. During differentiation, the state of the cells changes.

Furthermore, the cell state means that the cell is undifferentiated, has pluripotency, has differentiated, the tendency of undifferentiated cells to have differentiation, resistance to differentiation, and that the cells have become cancerous. , Including the tendency of canceration of cells, cell origin, cell tissue, viral infection of cells, and changes in cells due to genetic manipulation.

To measure the sugar chain of a cell, it is possible to extract sugar chain-containing molecules from the cell. In that case, the method includes a step of extracting a sugar chain-containing molecule from the cell, and the sugar chain-containing molecule includes glycoprotein, glycopeptide, glycolipid, and proteoglycan. Alternatively, the cell itself can be measured. The sugar chain-containing molecule extracted from the cell is preferably a sugar chain-containing molecule present on the cell surface, that is, a sugar chain-containing molecule extracted from the cell membrane, and particularly preferably a glycoprotein extracted from the cell membrane.

The means for obtaining the discrimination method in the present invention is preferably based on a statistical analysis method or machine learning. In particular, it is preferable to use discriminant analysis using teacher data. For example, when discriminating the state of two groups of cells, linear discriminant analysis is often sufficient. However, since linear discriminant analysis is based on the assumption that the population variance of each group is equal, its application is limited, so when discriminating using more complicated data or classifying multiple groups with high accuracy If necessary, a method suitable for the data to be used is required. As the method, nonlinear discriminant analysis or discriminant analysis by machine learning can be used. Nonlinear discriminant analysis includes discriminant analysis based on Mahalanobis distance. Discriminant analysis by machine learning includes k-nearest neighbor method, naive Bayes classifier, decision tree, neural network, support vector machine, and group learning such as bagging method, boosting method, and random forest method. Such a technique has been developed in the field of pattern recognition, and the knowledge can be applied to the present invention.

In addition, analysis methods that do not use teacher data, such as principal component analysis, hierarchical clustering, non-hierarchical clustering, and self-organizing maps, should be used as a discrimination method by performing analysis with cell data in a known state. You can also.

Lectins required for cell discrimination differ depending on the state of each cell. Therefore, it is possible to provide a lectin set and a discrimination method for discriminating whether or not the state is a specific cell. Furthermore, a discrimination kit can be provided.

It is also possible to create a lectin set containing lectins necessary for determining the state of a plurality of cells. Thereby, a lectin set and a discrimination method for discriminating a state of a plurality of cells or a change in the state can be provided. Furthermore, a discrimination kit can be provided.

For the extraction of the lectin set, lectins that show a difference between their states by statistical methods such as significant difference test, principal component analysis, etc. based on the results of measuring cells or cell extracts in a state to be distinguished with a plurality of lectins. According to the present invention, cells discriminated by the above discriminating method can be provided.

Further, it is possible to provide a method for maintaining undifferentiation by using the discriminated cell, a compound used for the method, a method for inducing differentiation and a screening method for the compound used for the method.

(Determination method when the sample is undifferentiated and after differentiation of human ES cells)
An example of the human undifferentiated cell discrimination method of the present invention when the sample is a cell extract will be described.

Extraction of proteins 1x10 ⁶ to 5x10 ⁶ human embryonic stem (hES) cells and embryoid bodies (EB) differentiated from hES cells were cryopreserved using CelLytic MEM protein extraction kit (manufactured by Sigma) Then, the hydrophobic fraction containing the membrane protein was separated and extracted. The extract was diluted 20-60 times, and the protein concentration was measured using Micro BCA protein assay kit (PIERCE).

Labeling of membrane protein About 0.2 μg of protein contained in the hydrophobic fraction, Cy3-NHS (manufactured by GE Healthcare) was reacted and fluorescently labeled. After the reaction, unreacted fluorescent labeling reagent was removed from the reaction solution using Sephadex G-25 column.

Reaction of Fluorescent Labeled Protein with Lectin Microarray Containing 1% Triton X-100, 0.5M Glycine, 1 mM CaCl ₂ , 1 mM MnCl ₂ to a final concentration of 0.5 μg / ml or 0.25 μg / ml Prepared with TBS, mounted on a lectin microarray and allowed to react overnight at 4 ° C. As a result, the fluorescently labeled glycoprotein in the sample binds depending on the specificity of each sugar and lectin.

