WO2006064983A1

WO2006064983A1 - Monoclonal antibody specific human embryonic stem cell

Info

Publication number: WO2006064983A1
Application number: PCT/KR2004/003285
Authority: WO
Inventors: Hyo-Jeong Hong; Chun-Jeih Ryu; Yeon-Sung Son; Jae-Hyun Park; Young-Kook Kang; Jin-Sung Park; Hong-Seo Choi; Hyun-Soo Yoon; Sung-Il Roh; Jeoung-Eun Lee; Jung-Bok Lee
Original assignee: Korea Research Institute Of Bioscience And Biotechnology; Mizmedi Hospital
Priority date: 2004-12-14
Filing date: 2004-12-14
Publication date: 2006-06-22

Abstract

Disclosed are a monoclonal antibody specifically binding to human embryonic stem cells and a hybridoma producing the monoclonal antibody. In detail, disclosed are a monoclonal antibody to HSPA8 (heat shock 70 kDa protein 8 isoform 1), which binds to human embryonic stem cells but does not bind to mouse embryonic stem cells, a hybridoma producing the monoclonal antibody, an assay kit comprising the monoclonal antibody, and a composition for removing human embryonic stem cells comprising the monoclonal antibody.

Description

MONOCLONAL ANTIBODY SPECIFIC TO HUMAN EMBRYONIC STEM CELL

Technical Field

The present invention relates to a monoclonal antibody that specifically bind to human embryonic stem cells, and a hybridoma producing the same.

Background Art

Stem cells have the ability to differentiate into a variety of cell types and they collectively mean undifferentiated cells prior to the differentiation that can be obtained from embryo, fetus and adult body tissues. Upon appropriate stimulus or in environment, stem cells are capable to differentiate into certain specific cell types. Unlike differentiated cells where cell division is stopped, stem cells can multiply by division, the process being called self-renewal or proliferation. Stem cells also exhibit plasticity and they are capable of differentiating into different cell types upon exposure to different environments or stimuli.

Stem cells are typically divided into two groups, pluripotent embryonic stem cells and multipotent adult stem cells. Embryonic ^stem cells are derived from the inner cell mass which will form the future fetus within the blastocyst during early embryogenesis, and have the potential to differentiate into every cell type of every tissue. Unlike adult stem cells embryonic stem cells are undifferentiated cells that are capable of proliferating indefinitely, differentiating into all cell types, and giving rise to germ cells, which are inherited by the next generation. When the fetus development progresses and reaches a stage at which each organ of the fetus is formed, tissue-specific stem cells (primary stem cells) are present in each organ and participate in the differentiation forming each organ. The tissue-specific stem cells are described as multipotent or unipotent because they usually have a restricted capacity for differentiation into cells specific to the tissue in which they are found. In adults, the tissue- specific stem cells remain in most tissues and replenish normally or pathologically lost cells.

Research into embryonic stem cells started in 1981 when a method of cultivating mouse embryonic stem cells was first established (Evans et al., Nature, 292: 151-156, 1981) . In 1996, a method of cultivating pluripotent embryonic stem cells from primates was developed (Thomson et al., Biol. Reprod., 55: 254-259, 1996). In 1998, Thomson et al. established a method of cultivating human embryonic stem cells (Thomson et al., Science, 282: 1145-1147, 1998). Since pluripotent embryonic stem cells have a special quality (the ability to develop into all cell types) , they may be used to replace specific cells or organs damaged by diseases or accidents after being induced to differentiate into the specific cells or cells specific to the organs. Thus, embryonic stem cells are receiving increasing attention as an effective treatment of various incurable diseases.

Human embryonic stem (ES) cells share common features with mouse embryonic stem (ES) cells, most of which have been well identified, but at the same time they are known to have many distinct differences from mouse ES cells. First, representative similarities are as follows. Since human ES cells are pluripotent like mouse ES cells, they differentiate into various cell types of all tissues when induced to form embryoid bodies (EBs) in vitro (Thomson et al., Science, 282: 1145-1147, 1998; Reubinoff, et al., Nat. Biotech. ,18:399-404; Park, et al., Biol. Reprod. 69:2007-2014, 2003). Human and mouse embryonic stem cells can both be cultivated in the presence of feeder layers or conditioned media from feeder layers. Also, human ES cells, like mouse ES cells, express Oct-4, which is known to participate in early-stage differentiation during embryonic development, show telomerase activity, which is expressed in cells having self- proliferative ability, and express alkaline phosphatase, which is expressed in high levels in mouse ES cells. However, these human ES cells also exhibit marked differences from their murine counterparts, as follows. Mouse ES cells can be cultivated using a single cell, but it is impossible to cultivate human embryonic stem cells from a single cell because they are mostly dead when a single cell is used.

Human and mouse ES cells are also different from each other in morphology and cytokine requirement for maintaining self- reproduction or pluripotency during cultivation. Microarray analysis revealed that human ES cells have a gene pool for maintaining sternness, which is different from that of mouse

ES cells (Bhattacharya, et al., Blood, 103:2956-2964, 2004).

Although to date there are few similarities between human and mouse ES cells, monoclonal antibodies prepared by injecting mouse embryonic or human embryonal carcinoma cells other than human embryonic stem cells into mice, for example, antibodies to cell-surface antigens TRA-1-60, TRA-1-81, SSEAl, SSEA3 and SSEA4, have been nonetheless used for identifying and defining human embryonic stem cells of undifferentiated states during cultivation. These antibodies recognize molecules that mostly possess carbohydrate epitopes and whose functions are unknown (Badcock, et al., Cancer Res. 59:4715-4719, 1999; Kannagi et al., EMBO. J. 2:2355-2361, 1983) . A protein marker, CD9, is known to be expressed on the surface of human ES cells, but is also expressed in mouse ES cells (Oka, et al., MoI. Biol. Cell 13:1274-1281, 2002; Carpenter, et al., Dev. Dyn. 229:243-258, 2004) . Thus, there is an urgent need to find many markers specific only to human ES cells for studies of undifferentiated human ES cells. In fact, it is expected that the direct injection of human ES cells cultured in undifferentiated states results in the finding of greater numbers of various cell surface molecules specific to human ES cells.

