HK1110335B - Anti-glypican 3 antibody having modified sugar chain - Google Patents

Abstract

An anti-glypican 3 antibody having a modified sugar chain, more specifically, an anti-glypican 3 antibody lacking fucose. This anti-glypican 3 antibody can be produced by a method for producing an antibody with the sugar chain modification as described above wherein a nucleic acid encoding an anti-glypican 3 antibody is transferred into host cells with lowered ability to add fucose (for example, YB2/0 cells) or fucose transporter-deficient cells and then the host cells are cultured. Because of having a high cytotoxic activity, this anti-glypican 3 antibody having a modified sugar chain is useful as a cell growth inhibitor such as an anticancer agent.

Description

Anti-glypican 3 antibody with modified sugar chain

Technical Field

The present invention relates to an antibody against glypican 3 antigen (i.e., an anti-glypican 3 antibody) having enhanced cytotoxic activity, particularly antibody-dependent cellular cytotoxicity (ADCC), and a method for producing the same.

Background

Glypican 3(GPC3) is a member of the heparan sulfate proteoglycan family present on the cell surface. It was suggested that GPC3 may be involved in cell division during development and cancer cell proliferation, but its function is still not well understood.

It has been found that various antibodies binding to GPC3 have an effect of inhibiting cell proliferation by ADCC (antibody-dependent cellular cytotoxicity) activity as well as CDC (complement-dependent cytotoxicity) activity (WO 2003/000883). GPC3 was cleaved in vivo and secreted into the blood as soluble GPC3, suggesting that cancer can be diagnosed using an antibody that detects the soluble GPC3 form. (WO2004/022739, WO2003/100429, WO 2004/018667).

In the development of an anticancer agent utilizing the cytotoxic activity of an antibody, it is preferable that the antibody used has a high level of ADCC activity, and therefore, an anti-GPC 3 antibody having a high level of cytotoxic activity is required.

It is known that modification of the sugar chain of an antibody can enhance ADCC activity of the antibody. For example, WO99/54342 describes that ADCC activity is increased by altering the glycosylation of an antibody. In addition, WO 00/61739 describes that ADCC activity is regulated by controlling the presence or absence of fucose in antibody sugar chains. WO02/31140 describes the production of antibodies having sugar chains that do not contain alpha-1, 6 core fucose by allowing YB2/0 cells to produce antibodies. WO 02/79255 describes an antibody having sugar chains containing bisecting (bisecting) GlcNAc. However, an anti-glypican 3 antibody that enhances ADCC activity due to the modification of sugar chains has not been disclosed so far.

Disclosure of Invention

The present invention provides an anti-glypican 3 antibody composition that enhances ADCC activity by changing the sugar chain component, and a method for producing the antibody.

After various studies, the present inventors found that an anti-GPC 3 antibody having an alpha-1, 6 core fucose-deficient sugar chain has a high level of cytotoxic activity. Accordingly, the present invention provides an anti-GPC 3 antibody composition in which the sugar chain component of the antibody is changed, more specifically, an antibody composition in which the proportion of fucose-deficient anti-GPC 3 antibody is large. The sugar chain-modified anti-GPC 3 antibody composition of the present invention has a high level of cytotoxic activity and is therefore useful as a cell growth inhibitor such as an anticancer agent.

The present invention also provides a method for producing an anti-GPC 3 antibody composition, wherein the sugar chain of the antibody is modified, comprising the steps of: a nucleic acid encoding an anti-GPC 3 antibody is introduced into a host cell having a reduced fucose addition ability such as YB2/0 cells, and the antibody is obtained by culturing the host cell. The cell having a reduced ability to add fucose to a sugar chain is preferably a cell lacking a fucose transporter.

Drawings

FIG. 1 shows the basic structure of an N-glycoside-linked sugar chain.

FIG. 2 shows ADCC activity of the chimeric antibody when human Peripheral Blood Mononuclear Cells (PBMC) and HepG2 cells were used as target cells.

FIG. 3 shows ADCC activity of the chimeric antibody when human PBMC were used and HuH-7 cells were used as target cells.

FIG. 4 shows ADCC activity of an antibody using human PBMC and HuH-7 cells as target cells.

FIG. 5 shows normal phase HPLC chromatograms of galactose-deficient 2-AB modified sugar chains prepared from antibodies (a, b, c) produced by FT-KO cells and antibodies produced by CHO cells.

FIG. 6 shows the estimated structures of the peaks G (0) and G (0) -Fuc shown in FIG. 5.

FIG. 7 shows DSC (differential scanning calorimetry) profiles of antibody (a) produced by FT-KO cells and antibody (b) produced by CHO cells.

Detailed Description

The present invention is characterized by providing an anti-GPC 3 antibody composition in which the sugar chain component of the antibody is changed. The structure of a sugar chain attached to an antibody is known to have a significant influence on the expression of cytotoxic activity of the antibody. The sugar chain linked to the antibody includes an N-glycoside-linked sugar chain linked to an N atom on the side chain of an asparagine residue of an antibody molecule, an O-glycoside-linked sugar chain linked to a hydroxyl group on the side chain of a serine or threonine residue of an antibody molecule, and the present invention focuses on the presence and absence of fucose in the N-glycoside-linked sugar chain.

FIG. 1 shows the basic structure of an N-glycoside-linked sugar chain linked to an antibody. As shown in IgG base sugar chains (1) and (3) of FIG. 1, the N-glycoside-linked sugar chain has a basic structure (core) [ -Man β 1-4GlcNAC β 1-4GlcNAc- ] in which 1 mannose (Man) and 2N-acetylglucosamine (GlcNAc) moieties are linked by β -1, 4 bonds. The GlcNAc on the right side of the structure is called the reducing end and the Man on the left side is called the non-reducing end. When fucose (Fuc) is linked to the reducing terminus, a linkage in a-linkage between the 6-position of N-acetylglucosamine at the reducing terminus and the 1-position of fucose is generally employed. On the other hand, in the sugar chains shown in the IgG base sugar chains (2) in FIG. 1, 1N-acetylglucosamine (GlcNAc) is linked by β 1, 4 bonds in addition to the above 2 sugar chains at the non-reducing end of the base structure (core). This type of N-acetylglucosamine (GlcNAc) is called "bisecting N-acetylglucosamine". The sugar chain having bisecting N-acetylglucosamine may be an O-glycoside-linked sugar chain or an N-glycoside-linked sugar chain, and is formed by transferring N-acetylglucosamine to the sugar chain by N-acetylglucosaminyltransferase III (GnTIII). The gene encoding the enzyme has been cloned, and the amino acid sequence thereof and the nucleotide sequence of the DNA encoding the enzyme are disclosed (NCBI database (accession number Dl 3789)).

In the present invention, an antibody composition with a modified or changed sugar chain component (sugar chain-modified antibody composition) refers to an antibody composition having a sugar chain component different from that of an antibody composition produced by a host cell as a reference standard.

In the present invention, whether or not the sugar chain component is changed is judged by using an antibody composition produced by a host cell as a reference standard. If the antibody composition has a sugar chain component different from that of the antibody composition as a reference standard, the antibody composition is considered to be an antibody composition in which the sugar chain component is changed.

In the present invention, the host cell used as a reference standard is CHO DG44 cell. CHODG44 cells are commercially available from, for example, Invitrogen.

Examples of the antibody composition having a modified sugar chain composition include, for example, an antibody composition in which the proportion of antibodies lacking fucose (e.g., α -1, 6 core fucose) in the antibody composition is increased, and an antibody composition in which the proportion of antibodies having linked bisecting N-acetylglucosamine (GlcNAc) in the antibody composition is increased.