The lectin microarray used (LecChip, manufactured by Sakai GP Bioscience) has 45 different lectins immobilized on each of three spots. The 45 types of lectins used are shown in FIG.

Measurement of lectin microarray After removing the reaction solution and washing with 1% Triton X-100-containing TBS, fluorescence measurement of the lectin microarray was performed with an evanescent field excitation scanner (GlycoStation Reader, GP Bioscience). The exposure time at the time of measurement was set to 199 msec, and the camera gain was measured at each stage of 70, 80, 90, 100, and 110.

Data Pretreatment The fluorescence intensity for each lectin was digitized, and the average of the obtained numerical data of the three spots of each lectin was used as the fluorescence intensity indicated by the lectin.

Since lectins with many proteins bound to them have a strong fluorescence intensity, this is used to measure the binding of sugar-containing molecules to multiple types of lectins, and to determine the amount of sugar-containing molecules bound to each lectin. It is what you do.

In this example, the fluorescence intensity was corrected by the following method, and the corrected data was used. That is, according to the method of Kuno, Itakura et al. (J. Proteomics Bioinform. Vol.1 68-72 (2008)), the fluorescence intensity was measured by combining the data measured with two appropriate gains and measured with the higher gain. The data corrected for the signal saturated with was used.

For each lectin signal value for each sample, after subtracting the background value, the relative value was determined with the UDA signal showing the maximum signal intensity among all lectins as the reference value (100). Processed.

Creation of discriminants for discriminating cell undifferentiation 1) Extraction of explanatory variables Using the values obtained by converting the above relative values into common logarithms, the lectin signal pattern is analyzed by cluster analysis and principal component analysis. It was confirmed that the cells were divided into differentiated cell groups (hES) and differentiated cell groups (EB) (FIGS. 2 and 3).

Next, as a result of extracting a lectin having a significant difference (p <0.01) in signal between the two groups using a t-test, three lectins of MAL, PHA (E), and EEL were obtained.

2) Creation of discriminant The following discriminant (1) using the lectin extracted in 1) as an explanatory variable was determined using a linear discriminant analysis method. MAL, PHA (E), and EEL indicate the relative values of each lectin.

When the score is positive, it is determined as a differentiated cell, and when it is negative, it is determined as an undifferentiated cell.

F = 2.64 × MAL-0.018 × PHA (E) -1.26 × EEL -1.17 …… (1)

3) Discriminant verification 1
The discriminant created in 2) above was applied to lectin microarray data obtained in the same manner as above using hES cells, EBs, human lung tissue-derived fibroblasts (MRC5) and MRC5-derived iPS cells as samples. Differentiation and differentiation discrimination were performed (FIG. 4).

From this result, hES cells and iPS cells scored negative, and EB and MRC5 scores showed positive values. Therefore, this expression is effective in distinguishing undifferentiated pluripotent stem cells. Was confirmed.

4) Verification of discriminant 2
The discriminant created in 2) above is applied to lectin microarray data obtained in the same manner as described above using human amnion-derived mesenchymal cells measured separately from 3) and iPS cells derived from amnion-derived mesenchymal cells as samples. Then, undifferentiation and differentiation were discriminated (FIG. 5).

From this result, the score of iPS cells showed a negative value, and the score of amnion-derived mesenchymal cells showed a positive value. Therefore, this formula is effective in distinguishing undifferentiated pluripotent stem cells. Was confirmed.

Discrimination by the k-nearest neighbor method Using the 3 lectin data extracted above, the hES cells and EB are used as learning data, the MRC5 and MRC5-derived iPS cells are used as test data, and discrimination is performed by the k-nearest neighbor method (k = 3). It was. As a result, all 21 data of iPS cells were classified into hES cell class, and all 3 data of MRC5 were classified into EB class. From the above, it was confirmed that this method is effective for discrimination of undifferentiated pluripotent stem cells.

Gene expression analysis of cells cDNA was synthesized from RNA extracted from the above ES cells and EB (each 3 strains), and the expression of undifferentiated marker genes in hES cells was measured by quantitative real-time PCR (qRT-PCR). The obtained results are shown as relative amounts with the expression level of hES cells being 1 (FIG. 6). From this result, it was confirmed that the undifferentiated markers exhibited by hES cells were significantly reduced in the EB used in this example.