At present, key issues in stem cell-based therapies include isolation of stem cells from humans and animals, establishment of stem cell cultivation techniques, induction of differentiation into specific functional cells under in vitro conditions and isolation of differentiated cells, insurance of efficacy and safety in animals before clinical trials, and techniques of suppressing the immune rejection of stem cell transplants in the body. The most important technique involves inducing differentiation into various specialized cells. Until today, studies that involved in mouse ES cells resulted in the establishment of differentiation methods into hematopoietic cells (Wiles et al., Development, 111: 259-267, 1991), cardiomyocytes (Klug et al., J. Clin. Invest., 98: 216-224, 1996), insulin- secreting cells (Soria et al., Diabetes, 49: 157-162, 2000), and neurons and glia (Bain et al., Dev. Biol., 168: 342-357, 1995; Okabe et al., Mech. Dev., 59: 89-102, 1996; Mujtaba et al., Dev. Biol., 214: 113-127, 1999; Brustle et al., Science, 285: 754-756, 1999; Brustle et al., Proc. Natl. Acad. Sci. USA, 94: 14809-14814, 1997). Many efforts have also been made to induce the differentiation of human ES cells into specialized functional cells and establish differentiation methods for human ES cells using previously established mouse ES cell cultivation methods. Human ES cells have been reported to have the potential to differentiate into various cell types, including trophoblasts (Xu et al., Nat. Biotechnol., 20: 1261-1264, 2002), cardiomyocytes (Kehat, et al., Circ. Res 91:659-661, 2002; Mummery, et al., Circulation 107:2733-2740, 2003), neural progenitors (Zhang et al., Nat. Biotechnol., 19:1129-1133; Reubinoff et al., Nat. Biotechnol., 19:1134-1140, 2001), endothelial cells (Levenberg, et al., PNAS. 99:4391-4396, 2002), and hematopoietic cells (Chadwick et al., Blood, 102: 906-915, 2003) . Many more methods of inducing differentiation of human ES cells into other various cell types are expected to develop.

As described above, when various specialized cells into which human ES cells are induced to differentiate are used for cell therapies, it is important to isolate the specialized cells with high purities and ensure efficacy and safety in animals or humans. Human ES cells themselves can be applied to cell therapies of various degenerative diseases, but this application results in the formation of tumors in mice (Thomson et al., Science, 282: 1145-1147, 1998; Reubinoff, et al., Nat. Biotech., 18:399-404; Park, et al., Biol. Reprod. 69:2007-2014, 2003) . Thus, specialized cells differentiated from human ES cells must be used for cell therapies after human ES cells have been completely removed. Current available antibodies used for identifying undifferentiated human ES cells are problematic in terms of not guaranteeing accurate analysis of features of undifferentiated human ES cells and their complete removal. Therefore, further development of antibodies specifically recognizing human ES cells may bring about the accurate analysis of features of human ES cells and the complete removal of human ES cells for cell therapies.

Based on this background, leading to the present invention, the intensive and thorough research conducted by the present inventors resulted in finding that, when human ES cells are cultivated, confirmed as human ES cells in an undifferentiated state and used in the preparation of monoclonal antibodies specific to human ES cells, a generated monoclonal antibody binds to undifferentiated human ES cells but does not bind to mouse-derived cells, recognizes HSPA8 as an antigen and is identified as a novel antibody recognizing an epitope different from epitopes of conventional antibodies to HSP70.

Disclosure of the Invention

It is therefore an object of the present invention to provide a monoclonal antibody to HSPA8 (heat shock 70 kDa protein 8 isoform 1) , which binds to human ES cells but does not bind to mouse ES cells.

It is another object of the present invention to provide a hybridoma that produces the monoclonal antibody.

It is a further object of the present invention to provide an assay kit for human ES cells comprising the monoclonal antibody. It is yet another object of the present invention to provide a composition for removing human ES cells comprising the monoclonal antibody.

It is still another object of the present invention to provide a method of removing human ES cells using the monoclonal antibody.

Brief Description of the Drawings

The above and other objects, features and other advantages of the present invention will be more clearly understood from the following detailed description taken in conjunction with the accompanying drawings, in which:

FIG. Ia photographically shows the morphology of cultured human ES cells (1) , and the results of the analysis of expression of ES cell surface markers: alkaline phosphatase (2) , SSEAl (3) , as a negative control marker, and SSEA3 (4) and SSEA4 (5), as positive control markers, wherein the expression of SSEAl, SSEA3 and SSEA4 is measured by an immunohistochemical assay;

FIG. Ib shows the results of electrophoresis of products of RT-PCR (Reverse transcriptase-Polymerase Chain Reaction) carried out to determine whether OCT4 (octamer binding protein 4) gene, not expressed in mouse embryonic fibroblasts (MEF) , is expressed in human ES cells (-RT: negative control not containing reverse transcriptase, MEF: mouse embryonic fibroblasts, and hES: human embryonic stem cells) ;

FIG. Ic shows the results of Southern blotting, displaying that MEF cells do not express telomerase and that human ES cells express heat-sensitive telomerase that leads to an increase in telomere length (P: positive control extract and its serial diluents, P+heat: inactivated telomerase by heat treatment of positive control (sample) , MEF: extract of mouse embryonic fibroblasts, hES: extract of human embryonic stem cells, and hES+heat: heat-treated extract of human ES cells) ;

FIG. 2 shows the results of fluorescent cell staining, displaying that a 20-202S monoclonal antibody according to the present invention bind to Miz-hESl and HSF6 human ES cells, wherein the solid line represents a monoclonal antibody, and the red background contains only a secondary antibody (SSEAl: antibody not binding to human ES cells (negative control), SSEA3 and SSEA4: antibodies binding to human ES cells (positive controls) ;

FIG. 3 shows the results of fluorescent cell staining, displaying that a 20-202S monoclonal antibody according to the present invention do not bind to mouse ES cells (mESC) , wherein the solid line represents a monoclonal antibody, and the red background contains only a secondary antibody (SSEA4: antibody not binding to mouse ES cells (negative control), SSEAl: antibody binding to mouse ES cells (positive control) ;

FIG. 4 shows the results of fluorescent cell staining, displaying that a 20-202S monoclonal antibody according to the present invention do not bind to mouse embryonic fibroblasts, wherein the solid line represents a monoclonal antibody, and the red background contains only a secondary antibody;

FIG. 5 shows the results of fluorescent cell staining, displaying that a 20-202S monoclonal antibody according to the present invention do not bind to mouse STO fibroblasts, used as supporting cells in the culture of human ES cells, wherein the solid line represents a monoclonal antibody, and the red background contains only a secondary antibody;

