JP7830838B2 - A method for separating cells with different degrees of undifferentiation using layered adsorbents. - Google Patents

Description

This invention relates to a cell separation method using an adsorbent in which a fucose-binding protein is immobilized on a water-insoluble carrier. The method involves sequentially contacting a cell population with a column containing multiple adsorbents with varying amounts of fucose-binding protein immobilized on the carrier, and then recovering the fraction of cells that leach out of the column, thereby separating cell populations with different degrees of undifferentiation.

BC2LCN, derived from the N-terminal domain of BC2L-C lectin produced by the Gram-negative bacterium *Burkholderia cenocepacia*, is a protein that has binding affinity to glycans containing fucose residues. For example, in addition to the H-type 1 glycans (Fucα1-2Galβ1-3GlcNAc) and H-type 3 glycans (Fucα1-2Galβ1-3GalNAc) known as undifferentiated glycan markers described in Non-Patent Document 1, Patent Document 1, and Patent Document 2, it is known to have high binding affinity to multiple types of glycans containing fucose residues, such as Lewis Y-type glycans (Fucα1-2Galβ1-4(Fucα1-3)GlcNAc) and Lewis X-type glycans (Galβ1-4(Fucα1-3)GlcNAc) (Non-Patent Document 2).

Furthermore, it is known that BC2LCN binds to undifferentiated human iPS cells and human ES cells that highly express H-type 1 and H-type 3 glycans, but does not bind to human somatic cells (Non-Patent Literature 3).

Furthermore, because BC2LCN has binding affinity to the aforementioned undifferentiated glycan marker, it is used, for example, in the detection of complex carbohydrates containing the undifferentiated glycan marker, and in the detection of undifferentiated cells such as human iPS cells and human ES cells (Patent Documents 1 and 2). Also, H-type 1 glycans are known to be highly expressed as SSEA-5 in certain cancer cells (Non-Patent Document 4).

Although BC2LCN possesses the same ability to detect undifferentiated stem cells as known antibodies for detecting undifferentiated cells, such as anti-Nanog antibodies (Patent Document 2), the binding of BC2LCN to the glycans of undifferentiated cells is due to electrostatic interactions, and the strength of the binding is affected by external environmental factors such as solvent and salt concentration. Therefore, depending on the experimental conditions, the binding affinity between the glycans and BC2LCN may be low when detecting the undifferentiated cells and/or complex carbohydrates containing the undifferentiated glycan marker. Thus, there is a need for BC2LCN with improved binding affinity to the undifferentiated glycan marker.

As a cell separation method using an adsorbent in which BC2LCN is immobilized on a water-insoluble carrier, a method for removing undifferentiated cells such as human iPS cells is known (Patent Document 3), and a method for detaching and recovering undifferentiated cells bound to the adsorbent is also known (Patent Document 4). However, while the methods disclosed in these patent documents could separate cells expressing H-type 1 and H-type 3 glycans from cells that did not express them, they could not individually separate heterogeneous cell populations with different levels of H-type 1 and H-type 3 glycan expression, or cells with different levels of H-type 1 and H-type 3 glycan expression from the same cell species population.

Generally, methods for separating cells by utilizing differences in the expression levels of undifferentiated markers on the cell surface and cell surface antigens characterized by the CD (Cluster of Difference) classification are known, including cell separation using magnetic beads (Non-Patent Literature 5), cell sorting using cell sorters (Non-Patent Literature 5), and cell separation using cell rolling columns (Non-Patent Literature 6). However, while cell separation using magnetic beads is suitable for processing large quantities of cells, it cannot separate cells with small differences in undifferentiated marker expression levels or CD antigen expression levels. Furthermore, there are quality control issues such as cell damage due to magnetic beads adhering to cells and contamination by foreign matter, and concerns have been raised about the impact on the proliferative and differentiation-oriented properties of the recovered cells. Cell sorting using cell sorters and cell separation using cell rolling columns require expensive equipment and take a considerable amount of time to process large quantities of cells, resulting in significant problems in terms of cell separation efficiency.

However, currently, there is a strong need to develop technologies for accurately separating and purifying large quantities of cells in a short time, based on differences in cell surface undifferentiated marker expression levels and CD antigen expression levels, for use in regenerative medicine.

WO2013/065302 WO2013/128914 Japanese Patent Publication No. 2018-134073 Japanese Patent Publication No. 2019-000063

Tateno, H et al., Stem Cells Transl Med. 2013, 2(4): 265-273. Sulak, O et al., Structure. 2010, 18(1):59-72. Tateno, H et al., J Biol Chem. 2011, 286(23):20345-20353. Tang, C et al., Nat Biotechnol. 2011, 29(9):829-835. Fong CY et al., Stem Cell Rev Rep. 2009, 5(1): 72-80. Mahara, A et al., J Biomater Sci Polym Ed. 2014, 25(14-15): 1590-601.

The object of the present invention is to provide a technology that enables the separation and purification of cell populations with different degrees of undifferentiation from a large number of cells in a short time when preparing cells for regenerative medicine applications, specifically a technology for separating cell populations with different levels of expression of undifferentiation markers present on the cell surface.
More specifically, the aim is to provide a technique for easily and efficiently separating cell populations with different degrees of undifferentiation by sequentially contacting a cell population with a column containing multiple adsorbents with varying amounts of fucose-binding protein immobilized on a carrier, and then collecting the fraction of cells that did not bind to the adsorbent.

The inventors of this invention, after diligent research to solve the aforementioned problems, discovered that it is possible to separate and recover cell populations with different levels of expression of undifferentiated markers present on the cell surface from the original cell population by first contacting cell populations with a low amount of fucose-binding protein immobilization with an adsorbent, and then progressively contacting them with adsorbents with a higher amount of fucose-binding protein immobilization, and then recovering the fraction of cells that did not adsorb with the adsorbent. This led to the completion of the present invention.

In other words, the present invention encompasses the inventions described in [1] to [8] below.
[1]
A method for separating cells using an adsorbent comprising a fucose-binding protein immobilized on a water-insoluble carrier, comprising the steps of: binding cell populations with different degrees of undifferentiation to multiple adsorbents with different amounts of immobilized fucose-binding protein; and obtaining cells that did not bind to the adsorbents.
[2]
A method for separating cells of different degrees of undifferentiation according to [1], comprising the steps of: binding cell populations of different degrees of undifferentiation to multiple adsorbents with different amounts of immobilized fucose-binding proteins, in a stepwise manner from an adsorbent with a low amount of immobilized fucose-binding proteins to an adsorbent with a high amount of immobilized fucose-binding proteins; and obtaining cells that did not bind to the adsorbents.
[3]
A method for separating cells of different degrees of undifferentiation according to [1] or [2], characterized in that the amount of fucose-binding protein immobilized ranges from 50 mg/L-adsorbent to 1000 mg/L-adsorbent, and multiple adsorbents with fucose-binding protein immobilization amounts differing by 100 mg/L-adsorbent or more.

[4]
A method for separating cells of different degrees of undifferentiation according to any one of [1] to [3], characterized in that the cell populations of different degrees of undifferentiation express a sugar chain containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc on the surface of the cells.
[5]
A method for separating cells with different expression levels of glycans containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc, using the method described in any of [1] to [4] above.
[6]
The method according to any one of the above [1] to [5], wherein the fucose-binding protein is any one of the following (a) to (d).
(a) A fucose-binding protein comprising the amino acid sequence from the first proline residue to the Xth amino acid residue of the amino acid sequence shown in Sequence ID No. 1, wherein X is an integer of 120 or more.
(b) A fucose-binding protein comprising an amino acid sequence in which one or more amino acids are deleted, substituted, or added in the amino acid sequence from the first proline residue to the Xth amino acid residue of the amino acid sequence shown in Sequence ID No. 1, and having a binding affinity to a sugar chain comprising a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc, wherein X is an integer of 120 or more.
(c) A fucose-binding protein comprising an amino acid sequence in which the amino acid sequence from the first proline residue to the Xth amino acid residue of the amino acid sequence shown in SEQ ID NO: 1 contains one or more of the amino acid substitutions described in (1) to (3) below, and X is an integer of 120 or more: (1) Substitution of the 39th glutamine residue of the amino acid sequence shown in SEQ ID NO: 1 with a leucine residue (2) Substitution of the 72nd cysteine residue of the amino acid sequence shown in SEQ ID NO: 1 with one amino acid residue selected from a glycine residue and an alanine residue Substitution (3) Substitution of the 65th glutamine residue in the amino acid sequence shown in Sequence ID No. 1 with a leucine residue (d) An amino acid sequence in which, in the amino acid sequence of the fucose-binding protein of (c) above, one or more amino acid residues are deleted, substituted, inserted or added in regions other than the 39th, 65th and 72nd positions of Sequence ID No. 1, and a fucose-binding protein having a binding affinity to a sugar chain containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc.