(Determination method when the sample is osteoblast differentiation of human mesenchymal stem cells)
An example of the human undifferentiated cell discrimination method of the present invention when the sample is a cell extract will be described.

Extraction of protein From cell pellets of 1x10 ⁶ to 5x10 ⁶ human mesenchymal stem cells and cells induced to induce osteoblast differentiation (LONZA, medium exchange every 3 days with osteoblast differentiation medium) for 2 weeks, CelLytic Using the MEM protein extraction kit, the hydrophobic fraction containing the membrane protein was separated and extracted. The extract was diluted 20-60 times, and the protein concentration was measured using Micro BCA protein assay kit.

Labeling of membrane protein About 0.2 μg of protein contained in the hydrophobic fraction, Cy3-NHS was reacted to perform fluorescence labeling. After the reaction, unreacted fluorescent labeling reagent was removed from the reaction solution using Sephadex G-25 column.

So that the reaction fluorescently labeled proteins with fluorescently labeled protein and lectin microarray final concentration 0.25 [mu] g / ml, was prepared in 1% Triton X-100, 0.5M Glycine, TBS containing 1mM CaCl _2, 1mM MnCl _2, It was mounted on the same lectin microarray (LecChip) used in Example 1 and allowed to react overnight at 4 ° C. As a result, the fluorescently labeled glycoprotein in the sample binds depending on the specificity of each sugar and lectin.

Measurement of Lectin Microarray After removing the reaction solution and washing with TBS containing 1% Triton X-100, fluorescence measurement of the lectin microarray was performed with an evanescent field excitation scanner (GlycoStation Reader). The exposure time at the time of measurement was set to 199 msec, and the camera gain was measured at each stage of 70, 80, 90, 100, and 110.

Data Pretreatment The fluorescence intensity for each lectin was digitized, and the average of the obtained numerical data of the three spots of each lectin was used as the fluorescence intensity indicated by the lectin.

Fluorescence intensity used data corrected by the same method as in Example 1.

For each lectin signal value for each sample, after subtracting the background value, the relative value was determined with the UDA signal showing the maximum signal intensity among all lectins as the reference value (100). Processed.

Creation of discriminant to discriminate cell undifferentiation 1) Extraction of explanatory variables Using the value obtained by converting the above relative value to the common logarithm, the signal pattern of lectin is analyzed by principal component analysis, and undifferentiated mesenchymal system It was confirmed that it was divided into a stem cell group and a bone differentiated cell group (FIG. 7).

Next, as a result of extracting lectins having a significant difference (p <0.01) in signal between the two groups using t-test, 5 lectins of MAL, SNA, SSA, TJA-I, and ECA were obtained.

2) Creation of discriminant The following discriminant (2) using the lectin extracted in 1) as an explanatory variable was determined using a linear discriminant analysis method. MAL, SNA, SSA, TJA-I and ECA show the relative values of the respective lectins.

When the score is positive, it is determined as a differentiated cell, and when it is negative, it is determined as an undifferentiated cell.

F = -10.81 × MAL + 5.87 × SNA-3.19 × SSA-1.49 × TJA-I + 16.20 × ECA + 5.97 (2)

3) Verification of discriminant formula The discriminant formula created in 2) above is applied to the lectin microarray data of mesenchymal stem cells and bone differentiated cells obtained in the same manner as in the above measurement method to discriminate cell undifferentiation and differentiation. (FIG. 8).

From this result, since the mesenchymal stem cell group showed a positive value, it was confirmed that this discriminant was effective in discriminating undifferentiated cells from differentiated cells.

(Method for distinguishing undifferentiation when the sample is a human ES cell)
An example of the human undifferentiated cell discrimination method of the present invention when the sample is a cell extract will be described.

From the extraction hES cells and EB proteins were stored frozen cell pellets (1x10 ⁶ ~ 5x10 ⁶ cells), using CelLytic MEM protein extraction kit, separated and extracted the hydrophobic fraction containing the membrane protein. The extract was diluted 20-60 times, and the protein concentration was measured using Micro BCA protein assay kit.

Labeling of membrane protein Cy3-NHS was reacted with 0.2 μg of protein contained in the hydrophobic fraction to carry out fluorescent labeling. After the reaction, unreacted fluorescent labeling reagent was removed from the reaction solution using Sephadex G-25 column.