FIG. 6 shows the results of fluorescent cell staining, displaying that a 20-202S monoclonal antibody according to the present invention have reduced binding affinity to embryonic stem cells in the presence of retinoic acid inducing the differentiation of human ES cells, wherein the expression of SSEAl, as a negative control, and SSEA3 and SSEA4, as positive controls, is also assessed; FIG. 7 shows the results of an immunoprecipitation assay for biotinylated human ES cells with a 20-202S monoclonal antibody according to the present invention, wherein precipitated proteins are assessed by SDS-PAGE followed by Western blotting (a sample not containing the 20- 2OS antibody is used as a negative control) ;

FIG. 8 shows the results of Q-TOF analysis of proteins co-immunoprecipitated with a 20-202S monoclonal antibody according to the present invention using lysates of human ES cells, displaying that the 20-202S monoclonal antibody recognizes HSPA8 (heat shock 70 kDa protein 8 isoform 1), wherein the proteins are subjected to Q-TOF analysis after being separated by SDS-PAGE and in-gel digested with trypsin to generate peptides; FIG. 9 shows the results of Western blotting, displaying that a 20-202S monoclonal antibody according to the present invention has binding specificity different from known antibodies to HSP70;

FIG. 10 shows the results of flow cytometry, displaying that a 20-202S monoclonal antibody according to the present invention binds to various cell lines including human ES cells;

FIG. 11 shows the results of Western blotting using cells expressing exogenous HSPA8 (A) , confirming that a 20- 202S monoclonal antibody according to the present invention recognizes HSPA8 protein, and the results of Western blotting in the presence and absence of ATP (B) , displaying that the 20-202S does not bind to HSPA8 on human ES cells in the presence of ATP; and

FIG. 12 shows the results of iimnunocytochemical analysis, displaying that HSPA8 is expressed on the surface of various human ES cell lines.

Best Mode for Carrying Out the Invention

In one aspect, the present invention relates to a monoclonal antibody specific to HSPA8 (heat shock 70 kDa protein 8 isoform 1) , which binds to human embryonic stem (ES) cells but does not bind to mouse embryonic stem (ES) cells.

In one detailed aspect, the monoclonal antibody of the present invention binds to undifferentiated human ES cells but does not bind to differentiated human ES cells.

In another detailed aspect, the monoclonal antibody of the present invention specifically recognizes an ATP- sensitive epitope of HSPA8.

In one preferred aspect, the monoclonal antibody of the present invention is a 20-202S monoclonal antibody that is produced by a hybridoma assigned accession number KCTC 10733BP.

The term "monoclonal antibody", as used herein, refers to a protein molecule that is directed by a single antigenic region (single epitope) and specifically binds thereto. With respect to the objects of the present invention, since the monoclonal antibody of the present invention specifically binds to a cell surface protein of human ES cells in an undifferentiated state, the monoclonal antibody is a protein molecule recognizing a cell surface protein of undifferentiated human ES cells.

The major regions of an antibody involved in the recognition of a specific epitope and the formation of antigen-antibody complexes are variable regions of heavy chain and light chain, and in particular, CDR (complementary determining region) attributes to the formation of antigen-antibody complexes. Thus, the present invention includes chimeric antibodies and humanized antibodies of the monoclonal antibody, which comprise variable regions of the monoclonal antibody of the present invention, especially CDR. The present invention further includes whole antibodies having two full-length light chains and two full-length heavy chains as well as functional fragments of antibody molecules, so long as they retain^"" the afάremehTfioned ^"T5inding^"" features?^~ "Functional fragments of antibody molecules" mean fragments retaining at least an antigen-binding function, and include Fab,

F(ab'), F(ab')₂ and Fv. In order to produce a monoclonal antibody specific to undifferentiated human ES cells, the present inventors cultivated human ES cells in a large scale using collagenase to facilitate the follow-up culture, analyzed the features of the human ES cells, confirmed that the cultured cells are human ES cells, and immunoinjected the cultured human ES cells into mice. In detail, human ES cells were cultured, subjected to hematoxylin and eosin staining, observed under a phase contrast microscope to determine the morphology of human ES cells and the expression of alkaline phosphatase (see, panels 1 and 2 of FIG. Ia) , and assayed telomerase activity (see, FIG. Ic) and Oct4 expression (see, FIG. Ib) using RT-PCR. This analysis confirmed that the cultured cells are human ES cells. Also, in order to further confirm that the cultured cells are human ES cells, an immunohistochemical assay through SSEA (stage-specific embryonic antigen) staining resulted in the finding that an antibody against SSEAl, a negative marker for human ES cells, does not bind to the cells, and antibodies against positive markers SSEA3 and SSEA4 bind to the cells (see, the panels 3, 4 and 5 of FIG. Ia) . These results allowed the preparation of monoclonal antibodies specifically binding to undifferentiated human ES cells using ES cells themselves as an antigen.

Then, the cultured human ES cells were inactivated and used to immunize mice. Splenocytes were isolated from the mice and fused with cancer cells to generate hybridomas. From the hybridomas, a hybridoma producing a monoclonal antibody 20-202S as the detailed aspects of the present invention was isolated. The monoclonal antibody was found to have binding affinity to human ES cells (see, FIG. 2), and not to bind to mouse ES cells, mouse embryonic fibroblasts and mouse fibroblasts (STO) (FIGS. 3 to 5) . The monoclonal antibody was found to have decreased binding affinity to cells differentiated from human ES cells by retinoic acid treatment (FIG. 6) . Also, 10% SDS-PAGE analysis resulted in the finding that the monoclonal antibody 20-202S recognizes a protein of human ES cells having a molecular weight of about 72 kDa, respectively

(FIG. 7) . Since the molecular weight of the human ES cell protein, recognized by the monoclonal antibody of the present invention, is determined by 10% SDS-PAGE analysis, they may increase or decrease within a certain range according to measurement conditions. Thus, the term "about" is unavoidably used to express the molecular weight of the protein, and is typically within a range of +2 kDa, preferably ±1 kDa. In addition, to determine a protein recognized by the 20-202S monoclonal antibody, a protein co-immunoprecipitated with the^~~monoclonal antibody was separated by SDS-PAGE, excised from the gel, digested with trypsin and subjected to Q-TOF analysis. The protein was found to be a HSP70 protein, HSPA8 (FIG. 8) . A heat shock protein HSPA8 (heat shock 70 kDa protein 8 isoform 1) (Tavaria, et al., Genomics 29:266-268, 1995), which is recognized by the monoclonal antibody of the present invention, was initially identified to function to prevent protein misfolding in cells, but, according to recent reports, is expressed on the surface of cancer cells, monocytes, and umbilical vein endothelial cells