As a preferred embodiment of the present invention, the proportion of fucose-deficient antibodies in the antibody composition is higher than in the antibody composition as a reference standard.

Since some antibodies have a plurality of N-glycoside sugar chains, the fucose-deficient antibodies of the present invention include not only antibodies having no fucose attached at all but also antibodies having a reduced number of fucose moieties attached to the antibodies (antibodies having no fucose in at least 1 or more sugar chains).

When an antibody with a modified sugar chain is produced using a host cell, it is generally difficult to obtain a composition containing homogeneous antibodies all having the same sugar chain. Therefore, if the antibody composition with a modified sugar chain composition of the present invention is, for example, an antibody composition with an increased proportion of fucose-deficient antibodies, the antibody composition with a modified sugar chain composition of the present invention may contain fucose-deficient antibodies and fucose-non-deficient antibodies, but the total proportion of fucose-deficient antibodies is higher than that in an antibody composition prepared from a host cell as a reference standard. In the antibody composition of the present invention having a high proportion of fucose-deficient antibodies, the proportion of fucose-deficient antibodies is not particularly limited, but is preferably 20% or more, more preferably 50% or more, and most preferably 90% or more.

In the antibody composition of the present invention having a high proportion of bisecting N-acetylglucosamine-added antibodies, the proportion of bisecting N-acetylglucosamine-added antibodies is not particularly limited, but is preferably 20% or more, more preferably 50% or more, and most preferably 90% or more.

The anti-GPC 3 antibody composition with a modified sugar chain component of the invention can be obtained by a method known to those skilled in the art.

For example, a fucose-deficient antibody can be prepared by expressing the anti-GPC 3 antibody in a host cell that does not have or has a reduced ability to add alpha-1, 6 core fucose.

The host cell having no or reduced fucose addition ability of the present invention is not particularly limited, but host cells lacking or reduced fucose transferase activity, host cells having reduced fucose concentration in Golgi apparatus, and the like can be used in the present invention. More specifically, examples of the host cells include rat myeloma YB2/3HL.P2.G11.16Ag.20 cells (abbreviated as YB2/0 cells) (deposited as ATCC CRL 1662), FTVIII knockout CHO cells (WO02/31140), Lec13 cells (WO03/035835), fucose transporter deficient cells (WO 2005/017155), and the like.

The term "fucose transporter deficient cell" as used herein refers to a cell in which the fucose transporter is present in a smaller amount in the cell than in a normal cell, or a cell in which the fucose transporter is structurally abnormal, resulting in a decrease in the function of the fucose transporter. Examples of the fucose transporter deficient cell include, for example, a cell in which the fucose transporter gene is knocked out (hereinafter referred to as "FT-KO cell"), a cell in which a part of the fucose transporter gene is deleted or mutated, a cell in which the expression system of the fucose transporter gene is defective, and the like. The nucleotide sequence of the gene encoding chinese hamster fucose transporter and the amino acid sequence thereof are shown in SEQ id no: 126 and 127.

In addition, using SEQ ID NO: 126, a fucose transporter deficient cell of the invention can be obtained using RNA interference (RNAi). RNAi refers to the following phenomenon: when double-stranded RNA (dsrna) is introduced into a cell, intracellular mRNA matching the RNA sequence is specifically degraded and cannot be expressed as a protein. RNAi typically uses double-stranded RNA, but the invention is not limited thereto, and for example, double-stranded RNA formed from self-complementary single-stranded RNA molecules may also be used. In the case of a region forming a double-stranded molecule, the molecule may be double-stranded in all regions, or may be single-stranded in a part of the region (for example, both ends or one end). The present invention is not particularly limited with respect to the length of the oligo-RNA used in RNAi. The length of the oligo-RNA of the present invention is, for example, 5 to 1000 bases (or 5 to 1000bp in a double-stranded molecule), preferably 10 to 100 bases (or 10 to 100bp in a double-stranded molecule), most preferably 15 to 25 bases (15 to 25bp in a double-stranded molecule), and particularly preferably 19 to 23 bases (19 to 23bp in a double-stranded molecule).

The above RNAi method utilizes the following phenomenon: dsrnas homologous to a particular gene, consisting of sense and antisense RNA, disrupt the homologous portions of the transcript (mRNA) of that gene. Double-stranded RNA corresponding to the entire sequence of the fucose transporter gene may be used, or shorter dsRNA (for example, 21 to 23bp) corresponding to a part of the sequence may be used (small interfering dsRNA; siRNA). The double-stranded RNA may be directly transferred into a cell, or may be introduced into a host cell after a vector for producing the double-stranded RNA is prepared, thereby producing the double-stranded RNA in the cell. For example, all or a portion of the DNA encoding the fucose transporter may be inserted into a vector so as to form an inverted repeat sequence, and then the vector may be introduced into a host cell. The RNAi method can be performed as described in the following documents: fire A. et al, Nature (1998), 391, 806-811, Montgomery M.K. et al, Proc. Natl. Acad. Sci. USA (1998), 95, 15502-15507, Timmons L. et al, Nature (1998), 395, 854, S a nchez A. et al, Proc. Natl. Acad. Sci. USA (1999), 96, 5049-5054, Misquitta L. et al, Proc. Natl. Acad. Sci. USA (1999), 96, 1451-1456, Kennerd J. R. et al, Cell (1998), 95, 1017-1026, Waterhouse P.M. et al, Proc. Natl. Acad. Sci. USA (1998), 95, 13959-13964, nanny F. et al, Nature 2-2000, Wil. 75 et al.

The fucose transporter deficient cells obtained by the RNAi method can be screened using the fucose transporter activity as an index. Screening can also be performed based on transcription and expression of fucose transporter genes expressed by western blotting or northern blotting.

An antibody having bisected N-acetylglucosamine (GlcNAc) added to the sugar chain can be produced by expressing the anti-GPC 3 antibody in a host cell having the ability to form a bisected N-acetylglucosamine (GlcNAc) structure in the sugar chain.

A method for producing an antibody having a bisected N-acetylglucosamine-added sugar chain has been known (WO 02/79255). The host cell having the ability to form a bisecting N-acetylglucosamine (GlcNAc) structure in a sugar chain of the present invention is not particularly limited, and may include, for example, a host cell having an expression vector containing a DNA encoding GnTIII. Therefore, an anti-GPC 3 antibody having a sugar chain to which bisected N-acetylglucosamine is added can be produced using a host cell having an expression vector containing DNA encoding GnTIII and an expression vector encoding anti-GPC 3 antibody. The DNA encoding GnTIII and the gene encoding anti-GPC 3 antibody may be present on the same vector or may be present on different vectors.

Another method for increasing the proportion of fucose-deficient antibodies or bisecting N-acetylglucosamine-added antibodies in an antibody composition is to increase the proportion of the above antibodies in the composition by purifying the fucose-deficient antibodies or bisecting N-acetylglucosamine-added antibodies.

The analysis of sugar chains can be performed by methods known to those skilled in the art. For example, by reacting an antibody with N-glycosidase F (Roche) or the like, a sugar chain can be released from the antibody. Then, the sugar chain can be desalted by a solid phase extraction method using a cellulose column (Shimizu Y. et al, Carbohydrate Research 332(2001), 381-388), concentrated and dried, and fluorescently labeled with 2-aminopyridine (Kondo A. et al, Agricultural and biological chemistry 54: 8(1990), 2169-2170). The reagent is removed from the pyridylaminochain (PA-sugar chain) by solid phase extraction using a cellulose column, and the sugar chain is then concentrated by centrifugation to obtain a purified PA-sugar chain. Then, the sugar chain can be determined by reverse phase HPLC analysis using an Octadecylsilane (ODS) column. The PA-sugar chains thus prepared can also be analyzed by two-dimensional mapping using a combination of reverse phase HPLC analysis using an ODS column and normal phase HPLC analysis using an amine column.