Reaction of fluorescently labeled protein with lectin microarray Prepare fluorescently labeled protein with TBS containing 1% Triton X-100, 0.5M Glycine, 1 mM CaCl ₂ , 1 mM MnCl ₂ to a final concentration of 0.5 μg / ml. It was mounted on the same lectin microarray (LecChip) used in Example 1 and allowed to react overnight at 20 ° C. As a result, the fluorescently labeled glycoprotein in the sample binds depending on the specificity of each sugar and lectin.

Measurement of Lectin Microarray After removing the reaction solution and washing with TBS containing 1% Triton X-100, fluorescence measurement of the lectin microarray was performed with an evanescent field excitation scanner (GlycoStation Reader). The exposure time at the time of measurement was set to 199 msec, and the camera gain was measured at each stage of 80, 90, 100, and 110.

Data Pretreatment The fluorescence intensity for each lectin was digitized, and the average of the obtained numerical data of the three spots of each lectin was used as the fluorescence intensity indicated by the lectin. As the fluorescence intensity, data corrected by the same method as in Example 1 was used.

The signal value of each lectin for each sample is calculated by subtracting the background value and then calculating the relative value with the lectin signal showing the maximum signal intensity as the reference value (100) among all lectins. Used for processing. Lectins showing the above relative value of less than 5 in all samples were not used for the following analysis.

Discrimination of cells by nonlinear discriminant analysis Using the value obtained by converting the above relative value to the common logarithm, the lectin signal pattern was analyzed by the cluster analysis method and the principal component analysis method, and it was confirmed that each cell group was divided (Fig. 9, FIG. 10).

As examples of lectins used for discrimination from the above data, 5 lectins of UEA-I, MAL, PHA (E), BPL, WFA are extracted, and the hES cell and EB data are used as learning data separately from the above. Discrimination analysis of undifferentiated cells by Mahalanobis distance method was performed using lectin array data of hES cells and EBs obtained by measurement by the method (FIG. 11). Here, D1 represents the distance between the sample cell and the center of the hES cell group, and D2 represents the distance between the sample cell and the center of the EB group, and each cell is identified as a cell having a smaller distance.

As a result, all the hES cell 5 data were discriminated as hES cells, and all EBs were discriminated as EBs. This showed that this method is effective in discriminating undifferentiated cells.

(Differentiation tendency discrimination method when the sample is human ES cells)
An example of the human ES cell differentiation tendency discrimination method of the present invention when the sample is a cell extract will be described.

Human ES cell lines three strains extraction differentiation trend was investigated proteins from (HUES 3, HUES 8, HUES 9) cell pellets (1x10 ⁶ ~ 5x10 ⁶ cells) which were stored frozen, using CelLytic MEM protein extraction kit, The hydrophobic fraction containing the membrane protein was separated and extracted. The extract was diluted 20-60 times, and the protein concentration was measured using Micro BCA protein assay kit.

Differentiation tendency of cell lines The differentiation tendency of each cell line is as follows (K. Osafune et al., Nature Biotechnology 2008, 26, 313-315). HUES 3: Has a tendency to differentiate into ectoderm and heart. HUES 8: Has a tendency to differentiate into mesoderm, endoderm, skin, fat, blood, endothelium, pancreas, liver and intestine. HUES 9: Has tendency to differentiate into ectoderm and nerve.

The labeling of the membrane protein, the reaction between the fluorescently labeled protein and the lectin microarray, the measurement of the lectin microarray, and the data pretreatment were performed in the same manner as in Example 3.

Discrimination of cells Using the value obtained by converting the relative value obtained above to the common logarithm, the lectin signal pattern was analyzed by cluster analysis and principal component analysis. As a result, it was confirmed that each cell line was separated ( 12 and 13).

(Determination method when the sample is human mesenchymal stem cells and human fetal cancer cells)
An example of the method for discriminating human mesenchymal stem cells and human fetal cancer cells of the present invention when the sample is a cell will be described.

Collection of cells Cultured cells of two human bone marrow-derived mesenchymal stem cells (UEET12, UET13) and one human embryonal cancer cell line (NCR G3) were detached and collected with PBS containing 5 mM EDTA / 2 mM EGTA. The cells were fixed with PBS containing 4% formaldehyde for 30 minutes at room temperature, and washed twice with PBS.