(Shin, et al., J. Biol. Chem. , 278:7607-7616, 2003; Asea, et al., Nature Med., 6:435-442, 2000; Triantafilou, et al.,

Nature Immunol. 2:338-345, 2001) . More recently, itiicroarray analysis revealed that HSPA8 gene is overexpressed in human ES cells (Abeyta, et al. Hum. MoI. Genet., 13:601-608, 2004; Zeng, et al., Stem Cells 22:292-312, 2004) . However, in this report, the overexpression of HSPA8 was observed only in RNA levels. Thus, the present invention is the first to describe the expression of HSPA8 on the surface of human ES cells, which was identified by using the specific monoclonal antibody of the present invention. Also, the expression of HSPA8 was found to decrease as human ES cells undergo transition from undifferentiated to differentiated stages, demonstrating that HSPA8 is a potential cell surface marker defining human ES cells in an undifferentiated state and is useful for isolating undifferentiated human ES cells at high purity.

In addition, Western blot analysis resulted in the finding that the monoclonal antibody of the present invention has epitope specificity different from those of conventional antibodies to HSP70 (W27, SPA810, 5GlO, SPA820) (FIG. 9) . ATP treatment revealed that the epitope recognized by the monoclonal antibody of the present invention is ATP-sensitive (FIG. 11) . These results indicate that the monoclonal antibody of the present invention is a novel antibody recognizing an epitope different from epitopes of the conventional antibodies to HSP70.

As described above, these binding specificities of the monoclonal antibody of the present invention, which specifically recognizes HSPA8 and binds to human ES cells but not to mouse-derived cells, mouse ES cells, mouse embryonic fibroblasts, and fibroblasts, demonstrate that the monoclonal antibody of the present invention is a novel antibody not identified prior to the present invention. In another aspect, the present invention relates to a hybridoma producing the monoclonal antibody of the present invention.

In one detailed aspect, the present invention provides a hybridoma producing a monoclonal antibody 20- 202S.

In one embodiment, the hybridoma of the present invention was prepared by irradiating human ES cells to inactivate them; intraperitoneally injecting the inactivated human ES cells into mice; isolating lymphocytes from the spleen of the mice; and fusing the lymphocytes with myeloma cells. Among the thus-produced hybridomas, a hybridoma secreting a monoclonal antibody 20-202S was designated as "hybridoma 20-202S", which was deposited at KCTC (Korean Collection for Type Cultures, Genetic Resources Center, KRIBB, 52, Oun-dong, Yusong-ku, Taejon, Korea) on Dec. 1, 2004 and assigned accession number KCTC 10733BP.

The hybridoma secreting a monoclonal antibody may be cultured in a large scale in vitro or in vivo.

The monoclonal antibody secreted by the hybridoma may be used without purification, but is preferably used after being highly purified (e.g., 95% or higher) by methods known in the art in order to obtain the best results.

Purification may be carried out using culture fluid or ascites fluid, for example, using gel electrophoresis, dialysis, salting out and chromatography.

In an embodiment of the present invention, for mass production of the monoclonal antibody of the present invention, hybridoma cells were intraperitoneally injected into mice to be . cultured in the peritoneal cavity, and ascites fluid was collected from the mice and subjected to protein G-sepharose column chromatography to isolate the monoclonal antibody.

In a further aspect, the present invention relates to a composition for removing undifferentiated human ES cells comprising the monoclonal antibody.

In yet another aspect, the present invention relates to a method of removing undifferentiated human ES cells using the monoclonal antibody.

The monoclonal antibody of the present invention may be used for removing ES cells present in cells to be transplanted for cell therapies or in transplanted cells.

In order to selectively remove embryonic stem cells, the monoclonal antibody of the present invention may be linked to a known therapeutic agent by direct or indirect coupling (e.g., covalent bonding) through a linker. Non- limiting examples of therapeutic agents capable of being linked to the antibody include radionuclides, drugs, lymphokines, toxins and heterologous antibodies.

In order to remove ES cells present in transplanted cells, the antibody may be administered as it is, or a composition comprising the antibody may be administered. The composition comprising the antibody may include an acceptable carrier according to administration methods and may be formulated into suitable pharmaceutical preparations. Suitable pharmaceutical preparations according to administration methods are known in the art. These pharmaceutical preparations may be administered by suitable methods including parenteral, subcutaneous, intraperitoneal, intrapulmonary and intranasal administration, and, if desired, intralesional administration for local immunosuppressive treatment. Parenteral injections include intramuscular, intravenous, intraarterial, intraperitoneal and subcutaneous administration. Preferred administration methods and pharmaceutical preparations include intravenous injectable preparations, subcutaneous injectable preparations, transdermal injectable preparations, intramuscular injectable preparations and dropping injectable preparations. The composition comprising the antibody of the present invention may be administered in an amount pharmaceutically effective for removing ES cells. Typical dosage levels may be optimized using a standard clinical technique.

In still another aspect, the present invention relates to an assay kit for undifferentiated human ES cells comprising the monoclonal antibody. The monoclonal antibody of the present invention may be used for specifically detecting undifferentiated human ES cells through an antigen-antibody complex reaction, as well as removing embryonic stem (ES) cells from cells to be transplanted or transplanted cells. In addition to the monoclonal antibody of the present invention, this assay kit may include tools and reagents, which are generally used in the art for immunological analysis. These tools/reagents include, but are not limited to, suitable carriers, labeling substances capable of generating detectable signals, solubilizing agents, detergents, buffering agents, and stabilizing agents. When the labeling substance is an enzyme, the assay kit may include a substrate allowing the measurement of enzyme activity and a reaction terminator. Suitable carriers include, but are not limited to, soluble carriers, for example, physiologically acceptable buffers known in the art, for example, PBS, insoluble carriers, for example polymers such as polystylene, polyethylene, polypropylene, polyesters, polyacrylnitrile, fluorocarbon resin, crosslinked dextran, polysaccharides and magnetic microparticles composed of latex plated with metals, papers, glasses, metals, agarose, and combinations thereof.

Antigen-antibody complex formation may be detected by using histoimmunological staining, radio-immunoassay (RIA) , enzyme-linked immunosorbent assay (ELISA) , Western blotting, immunoprecipitation assay, immunodiffusion assay, complement fixation assay, FACS and protein chips, but the present invention is not limited to these examples.