[7]
The method according to any one of the above [1] to [5], wherein the fucose-binding protein is any of the following (e) to (h).
(e) A fucose-binding protein having an amino acid sequence in which the amino acid sequence from the first proline residue to the Xth amino acid residue of the amino acid sequence shown in Sequence ID No. 1 is further modified by adding an oligopeptide containing a polyhistidine sequence to the N-terminus and an oligopeptide containing cysteine to the C-terminus, and where X is an integer of 120 or more.
(f) A fucose-binding protein having an amino acid sequence in which one or more amino acids are deleted, substituted, or added in the amino acid sequence from the first proline residue to the Xth amino acid residue of the amino acid sequence shown in Sequence ID No. 1, and further having a polyhistidine sequence added to the N-terminus and a cysteine-containing oligopeptide added to the C-terminus, and having a binding affinity to a sugar chain containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc, wherein X is an integer of 120 or more.
(g) An amino acid sequence in which an oligopeptide containing a polyhistidine sequence is added to the N-terminus and an oligopeptide containing cysteine is added to the amino acid sequence from the first proline residue to the Xth amino acid residue of the amino acid sequence shown in SEQ ID NO: 1, and the amino acid sequence includes one or more of the amino acid substitutions described in (4) to (6) below, and X is an integer of 120 or more, and is a fucose-binding protein (4) Substitution of the 39th glutamine residue of the amino acid sequence shown in SEQ ID NO: 1 to a leucine residue (5) Substitution of the 72nd cysteine residue of the amino acid sequence shown in SEQ ID NO: 1 to one amino acid residue selected from a glycine residue and an alanine residue (6) Substitution of the 65th glutamine residue in the amino acid sequence shown in Sequence ID No. 1 with a leucine residue (h) In the amino acid sequence of the fucose-binding protein of (g) above, one or more amino acid residues are deleted, substituted, inserted or added in regions other than the 39th, 65th and 72nd positions of Sequence ID No. 1, and an oligopeptide containing a polyhistidine sequence is further added to the N-terminus and an oligopeptide containing cysteine is further added to the C-terminus, and the fucose-binding protein has a binding affinity to a sugar chain containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc.

[8]
The method according to any one of [1] to [7], characterized by using a column packed with multiple adsorbents having different amounts of fucose-binding protein immobilized on it.

The present invention will be described in detail below.

The present invention provides a method for separating cells with different degrees of undifferentiation using stacked adsorbents (hereinafter sometimes referred to as the cell separation method of the present invention). This method utilizes an adsorbent in which a fucose-binding protein is immobilized on a water-insoluble carrier. The method is characterized by including a step of sequentially contacting a cell population with a column containing multiple adsorbents stacked with different amounts of fucose-binding protein immobilized on the carrier, and then obtaining cells that did not bind to the adsorbent as a fraction.

In the cell separation method of the present invention, the undifferentiated marker is a molecule present on the cell surface that serves as an indicator of the degree of undifferentiation. Specifically, examples include a glycan containing a structure made of Fucα1-2Galβ1-3GlcNAc, and/or a glycan containing a structure made of Fucα1-2Galβ1-3GalNAc, which has a glycan structure including H-type 1 glycan structure, H-type 3 glycan structure, Lewis Y-type glycan structure, and/or Lewis b-type glycan structure, all of which are known as fucose-containing glycans to which the above-mentioned fucose-binding protein can bind. Furthermore, the undifferentiated marker in the present invention is not limited to the above-mentioned glycan structures, but may also be TRA-1-60, TRA-1-81, SSEA-3, SSEA-4, SSEA-5, and other glycans that are generally known as undifferentiated markers on the cell surface.

In the cell isolation method of the present invention, cells with different degrees of undifferentiation refer to cells with different levels of expression of the undifferentiation marker on their cell surface. Cells with different degrees of undifferentiation primarily refer to cells with clearly different levels of expression of the undifferentiation marker. In this case, the level of expression of the undifferentiation marker is determined by the proportion of cells in the cell population that express the undifferentiation marker molecule on their cell surface, or by the number and density of undifferentiation marker molecules present on the surface of a single cell. Taking a sugar chain to which the fucose-binding protein can bind as an example of an undifferentiation marker, it is known that, for example, K562 cells (human chronic myeloid leukemia) and NHDF cells (human dermal fibroblasts) do not contain cells in the cell population that have a sugar chain that binds to the fucose-binding protein. Furthermore, it is known that in 2102Ep cells (human embryonic tumor cells), approximately 50% to 80% of the cell population expresses glycans that bind to the aforementioned fucose-binding protein, and in human pluripotent stem cells, such as human ES cells and human iPS cells, nearly 100% of the cells express glycans that bind to the aforementioned fucose-binding protein.

Thus, it is thought that the proportion of cells possessing undifferentiated marker molecules differs among different cell types (hereinafter referred to as heterogeneous cells). Furthermore, when comparing the degree of undifferentiation of heterogeneous cells, even if each cell type expresses the same proportion of glycans that bind to the fucose-binding protein, the number and density of undifferentiated marker molecules present on the cell surface are likely to differ among heterogeneous cells.

Secondly, a cell population with different degrees of undifferentiation, even within a single cell type, refers to a case where the number and density of undifferentiation marker molecules present on the cell surface differ from cell to cell. For example, if the undifferentiation marker is a glycan capable of binding to the fucose-binding protein, then in 2102Ep cells, the expression level of this glycan varies widely, from low to high. Therefore, even within a single cell type, it can be considered a mixture of cell populations with different levels of undifferentiation marker expression, and this case is also included in the definition of cells with different degrees of undifferentiation in this invention.

Furthermore, in the case of human iPS cells, as mentioned above, it is known that nearly 100% of the cells express glycans that bind to the fucose-binding protein. However, depending on the culture conditions, it is known that less than 1% of the cells may be undifferentiated offshoots with a low degree of undifferentiation. In this case as well, in the present invention, human iPS cells and undifferentiated offshoot human iPS cells can be considered as cells with different degrees of undifferentiation, even though they are the same cell type. A third form refers to a mixture of heterogeneous cells with different levels of expression of the undifferentiation markers described in the first form above, and all cell populations containing one or more glycans that bind to the fucose-binding protein, where the number and density of undifferentiation marker molecules on the cell surface differ for each individual cell, as described in the second form.

In this invention, all of these first to third forms of cell mixtures can be used as the target for separation.

In the cell separation method of the present invention, the cell mixture to be separated may be a mixture of heterogeneous cells with different levels of undifferentiated marker expression, a single cell type which is an aggregate of cell populations with different levels of undifferentiated marker expression, or a cell mixture containing both. In the present invention, if heterogeneous cells with different levels of undifferentiated marker expression, or cell populations within a single cell type with different levels of undifferentiated marker expression, are used, it is possible to separate cells with different degrees of undifferentiation based on the differences in cell type and undifferentiated marker expression levels, respectively. Furthermore, in this case, the number of heterogeneous cell types with different levels of undifferentiated marker expression, and the number of cell populations within a single cell type with different levels of undifferentiated marker expression, may be two or more.

Furthermore, the indicator of differences in undifferentiated marker expression levels can be considered, for example, as the degree of shift in the fluorescence intensity of cell populations detectable by a cell sorter such as the BD FACSAria (manufactured by Becton Dickinson) when a certain number and volume of cell suspensions are treated with the same concentration and volume of fluorescent antibody solution.