Cell staining with lectin The immobilized cells were reacted with 1 μg / mL biotinylated lectin for 60 minutes. The cells were washed twice with PBS and reacted with Alexa488-labeled streptavidin for 60 minutes. Further, the cells were washed twice with PBS, and the aggregated cells were removed with a cell strainer. Lectins are WGA, SSA, SNA, MAH, PNA, BPL, ABA, ACA, PWM, MAL, AAL, LEL, DSA, ConA, PHA (E), Jacalin, HPA, DBA, SBA, PHA (L), PSA 21 types were used.

Flow cytometry measurement Flow cytometry measurement was performed with a Beckman FC500 flow cytometer. The value obtained by subtracting the signal detected in the unstained cells from the average value of the fluorescence signal detected in the cells subjected to lectin staining was used as the measurement value. Of the 21 lectin signals, the highest lectin signal was normalized to 1.0, and the other lectin signals were normalized to relative values.

Discrimination of cells Using the value obtained by converting the measured value obtained above to the common logarithm, out of 21 lectins, MAL and PHA (L) were analyzed by cluster analysis using two types of values. The result was that the sex cancer cell line and the two mesenchymal stem cell lines were separated (FIG. 14).

From the above, it was confirmed that this method is effective in distinguishing cell types.

(Determination method when the sample is a human mesenchymal cell of different origin)
An example of a method for discriminating human mesenchymal stem cells from different origin tissues of the present invention when the sample is a cell extract is shown.

Cell types Five human mesenchymal stem cell lines derived from umbilical cord blood (UCB302), bone marrow (UET13, UBET7, 3F0664) and endometrium (EPC100) were used as samples.

Discrimination of cells by gene expression analysis The data registered in NCBI's GEO database (http://www.ncbi.nlm.nih.gov/geo/) was used for gene expression analysis data of each cell line. The ID of each cell line is 3F0664: GSM201145, UET13: GSM210397, UBET7: GSM210386, UCB302: GSM210404, EPC100: GSM210413.

Analysis by cluster analysis using the above gene expression data resulted in a grouping different from the cell-derived tissue (FIG. 15).

Extraction of Protein A hydrophobic fraction containing a membrane protein was separated and extracted from a cell pellet obtained by cryopreserving each cell using a CelLytic MEM protein extraction kit. The extract was diluted 20-60 times, and the protein concentration was measured using Micro BCA protein assay kit.

Labeling of membrane protein About 0.2 μg of protein contained in the hydrophobic fraction, Cy3-NHS was reacted to perform fluorescence labeling. After the reaction, unreacted fluorescent labeling reagent was removed from the reaction solution using Sephadex G-25 column.

So that the reaction fluorescently labeled proteins with fluorescently labeled protein and lectin microarray final concentration 0.25 [mu] g / ml, was prepared in 1% Triton X-100, 0.5M Glycine, TBS containing 1mM CaCl _2, 1mM MnCl _2, It was mounted on the same lectin microarray (LecChip) used in Example 1 and allowed to react overnight at 4 ° C. As a result, the fluorescently labeled glycoprotein in the sample binds depending on the specificity of each sugar and lectin.

Measurement of Lectin Microarray After removing the reaction solution and washing with TBS containing 1% Triton X-100, fluorescence measurement of the lectin microarray was performed with an evanescent field excitation scanner (GlycoStation Reader). The exposure time at the time of measurement was set to 199 msec, and the camera gain was measured at each stage of 70, 80, 90, 100, and 110.

Data Pretreatment The fluorescence intensity for each lectin was digitized, and the average of the obtained numerical data of the three spots of each lectin was used as the fluorescence intensity indicated by the lectin.

Fluorescence intensity used data corrected by the same method as in Example 1.

The signal value of each lectin for each sample was obtained by subtracting the background value and then determining the relative value with the UDA signal showing the maximum signal intensity as the reference value (100) among all the lectins.

Discrimination of cells When the signal was analyzed by cluster analysis using the value obtained by converting the relative value obtained above into the common logarithm, it was confirmed that grouping reflecting the cell-derived tissue was performed (FIG. 16). ).

In the cluster analysis based on the gene analysis results, cell lines derived from the same tissue were not necessarily discriminated into the same group, but in the cluster analysis using the lectin microarray, cell lines derived from the same tissue were discriminated into one group. From the above, it was confirmed that the lectin microarray is effective for discrimination of the cell state. However, since this will vary depending on the target cell, it is more desirable to perform the two approaches simultaneously.