Labels allowing qualitative or quantitative analysis of the formation of antigen-antibody complexes include, but are not limited to, enzymes, fluorescent substances, ligands, luminescent substances, microparticles, redox molecules and radioactive isotopes. Examples of enzymes available as detection labels include, but are not limited to, β-glucuronidase, β-D-glucosidase, .β-D-galactosidase, urase, peroxidase, alkaline phosphatase, acetylcholinesterase, glucose oxidase, hexokinase and GDPase, RNase, glucose oxidase and luciferase, phosphofructokinase, phosphoenolpyruvate carboxylase, aspartate aminotransferase, phosphenolpyruvate decarboxylase, and β-latamase. Examples of the fluorescent substances include, but are not limited to, fluorescin, isothiocyanate, rhodamine, phycoerythrin, phycocyanin, allophycocyanin, o-phthaldehyde and fluorescamin. Examples of the ligands include, but are not limited to, biotin derivatives. Examples of luminescent substances, but are not limited to, include acridinium esters, luciferin and luciferase. Examples of the microparticles include, but are not limited to, colloidal gold and colored latex. Examples of the redox molecules include, but are not limited to, ferrocene, ruthenium complexes, viologen, quinone, Ti ions, Cs ions, diimide, 1, 4-benzoquinone, hydroquinone, K₄ W(CN)₈ , [Os(bpy)₃]²⁺ , [RU(bpy)₃]²⁺, and [MO(CN)₈]⁴-. Examples of the radioactive isotopes include, but are not limited to, ³H, ¹⁴C, ³²P, ³⁵S, ³⁶Cl, ⁵¹Cr, ⁵⁷Co, ⁵⁸Co, ⁵⁹Fe, ⁹⁰Y, ¹²⁵I, ¹³¹I, and ¹⁸⁶Re. A better understanding of the present invention may be obtained through the following examples which are set forth to illustrate, but are not to be construed as the limit of the present invention.

EXAMPLE 1: Culture of human ES cells and analysis of their features <1-1> Culture of human ES cells

In order to prepare a novel monoclonal antibody capable of specifically recognizing human ES cells, first, three human ES cell lines, Miz-hESl (Park, et al., Biol. Reprod. 69:2007-2014, 2003), Miz-hES4 (Kim, et al Mole &

Cells 2004 In press) and HSFβ (Abeyta, et al., Human MoI.

Genet 13:601-608, 2004), were obtained from MizMedi

Hospital of the Sungsam Medical Foundation (701-4,

Naebalsan-dong, Gangseo-gu, Seoul, Korea) . The human ES cells were cultured in DMEM (Dulbecco's modified Eagle's medium) /F12 (Gibco, Rockville, MD, USA) containing 20% knockout SR (Gibco) supplemented with 0.1 mM β- mercaptoethanol (Sigma, St Luis, MO, USA) , 2 mM glutamine

(Gibco), 0.1 mM non-essential amino acids (Gibco), 100 U/ml penicillin G (Sigma) , 100 μg/ml streptomycin (Sigma) and 4 ng/ml bFGF (Gibco Invitrogen) , and were subcultured every 6 days.

In detail, 12-well tissue culture plates (Nunclon) were coated with 0.1% gelatin at 37°C for 10 min. Then, gamma-irradiated (3000 rad) MEF (mouse embryonic fibroblasts (Laboratory of Animal Model Evaluation, Korean Research Institute of Bioscience & Biotechnology (KRIBB) ,

Korea) were seeded onto the plates at a density of 6.5xlO⁴ cells per well. The irradiated MEF did not grow but supported the growth of human ES cells. 24 hrs after MEF culture, 5-7 day-cultured human ES cell tissues were treated with 1 mg/ml collagenase type IV (Gibco) at 37⁰C for 1 hr, cut into a suitable size and transferred to the tissue culture plates containing the irradiated MEF feeder layer. After 48 hrs, the culture medium was exchanged with a fresh medium every day.

<l-2> Evaluation of human ES cells by hematoxylin & eosine staining

The human ES cells cultured for 6-7 days according to the same method as in the above <1-1> were washed with PBS

(phosphate-buffered saline) and fixed with 4% paraformaldehyde (Roche, NY, USA) for 30 min. After three washings with PBS, the human ES cells were treated sequentially with 100%, 95%, 80% and 70% ethanol for 1 min for each treatment. Then, the cells were washed with running water and treated with hematoxylin (Sigma) for 5 min. Hematoxylin was completely removed using ammonia water, and the cells were treated with 0.5% eosine (Sigma) for 5 min. After being washed with ammonia water, the cells were treated sequentially with 70%, 80%, 95% and 100% ethanol for 1 min for each treatment and observed by phase contrast microscopy (the panel 1 of FIG. Ia) . As a result, the human ES cells grew while forming the distinct boundary with MEF feeder cells and were closely connected with each other to form flat spherical clumps, indicating that they have characteristic morphologies of human ES cells.

<l-3> Evaluation of alkaline phosphatase expression

Human ES cells cultured according to the same method as in the <1-1> of Example 1 were fixed with 4% paraformaldehyde for 30 min. After being washed with PBS, the human ES cells were treated with 2% Tween 20 for 30 min. After being washed again with distilled water, the cells were stained using an AP staining kit (Sigma) for 15 min according to the manufacturer's protocol. After being finally washed with distilled water, the cells were counterstained with hematoxylin for 2 min and observed under a phase contrast microscope (the panel 2 of FIG. Ia) . Nuclei are shown in blue because they were counterstained with hematoxylin, and cell masses are shown in red, indicating that the cells express alkaline phosphatase.

<l-4> SSEA staining

Human ES cells cultured according to the same method as in the <1-1> of Example 1 were fixed with 4% paraformaldehyde for 30 min. After being washed with PBS, the cells were blocked in typical horse serum for 1 hr.

Subsequently, the cells were incubated with antibodies to stage specific embryonic antigen (SSEA) 1, SSEA3 and SSEA4

(DSHB, the University of Iowa, USA) for 1 hr. After being washed with PBS, the cells were incubated with a secondary antibody to SSEA for 1 hr. Then, the cells were stained using Vectastain ABC reagents (DAP staining kit, Sigma) for

20 min, washed with PBS and treated with a substrate solution. The cells were immunohistochemically observed under a phase contrast microscope. The human ES cells were negative for SSEAl as a negative control marker and positive for SSEA3 and SSEA4 as positive control markers

(the panels 3, 4 and 5 of FIG. Ia) .