The sugar chain-modified anti-GPC 3 antibody of the present invention is not particularly limited as long as it can bind to GPC 3. The binding to GPC3 is preferably specific. Preferred anti-GPC 3 antibodies of the present invention include antibodies having Complementarity Determining Region (CDR) sequences shown in Table 1 below.

TABLE 1

Antibodies	CDR	Amino acid sequence	SEQ ID NO：
				M13B3(H)	CDR1	NYAMS	5
	CDR2	AINNNGDDTYYLDTVKD	6
					CDR3	QGGAY	7
M3B8(H)	CDR1	TYGMGVG	8
					CDR2	NIWWYDAKYYNSDLKS	9
	CDR3	MGLAWFAY	10
				M11F1(H)	CDR1	IYGMGVG	11
	CDR2	NIWWNDDKYYNSALKS	12
					CDR3	IGYFYFDY	13
M5B9(H)	CDR1	GYWMH	14
					CDR2	AIYPGNSDTNYNQKFKG	15
	CDR3	SGDLTGGLAY	16
				M6B1(H)	CDR1	SYAMS	17
	CDR2	AINSNGGTTYYPDTMKD	18
					CDR3	HNGGYENYGWFAY	19
M10D2(H)	CDR1	SYWMH	20
					CDR2	EIDPSDSYTYYNQKFRG	21
	CDR3	SNLGDGHYRFPAFPY	22
				L9G11(H)	CDR1	SYWMH	20
	CDR2	TIDPSDSETHYNLQFKD	23
					CDR3	GAFYSSYSYWAWFAY	24
GC33(H)	CDR1	DYEMH	25
					CDR2	ALDPKTGDTAYSQKFKG	26
	CDR3	FYSYTY	27
				GC179(H)	CDR1	INAMN	28
	CDR2	RIRSESNNYATYYGDSVKD	29
					CDR3	EVTTSFAY	30
GC194(H)	CDR1	ASAMN	31
					CDR2	RIRSKSNNYAIYYADSVKD	32
	CDR3	DPGYYGNPWFAY	33
				GC199(H)	CDR1	DYSMH	34
	CDR2	WINTETGEPTYADDFKG	35
					CDR3	LY	36

TABLE 1 (continuation)

GC202(H)	CDR1	TYGMGVG	8
					CDR2	NIWWHDDKYYNSALKS	37
	CDR3	IAPRYNKYEGFFAF	38
				M13B3(L)	CDR1	KSSQSLLDSDGKTYLN	39
	CDR2	LVSKLDS	40
					CDR3	WQGTHFPLT	41
M3B8(L)	CDR1	KASQDINNYLS	42
					CDR2	RANRLVD	43
	CDR3	LQCDEFPPWT	44
				M11F1(L)	CDR1	RSSQSLVHSNGNTYLH	45
	CDR2	KVSNRFS	46
					CDR3	SQSTHVPWT	47
M5B9(L)	CDR1	RSSKSLLHSNGITYLY	48
					CDR2	QMSNLAS	49
	CDR3	AQNLELPYT	50
				M6B1(L)	CDR1	KASQDINKNII	51
	CDR2	YTSTLQP	52
					CDR3	LQYDNLPRT	53
M10D2(L)	CDR1	RASHSISNFLH	54
					CDR2	YASQSIS	55
	CDR3	QQSNIWSLT	56
				L9G11(L)	CDR1	RASESVEYYGTSLMQ	57
	CDR2	GASNVES	58
					CDR3	QQSRKVPYT	59
GC33(L)	CDR1	RSSQSLVHSNGNTYLH	45
					CDR2	KVSNRFS	46
	CDR3	SQNTHVPPT	60
				GC179(L)	CDR1	KSSKSLLHSNGNTYLN	61
	CDR2	WMSNLAS	62
					CDR3	MQHIEYPFT	63
GC194(L)1	CDR1	RSSKSLLHSYDITYLY	64
					CDR2	QMSNLAS	49
	CDR3	AQNLELPPT	65
				GC194(L)2	CDR1	SASSSVSYMY	66
	CDR2	DTSNLAS	67
					CDR3	QQWSSYPLT	68
GC199(L)	CDR1	KSSQSLLHSDGKTFLN	69
					CDR2	LVSRLDS	70
	CDR3	CQGTHFPRT	71
				GC202(L)	CDR1	RSSQSIVHSNGNTYLE	72
	CDR2	KVSNRFS	46
					CDR3	FQGSHVPWT	73

Antibodies having the CDR sequences listed in the above tables have high levels of cytotoxic activity. Antibodies having the CDR sequences listed in the above tables recognize the epitope at amino acids 524-563 of GPC 3. Since the antibody recognizing the epitope of amino acids 524-563 has a high level of cytotoxic activity, it is preferable as the anti-GPC 3 antibody of the present invention.

In a preferred embodiment of the present invention, the antibody composition having a modified sugar chain component of the present invention is characterized by having enhanced ADCC activity. In the present invention, whether or not the ADCC activity is enhanced can be determined by comparing with an antibody composition as a reference standard. If the antibody composition of the present invention shows an ADCC activity higher than the reference standard, the ADCC activity is judged to be enhanced.

ADCC activity can be measured by a method known to those skilled in the art, and for example, ADCC can be measured by mixing effector cells, target cells, and an anti-GPC 3 antibody and then detecting the degree of ADCC. More specifically, for example, mouse spleen cells, monocytes isolated from human Peripheral Blood (PBMC) or bone marrow, etc. may be used as effector cells, and human cells expressing GPC3, such as human hepatocellular carcinoma cell line HuH-7, etc. may be used as target cells. The target cells were labeled with 51Cr in advance, and then cultured after adding an anti-GPC 3 antibody thereto, and then cultured together with effector cells added in an appropriate ratio to the target cells. After the culture, the supernatant was collected, and the ADCC activity was measured by counting radioactivity in the supernatant.

anti-GPC 3 antibody

anti-GPC 3 antibodies can be prepared by methods well known to those skilled in the art. For example, the antibody can be prepared by immunizing according to a conventional immunization method using GPC3 as a sensitizing antigen, fusing the resulting immunocytes with known parent cells by a conventional cell fusion method, and then screening monoclonal antibody-producing cells by a conventional screening method. Specifically, monoclonal antibodies can be prepared as follows. First, GPC3 used as a sensitizing antigen for obtaining an antibody was obtained by expressing GPC3(MXR7) based on the Gene/amino acid sequence disclosed in Lage, H.et al, Gene 188(1997), 151-156. That is, the gene sequence encoding GPC3 is inserted into a known expression vector, an appropriate host cell is transformed with the vector, and the target human glypican 3 protein is purified from the host cell or culture supernatant by a known method. Then, the purified GPC3 protein was used as a sensitizing antigen. A partial peptide of GPC3 can also be used as a sensitizing antigen. In this case, a partial peptide can be chemically synthesized based on the amino acid sequence of human GPC 3. The epitope on the GPC3 molecule recognized by the anti-GPC 3 antibody of the present invention is not limited as long as the anti-GPC 3 antibody of the present invention can recognize any epitope present on the GPC3 molecule. This is because the anti-GPC 3 antibody exhibits cell growth inhibitory activity by its ADCC activity, CDC activity, or growth factor inhibitory activity, and furthermore, the cell growth is inhibited by the action of a cytotoxic substance such as a radioisotope, a chemotherapeutic agent, a bacterial toxin, or the like linked to the anti-GPC 3 antibody. Therefore, the antigen used for preparing the anti-GPC 3 antibody of the present invention may be any fragment of GPC3 as long as it contains an epitope present on the GPC3 molecule.