(Determination method when sample is undifferentiated and post-differentiation state of human iPS cells)
An example of the discriminating method of the present invention when the sample is a cell extract and the discriminating state is the undifferentiation of human undifferentiated cells and the degree of cell differentiation will be described.

Protein extraction Human iPS cells and two embryoid bodies (EBs) differentiated from iPS cells (4 and 16 days after induction of differentiation) were cryopreserved using the CelLytic MEM protein extraction kit and hydrophilic. Fractions were separated and extracted. The extract was diluted 20-60 times, and the protein concentration was measured using Micro BCA protein assay kit.

Labeling of protein Cy3-NHS (manufactured by GE Healthcare) was reacted with 1 μg of protein contained in the hydrophilic fraction to perform fluorescence labeling. After the reaction, unreacted fluorescent labeling reagent was removed from the reaction solution using Sephadex G-25 column.

Reaction of fluorescently labeled protein with lectin / antibody microarray 1% Triton X-100, 0.5M Glycine, 1 mM CaCl ₂ , 1 mM MnCl ₂ so that the final concentration of fluorescently labeled protein is 0.5 μg / ml or 0.25 μg / ml Was prepared with TBS containing and placed on a lectin-antibody microarray and allowed to react overnight at 4 ° C. As a result, the fluorescence-labeled glycoprotein in the sample binds according to the specificity of each sugar and lectin, and the marker substance in the sample binds according to the specificity of each antibody.

The lectin / antibody microarray used here is TRA-1-60, TRA-1-81, SSEA-3, which is an undifferentiated marker of 45 lectins and human ES cells and human iPS cells shown in FIG. SSEA-4, Sox2 Oct3 / 4, and Nanog antibodies each immobilized on two spots were used.

Measurement of lectin microarray / antibody microarray After removing the reaction solution and washing with TBS containing 1% Triton X-100, perform fluorescence measurement of the lectin microarray using an evanescent field excitation scanner (GlycoStation Reader, GP Biosciences) It was. The exposure time at the time of measurement was set to 199 msec, and the camera gain was measured at each stage of 70, 80, 90, 100, and 110.

Pretreatment of lectin / antibody microarray data The fluorescence intensity for each lectin and antibody was digitized, and the average of the obtained numerical data of the two spots of each lectin was used as the fluorescence intensity indicated by the lectin and antibody. As the fluorescence intensity, data corrected by the same method as in Example 1 was used.

Pretreatment of signal obtained from lectin microarray The signal value of each lectin for each sample was subtracted from the background value, and then the lectin signal showing the maximum signal intensity among all lectins was defined as the reference value (100). The defined relative values were used for the following statistical treatments. In all samples, lectin signals showing the above relative values of less than 1 were all set to 1.

Signal obtained from antibody microarray The signal obtained for each sample was shown as a relative value with the iPS cell signal as 1 (FIG. 17). As a result, in the EB used in this example, it was confirmed that the undifferentiated marker was lowered, and it was confirmed that the cells had lost undifferentiated properties. Moreover, there was almost no difference between the cells 4 days after differentiation induction and the cells after 16 days.

Discrimination of cells by lectin Using the value obtained by converting the relative value obtained above to the common logarithm, the lectin signal pattern was analyzed by cluster analysis and principal component analysis. This was confirmed (FIGS. 18 and 19). Further, FIG. 19 shows that EB 4 days after differentiation induction and EB 16 days after differentiation induction can also be discriminated by the main component 1 (PC1) capable of discriminating iPS cells and EBs. That is, it was confirmed by lectin microarray analysis that the state after differentiation induction can also be discriminated.

Based on the above, the data obtained by the lectin / antibody microarray is effective for discrimination of the state of cells, and the discrimination of undifferentiation can be clearly discriminated by combining the results of the lectin microarray and the antibody microarray. It was confirmed that the analysis by the lectin microarray is effective for discriminating the state in more detail.

(Determination method when the sample is undifferentiated and after differentiation of human ES cells)
Another example of the human undifferentiated cell discrimination method of the present invention when the sample is a cell extract will be described.

In the same manner as in Example 1, protein extraction, membrane protein labeling, reaction of fluorescently labeled protein with lectin microarray, and measurement of lectin microarray were performed. In addition, data preprocessing was performed in the same manner as in Example 1, and a relative value was determined for the signal value of each lectin for each sample.