<l-5> Detection of telomerase activity MEF (mouse embryonic fibroblasts) do not express telomerase, but human ES cells express heat-sensitive telomerase that leads to an increase in telomere length. This increased telomere length was detected by Southern blotting. In detail, human ES cells cultured according to the same method as in the <1-1> of Example 1 were treated with collagenase to dissociate colonies. To measure telomerase activity, cell lysis and PCR were carried out using a telomerase activity assay kit (Intergen, NY, USA) according to the manufacturer's protocol. A cell lysate was mixed with a TS primer, and telomerase in the cell lysate was allowed to elongate the TS primer. Then, PCR was carried out using an RP primer and Taq polymerase. PCR conditions included 36 cycles of 94°C for 30 sec, 59°C for 30 sec and 72°C for 1 min. PCR products were electrophoresed on a 2% agarose gel and transferred onto a membrane. Southern blotting was carried out using a ³²P-labeled-primer having the oligonucleotide sequence represented by SEQ ID NO. 5 to measure telomerase activity. The human ES cells were found to express heat-sensitive telomerase (FIG. Ic) . In FIG. Ic, the lanes represent the following: P: positive control extract and its diluents, P+heat: telomerase inactivated by heat treatment of positive control (sample) , MEF: extract of mouse embryonic fibroblasts, hES: extract of human embryonic stem cells, and hES+heat: heat-treated extract of human ES cells.

<l-6> Evaluation of Oct-4 expression

Total RNA was isolated from human ES cells cultured according to the same method as in the <1-1> of Example 1 using an RNA isolation kit (Roche) . In order to determine whether Oct4 gene, not expressed in MEF cells, is expressed in the human ES cells, RT-PCR was carried out with Oct4- specific primers represented by SEQ ID NOS. 1 and 2 and β- actin primers represented by SEQ ID NOS. 3 and 4 for RNA quantification. Then, PCR products were electrophoresed on a 1.5% agarose gel. The human ES cells were found to express 0ct4 (FIG. Ib) . In FIG. Ib, the lanes represent the following: -RT: negative control not containing reverse transcriptase, MEF: mouse embryonic fibroblasts, and hES: human embryonic stem cells. EXAMPLE 2: Preparation of mouse hybridomas

<2-l> Irnmunoinjection of human ES cells into mice

Miz-hESl human ES cells cultured according to the same method as in the <1-1> of Example 1 were treated with collagen type IV. About 2xlO⁶ cells were suspended in 100 μl of PBS, gamma-irradiated to be inactivated, and intraperitoneally injected into Balb/c mice (Laboratory of

Animal Model Evaluation, KRIBB, Korea) . Injection was repeated three times at 3-week intervals, and a final injection was carried out 3 days before cell fusion.

<2-2> Preparation of mouse hybridomas producing monoclonal antibodies

To prepare feeder cells, 20 ml of DMEM (GIBCO) was injected into the peritoneal cavity of healthy mice one day before cell fusion, and cells in the peritoneal cavity were suctioned and centrifuged. Splenocytes were prepared by grinding normal spleen and isolating cells from the spleen.

After the feeder cells were mixed with the isolated splenocytes, the cell mixture was supplemented with 20% fetal bovine serum (FBS) , plated onto a 96-well plate at a density of 10⁵ cells per well, and cultured in a CO₂ incubator at 37°C. Two weeks before cell fusion, NSl myeloma cells (ATCC, USA) to be fused with the splenocytes were cultured in a 10% FBS-containing medium.

The spleen was excised from mice immunized with human

ES cells according to the same method as in the <2-l> of

Example 2, washed with RPMI1640 (GIBCO), ground well in a petri dish using a glass bar, and transferred to a 15-ml tube. The tube was allowed to stand until debris precipitated. When the debris had precipitated, the supernatant was transferred to a new tube and centrifuged to recover NSl cells. The cell pellet was suspended in 10 ml of RPMI1640 and counted. The splenocytes were also counted. 10⁷ NSl cells were mixed with 10⁸ splenocytes in a

50-ml tube and centrifuged at 200xg for 5 min. After the supernatant was discarded, the tube was incubated in a beaker containing water at 37° for 2 min. The tube was then tapped to break up the cell pellet, and 1 ml of PEG (GIBCO) was added to the tube over one minute while the tube was gently shaken in the beaker. The cells were spun down at lOOxg for 2 min. 5 ml of RPMI1640 was slowly added to the tube for 3 min, and 5 ml of RPMI1640 was again added slowly to the tube for 2 min. After centrifugation at 200xg, the recovered cells were carefully resuspended in 30 ml of a normal medium (RPMI1640 +20% FBS) . After being incubated in a CO₂ incubator at 37° for 30 min, 70 μl of the cell suspension was aliquotted onto the 96-well plate containing the MEF feeder cells at a density of 10⁵ cells per well and cultured in a CO₂ incubator at 37°. The next day, 70 μl of HAT was added to each well, and the HAT medium was changed every three days for over two weeks. During this culture period, emerged colonies were observed.

Clones expressing antibodies were selected using sandwich ELISA (Enzyme-Linked Immunosorbent Assay) . 100 μl of a hybridoma culture was added to a plate coated with 2 μg/ml of an anti-mouse IgG or IgM antibody, incubated at 37°C for 1 hr, and then incubated with a 1:5,000 dilution of an anti-mouse IgG or IgM HRP (horseradish peroxidase, Sigma) conjugate for 1 hr. The plate was washed with phosphate buffer containing 0.05% Tween 20, and a substrate solution containing OPD and H₂O₂ was added to the plate. Absorbance was measured at 492 nm to primarily select clones producing antibodies.

EXAMPLE 3: Preparation of monoclonal antibody binding to human ES cells

<3-l> Selection of hybridoma clones

Among the clones prepared in Example 2, hybridoma supernatants relatively stably secreting antibodies were evaluated for the ability to bind to human ES cells. In detail, cultured ^' human ES cells were split using collagenase type IV and dissociated into single cells by incubation with cell dissociation buffer (GIBCO) at 37°C for 20 min. The single-cell suspension was passed through a 40- μm cell strainer, and 2xlO⁵ cells were used for flow cytometry. First, the dissociated human ES cells were suspended in PBA (1% BSA in PBS) and allowed to react with an antibody supernatant at 4°C for 30 min. The cells were centrifuged at 1200 rpm at 4°C for 5 min, and 100 μl of the antibody supernatant was discarded. Then, the cells were incubated with a 1:200 dilution of anti-mouse Ig-FITC (BD) at 4°C for 30 min. After washing with PBA twice, only propidium iodide (PI) -negative cells were selected, and analyzed for their ability to bind to human ES cells using a FACS caliber flow cytometer.