In a particularly preferred embodiment, in order to obtain an antibody recognizing an epitope of amino acids 524-563 of GPC3, a peptide containing amino acids 524-563 may be used as the sensitizing antigen.

The mammal immunized with the sensitizing antigen in the present invention is not particularly limited, but preferably, the selected animal includes rodents, for example, mice, rats or hamsters, or rabbits, monkeys, etc., in view of compatibility with the parent cell used in cell fusion. Animals are immunized with the sensitizing antigen according to well known methods. For example, the sensitizing antigen is typically injected intraperitoneally or subcutaneously into a mammal. Specifically, the sensitizing antigen is diluted or suspended with an appropriate amount of Phosphate Buffered Saline (PBS) or physiological saline, and if necessary, mixed with an appropriate amount of a conventional adjuvant, for example, freund's complete adjuvant, emulsified, and administered to the mammal several times per 4 to 21 days. In addition, a suitable carrier may be used for immunization with a sensitizing antigen.

After immunization of the mammal as described above and detection of the desired antibody levels in the serum, immune cells are harvested from the mammal and subjected to cell fusion. Spleen cells are particularly preferred as immune cells for cell fusion. As the parent cell to be fused with the above immune cell, a mammalian myeloma cell is used. Well-known Cell lines suitable for use as myeloma cells include, for example, P3(P3x63Ag8.653) (J.Imnolol. (1979)123, 1548-1550), P3x63Ag8U.1(Current Topics in Microbiology and Immunology (1978)81, 1-7), NS-1(Kohler.G.and Milstein, C.Eur.J.Immunol. (1976)6, 511-519), MPC-11 (Margulies.D.H.et al, Cell (1976)8, 405-415), SP2/0(Shulman, M. et al, Nature (1978)276, 269-270), FO (St.Groth, S.F. et al, J.Immunol.35, 1-21), S194, S.194, J.42-55, Nature J.11-11 (19723, 11, 1979-11, Nature J.J.Immunol.R.323, 1978, 1979-210). Cell fusion of the above immunocytes and myeloma cells can be carried out basically according to a known method, for example, the method of Kohler and Milstein (Kohler.G. and Milstein, C., Methods Enzymol (1981)73, 3-46). More specifically, for example, the above cell fusion is carried out in a conventional culture medium containing a cell fusion promoter. Examples of the cell fusion-promoting chemical include polyethylene glycol (PEG), Sendai virus (HVJ), and the like. If necessary, an auxiliary agent such as dimethyl sulfoxide may be added to improve the fusion efficiency. The ratio of the immune cells to the myeloma cells can be arbitrarily set, and for example, it is preferable that the ratio of the immune cells to the myeloma cells is 1 to 10 times. As the culture solution used in the above cell fusion, a conventional liquid medium suitable for culturing these cell types, for example, RPMI1640 culture solution, MEM culture solution, and other culture solutions suitable for the culture of myeloma cells can be used. Serum additives such as Fetal Calf Serum (FCS) may also be used in combination. In the cell fusion procedure, prescribed amounts of immune cells and myeloma cells are mixed well in the above-mentioned culture solution, and then a PEG solution (for example, one having an average molecular weight of about 1000-6000) previously warmed to 37 ℃ is added thereto, usually at a concentration of 30-60% (w/v), and mixed to form fused cells (hybridomas). Then, an appropriate culture medium was added thereto, and the supernatant was centrifuged off. This step is repeated, thereby removing all cell fusion chemicals that are detrimental to hybridoma growth. The hybridoma thus obtained is selected by culturing in a conventional selection medium, for example, HAT medium (medium containing hypoxanthine, aminopterin, and thymidine). The culture in the HAT medium is continued for a sufficient period of time (usually several days to several weeks) until the cells other than the target hybridoma (non-fused cells) are killed. Then, a conventional limiting dilution is performed, followed by screening of target antibody-producing hybridomas and monoclonality. In addition to the above hybridomas obtained by immunizing a non-human animal with an antigen, a desired human antibody having a GPC3 binding activity can be obtained by sensitizing human lymphocytes with GPC3 in vitro and then fusing the sensitized lymphocytes with immortalized human myeloma cells (see Japanese patent publication No. Hei 1-59878). Furthermore, GPC3 as an antigen can be administered to a transgenic animal having all the components of a human antibody gene to obtain cells producing anti-GPC 3 antibody, and human antibody against GPC3 can be collected from immortalized cells (see International patent application publication Nos. WO 94/25585, WO 93/12227, WO92/03918, and WO 94/02602). The monoclonal antibody-producing hybridoma thus prepared can be subcultured in a conventional culture medium, and can be stored in liquid nitrogen for a long period of time.

Recombinant antibodies

The monoclonal antibody used in the present invention may be a recombinant monoclonal antibody prepared by cloning an antibody gene from a hybridoma, inserting the gene into an appropriate vector, and introducing into a host cell (see, for example, Vandamm, A.M., et al, Eur.J.biochem. (1990)192, 767. SP. 775, 1990). Specifically, mRNA encoding the variable (V) region of the anti-GPC 3 antibody was isolated from a hybridoma producing the anti-GPC 3 antibody. Isolation of mRNA can be carried out by a known method, for example, guanidine ultracentrifugation (Chirgwin, J.M. et al, Biochemistry (1979)18, 5294-. mRNA can also be prepared directly by using QuickPrep mRNA purification kit (Pharmacia). cDNA of the antibody V region is synthesized from the mRNA thus obtained using reverse transcriptase. cDNA can be synthesized using an AMV reverse transcriptase first strand cDNA Synthesis kit (available from Biochemical industries, Ltd.). For the synthesis and amplification of cDNA, 5 '-RACE method using 5' -Ampli FINDER RACE kit (Clontech) and PCR (Frohman, M.A. et al, Proc. Natl. Acad. Sci. USA (1988)85, 8998-. The target DNA fragment is purified from the resulting PCR product and ligated with the vector DNA. The desired recombinant vectors are prepared by inserting these vectors. The recombinant vector is introduced into E.coli, and a desired colony is selected to prepare a desired recombinant vector. The nucleotide sequence of the target DNA is confirmed by a known method, for example, the dideoxynucleotide chain termination method. After obtaining the DNA encoding the V region of the target anti-GPC 3 antibody, it was integrated into an expression vector containing the DNA encoding the desired antibody constant region (C region). To prepare the anti-GPC 3 antibody for use in the present invention, the antibody gene is incorporated into an expression vector and expressed under the control of an expression control region, such as an enhancer, promoter, or the like. Then, the expression vector is used to transform a host cell and express the antibody. Expression of antibody genes in host cells can be carried out by integrating DNAs encoding the heavy chain (H chain) and light chain (L chain) of an antibody into separate expression vectors and simultaneously transforming the host cells, or by integrating DNAs encoding the H chain and L chain into a single expression vector and transforming the host cells (see WO 94/11523). In addition, the recombinant antibody can be produced using not only the above-described host cell but also a transgenic animal. For example, a fusion gene is prepared by inserting an antibody gene into the middle of a gene encoding a protein inherently produced in milk (e.g., goat. beta. casein). Then, a DNA fragment containing a fusion gene comprising an antibody gene is injected into a goat embryo, and the embryo is implanted into a female goat. The desired antibody can be obtained from the milk of a transgenic goat produced from the goat implanted with the embryo or its offspring. In addition, in order to increase the amount of milk containing the desired antibody produced by the transgenic goat, an appropriate hormone may be administered to the transgenic goat (Ebert, K.M., et al, Bio/Technology (1994)12, 699-702).