Then, in the same manner as in Example 1, using the value obtained by converting the relative value to the common logarithm, the lectin signal pattern was analyzed by the cluster analysis method and the principal component analysis method, and differentiated from the undifferentiated cell group (hES). After confirming that the cells were divided into cell groups (EB), we extracted the lectins that had a significant difference in signal (p <0.01) between the two groups using t-test, and as a result, four lectins MAL, PHA (L), EEL , PHA (E) was extracted.

Next, the combination of the extracted lectins and the discrimination accuracy were examined. Using the discriminant described above, the correctness of discrimination based on the discriminant of data obtained by separately measuring iPS cells derived from human lung tissue-derived fibroblasts (MRC5) using a lectin microarray was quantified. The result was as shown in FIG.

In this way, PHA (L) and EEL were selected as the optimal combination. The clustering result with the selected PHA (L) and EEL is as shown in FIG.

From the above, it was confirmed that the discrimination accuracy was improved by identifying the state to be discriminated and selecting the lectins.

(Determination method when the sample is human mesenchymal stem cells and human fetal cancer cells)
An example of a discrimination method in which the sample is a cell extract and the discrimination state is the cell type and the cell-derived tissue will be described.

Protein extraction 1x10 ⁶ to 5x10 ⁶ human mesenchymal stem cells (excess finger-derived cells (Yub), surplus finger bone marrow-derived cells (Yub_BMC), cord blood-derived cells (UCB), placenta-derived cells (PL), amnion-derived Cells (AM), bone marrow-derived cells (2F, 3F, UET, UEET, UBET), endometrial-derived cells (UtE, EPC), menstrual blood-derived cells (Edom)), and human fetal cancer cells (NCRG3) From the cell pellet, a hydrophobic fraction containing membrane protein was separated and extracted using CelLytic MEM protein extraction kit. The extract was diluted 20-60 times, and the protein concentration was measured using Micro BCA protein assay kit.

Labeling of membrane protein About 0.2 μg of protein contained in the hydrophobic fraction, Cy3-NHS was reacted to perform fluorescence labeling. After the reaction, unreacted fluorescent labeling reagent was removed from the reaction solution using Sephadex G-25 column.

So that the reaction fluorescently labeled proteins with fluorescently labeled protein and lectin microarray final concentration 0.25 [mu] g / ml, was prepared in 1% Triton X-100, 0.5M Glycine, TBS containing 1mM CaCl _2, 1mM MnCl _2, It was mounted on the same lectin microarray (LecChip) used in Example 1 and allowed to react overnight at 4 ° C. As a result, the fluorescently labeled glycoprotein in the sample binds depending on the specificity of each sugar and lectin.

Measurement of Lectin Microarray After removing the reaction solution and washing with TBS containing 1% Triton X-100, fluorescence measurement of the lectin microarray was performed with an evanescent field excitation scanner (GlycoStation Reader). The exposure time at the time of measurement was set to 199 msec, and the camera gain was measured at each stage of 70, 80, 90, 100, and 110.

Data Pretreatment The fluorescence intensity for each lectin was digitized, and the average of the obtained numerical data of the three spots of each lectin was used as the fluorescence intensity indicated by the lectin.

Fluorescence intensity used data corrected by the same method as in Example 1.

For each lectin signal value for each sample, after subtracting the background value, the relative value was determined with the UDA signal showing the maximum signal intensity among all lectins as the reference value (100). Processed.

Using the value obtained by converting the relative value to the common logarithm, the lectin signal pattern was analyzed by the cluster analysis method. As shown in FIG. 21, the mesenchymal stem cells and fetal cancer cells were discriminated and the origin of the cells was determined. I was able to determine. That is, when the signal patterns of 45 types of lectins shown in FIG. 1 were analyzed by cluster analysis, mesenchymal stem cells and fetal cancer cells could be discriminated as shown in FIG.

Selection of lectins according to purpose Four lectins (GSL I B4, PTL-I, GSL I A4, PNA) were selected for Yub discrimination.

As a result of discrimination by hierarchical clustering analysis, it was confirmed that Yub-derived mesenchymal stem cells and other cells were clearly discriminated (FIG. 23).