As a result, various hybridoma clones secreting antibodies binding to human ES cells were selected and continuously subcultured for subcloning. Finally, a hybridoma clone, secreting a 20-202S antibody and reliably maintaining their stability and specificity for human ES cells, were selected.

The hybridoma secreting a monoclonal antibody 20-202S was designated as "hybridoma 20-202S", deposited at KCTC (Korean Collection for Type Cultures, Genetic Resources Center, KRIBB, 52, Oun-dong, Yusong-ku, Taejon, Korea) on Dec. 1, 2004, and assigned accession number KCTC 10733BP.

<3-2> Purification of the monoclonal antibody

A monoclonal antibody 20-202S was isolated from the hybridoma 20-202S, respectively, selected in the <3-l> of Example 3.

In detail, to purify the 20-202S antibody, before one week, IxIO⁷ hybridoma cells were suspended in 0.5 ml of PBS and intraperitoneally injected into Balb/c mice primed with an injection with 0.5 ml of pristine. After 10 to 14 days, ascites fluid was collected using a syringe and centrifuged, and the supernatant was recovered. 1 ml of the ascites fluid was diluted with PBS to give a volume of 2 ml. The ascites fluid was then mixed with 1 nM EDTA and 0.02% NaN₃ and passed through a 0.22-μm filter. A Protein G-sepharose column (Pharmacia, Sweden) was allowed to bind to antibodies by rotation at 4°C for 2 hrs. After the column was stood up vertically, the wall of the column was washed with washing buffer (0.5 M NaCl, 0.1 M Tris, pH 8.0) using a serum separator, and the column was sufficiently washed using a peristaltic pump. Then, antibodies were eluted with 0.2 M glycin-HCl (pH 2.7) and neutralized using 1 M Tris (pH 9.0) .

EXAMPLE 4: Evaluation of binding specificity of the monoclonal antibody

<4-l> Binding specificity to human- and mouse-derived cells

The 20-202S antibody purified in the <3-2> of Example 3 were assessed for the binding affinity to human ES cells by fluorescent cell staining according to the same method as in the <3-l> of Example 3 (FIG. 2) . In FIG. 2, the solid line represents a monoclonal antibody, and the red background contains only a secondary antibody. SSEAl indicates an antibody as a negative control that does not bind to human ES cells, and SSEA 4 indicates an antibody as a positive control that binds to human ES cells.

In addition, mouse ES cells (Jl) (Li. et al., Cell, 69:906-915, 1992), mouse embryonic fibroblasts (MEF), and mouse STO fibroblasts (ATCC 56-X) were cultured in DMEM (GIBCO) supplemented with 10% FBS and split using collagenase type IV. To determine whether the 20-202S antibody has the capacity to bind to mouse ES cells, MEF cells and mouse STO fibroblasts, according to the same method as described above, fluorescent cell staining was carried out for flow cytometry (FIGS. 3, 4 and 5) . The 20- 202S antibody was found not to bind to the above cell lines.

<4-2> Specificity to differentiated and undifferentiated ES cells Human ES cells undergo the transition from an undifferentiated state to a differentiated state in the presence of retinoic acid (Henderson, et al., Stem Cells 20:329-337, 2002) . Based on this report, 4-day-cultured Miz-hESl cells were treated with 10^~5 M retinoic acid for 6 days, and additional cells were not treated. Then, the cells were detached and subjected to FACS analysis using the monoclonal antibody according to the same method as in the <3-l> of Example 3 (FIG. 6) . The monoclonal antibody- exhibited significantly reduced binding affinity to the differentiated cells, indicating that the monoclonal antibody of the present invention specifically bind to undifferentiated human ES cells.

<4-3> Specificity to various human ES cell lines

The 20-202S antibody was evaluated for binding to the surface of various human ES cell lines, Miz-hESl, Miz-hES4, Miz-hES6 and HSF6 by immunocytochemical analysis. Human ES cells were washed with Ca²⁺-Mg²⁺-PBS and fixed with 4% paraformaldehyde. A plate was blocked with 1.5% horse serum, incubated with a primary antibody at room temperature for 1 hr, and incubated with a biotin-labeled secondary antibody at room temperature for 1 hr. Thereafter, the plate was allowed to react with a Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA) , and positive colonies were developed using a DAB substrate kit (Vector Laboratories) . Immunofluorescent staining was carried out as follows. Human ES cells were incubated in a blocking solution (10% normal horse serum and 0.1% bovine serum albumin in PBS) at room temperature for 1 hr, and were incubated with the 20-202S antibody or anti-TRA-1-60 (Chemicon) at 4°C overnight. After being washed six times, the cells were allowed to react with FITC-conjugated anti-mouse IgG (Vector) or FITC-conjugated anti-mouse IgM (Sigma), and were counterstained with 4,6 diamidino-2-phenylindole (DAPI) . After being washed four times, the cells were mounted in Vectashield (Vector) and observed under a Zeiss 510LSM META laser-scanning microscope. As shown in FIG. 12, like positive control antibodies to SSEA3 and SSEA4, the 20-202S antibody binds to Miz-hESl, Miz-hES4, Miz-hES6 and HSF6 cells (A) . When Miz-hESl cells were more closely observed under a confocal microscope, like anti-Tra-1-60, the 20-202S antibody was found to have stained the cell surface (the B panel of FIG. 12) .

EXAMPLE 5: Immunoprecipitation assay for determining antigens recognized by the monoclonal antibody

In order to isolate cell surface markers on human ES cells, recognized by the 20-202S monoclonal antibody, first, cultured human ES cells were washed with PBS and biotinylated using EZ-Link Sulfo-NHS-LC-Biotin (Pierce, Rockford, IL) . Then, the cells were lysed with lysis buffer

(25 mM Tris-HCl, pH 7.5, 250 mM NaCl, 5 mM EDTA, 1% Nonidet

P-40, 2 μg/ml aprotinin, 100 μg/ml phenylmethylsulfonyl fluoride, 5 μg/ml leupeptin) at 4°C for 20 min, and were centrifuged to remove the nuclei. Protein concentrations were determined using a BCA (bicinchoninic acid) protein assay kit (Pierce) . Proteins nonspecifically binding to Protein G plus-Sepharose (Santa Cruz Biotechnology, Santa Cruz) were prepared as follows. The cell lysate was allowed to react with 20 μl of Protein G plus-sepharose at 4°C for 2 hrs and centrifuged. The supernatant was recovered and incubated with about 1 μg of an antibody at 4°C for 12 hrs. The cell lysate was then mixed with 20 μl of Protein G plus-sepharose, incubated at 4°C for 2 hrs and centrifuged. The pellet was recovered and washed with lysis buffer ten times or more. The remaining proteins were separated on 10% SDS-PAGE, transferred onto a nitrocellulose membrane and subjected to Western blotting. The nitrocellulose membrane was blocked in 5% skim milk in PBST (PBS + 0.1% Tween 20) for 1 hr. After being washed with PBST twice or more, the blot was incubated with a Streptavidin-HRP (horseradish peroxidase) conjugate (1:1,500, Amersham Biosciences) for 1 hr. After being washed with PBST five times, the blot was developed using ECL detection reagents (Amersham Biosciences) to detect biotinylated proteins (A panel of FIG. 7) . As a result, a protein of about 72 kDa was found to bind to the 20-202S antibody.