Altered antibodies

In the present invention, in addition to the above-mentioned antibodies, artificially altered recombinant antibodies, such as chimeric antibodies, humanized antibodies, and the like, can be used in order to reduce xenoantigenicity to humans. The above-described altered antibodies can be prepared using known methods. Chimeric antibodies can be obtained as follows: the DNA encoding the antibody V region obtained as described above is ligated with the DNA encoding the human antibody C region, and the DNA is inserted into an expression vector. The vector into which the DNA is inserted is integrated into a host cell to produce an antibody. Using such a conventional method, a chimeric antibody suitable for use in the present invention can be obtained. Humanized antibodies, also known as reshaped (reshaped) human antibodies, comprise CDRs of an antibody derived from a non-human mammal, e.g., a mouse, grafted onto CDRs of a human antibody. General genetic engineering methods for obtaining humanized antibodies are also known in the art (see European patent application publication Nos. EP125023 and WO 96/02576). Specifically, a DNA sequence designed to link a CDR of a mouse antibody and a Framework Region (FR) of a human antibody is synthesized by PCR using a plurality of oligonucleotides prepared so as to have overlapping CDR and FR terminal regions as primers (see the method described in WO 98/13388). The framework regions of the human antibody connected by the CDRs are selected so that the CDRs form a suitable antigen-binding site. If desired, the complementarity determining regions of a reshaped human antibody may be substituted for amino acids in the framework regions in the variable regions of the antibody so that the CDRs form the appropriate antigen binding sites (Sato, K., et al, Cancer Res. (1993)53, 851 856). The C region of the chimeric antibody and the humanized antibody may be the C region of a human antibody. For example, Cy1, Cy2, Cy3 and Cy4 may be used for the H chain, and ck and C λ may be used for the L chain. In addition, the human antibody C region may be modified in order to improve the stability or productivity of the antibody. Chimeric antibodies comprise the variable region of an antibody derived from a non-human mammal and the constant region derived from a human antibody. In another aspect, a humanized antibody comprises complementarity determining regions of an antibody derived from a non-human mammal and framework regions and C regions derived from a human antibody. The humanized antibody has reduced antigenicity in humans, and is therefore useful as an active ingredient of the therapeutic agent of the present invention.

Altered antibodies

The antibody used in the present invention is not limited to the whole molecule of the antibody, and may be a fragment of the antibody or a modified form of the antibody as long as it can bind to GPC3 and inhibit the activity of GPC 3. The invention also includes bivalent and monovalent antibodies. Examples of antibody fragments include Fab, F (ab') 2, Fv, Fab/c with 1 Fab and a complete Fc, or single chain Fv (scFv) in which the Fv of the H or L chain are linked by a suitable linker. Specifically, for the preparation of antibody fragments, the antibody may be treated with an enzyme such as papain or pepsin, or a gene encoding the above antibody fragment may be constructed and expressed by an appropriate host cell after insertion into an expression vector (see, for example, Co, M.S. et al, J.Immunol. (1994)152, 2968-2976, Better, M.H.methods in Enzymology (1989)178, 476-496, Academic Press, Inc., Plueckthun, A.Skerra, A.methods in Enzymology (1989)178, 476-496, Academic Press, Inc., Lamoyi, E.E., Methods in Enzymology (1989)121, 652-663, Rousaux, J.et al, Methods in Enzymology (1989) 137, ECH.199137, BTE.132). The scFv can be obtained by linking the H chain V region and the L chain V region of an antibody. In this scFv, the H chain V region and the L chain V region are connected by a linker, preferably a peptide linker (Huston, J.S. et al, Proc. Natl.Acad.Sci.U.S.A. (1988)85, 5879-. The H chain V region and the L chain V region in the scFv may be derived from any of the antibodies described in the present specification. As the peptide linker for linking the V region, for example, any single-chain peptide having 12 to 19 amino acid residues can be used. DNA encoding scFv can be obtained as follows: a single fragment is amplified by PCR using a DNA portion encoding all or a desired amino acid sequence of the DNA encoding the H chain or H chain V region of the antibody and the DNA sequence encoding the L chain or L chain V region of the antibody as templates, and a primer set for defining both ends of the DNA portion. The fragment is then amplified with a combination of DNA encoding the peptide linker portion and a primer pair defining the two ends to be ligated to the H and L chains. Once the DNA encoding scFv is prepared, an expression vector containing the above DNA, and a host cell transformed with the expression vector can be obtained by a conventional method. The scFv can be obtained from the host according to a conventional method. The above antibody fragment can be produced in a host by obtaining and expressing a gene in the same manner as described above. The term "antibody" in the present invention also includes antibody fragments. As the modified antibody, GPC3 antibody linked to various molecules such as polyethylene glycol (PEG) can be used. The term "antibody" in the present invention also includes the above-mentioned altered antibodies. The modified antibody can be obtained by chemically modifying the antibody obtained above. Methods for modifying antibodies have been established in the art.

Furthermore, the antibody used in the present invention may be a bispecific antibody. The bispecific antibody may be a bispecific antibody having antigen binding sites that recognize different epitopes on the GPC3 molecule, or may be an antibody in which one antigen binding site recognizes GPC3 and the other antigen binding site recognizes a cytotoxic substance such as a chemotherapeutic agent, a cell-derived toxin, or the like. In this case, the cytotoxic substance acts directly on the cells expressing GPC3, specifically damages tumor cells, and inhibits the growth of tumor cells. Bispecific antibodies can be prepared by linking HL pairs of 2 types of antibodies. Alternatively, a bispecific antibody can be obtained by fusing hybridomas producing different monoclonal antibodies to prepare a bispecific antibody-producing fused cell. Bispecific antibodies can also be prepared by genetic engineering methods.

Expression and production of recombinant or altered antibodies

The antibody gene constructed as described above can be expressed by a known method to obtain an antibody. In the case of mammalian cells, a common useful promoter, a gene to be expressed, and a poly-A signal sequence downstream of the 3' end of the gene may be functionally linked together and expressed. For example, a human cytomegalovirus immediate early promoter/enhancer may be used as the promoter/enhancer. In addition, other promoters/enhancers that may be used to express the antibodies of the invention include promoters/enhancers of retroviruses, polyomaviruses, adenoviruses, simian virus 40(SV40), or promoters/enhancers of mammalian cell origin such as human elongation factor 1 α (HEF-1 α). Antibodies are readily expressed by the method of Mullgan et al (Nature (1979)277, 108) when the SV40 promoter/enhancer is used, and by the method of Mizushima et al (Nucleic Acids Res. (1990)18, 5322) when the HEFl α promoter/enhancer is used.

In the production of the antibody of the present invention, a eukaryotic cell expression system having the ability to add a sugar chain to an expressed antibody can be used. Eukaryotic cells include, for example, established mammalian cell lines, insect cell lines, animal cells, fungal cells, and yeast cells.

The antibody of the present invention is preferably expressed in mammalian cells, such as CHO, COS, myeloma cells, BHK, Vero, and HeLa cells. The target antibody is prepared by culturing the transformed host cell in vitro or in vivo. The host cell can be cultured by a known method. For example, DMEM, MEM, RPMI1640, or IMDM may be used as the culture medium, and a serum supplement such as Fetal Calf Serum (FCS) may be supplemented.