The present invention provides a cell separation method that involves first contacting cell populations with different degrees of undifferentiation with an adsorbent that has a low amount of fucose-binding protein immobilized, and then progressively contacting them with adsorbents that have a higher amount of fucose-binding protein immobilized, thereby collecting the fraction of cells that did not bind to the adsorbent.

In this case, at least two or more adsorbents prepared with different amounts of fucose-binding protein immobilized are sufficient. The number of adsorbent types used is not limited, as long as the order in which the cell population is brought into contact with the adsorbents progresses from those with a low amount of fucose-binding protein immobilized to those with a high amount. Furthermore, the fucose protein immobilized on the adsorbent can be either one with the same amino acid sequence or one whose function has been improved through amino acid substitutions as described later. It is also possible to prepare multiple adsorbents with different amounts of fucose-binding protein immobilized with different amino acid sequences, and then arbitrarily select and combine two or more of these adsorbents.

In the cell separation method of the present invention, by first contacting cell populations with different degrees of undifferentiation with an adsorbent that has a low amount of fucose-binding protein immobilization, the cell population with a high degree of undifferentiation within the cell population can be selectively captured by the adsorbent.

Next, by gradually bringing the remaining cell population into contact with adsorbents containing higher amounts of fucose-binding protein, the less differentiated cell populations within the cell population can be captured by the adsorbent. Therefore, for example, when the cell separation method of the present invention is carried out using a column packed with adsorbent, by layering an adsorbent with a low amount of fucose-binding protein at the top of the column and an adsorbent with a high amount of fucose-binding protein at the bottom of the column, when cell populations with different degrees of undifferentiation are brought into contact, the more differentiated cell populations will be selectively adsorbed to the adsorbent with the low amount of fucose-binding protein at the top of the column, and the less differentiated cell populations will be selectively adsorbed to the adsorbent with the high amount of fucose-binding protein at the bottom of the column.

In this case, based on the differences in the expression levels of undifferentiated glycan markers on the surface of individual cells, it is thought that a cell distribution is formed in the column from the top to the bottom, with cells expressing high levels of undifferentiated glycan markers (i.e., cells with a high degree of undifferentiation) and cells expressing low levels of undifferentiated glycan markers (i.e., cells with a low degree of undifferentiation). Therefore, the cells that first flow out from the bottom of the column are thought to be a population of cells with a low degree of undifferentiation, and the cells that flow out later are thought to be a population of cells with a high degree of undifferentiation. By carefully collecting fractions of each cell population with a different degree of undifferentiation that flow out from the bottom of the column, it is possible to separate cell populations with different degrees of undifferentiation.

Furthermore, in this case, the binding of undifferentiated glycan markers on the cell surface to the fucose-binding protein of the adsorbent is reversible, and since it is affected by the gravity of the cells themselves and the shear force of the solution flow introduced from the top of the column, it is thought that the cells repeatedly bind to and detach from the adsorbent. Therefore, in the cell separation method of the present invention, among the multiple adsorbents packed into the column, it is thought that a cell distribution is formed such that the cell population closer to the top of the column is more undifferentiated, and the cell population further from the top is less undifferentiated.

This phenomenon is thought to be based on the same principle as, for example, the phenomenon demonstrated in rolling columns of anti-CD34 antibody-immobilized cells, where cells with lower CD34 antigen expression levels migrate and leach faster, while cells with higher CD34 antigen expression levels migrate and leach slower.

In the cell separation method of the present invention, it is possible to use adsorbents with fucose-binding protein immobilization amounts ranging from 50 mg/L to 1000 mg/L, and it is preferable to use multiple adsorbents such that the fucose-binding protein immobilization amounts of each adsorbent differ by 100 mg/L or more. An example of the cell separation method of the present invention is a procedure in which a cell population is first brought into contact with an adsorbent with a fucose-binding protein immobilization amount of 100 mg/L, then the cell population is brought into contact with an adsorbent with a fucose-binding protein immobilization amount of 200 mg/L, and then with an adsorbent with a fucose-binding protein immobilization amount of 300 mg/L. The type of fucose-binding protein used in the multiple adsorbents, the amount of fucose-binding protein immobilization, the combination of adsorbents, and the amount of each adsorbent used (adsorbent ratio) can be appropriately adjusted according to the degree of undifferentiation and range (variability) of the undifferentiated cells to be separated.

The best embodiment of the cell separation method of the present invention involves stacking and packing multiple adsorbents in a vertically positioned column, with adsorbents having a low fucose-binding protein immobilization rate at the top of the column and adsorbents having a high fucose-binding protein immobilization rate at the bottom. In this case, the method of packing the adsorbents is not particularly limited; it is sufficient that a concentration gradient is created in the amount of fucose-binding protein immobilized on the surface of the packed adsorbents. Multiple adsorbents with different fucose-binding protein immobilization rates may be stacked and packed, or the adsorbents may be packed so that the concentration gradient of fucose-binding protein immobilized on the adsorbents is linear.

Next, the fucose-binding proteins in the cell separation method of the present invention will be described. The fucose-binding proteins in the cell separation method of the present invention are proteins that have the ability to bind to fucose-containing glycans such as H-type I glycans (Fucα1-2Galβ1-3GlcNAc), H-type III glycans (Fucα1-2Galβ1-3GalNAc), Lewis Y-type glycans (Fucα1-2Galβ1-4(Fucα1-3)GlcNAc), and Lewis b-type glycans (Fucα1-2Galβ1-3(Fucα1-4)GlcNAc). The recombinant BC2LCN lectin mentioned above is also included in the fucose-binding proteins in the cell separation method of the present invention. The fucose-binding protein in the cell isolation method of the present invention is specifically (a) a protein containing the amino acid sequence from the 1st proline to the Xth amino acid of the recombinant BC2LCN lectin shown in SEQ ID NO: 1 (which matches the amino acid sequence from the 2nd to the 156th amino acid of the amino acid sequence registered in GenPept under registration number WP_006490828), where X is an integer of 120 or more, or (b) a protein consisting of an amino acid sequence in which one or more amino acids are deleted, substituted, or added in the amino acid sequence from the 1st proline to the Xth amino acid of the amino acid sequence shown in SEQ ID NO: 1, and which has binding ability to H-type 1 glycans and/or H-type 3 glycans, and where X is an integer of 120 or more, expressed as a recombinant protein in a transformed E. coli.

In the cell separation method of the present invention, the fucose-binding protein has binding ability to the fucose-containing glycans, particularly glycans containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc. In this case, one or more amino acids may be deleted, substituted, or added in the amino acid sequence from the first proline to the Xth amino acid of the amino acid sequence shown in SEQ ID NO: 1. For example, 15 or fewer amino acids, preferably 10 or fewer, may be deleted, substituted, or added. Furthermore, X may be between 120 and 155, or between 125 and 155.

As disclosed in Japanese Patent Publication No. 2020-25535, the fucose-binding protein in the cell isolation method of the present invention can have its productivity (expression level) improved when produced in E. coli transformants compared to when the amino acid residues shown in SEQ ID NO: 1 are not deleted, by deleting multiple amino acid residues at the C-terminal end of the amino acid sequence.

Furthermore, the fucose-binding protein in the cell separation method of the present invention may include one or more of the following to improve thermal stability: (i) substitution of the 39th glutamine residue in the amino acid sequence shown in SEQ ID NO: 1 with a leucine residue; (ii) substitution of the 72nd cysteine residue in the amino acid sequence shown in SEQ ID NO: 1 with one amino acid residue selected from a glycine residue and/or an alanine residue; and (iii) substitution of the 65th glutamine residue in the amino acid sequence shown in SEQ ID NO: 1 with a leucine residue.

As disclosed in Japanese Patent Application Publication No. 2020-25535, the thermal stability of the fucose-binding protein in the cell separation method of the present invention can be improved by performing the amino acid substitutions described in (i) to (iii) above. While the amino acid substitutions described in (i) to (iii) above are effective in improving thermal stability whether performed individually or in combination, it is more preferable to combine multiple amino acid substitutions described in (i) to (iii) above in order to further improve thermal stability.