Separately, in order to collect proteins co- immunoprecipitated with the 20-202S antibody, an immunoprecipitation assay was carried out using lysates of IxIO⁸ Miz-hESl cells according to the same method as described above. Also, lysates of Choi-CK cancer cells, to which the 20-202S antibody showed high binding specificity, were immunoprecipitated according to a method similar to the above method. Proteins co-immunoprecipitated with the 20-202S antibody were analyzed by SDS-PAGE, and the gel was stained with Coomassie G250 (Biorad) (B panel of FIG. 7) .

EXAMPLE 6: Analysis of antigen recognized by the 20-202S monoclonal antibody

A SDS-PAGE gel, on which proteins co- immunoprecipitated with the 20-202S antibody were separated, was stained with Coomassie G250 (Biorad) according to the manufacturer's protocol. Protein bands were excised from the gel, washed in 30% methanol for 5 min, and cut into small sizes. The gel pieces were completely decolored with 30% methanol. The gel pieces were treated with 100% acetonitrile for 10 min to remove all remaining liquid and dried in a vacuum centrifuge for 30 min. Then, 300 ng of trypsin (Promega) and 50 mM NH₄HCO₃ were added to the gel pieces to digest proteins, followed by incubation at 37°C for 16 hrs. The resulting peptides were extracted with 100 μl of 50 mM NH₄HCO₃ three times and dried in a vacuum centrifuge. The peptide mixture was analyzed by ESI Q-TOF MS/MS (electrospray quadrupole time of flight tandem mass spectrometry) using a MicroMass Q-TOF micro mass spectrometer (FIG. 8) . As a result, a protein co-immunoprecipitated with the 20-202S antibody was found to be HSPA8 (heat shock 70 kDa protein 8 isoform 1) .

EXAMPLE 7: Comparison of the 20-202S monoclonal antibody with other anti-HSP70 antibodies

HSPA8 is a member of the heat shock protein (HSP) 70 family. HSP70 family members have high similarity in amino acid sequence, and antibodies to HSP70 are commercially available. Available anti-HSP70 antibodies include W27 (Santa Cruz) , SPA810 (Stressgen) , 5G10 (BD Pharmingen) and SPA820 (Stressgen) . To compare the 20-202S monoclonal antibody with the conventional anti-HSP70 antibodies, Western blotting was carried out using lysates of various carcinoma cell lines (cholangiocarcinoma Choi-CK, sarcomatoid cholangiocarcinoma SCK, hepatocellular carcinoma HepG2, sarcomatoid hepatocellular carcinoma SH-Jl and cervical carcinoma HeLa) , mouse-derived cell lines

(MEF, STO and mESC) , Miz-hESl human ES cells and hNPSTl neural progenitor cells differentiated from human ES cells (Korean Pat. Application No. 10-2004-0011705) . As shown in FIG. 9, the 20-202S antibody recognized Miz-hESl cells expressing HSPA8 protein. W27 and SPA820 antibodies also recognized Miz-hESl cells, but, unlike the 20-202S antibody, displayed positive responses for mouse-derived cells (STO, MEF and mESC) and cancer cells (SCK and HepG2) . These results indicate that the 20-202S antibody is distinct from the known antibodies to HSPA8 (W27, SPA810, 5G10 and SPA820) . These results were consistent with FACS results. As apparent from data in FIGS. 3, 4, 5 and 10, the 20-202S antibody bound only to Miz-hESl, Choi-CK, SH-Jl and HeLa cells but did not bind to MEF, STO, mESC, hNPSTl, SCK, HepG2 and PBL cells. In order to confirm that the 20-202s antibody recognizes HSPA8, SCK carcinoma cells, which were negative in Western blotting and FACS analysis using the 20-202S antibody, were transfected with an HSPA8 expression vector, HSPA8/pCMV-SPORT6. As shown in FIG. 11, Western blotting resulted in the finding that the 20-202s antibody binds to SCK cells transfected with the HSPA8 expression vector (A) . Also, an epitope of the HSPA8 antigen, recognized by the 20-202S antibody, was assessed for sensitivity to ATP by performing Western blotting in the presence and absence of 2.5 IDM ATP and 10 mM MgCl₂. As a result, the 20-202S antibody recognized an ATP-sensitive epitope (B) .

Industrial Applicability

As described hereinbefore, the monoclonal antibody of the present invention specifically recognizes a cell surface protein of human ES cells. Thus, the monoclonal antibody provides a tool for research into the difference between mice and humans as a higher animal species in the early embryonic development and thus is useful for the analysis of human ES cells. Also, the monoclonal antibody is useful for the removal of undifferentiated human ES cells for cell therapies.

Claims

1. A monoclonal antibody specific to HSPA8 (heat shock 70 kDa protein 8 isoform 1) , which binds to human embryonic stem cells but does not bind to mouse embryonic stem cells.

2. The monoclonal antibody as set forth in claim 1, which binds to undifferentiated human embryonic stem cells but does not bind to differentiated human embryonic stem cells.

3. The monoclonal antibody as set forth in claim 2, which specifically recognizes an ATP-sensitive epitope of the HSPA8.

4. The monoclonal antibody as set forth in claim 3, which is a 20-202S monoclonal antibody produced by a hybridoma assigned accession number KCTC 10733BP.

5. A 20-202S hybridoma assigned accession number KCTC 10733BP.

6. A composition for removing undifferentiated human embryonic stem cells, comprising the monoclonal antibody of claim 1.

42

7 . An assay kit for undifferentiated human embryonic stem cells, comprising the monoclonal antibody of claim 1.

43