Isolation and purification of antibodies

The antibody expressed and prepared in the above manner is isolated from the host cell or host animal and purified to homogeneity. The antibodies of the invention can be isolated and purified using affinity columns, for example protein A columns such as Hyper D, POROS, Sepharose F.F (Pharmacia). In addition, any conventional protein isolation and purification method can be used in the present invention. For example, Antibodies can be isolated and purified by appropriately selecting and combining an affinity column such as a protein A column with a chromatography column, a filter, ultrafiltration, salting out, and dialysis methods (Antibodies A Laboratory Manual. Ed Harbor, David Lane, Cold Spring Harbor Laboratory, 1988). The antibody having the desired sugar chain can be separated by a method known to those skilled in the art using a lectin column, for example, according to the method described in WO 02/30954.

Detection of antibody Activity

The antigen binding activity of the Antibodies used in the present invention (Antibodies A Laboratory Manual, Ed Harlow, David Lane, Cold spring harbor Laboratory, 1988), ligand-receptor binding inhibitory activity (Harada, A. et al, International Immunology (1993)5, 681-) 690) can be determined using well known methods. The antigen-binding activity of the anti-GPC 3 antibody of the present invention can be measured by ELISA (enzyme-linked immunosorbent assay), EIA (enzyme immunoassay), RIA (radioimmunoassay) or fluorescent antibody method. For example, EIA proceeds as follows: to the GPC 3-coated plate, a sample containing an anti-GPC 3 antibody, a culture supernatant of anti-GPC 3 antibody-producing cells, or a purified antibody was added. A second antibody labeled with an enzyme such as alkaline phosphatase is added, the plate is incubated, washed, and then an enzyme substrate such as p-nitrophenyl phosphate is added to measure absorbance, thereby evaluating the antigen binding activity.

Pharmaceutical composition

The present invention provides a pharmaceutical composition comprising the GPC3 antibody with a modified sugar chain component of the present invention.

The pharmaceutical composition containing the antibody composition of the present invention is useful for the treatment and/or prevention of diseases associated with cell proliferation, such as cancer, and particularly useful for the treatment and/or prevention of liver cancer. Pharmaceutical compositions containing the antibodies of the invention may be formulated using methods well known to those skilled in the art. The pharmaceutical compositions may be administered parenterally in the form of injections comprising sterile solutions or suspensions in water or other pharmaceutically acceptable liquids. For example, the pharmaceutical composition can be formulated by appropriately combining the antibody with a pharmaceutically acceptable carrier or vehicle such as sterile water or physiological saline, vegetable oil, emulsifier, suspension, surfactant, stabilizer, flavoring agent, excipient, diluent, carrier, preservative, binder, etc., and then mixing in a unit dosage form required for generally accepted formulation. The content of the active ingredient in the pharmaceutical preparation is set to obtain an appropriate dose within the specified range.

The sterile composition for injection can be formulated according to a conventional pharmaceutical method using distilled water for injection as a carrier.

As an aqueous solution for injection, for example, physiological saline, or an isotonic solution containing glucose or other adjuvants such as D-sorbitol, D-mannose, D-mannitol, sodium chloride, etc., optionally in combination with an appropriate cosolvent, for example, an alcohol such as ethanol; and polyols, such as propylene glycol or polyethylene glycol; and nonionic watchSurfactants, e.g. polysorbate 80^TMHCO-50, and the like.

Examples of the oily liquid include sesame oil, soybean oil, and benzyl benzoate or benzyl alcohol may also be used in combination as a cosolvent. Other ingredients that may be included are buffers such as phosphate buffer, sodium acetate buffer; soothing agents, such as procaine hydrochloride; stabilizers, such as benzyl alcohol or phenol; an antioxidant. The formulated injection is usually filled into suitable ampoules.

The route of administration is preferably parenteral, such as by injection, nasal, pulmonary, transdermal, and the like. Systemic or local administration can be carried out by intravenous injection, intramuscular injection, intraperitoneal injection, subcutaneous injection, etc.

The appropriate method of administration may be selected according to the age and condition of the patient. As a single dose of the pharmaceutical composition containing the antibody or the oligonucleotide encoding the antibody, a dose in the range of 0.0001 to 1000mg/kg body weight can be selected. On the other hand, the dose may be selected from the range of 0.001 to 100000mg/kg body weight, but the present invention is not limited to the above numerical range. The dose and the administration method are changed depending on the body weight, age, condition, etc. of the patient, and those skilled in the art can appropriately select them as needed.

The contents of all patents and references cited explicitly in the specification of this application are incorporated herein by reference in their entirety. The contents described in the specification of japanese patent application No. 2004-311356 and the drawings attached thereto, which are the basis of the claims of the present application, are all included as a part of the present specification.

Examples

The present invention will be described in detail with reference to examples, but the present invention is not limited to these examples.

Example 1

Mouse anti-GPC 3 antibodyPreparation of

The soluble GPC3 protein lacking the C-terminal hydrophobic region (amino acids 564-580) was prepared as an immunoprotein for preparing an anti-GPC 3 antibody, and immunization was performed. An autoimmune disease mouse MRL/MpJUmmCrj-lpr/lpr mouse (hereinafter referred to as MRL/lpr mouse, purchased from Charles River, Japan) was used as an immunized animal. Immunization was initiated when mice were 7 weeks or 8 weeks old and the formulation used for the initial immunization was adjusted to a dose of 100 μ g/mouse of soluble GPC 3. Emulsions were prepared using Freund's complete adjuvant (FCA, Becton Dickinson) and injected subcutaneously. After 5 immunizations, the last immunization dose was diluted to 50 μ g/mouse with PBS and administered via tail vein. On day 4 after the final immunization, spleen cells were taken out, mixed with mouse myeloma cells P3-X63Ag8U1 (hereinafter referred to as P3U1, purchased from ATCC) at a ratio of 2: 1, and cell fusion was performed by slowly adding PEG1500(Roche Diagnostics). Hybridomas were screened by ELISA using an immunoplate immobilized with soluble GPC3 core protein. Positive clones were monocloned by limiting dilution. Thus, 11 antibody clones (M3C11, M13B3, M1E7, M3B8, M11F1, L9G11, M19B11, M6B1, M18D4, M5B9, and M10D2) having a strong binding activity to GPC3 were obtained.

Among the anti-GPC 3 antibodies obtained, M11F1 and M3B8 showed particularly strong CDC activity, and therefore 3 Balb/c mice (Charles River, Japan) and 3 MRL/lpr mice were immunized with GST fusion proteins containing M11F1, M3B8 epitope (GC-3) and GST, which are peptides containing Lys from Ala at position 524 to position 563 of GPC3, as immunogens. At the time of initial immunization, GC-3 formulations at 100. mu.g/mouse were emulsified with FCA and injected subcutaneously. After 2 weeks 50. mu.g/mouse of GC-3 preparation were emulsified with Freund's Incomplete Adjuvant (FIA) and injected subcutaneously. After 5 immunizations, the final immunization (50. mu.g/mouse) was performed by tail vein injection and cell fusion to all mice. Hybridomas were screened by ELISA using an immunoplate on which a core protein of the soluble GPC3 protein lacking the C-terminal hydrophobic region (amino acids 564-580) was immobilized. Positive clones were monocloned by limiting dilution. As a result, 5 antibody clones (GC199, GC202, GC33, GC179 and GC194) having a strong binding activity to GPC3 were obtained.