Furthermore, as long as the fucose-binding protein in the cell separation method of the present invention has the ability to bind to a sugar chain containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc, one or more amino acid residues may be deleted, substituted, or inserted in the amino acid sequence from the first proline to the Xth amino acid of the amino acid sequence shown in SEQ ID NO: 1, in addition to the positions substituted by the substitutions in (i) to (iii). For example, 15 or fewer, preferably 10 or fewer, amino acid residues may be deleted, substituted, or inserted.

Specific examples of fucose-binding proteins in the cell separation method of the present invention include: SEQ ID NO: 1, SEQ ID NO: 2 (amino acid sequence from the 1st to the 127th amino acid sequence shown in SEQ ID NO: 1), SEQ ID NO: 3 (amino acid sequence in which the 72nd cysteine residue of SEQ ID NO: 2 is replaced with a glycine residue), SEQ ID NO: 4 (amino acid sequence in which the 39th glutamine residue of SEQ ID NO: 2 is replaced with a leucine residue, and the 72nd cysteine residue of SEQ ID NO: 2 is replaced with a leucine residue, the 65th glutamine residue of SEQ ID NO: 2 is replaced with a leucine residue, and the 72nd cysteine residue of SEQ ID NO: 2 is replaced with a glycine residue), and SEQ ID NO: 6 (amino acid sequence in which an oligopeptide containing a polyhistidine sequence is added to the N-terminus of the amino acid sequence shown in SEQ ID NO: 1, and an oligopeptide containing a cysteine residue is added to the C-terminus). Examples of fucose-binding proteins include those represented by SEQ ID NO: 1, 7 (an amino acid sequence in which a polyhistidine sequence oligopeptide is added to the N-terminus of the amino acid sequence shown in SEQ ID NO: 2, and a cysteine residue oligopeptide is added to the C-terminus), 8 (an amino acid sequence in which a polyhistidine sequence oligopeptide is added to the N-terminus of the amino acid sequence shown in SEQ ID NO: 3, and a cysteine residue oligopeptide is added to the C-terminus), 9 (an amino acid sequence in which a polyhistidine sequence oligopeptide is added to the N-terminus of the amino acid sequence shown in SEQ ID NO: 4, and a cysteine residue oligopeptide is added to the C-terminus), and 10 (an amino acid sequence in which a polyhistidine sequence oligopeptide is added to the N-terminus of the amino acid sequence shown in SEQ ID NO: 5, and a cysteine residue oligopeptide is added to the C-terminus).

In the cell separation method of the present invention, the fucose-binding protein may have additional amino acid sequences useful for detecting the fucose-binding protein at its N-terminus and/or C-terminus, as long as it has the ability to bind to a sugar chain containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc. Examples of such additional amino acid sequences include oligopeptides containing polyhistidine sequences, glutathione S-transferase (hereinafter referred to as GST), maltose-binding proteins, cellulose-binding domains, myc tags, FLAG tags, and the like.

Among these additional amino acid sequences, oligopeptides containing polyhistidine sequences or GST are preferred, and more preferably, oligopeptides containing polyhistidine sequences, because they offer high productivity when manufactured using E. coli and allow for easy detection of fucose-binding proteins using fluorescently labeled anti-polyhistidine antibodies or anti-GST antibodies.

There are no particular restrictions on the number of histidine repeats in the oligopeptide containing the polyhistidine sequence. However, if the histidine repeats are short, detection by anti-polyhistidine antibodies becomes difficult, and if they are long, the binding ability of the fucose-binding protein to the aforementioned sugar chain may be impaired. Therefore, the length of the histidine repeats in the oligopeptide containing the polyhistidine sequence is preferably such that the repeat consists of 5 to 15 histidine molecules, and more preferably such that it consists of 5 to 10 molecules.

There are no particular restrictions on the position where the oligopeptide containing the polyhistidine sequence is attached to the fucose-binding protein; it may be attached to both the N-terminus and the C-terminus, or to either the N-terminus or the C-terminus. However, for efficient detection using an anti-polyhistidine antibody, it is preferable that the oligopeptide containing the polyhistidine sequence is attached to the N-terminus of the fucose-binding protein.

Furthermore, the fucose-binding protein in the cell separation method of the present invention may have an additional amino acid sequence (hereinafter referred to as a carrier immobilization tag) consisting of an oligopeptide containing a cysteine residue or a lysine residue at its N-terminal and/or C-terminal end, which is useful for immobilizing the fucose-binding protein on a water-insoluble carrier. By immobilizing the fucose-binding protein on a carrier, for example, an undifferentiated cell adsorbent for removing undifferentiated cells such as human iPS cells described in Patent Document 3 can be produced. The length of the carrier immobilization tag is not particularly limited, as long as the fucose-binding protein has the ability to bind to a glycan containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc. As a tag for carrier immobilization, an oligopeptide consisting of 2 to 10 amino acid residues containing one or more cysteine residues is preferred because it can be immobilized on a water-insoluble carrier with high selectivity and efficiency. Specifically, examples include an oligopeptide consisting of 3 amino acid residues of "Gly-Gly-Cys", an oligopeptide consisting of 5 amino acid residues of "Ala-Ser-Gly-Gly-Cys", and an oligopeptide consisting of 7 amino acid residues of "Gly-Gly-Gly-Ser-Gly-Gly-Cys". There are no particular restrictions on the position where the oligopeptide containing one or more cysteine residues is attached to the fucose-binding protein. It may be attached to both the N-terminus and the C-terminus, or to either the N-terminus or the C-terminus. However, it is preferable that the oligopeptide containing one or more cysteine residues is attached to the C-terminus of the fucose-binding protein, as this allows for efficient immobilization of the fucose-binding protein to the carrier and, because it is far from the active site of the fucose-binding protein, is less likely to inhibit its binding activity.

In the cell isolation method of the present invention, a signal peptide may be added to the N-terminus of the fucose-binding protein to promote efficient expression in the host. Examples of such signal peptides when the host is E. coli include those that induce protein secretion into the periplasm, such as Pelb, DsbA, MalE, and TorT. The DNA encoding the fucose-binding protein in the cell isolation method of the present invention can be prepared by known methods. Examples of DNA preparation methods include converting the amino acid sequence of the fucose-binding protein in the cell isolation method of the present invention into a base sequence and artificially synthesizing DNA containing that base sequence; directly and artificially preparing the DNA encoding the fucose-binding protein in the cell isolation method of the present invention; or preparing it from genomic DNA of Burkholderia cenocepacia using DNA amplification methods such as PCR.

Furthermore, in the preparation method described above, when designing the base sequence, it is preferable to consider the frequency of codon use in the E. coli to be transformed. For example, in arginine (Arg), AGA, AGG, CGG, or CGA are less frequently used codons (rare codons); in isoleucine (Ile), ATA is less frequently used; in leucine (Leu), CTA is less frequently used; in glycine (Gly), GGA is less frequently used; and in proline (Pro), CCC is less frequently used. Therefore, it is preferable to select and convert codons other than these. Analysis of codon usage frequency can also be performed using public databases (for example, the Codon Usage Database on the Kazusa DNA Research Institute website, http://www.kazusa.or.jp/codon/, accessed May 7, 2020).

To transform Escherichia coli using the DNA encoding the fucose-binding protein prepared by the above method, transformation may be performed using the DNA itself. However, it is preferable to insert the DNA into an appropriate position in a vector based on a bacteriophage, cosmid, or plasmid, which is commonly used for the transformation of prokaryotic or eukaryotic cells, to create an expression vector, and then use that to perform transformation. This is preferable because it allows for stable transformation. Here, an appropriate position means a position that does not disrupt the replication function of the expression vector, the desired antibiotic marker, and the regions involved in transduction. When inserting the DNA into the vector, it is preferable to insert it into the vector in a state where it is linked to functional DNA such as a promoter necessary for expression. The vector used as the expression vector is not particularly limited as long as it can exist stably and replicate in the host, and examples include pET vectors, pUC vectors, pTrc vectors, pCDF vectors, pBBR vectors, etc. Furthermore, examples of the promoters include the trp promoter, tac promoter, trc promoter, lac promoter, T7 promoter, recA promoter, lpp promoter, and even the λPL promoter and λPR promoter of λ phage. Transformation of the host E. coli using the expression vector can be carried out using methods commonly used by those skilled in the art. For example, when selecting E. coli strain JM109, E. coli strain BL21 (DE3), E. coli NiCo21 (DE3), or E. coli strain W3110 as the host, methods described in known literature (e.g., Molecular Cloning, Cold Spring Harbor Laboratory, 256, 1992) can be used.