Next, 2 lectins (MAL, PHA (L)) were selected for discrimination of fetal cancer cells.

As a result of discrimination by hierarchical clustering analysis, it was confirmed that fetal cancer cells and mesenchymal stem cells were clearly discriminated (FIG. 24).

As mentioned above, although embodiment of this invention and the preferable example were described, this invention is not limited to these, It can change into various forms in the technical range grasped | ascertained from description of a claim. .

Claims (21)

Obtained by measuring a sugar chain of a cell to be discriminated or a sugar chain-containing sample containing the sugar chain while bound to a plurality of types of lectins in a liquid phase, the plurality of types A method of discriminating the state of a cell by using the amount of the cell to be discriminated or the amount of a sugar chain-containing sample, each bound to a lectin. The method according to claim 1, wherein the cells are stem cells or progenitor cells. The method according to claim 2, wherein the stem cells are somatic stem cells, embryonic stem cells, nuclear transplant stem cells, or induced pluripotent stem cells. The method according to claim 3, wherein the somatic stem cells are hematopoietic stem cells or mesenchymal stem cells. 5. The method for discriminating the state of a cell according to any one of claims 2 to 4, wherein the cell is derived from a mammal or a human. 6. The method for discriminating a cell state according to any one of claims 1 to 5, wherein the cell state to be discriminated includes an undifferentiated state of the cell and a differentiation state of the cell. 6. The method for discriminating a cell state according to any one of claims 1 to 5, wherein the cell state to be discriminated is a cell differentiation tendency or a resistance to cell differentiation. 6. The method for discriminating a cell state according to any one of claims 1 to 5, wherein the cell state to be discriminated is a cell canceration or a tendency of cell canceration. The method for discriminating the state of a cell according to any one of claims 1 to 8, wherein the sugar-containing molecule is any one or a plurality of glycoproteins, glycopeptides, glycolipids, and proteoglycans. . 10. The state of a cell according to any one of claims 1 to 9, wherein the plurality of types of lectins are fixed to any one of a lectin array, an ELISA plate, a magnetic bead, and a latex bead. how to. The method for determining the state of a cell according to any one of claims 1 to 10, wherein the amount of the cell or sugar chain-containing sample is measured by fluorescence measurement or luminescence measurement. 12. The method for determining the state of a cell according to claim 11, wherein the fluorescence measurement is an evanescent excitation method or a confocal fluorescence measurement method. The method according to claim 11, wherein the fluorescence measurement is flow cytometry. The cell state discrimination method is performed in combination with one or both of gene expression analysis and binding measurement with a specific marker antibody of a cell in a target state. To determine the state of cells. The method for discriminating a cell state according to any one of claims 1 to 13, wherein the cell state discrimination method is performed by linear discriminant analysis or non-linear discriminant analysis. The method according to claim 15, wherein the nonlinear discriminant analysis is discriminant analysis based on Mahalanobis distance. 14. The method for discriminating a cell state according to any one of claims 1 to 13, wherein the cell state discrimination method is a supervised machine learning method or a semi-supervised machine learning method. The machine learning method is performed by any method selected from a k-nearest neighbor method, a naive Bayes classifier, a decision tree, a neural network, a support vector machine, a bagging method, a boosting method, and a random forest method. Item 18. A method for determining the state of a cell according to Item 17. The cell state determination method according to any one of claims 1 to 13, wherein the cell state determination method is performed by any one method selected from principal component analysis, cluster analysis, and self-organizing map. how to. The state of the cell to be identified is undifferentiated and differentiated human embryonic stem cells or human induced pluripotent stem cells, and the lectins used for the identification are canine endlectin I, kidney bean lectin (L), kidney bean lectin (E), spindle tree 2. One or a plurality of lectins selected from the group consisting of molecules each having a similar sugar-binding specificity to any one of these lectins or to any one of these lectins. 20. A method for determining the state of a cell according to any one of items 1 to 19. The state of the cell to be identified is undifferentiation and bone differentiation of human mesenchymal stem cells, and the lectins used for the discrimination are canine endlectin I, elder chicken collectin, Japanese elder chicken collectin, kirasouri lectin-I, and deigo bean lectin. 20. One or a plurality of molecules selected from the group consisting of molecules each having a sugar-binding specificity similar to any one or any of these lectins. A method for determining the state of a cell according to claim 1.

Images