The H chain, L chain variable regions were cloned and sequenced by standard methods. In addition, the CDR regions were determined by comparison with a database of amino acid sequences of known antibodies and investigation of homology. The sequences of the CDR regions are shown in tables 1 and 2.

Example 2

Preparation of mouse-human chimeric antibody against GPC3 antibody

The H chain and L chain variable region sequences of anti-GPC 3 antibody GC33 were linked to the constant region sequences of human IgG1 and the kappa chain. PCR was performed using a synthetic oligonucleotide complementary to the 5 'terminal nucleotide sequence of the antibody H chain variable region having a Kozak sequence and a synthetic oligonucleotide having a NheI site complementary to the 3' terminal nucleotide sequence. The resulting PCR product was cloned into a pB-CH vector obtained by inserting the constant region of human IgG1 into a pBluescript KS + vector (Toyo Boseki). The mouse H chain variable region and the human H chain (γ 1 chain) constant region are linked by a NheI site. The prepared H chain gene fragment was cloned into an expression vector pCXND 3. In addition, PCR was performed using a synthetic oligonucleotide complementary to the 5 '-terminal nucleotide sequence of the variable region of the antibody L having a Kozak sequence and a synthetic oligonucleotide having a BsiWI site complementary to the 3' -terminal nucleotide sequence. The resulting PCR product was cloned into pB-CL vector with human kappa chain constant region inserted into pBluescript KS + vector (Toyo Boseki). The human L variable and constant regions were connected by BsiWI sites. The prepared L chain gene fragment was cloned into an expression vector pUCAG. This vector pUCAG was obtained by digesting pCXN (Niwa et al, Gene 1991; 108: 193-200) with a restriction enzyme BamHI to obtain a 2.6kbp fragment and ligating the fragment to the restriction enzyme BamHI site of pUC 19 vector (Toyo Boseki).

To prepare a mouse-human chimeric antibody expression vector against GPC3, a gene fragment was obtained by digesting the pUCAG vector into which an L chain gene fragment was inserted with a restriction enzyme HindIII (Takara Shuzo). The gene fragment was ligated with the restriction enzyme HindIII site of pCXND3 containing the H chain gene, followed by cloning. The plasmid thus obtained expressed a neomycin resistance gene, DHFR gene, anti-GPC 3 mouse-human chimeric antibody gene (the amino acid sequence of the H chain variable region is shown in SEQ ID NO: 3, and the amino acid sequence of the L chain variable region is shown in SEQ ID NO: 4) in animal cells.

Example 3

Preparation of Low fucose-type anti-GPC 3 chimeric antibody

YB2/0(ATCC, CRL-1662) cells were first used as host cells and cultured in RPMI1640 medium containing 10% FBS. Then electroporated at 25 μ F at 1.4kV and 7.5X 10⁶Mu.g of the chimeric antibody expression vector against GPC3 prepared in example 2 was introduced into YB2/0 cells (ATCCRT-1662) at a concentration of cells/0.75 mL of PBS (-). After a recovery period of 10 minutes at room temperature, the cells treated with electroporation were suspended in 40mL RPMI1640 medium containing 10% FBS. A10-fold dilution was prepared using the same medium, and the cells were dispensed into 96-well culture plates at 100. mu.L/well. In CO₂Incubator (5% CO)₂) After 24 hours of medium culture, geneticin (Invitrogen) was added thereto at a concentration of 0.5mg/mL, and the cells were cultured for 2 weeks. Cell lines expressing the chimeric antibody at a high level were selected by sandwich ELISA using an anti-human IgG antibody, and cell lines expressing the antibody stably were established. Each mouse-human chimeric antibody against GPC3 was purified using Hi Trap protein G HP (Amersham).

Example 4

ADCC Activity assay Using PBMCs from human peripheral blood

Preparation of human PBMC solution

Heparin-added peripheral blood collected from healthy adults was diluted 2-fold with PBS (-) and Ficoll-Paque^TMPLUS (Amersham) layering. After centrifugation (500 Xg, 30 min, 20 ℃ C.), the intermediate layer containing monocytes was separated. After washing the layer 3 times, the cells were suspendedHuman PBMC solutions were prepared in 10% FBS/RPMI.

Preparation of target cells

HepG2 cells (ATCC) and HuH-7 cells (health scientific research resource bank) cultured in 10% FBS/RPMI1640 medium were separated from the culture dish using cell isolation buffer (Invitrogen) at 1X 10⁴The concentration of cells/well was distributed to each well of a 96-well U-bottom plate (Falcon) and cultured for 1 day. After incubation, 5.55MBq of⁵¹Cr, cells were cultured in a 5% carbon dioxide incubator at 37 ℃ for 1 hour. The cells were washed 1 time with the medium, and 50. mu.L of 10% FBS/RPMI1640 medium was added to prepare target cells.

Chromium Release assay (ADCC Activity)

To the target cells, 50. mu.L of antibody solutions prepared at various concentrations were added and reacted for 15 minutes on ice. Then 100. mu.L (5X 10) of human PBMC solution was added⁵Cells/well), cells were cultured in a 5% carbon dioxide incubator at 37 ℃ for 4 hours, after which the plates were centrifuged and radioactivity in 100. mu.L of culture supernatant was measured using a gamma counter. The specific chromium release rate was calculated according to the following formula.

Specific chromium release rate (%) - (A-C). times.100/(B-C)

In the formula, a represents an average value of radioactivity (cpm) of each well; b represents the average value of the radioactivity (cpm) in wells to which 100. mu.L of 2% NP-40 aqueous solution (Nonidet P-40, Code No.252-23, Nacalai Tesque) and 50. mu.L of 10% FBS/RPMI medium were added to target cells; c represents the average value of radioactivity (cpm) in wells containing 150. mu.L of 10% FBS/RPMI medium to target cells. The assay was performed in triplicate, and the mean value and standard deviation of ADCC activity (%) were calculated.

Figures 2 and 3 show ADCC activity of the anti-GPC 3 chimeric antibodies determined using PBMCs. In the figure, the vertical axis represents the cytotoxic activity (%), and the horizontal axis represents the concentration of the added antibody (. mu.g/mL). FIG. 2 shows the results obtained when HepG2 was used as the target cell, and FIG. 3 shows the results obtained when HuH-7 was used as the target cell. Open circles indicate the activity of the chimeric GC33 antibody produced by CHO cells and black circles indicate the activity of the chimeric GC33 antibody produced by YB2/0 cells. The low fucose type GC33 chimeric antibody produced by YB2/0 cells showed stronger ADCC activity compared to the GC33 chimeric antibody produced by CHO cells, clearly indicating that ADCC activity of the anti-GPC 3 antibody was enhanced by the modification of the sugar chain.

Example 5

Construction of antibody-producing cells

Hygromycin B was added to SFMII (+) medium to a final concentration of 1mg/ml, and the fucose transporter deficient cell strain (clone 3F2) was subcultured in the medium. A suspension of 3F2 cells (8X 10) was prepared in Dulbecco's phosphate buffer⁶Individual cells/0.8 mL). To the cell suspension, 25. mu.g of the antibody expression vector (refer to examples 1 and 2) was added, and the cell suspension was transferred into a Gene Pulser cuvette. After the cup was left on ice for 10 minutes, the vector was introduced into the cells by electroporation using GENE-PULSER II under the conditions of 1.5kV and 25. mu. FD. Cells were suspended in 40mL of SFMII (+) medium and transferred to a 96-well flat-bottom plate (Iwaki) at 100. mu.l/well. Placing the plate on CO₂After incubation at 37 ℃ for 24 hours in an incubator, geneticin (Invitrogen, Cat. No. 10131-027) was added to a final concentration of 0.5 mg/mL. The amount of antibody produced by the drug-resistant cells was measured to establish a humanized anti-GPC 3 antibody-producing cell line.