Next, the method for producing the fucose-binding protein in the present invention will be described. The fucose-binding protein in the present invention can be produced by a process comprising two steps: a step of producing the fucose-binding protein by culturing the transformant (hereinafter referred to as the first step), and a step of recovering the fucose-binding protein from the obtained culture (hereinafter referred to as the second step). In this specification, the culture includes not only the cells of the cultured transformant itself and cell secretions, but also the culture medium used for cultivation. In the first step, the transformant should be cultured in a medium suitable for its cultivation. For example, when Escherichia coli is used as the host, it is preferable to use Terrifi Blossom (TB) medium, Luria-Bertani (LB) medium, etc., supplemented with the necessary nutrients. If the expression vector contains a drug resistance gene, selective proliferation of the transformant can be achieved by adding a drug corresponding to that gene to the culture medium and carrying out the first step. For example, if the expression vector contains a kanamycin resistance gene, it is preferable to add kanamycin to the culture medium. The culture temperature can be any temperature generally known for the host organism being used. For example, if the host is E. coli, the temperature should be between 10°C and 40°C, preferably between 20°C and 37°C, and should be determined appropriately while considering factors such as the amount of fucose-binding protein to be produced.

Furthermore, the pH of the culture medium should be within the generally known pH range for the host organism being used. For example, if the host is E. coli, the pH should be in the range of 6.8 to 7.4, preferably around 7.0, and should be determined appropriately while considering the amount of fucose-binding protein to be produced. If an inducible promoter is introduced into the expression vector, the expression should be induced by adding an inducer to the culture medium under conditions that allow for good production of the fucose-binding protein. A preferred inducer is isopropyl-β-D-thiogalactopyranoside (IPTG) when using the tac promoter or lac promoter, with an addition concentration in the range of 0.005 mM to 1.0 mM, preferably 0.01 mM to 0.5 mM. Expression induction by adding IPTG should be carried out under conditions generally known for the host organism being used.

In the second step, the fucose-binding protein is recovered from the culture obtained in the first step using a generally known recovery method. For example, if the fucose-binding protein is secreted into the culture medium, the cells can be separated by centrifugation, and the fucose-binding protein can be recovered from the resulting culture supernatant. If it is expressed intracellularly (including the periplasm in prokaryotes), the cells can be collected by centrifugation, then lysed by adding an enzyme treatment agent or surfactant, and the fucose-binding protein can be recovered from the cell lysate. Furthermore, if it is desired to improve the purity of the fucose-binding protein, methods known in the relevant art can be used; one example is separation and purification using liquid chromatography.

For liquid chromatography, ion exchange chromatography, hydrophobic interaction chromatography, gel filtration chromatography, affinity chromatography, etc., are preferred, and a combination of these chromatography methods is even more preferable. Furthermore, the purity and molecular weight of the fucose-binding protein purified by the above chromatography methods can be determined using methods known in the art. Examples include SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and gel filtration chromatography.

The binding affinity of the fucose-binding protein to the glycan in this invention can be evaluated by methods such as the Enzyme-linked immunosorbent assembly method or surface plasmon resonance (SPL) spectroscopy. As an example, SPL will be described.

Surface plasmon resonance (SPL) spectroscopy can be used to evaluate binding affinity, for example, with the Biacore T200 instrument (GE Healthcare), where the analyte is a fucose-binding protein and the solid phase is a glycan (a glycan containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc). Sensor chips with immobilized glycans can be prepared using biotin-labeled glycans, either by using a streptavidin-coated sensor chip (Sensor Chip SA, GE Healthcare) or a dextran-coated sensor chip (Sensor Chip CM5, GE Healthcare) with streptavidin pre-immobilized. Furthermore, binding affinity evaluation can be performed using the kinetics analysis program included with the instrument.

Next, the adsorbent in the cell separation method of the present invention will be described. There are no particular restrictions on the raw materials of the water-insoluble carrier used as the adsorbent in the cell separation method of the present invention. Examples include inorganic carriers such as silica gel and glass coated with a gold thin film, water-insoluble polysaccharide carriers made from polysaccharides such as agarose, cellulose, chitin, and chitosan, and crosslinked polysaccharide carriers obtained by crosslinking them with a crosslinking agent, crosslinked polysaccharide carriers obtained by crosslinking water-soluble polysaccharides such as dextran, pullulan, starch, alginate, and carrageenan with a crosslinking agent, synthetic polymer carriers such as poly(meth)acrylate, polyvinyl alcohol, polyurethane, and polystyrene, and crosslinked synthetic polymer carriers obtained by crosslinking them with a crosslinking agent. Among these carriers, uncharged polysaccharide carriers such as agarose, cellulose, dextran, and pullulan, and cross-linked polysaccharide carriers obtained by cross-linking them with a cross-linking agent, are preferred because they have hydroxyl groups and can be easily modified with hydrophilic polymers as described later. Hydrophilic synthetic polymer carriers such as poly(meth)acrylate and polyurethane, and cross-linked hydrophilic synthetic polymer carriers obtained by cross-linking them with a cross-linking agent are also preferred. Furthermore, for water-insoluble carriers used as adsorbents, it is preferable that the surface of the water-insoluble carrier is modified with a hydrophilic polymer, and more preferably that the hydrophilic polymer is covalently fixed to the water-insoluble carrier, in order to suppress non-specific adsorption of cells.

Examples of hydrophilic polymers used to modify the surface of water-insoluble carriers include neutral polysaccharides such as agarose, cellulose, dextran, pullulan, and starch, as well as synthetic polymers having hydroxyl groups such as poly(2-hydroxyethyl (meth)acrylate) and polyvinyl alcohol. Among these hydrophilic polymers, neutral polysaccharides such as dextran, pullulan, and starch are preferred due to their high hydrophilicity and ease of covalent fixation to the surface of the insoluble carrier, with dextran and pullulan being more preferred.

There are no particular restrictions on the molecular weight of dextran and pullulan, but those with a number-average molecular weight of 10,000 to 1,000,000 are preferred in that they allow for sufficient hydrophilic modification of the insoluble carrier surface. There are no particular restrictions on the shape of the water-insoluble carrier used as the adsorbent; it may be particulate, spongy, flat film-like, plate-like, hollow, or fibrous. However, a particulate carrier is preferred in that it allows for efficient cell adsorption to the adsorbent, and a perfectly spherical particulate carrier is more preferred.

The average particle size (median diameter) of the water-insoluble carrier used as an adsorbent in the cell separation method of the present invention, when swollen in water, is preferably 100 μm to 1000 μm, more preferably 100 μm to 500 μm, and even more preferably 150 μm to 300 μm, such that when the adsorbent produced from the carrier is packed into a column, the cells to be separated make sufficient contact with the adsorbent surface, and cells that do not bind to the adsorbent can pass smoothly through the gaps between the adsorbents. If the particle size is less than 100 μm, cells that do not bind to the adsorbent have difficulty passing through the gaps between the adsorbents, reducing the cell recovery rate. If the particle size is greater than 1000 μm, contact between cells that bind to the adsorbent and the adsorbent surface becomes insufficient, reducing the separation efficiency between cells that bind to the adsorbent and those that do not. The particle size of the insoluble carrier can be measured using, for example, a precision particle size distribution analyzer (product name "Multisizer 3") manufactured by Beckman Coulter, Inc.

Alternatively, the particle size can be determined by taking an image of a graduated slide glass using an optical microscope, then taking images of multiple particles of the target at the same magnification, measuring the particle size of the multiple carriers photographed using a ruler, and calculating the average value. There are no particular restrictions on the presence or absence of pores in the water-insoluble carrier used as the adsorbent; it may be either porous or non-porous. Furthermore, the water-insoluble carrier used in the adsorbent of the present invention is preferably a particulate carrier having hydroxyl groups, as this facilitates the introduction of active functional groups for immobilizing the fucose-binding protein used in the adsorbent of the present invention onto the carrier. In addition, commercially available water-insoluble carriers may be used in the adsorbent of the present invention; for example, Toyopearl (manufactured by Tosoh), made from poly(meth)acrylate, Sepharose (manufactured by GE Healthcare), made from agarose, or Cellfia (manufactured by Asahi Kasei), made from cellulose, can be used.

The adsorbent in the cell separation method of the present invention can be manufactured by a process comprising two steps: a step of producing a reactive water-insoluble carrier from a water-insoluble carrier (hereinafter referred to as step X), and a step of immobilizing the reactive water-insoluble carrier by reacting it with a fucose-binding protein (hereinafter referred to as step Y). The details of steps X and Y are described below.

Step X is a step in producing a reactive water-insoluble carrier by introducing a reactive functional group for immobilizing a fucose-binding protein onto a water-insoluble carrier. The reactive functional group used to immobilize the fucose-binding protein of the present invention onto the water-insoluble carrier is not particularly limited as long as it is a general functional group for protein immobilization, and examples include epoxy groups, formyl groups, carboxyl groups, active ester groups, amino groups, maleimide groups, haloacetyl groups, etc. The method for introducing the functional group into the water-insoluble carrier is not particularly limited as long as it is a general method for introducing functional groups, and can be carried out, for example, according to the method described in Patent Document 3.

Step Y is a step in which the fucose-binding protein of the present invention is immobilized on the reactive water-insoluble carrier produced in Step X. The method for immobilizing the fucose-binding protein on the reactive water-insoluble carrier obtained in Step X is not particularly limited as long as it is a general method for immobilizing proteins; for example, it may be carried out according to the method described in Patent Document 3.

The amount of fucose-binding protein immobilized on the water-insoluble carrier can be appropriately determined considering the binding affinity between the cells to be separated and the fucose-binding protein in the cell separation method of the present invention. Preferably, the amount is 0.01 mg to 50 mg per 1 mL of water-insoluble carrier, and more preferably 0.05 mg to 30 mg.

Furthermore, the amount of fucose-binding protein immobilized on the water-insoluble carrier can be adjusted by controlling the amount of protein used during the immobilization reaction and the amount of active functional groups introduced into the water-insoluble carrier. The amount of fucose-binding protein immobilized on the water-insoluble carrier can be calculated by recovering the immobilization reaction solution and the subsequent washing solution to determine the amount of unreacted fucose-binding protein, and then subtracting the amount of unreacted fucose-binding protein of the present invention from the amount of fucose-binding protein used in the immobilization reaction.

Furthermore, as mentioned above, since the water-insoluble carrier used in the adsorbent of the present invention preferably has a hydrophilic polymer covalently immobilized on it in order to suppress non-specific adsorption of cells, when manufacturing the adsorbent, the hydrophilic polymer can be covalently immobilized on the water-insoluble carrier before introducing the functional groups for immobilizing the fucose-binding protein of the present invention in step X. The method for covalently immobilizing the hydrophilic polymer on the water-insoluble carrier is not particularly limited as long as it is a general covalent bond formation reaction. For example, one method involves introducing epoxy groups to the water-insoluble carrier by reacting the hydroxyl groups on the surface of the water-insoluble carrier with an epoxy group-containing compound such as epichlorohydrin, ethylene glycol diglycidyl ether, or 1,4-butanediol diglycidyl ether under basic conditions, and then reacting the epoxy groups with the hydroxyl groups of the hydrophilic polymer under basic conditions.

The present invention provides an effective cell separation method for separating and purifying cells containing glycans comprising the structure Fucα1-2Galβ1-3GlcNA and/or Fucα1-2Galβ1-3GalNAc, based on differences in glycan expression levels. This is achieved by using multiple adsorbents with different amounts of fucose-binding protein immobilized and sequentially contacting cell populations with varying degrees of undifferentiation. Because this method allows for the separation and recovery of cells based on differences in the expression levels of the undifferentiation marker glycan, it is particularly useful for separating cells with low degrees of undifferentiation according to their marker expression levels, which was difficult with conventional methods. Therefore, it can efficiently remove and separate cells with particularly low degrees of undifferentiation, i.e., undifferentiated cells, from undifferentiated cells such as human iPS cells. This method is extremely effective for preparing high-quality, highly undifferentiated cells for regenerative medicine and for functional analysis of undifferentiated cells.

Furthermore, compared to conventional cell separation methods using magnetic beads, cell sorting using cell sorters, and cell separation methods using cell rolling columns, the cell separation method of the present invention can process large quantities of cells easily and is an extremely effective means for the large-scale purification of cell populations with varying degrees of undifferentiation.

This graph shows the 2102Ep cell recovery rate for each cell fraction in Example 1, Comparative Example 1, and Reference Example 1. This graph shows the BC2LCN positivity rate of 2102Ep cells in each cell fraction in Example 1, Comparative Example 1, and Reference Example 1.

The present invention will be described in further detail below with reference to examples of preparation, embodiments, comparative examples, and reference examples, but the present invention is not limited to these.

Preparation Example 1 Preparation of Adsorbents 156GGC-A, B, C, and D Following the methods described in Examples 1 to 3 of Japanese Patent Application Publication No. 2019-000063, four types of adsorbents (adsorbents 156GGC-A, B, C, and D) with different amounts of immobilized fucose-binding protein 156GGC (hereinafter referred to as BC2LCN) were prepared by changing the reaction amount of immobilized BC2LCN during adsorbent preparation. The immobilized amounts of BC2LCN for each adsorbent were 146 mg/L for adsorbent 156GGC-A, 307 mg/L for adsorbent 156GGC-B, 412 mg/L for adsorbent 156GGC-C, and 682 mg/L for adsorbent 156GGC-D (Table 1).

Example 1: Separation of cell populations with different degrees of undifferentiation from 2012Ep cells using a column stacked with adsorbents 156GGC-A, B, C, and D. Example 1 describes a method for separating cell populations with different degrees of undifferentiation from 2102Ep cells (Embryonal Carcinoma Cells Cl. 4/D3 cells (obtained from CosmoBio)), which are human embryonic cancer cells having a sugar chain containing a structure consisting of "Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc," using a column stacked with adsorbents with different amounts of immobilized fucose-binding protein.
(1) Preparation of a column packed with adsorbent One column was prepared by attaching a 40 μM hydrophilic nylon mesh filter (Corning) between a 2.5 mL syringe (Terumo) and an injection needle (Terumo, 22 G). Next, the adsorbent 156GGC-D prepared in Preparation Example 1 was replaced with MACS buffer (Milteny Biotech), and a 50% suspension of the adsorbent was prepared by adjusting the amount of adsorbent sedimentation after standing for 12 hours or more so that the sedimentation volume of the adsorbent was 50%. 1.0 mL of this suspension was added to the prepared column, and the adsorbent 156GGC-D was packed into the column (adsorbent volume: 0.5 mL). Next, using the same procedure, the adsorbent 156GGC-C prepared in Preparation Example 1 was packed on top of the adsorbent 156GGC-D with an adsorbent volume of 0.5 mL. Next, following the same procedure, adsorbent 156GGC-B, prepared in Example 1, was packed on top of adsorbent 156GGC-C with an adsorbent volume of 0.5 mL. Then, following the same procedure, adsorbent 156GGC-A, prepared in Example 1, was packed on top of adsorbent 156GGC-B with an adsorbent volume of 0.5 mL. In this way, a column was prepared in which adsorbents 156GGC-A, 156GGC-B, 156GGC-C, and 156GGC-D were stacked in this order from the top of the column, each with a packing volume of 0.5 mL. Hereafter, this prepared column will be referred to as column X.
(2) Culture of 2102Ep cells and preparation of cell suspension 2102Ep cells were cultured in D-MEM medium (High Glucose, Fujifilm Wako Pure Chemical Industries) supplemented with 10% FBS (Biological Industries) and an antibiotic solution (penicillin-streptomycin solution, Fujifilm Wako Pure Chemical Industries), seeded in 6 cm diameter adhesive culture dishes (Corning) or 10 cm diameter adhesive culture dishes (Corning), and cultured at 37°C under a 5% CO2 atmosphere. Next, the 2102Ep cells were fluorescently stained using Cell Tracker Orange (Thermo Fisher Scientific) in the following manner. After discarding the culture medium in the petri dish containing the 2102Ep cells, the cells were washed with serum-free RPMI 1640 medium (manufactured by Fujifilm Wako Pure Chemical Industries, Ltd.), and this medium was then discarded. Next, a solution of Cell Tracker Orange dissolved in serum-free RPMI 1640 medium to a final concentration of 10 μM was added, and the cells were cultured at 37°C for 1 hour under a 5% CO2 atmosphere. After discarding the fluorescent reagent solution, the D-MEM medium containing 10% FBS and antibiotic solution was added, and the cells were cultured at 37°C for 1 hour under a 5% CO2 atmosphere. Next, the D-MEM medium containing 10% FBS and antibiotic solution was discarded, and a fresh D-MEM medium containing 10% FBS and antibiotic solution was added again, and the cells were cultured overnight at 37°C under a 5% CO2 atmosphere.

Next, cell collection and cell suspension preparation were carried out using the following method. The D-MEM medium containing 10% FBS and antibiotic solution in the petri dish during cell culture was discarded, and D-PBS(-) (manufactured by Cell Science Institute) was added. After washing the cells, the D-PBS(-) was discarded. Next, Accutase (manufactured by Innovative Cell Technologies) was added and left for several minutes to detach the 2102Ep cells, which were then collected in a 50 mL tube. After centrifuging the cells to settle them, the cells were suspended in the aforementioned MACS buffer, and the cells were washed by centrifugation again and discarding the supernatant. After repeating the cell washing procedure twice, the cells were suspended in MACS buffer and filtered using a cell strainer (Corning, 40 μm mesh) to prepare a cell suspension of 2102Ep cells stained with Cell Tracker Orange. A portion of the obtained 2102Ep cell saturation was taken, diluted 10-fold, and the cell density was calculated using a hemocytometer. Based on this cell density, the number of cells added to the column was calculated from the value of the volume of cell saturation applied to the column multiplied by the cell density.
(3) Separation of populations with different BC2LCN positivity rates from 2012Ep cells using a column packed with adsorbent With the column X prepared in (1) above in an upright position, the 2012Ep cell suspension prepared in (2) above was added to the top of the column in a mixed cell volume of 0.1 mL so that the number of cells added per column was 7.8 × ^10⁶ . Next, 1.0 mL of MACS buffer was gently added from the top of the column, and a total of 1.1 mL of effluent from the needle was collected in a separate container (hereinafter referred to as cell solution Fr. 1). Next, 1.0 mL of MACS buffer was gently added from the top of the column, and a total of 1.0 mL of effluent from the needle was collected in a separate container (hereinafter referred to as cell solution Fr. 2). Next, 1.0 mL of MACS buffer was gently added from the top of the column, and a total of 1.0 mL of effluent from the needle was collected in a separate container (hereinafter referred to as cell solution Fr. 3). Next, 1.0 mL of MACS buffer was gently added from the top of the column, and a total of 1.0 mL of effluent from the needle was collected in a separate container (hereinafter referred to as cell solution Fr. 4). Next, 1.0 mL of MACS buffer containing 0.2 M fucose and 10 mM EDTA (hereinafter referred to as cell detachment solution) was gently added from the top of the column, and a total of 1.0 mL of effluent from the needle was collected in a separate container (hereinafter referred to as cell solution Fr. 5). Next, 1.0 mL of cell detachment solution was gently added from the top of the column, and a total of 1.0 mL of effluent from the needle was collected in a separate container (hereinafter referred to as cell solution Fr. 6). Next, 1.0 mL of cell detachment solution was gently added from the top of the column, and a total of 1.0 mL of effluent from the needle was collected in a separate container (hereinafter referred to as cell solution Fr. 7). Next, 1.0 mL of cell detachment solution was gently added from the top of the column, and a total of 1.0 mL of effluent from the needle was collected in a separate container (hereinafter referred to as cell solution Fr. 8).
(4) Measurement of cell recovery rate and BC2LCN positivity rate of 2012Ep cells in each cell saturation Fr. For cell saturations Fr. 1 to 8 obtained by the above procedure, 0.5 mL of the cell saturation was diluted with 2 mL of MACS buffer, and then dispensed into 5 mL polystyrene round tubes (Corning) with cell strainer caps. 50 μL of CountBright Absolute Counting Beads (Invitrogen) were added as internal standard beads for cell counting, and 50 μL of 7-AAD was added as a cell viability determination reagent. Then, the cell count was measured using a cell sorter BD FACSAria (Becton Dickinson). The number of cells contained in each cell saturation was calculated by proportional calculation based on the number of internal standard beads obtained from the dot plot. The cell recovery rate for each cell saturation was calculated for each cell saturation. The number of cells contained in each was calculated by dividing each by the number of added cells (7.8 × ^10⁶ cells) described in (3) above.

Furthermore, for cell saturation Fr. 1 to 8, the BC2LCN positivity rate of cells contained in each cell saturation Fr. was measured as follows: 0.5 mL of the cell saturation from each Fr. was transferred to a 1.5 mL Eppendorf tube, centrifuged, and the supernatant was discarded. The cells were then washed by resuspending them in 1.0 mL of MACS buffer. After centrifugation, the supernatant was discarded, and the resulting cell pellet was suspended in 1 mL of MACS buffer containing 0.5% (vol./vol.) BC2LCN-FITC (manufactured by Fujifilm Wako Pure Chemical Industries, Ltd.) and allowed to stand in the dark at room temperature for 30 minutes. After 30 minutes, the cell washing procedure with MACS buffer described above was performed in the Eppendorf tube, and the resulting cell pellet was suspended in 2 mL of MACS buffer. After transferring the cell suspension to a 5 mL polystyrene round tube (Corning), 50 μL of CountBright Absolute Counting Beads (Invitrogen) were added as internal standard beads for cell counting, and 50 μL of 7-AAD was added as a cell viability test reagent. The fluorescence intensity of the cell population was then analyzed using a cell sorter BD FACSAria (Becton Dickinson). The BC2LCN positivity rate of the cells was calculated as follows: BC2LCN positivity rate = (number of positive viable cells in each cell saturation frame) ÷ (total number of viable cells in each cell saturation frame). 2102Ep cells that were not stained with BC2LCN-FITC were considered the negative cell population, and cell populations whose FITC fluorescence intensity increased upon reaction with the fluorescent reagent were considered the positive cell population. In other words, the BC2LCN positivity rate indicates the proportion of cells possessing a glycan structure that exhibits BC2LCN reactivity, which is one of the undifferentiated cell markers.

FACS analysis revealed that the recovery rates of 2102p cells in each cell saturation Fr. were as follows: Fr. 1 = 1.2%, Fr. 2 = 2.4%, Fr. 3 = 4.3%, Fr. 4 = 5.8%, Fr. 5 = 33.3%, Fr. 6 = 20.5%, Fr. 7 = 5.6%, and Fr. 8 = 7.1%. Furthermore, the BC2LCN positivity rates of 2102Ep cells in each cell saturation Fr. were as follows: Fr. 1 = 31.9%, Fr. 2 = 38.9%, Fr. 3 = 58.6%, Fr. 4 = 60.0%, Fr. 5 = 82.3%, Fr. 6 = 81.0%, Fr. 7 = 80.1%, and Fr. 8 = 76.7%. Table 2 and Figure 1 show the recovery rates of 2102p cells in each cell saturation Fr. of Example 1, while Table 3 and Figure 2 show the BC2LCN positivity rates of cells contained in each cell saturation Fr.

From these results, in a column (Column X) where four types of adsorbents with different amounts of fucose-binding protein immobilized were stacked in the following order from the top of the column: adsorbent 156GGC-A (146 mg BC2LCN immobilized/mL-adsorbent), adsorbent 156GGC-B (307 mg BC2LCN immobilized/mL-adsorbent), adsorbent 156GGC-C (412 mg BC2LCN immobilized/mL-adsorbent), and adsorbent 156GGC-D (682 mg BC2LCN immobilized/mL-adsorbent), the degree of undifferentiation (BC2LCN positivity rate) of efflux cells Fr. obtained by passing MACS buffer through the column was: Fr. 1 = 31.9%, Fr. 2 = 38.9%, Fr. 3 = 58.6%, Fr. The result was 4 = 60.0%, indicating that 2102Ep cell populations with different degrees of undifferentiation can be separated into cell Fr. groups, from the lower-undifferentiated cell population to the higher-undifferentiated cell population.

This result can be interpreted as follows: Because the adsorbent packed from the top to the bottom of the column has a concentration gradient from low to high fucose-binding protein immobilization, highly undifferentiated cell populations selectively bind to the packing material at the top of the column, while less undifferentiated cell populations bind to the packing material at the bottom. This results in a distribution of cells based on their degree of undifferentiation. Therefore, when cells are leached out of the column using MACS buffer, less undifferentiated cells are initially leached out, and with each subsequent cell leaching, more undifferentiated cells are gradually leached out. Thus, it was confirmed that by stacking adsorbents with different amounts of fucose-binding protein immobilization on the column and gradually exposing cell populations with different degrees of undifferentiation to adsorbents with higher fucose-binding protein immobilization, cells with different degrees of undifferentiation can be separated as cell leaching.

Comparative Example 1: Separation of cell populations with different degrees of undifferentiation from 2012Ep cells using a column packed with adsorbent 156GGC-B. A 50% suspension of adsorbent was prepared by replacing adsorbent 156GGC-B (307 mg/L - adsorbent) with MACS buffer (Milteny Biotech), and then adjusting the suspension so that the sedimentation volume of the adsorbent after standing for 12 hours or more was 50%. 4.0 mL of this suspension was added to the prepared column, and the column was packed with adsorbent 156GGC-B (adsorbent volume: 2.0 mL). This prepared column will be referred to as column Y below. Except for the packing of the column with adsorbent, the preparation of 2102Ep cells, passing the solution through the column, and FACS analysis of the obtained cells Fr. were performed in the same manner as in Example 1.

FACS analysis revealed that the recovery rates of 2102p cells in each cell Fr. were as follows: Fr. 1 = 13.4%, Fr. 2 = 14.3%, Fr. 3 = 3.2%, Fr. 4 = 3.3%, Fr. 5 = 10.3%, Fr. 6 = 8.1%, Fr. 7 = 2.3%, and Fr. 8 = 1.2%. Furthermore, the BC2LCN positivity rates of 2102Ep cells in each cell Fr. were as follows: Fr. 1 = 71.2%, Fr. 2 = 74.5%, Fr. 3 = 80.8%, Fr. 4 = 85.6%, Fr. 5 = 88.5%, Fr. 6 = 83.5%, Fr. 7 = 80.2%, and Fr. 8 = 87.3%. Table 2 and Figure 1 show the recovery rates of 2102p cells in each cell saturation Fr. of Comparative Example 1, while Table 3 and Figure 2 show the BC2LCN positivity rates of cells contained in each cell saturation Fr.

From these results, it was found that in a column (Column Y) packed with only one adsorbent, 156GGC-B (307 mg/L - adsorbent), the degree of undifferentiation (BC2LCN positivity rate) of eluted cells Fr. obtained by passing MACS buffer through the column was Fr. 1 = 71.2%, Fr. 2 = 74.5%, Fr. 3 = 80.8%, and Fr. 4 = 85.6%. Although the rate gradually increased, the difference in BC2LCN positivity rate between Fr. was smaller than in Example 1, and it was not possible to individually separate cell populations with a low degree of undifferentiation, such as those with a BC2LCN positivity rate of approximately 30% to 60%, as in Example 1. Therefore, it was confirmed that when a column is packed with an adsorbent containing only one adsorbent and cell populations with different degrees of undifferentiation are brought into contact, cells with significantly different degrees of undifferentiation cannot be separated as cell Fr.

Reference Example 1: Separation of cell populations with different degrees of undifferentiation from 2012Ep cells using a column packed with Toyopal HW-40EC. A 50% suspension of adsorbent was prepared by replacing BC2LCN-unimmobilized Toyopal HW-40EC (manufactured by Tosoh) with MACS buffer (manufactured by Milteny Biotech), and then adjusting the adsorbent's sedimentation volume to 50% after standing for 12 hours or more. 4.0 mL of this suspension was added to the prepared column, and the column was packed with Toyopal HW-40EC adsorbent (adsorbent volume: 2.0 mL). Hereafter, this prepared column will be referred to as column Z. Except for the packing of the column with adsorbent, the preparation of 2102Ep cells, the passage of the column, and FACS analysis of the obtained cells Fr. were performed in the same manner as in Example 1.

FACS analysis revealed that the recovery rates of 2102p cells in each cell fluid Fr. were as follows: Fr. 1 = 22.5%, Fr. 2 = 33.0%, Fr. 3 = 3.6%, Fr. 4 = 0.8%, Fr. 5 = 1.7%, Fr. 6 = 2.9%, Fr. 7 = 0.9%, and Fr. 8 = 1.9%. Furthermore, the BC2LCN positivity rates of 2102Ep cells in each cell fluid Fr. were as follows: Fr. 1 = 79.5%, Fr. 2 = 74.0%, Fr. 3 = 84.0%, and Fr. 4 = 82.7% (Fr. 5-8 were not measured). Table 2 and Figure 1 show the recovery rates of 2102p cells in each cell fluid Fr. of Reference Example 1, while Table 3 and Figure 2 show the recovery rates for each cell fluid Fr. This shows the BC2LCN positivity rate of cells contained within.

From these results, it was found that in a column (Column Z) packed with the adsorbent Toyopal HW-40EC, which does not have fucose-binding proteins immobilized, the degree of undifferentiation (BC2LCN positivity rate) of eluted cells Fr. obtained by passing MACS buffer through the column was Fr. 1 = 79.5%, Fr. 2 = 74.0%, Fr. 3 = 84.0%, and Fr. 4 = 82.7%. Since 2102Ep cells did not adsorb to the adsorbent and passed through, it was not possible to individually separate cell populations with different degrees of undifferentiation. Therefore, it was confirmed that when an adsorbent without fucose-binding proteins is packed into a column and cell populations with different degrees of undifferentiation are brought into contact, cells with different degrees of undifferentiation cannot be separated as cell Fr.

Claims (4)

A method comprising the steps of: binding a population of undifferentiated cells of different degrees, on the surface of the cells, a glycan containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc, to multiple adsorbents with different amounts of immobilized fucose-binding protein, stepwise from an adsorbent with a low amount of immobilized fucose-binding protein to an adsorbent with a high amount of immobilized fucose-binding protein; and obtaining cells that did not bind to the adsorbent, wherein the fucose-binding protein is (e) below.
A method for separating cells with different degrees of undifferentiation.
(e) A fucose-binding protein comprising the amino acid sequence shown in Sequence ID No. 6, wherein the amino acid sequence from the first proline residue to the 155th amino acid residue of the amino acid sequence shown in Sequence ID No. 1 is further modified by adding an oligopeptide containing a polyhistidine sequence to the N-terminus and an oligopeptide containing cysteine to the C-terminus. The method for separating cells of different degrees of undifferentiation according to claim 1, characterized in that the amount of fucose-binding protein immobilized ranges from 50 mg/L-adsorbent to 1000 mg/L-adsorbent, and a plurality of adsorbents are used, each having an immobilization amount of fucose-binding protein that differs by 100 mg/L-adsorbent or more. A method for separating cells with different expression levels of glycans containing a structure consisting of Fucα1-2Galβ1-3GlcNAc and/or Fucα1-2Galβ1-3GalNAc, using the method according to either claim 1 or 2. The method according to any one of claims 1 to 3, characterized by using a column packed with multiple adsorbents having different amounts of the fucose-binding protein immobilized on it.