Example 6

Purification of antibodies

The supernatant of the antibody-expressing strain was collected, and the sample was added to Hitrap using a P-1 pump (Pharmacia)^TMrProtein A column (Pharmacia, Cat. No. 17-5080-01). After washing the column with binding buffer (20mM sodium phosphate (pH7.0)), the column was eluted with elution buffer (0.1M glycine-HCl (pH 2.7)). The eluate was immediately neutralized with a neutralization buffer (1M Tris-HCl (pH9.0)). Using the DC protein assay (BIO-RAD, Cat. No.)500-0111) elution fractions of the selected antibodies were pooled and concentrated to around 2mL using Centriprep-YM10(Millipore, Cat. 4304). Then, using a Superdex20026/60 column (Pharmacia) equilibrated with 20mM acetate buffer solution (pH6.0) containing 150mM NaCl, the antibody was isolated by gel filtration. The peaks of the monomer fraction were collected, concentrated using Centriprep-YM10, filtered through a MILLEX-GW 0.22 μm filtration unit (Millipore, Cat. No. SLGV 013SL), and stored at 4 ℃. The absorbance of the purified antibody at 280nm was measured, and the concentration of the antibody was calculated from the molar absorption coefficient.

Example 7

ADCC Activity of humanized anti-GPC 3 antibodies produced by FT-KO cells in vitro

FIG. 4 shows ADCC activity in vitro of anti-GPC 3 antibodies produced by FT-KO cells when human PBMCs are used. The procedure is as described in example 4. In the figure, the vertical axis represents the cytotoxic activity (%), and the horizontal axis represents the concentration of the added antibody (. mu.g/mL). HuH-7 was used as the target cell. Open circles indicate the activity of the anti-GPC 3 antibody produced by wild-type CHO cells, and black circles indicate the activity of the anti-GPC 3 antibody produced by FT-KO cells. The low fucose anti-GPC 3 antibody produced by FT-KO cells showed stronger ADCC activity compared to the anti-GPC 3 antibody produced by wild-type CHO cells, clearly indicating that the ADCC activity of the anti-GPC 3 antibody produced by FT-KO cells was enhanced.

Example 8

Analysis of sugar chain of humanized anti-GPC 3 antibody produced by FT-KO cells

1.2 preparation of amino benzamide-labeled sugar chain (2-AB-labeled sugar chain)

The antibody produced by FT-KO cells of the present invention and the antibody produced by CHO cells as a control sample were treated with N-glycosidase F (Roche diagnostics) to release the sugar chain from the protein (Weitzhandler M. et al, Journal of Pharmaceutical Sciences 83: 12(1994), 1670-. After removal of the protein with ethanol (Schenk B. et al, the journal of Clinical Investigation 108: 11(2001), 1687-. The reagent was removed from the 2-AB-labeled sugar chains by solid phase extraction with a cellulose column, and after centrifugal concentration, purified 2-AB-labeled sugar chains were obtained. Then, the purified 2-AB-labeled sugar chain was treated with beta-galactosidase (Biochemical industry) to obtain a galactose-deficient 2-AB-labeled sugar chain.

2. Analysis of galactose-deficient 2-AB-labeled sugar chain by Normal phase HPLC

The antibody produced by FT-KO cells of the present invention and the antibody produced by CHO cells as a control sample were prepared as galactose-deficient 2-AB-labeled sugar chains by the above-described method, and subjected to normal phase HPLC analysis using an Amide column (TSKgel Amide-80 manufactured by Tosoh), followed by comparison of chromatograms. In the antibody produced by CHO cells, G (0) was the main component, and G (0) -Fuc accounted for about 4% of the peak area. On the other hand, in the antibody produced by FT-KO cells, G (0) -Fuc was the main component, and accounted for 90% or more of the peak area in both cell lines (fig. 5 and table 2). FIG. 6 shows the structure of the G (0) peak and the G (0) -Fuc peak.

TABLE 2

Relative ratio of sugar chains deduced from Normal phase HPLC analysis of galactose-deficient 2-AB sugar chains

Sugar chain	CHO	FT-KO-a	FT-KO-b	FT-KO-c
					G(0)-Fuc	4.0％	92.4％	92.5％	93.2％
G(0)	96.0％	7.6％	7.5％	6.8％

Example 9

Thermal stability analysis of humanized anti-GPC 3 antibody produced by FT-KO cells

Preparation of sample solution for DSC measurement

The external dialysate was 20mol/L sodium acetate buffer (pH6.0) containing 200mmol/L sodium chloride. The dialysis membrane containing 700. mu.g equivalent of the antibody solution was immersed in the dialysate and dialyzed overnight to prepare a sample solution.

2. Measurement of thermal degradation temperature by DSC

After the sample solution and the reference solution (external dialysate) were sufficiently degassed, they were added to a calorimeter respectively, and heat-equilibrated at 20 ℃. Then, DSC measurement was carried out at a scanning speed of about 1K/min at 20 ℃ to 100 ℃. The results are shown as the peak of the degradation peak as a function of temperature (FIG. 7). The thermal degradation temperatures of the antibody produced by CHO cells and the antibody produced by FT-KO cells were found to be the same.

Reference example 1

Humanization of GC33

Antibody sequence data were obtained from published Kabat databases (ftp:// ftp. ebi. ac. uk/pub/databases/Kabat /) and ImmunoGeneTiCs database (IMGT), and homology searches were performed for the H chain variable region and L chain variable region, respectively. As a result, it was found that the H chain variable region has high homology with DN13(Smithson et al, mol. Immunol.1999; 36: 113-124). The L chain variable region was also found to have high homology to IGK mRNA from part of the cds clone K64 from human immunoglobulin kappa light chain VLJ region accession number AB 064105. The signal sequence of accession number S40357 having high homology to AB064105 was used as the signal sequence of the L chain. The humanized antibody was prepared by grafting CDRs into the framework regions of the above antibody.

Specifically, synthetic oligo DNAs of about 50 bases were designed, about 20 bases were hybridized with each other, and the synthetic oligo DNAs were assembled by PCR to prepare genes encoding the respective variable regions. The synthetic oligo DNA was cloned into an expression vector HEFg.gamma.1 into which a constant region of human IgG1 had been cloned or into an expression vector HEFg.kappa.into which a constant region of a human kappa chain had been cloned by performing digestion at the HindIII sequence inserted into the 5 '-terminus of the synthetic oligo DNA and at the BamHI sequence inserted into the 3' -terminus of the synthetic oligo DNA (Sato et al, Mol Immunol.1994; 371-. The H chain and L chain of humanized GC33 constructed as above were designated as ver.a, respectively. Humanized GC33(ver.a/ver.a) in which both H chain and L chain were ver.a had reduced binding activity compared to an antibody (mouse/mouse) with the mouse GC33 variable region. Chimeric antibodies (mouse/ver.a, ver.a/mouse) were prepared by combining the mouse GC33 sequence and ver.a sequence as H chain and L chain, and their binding activity was evaluated. The binding activity of the ver.a/mouse antibody was found to decrease, indicating that the decrease in binding activity due to amino acid substitution was due to the H chain. Modified H chains were then prepared and named ver.c, ver.f, ver.h, ver.i, ver.j, ver.k. All humanized GCs 33 showed binding activity equivalent to that of a chimeric antibody having the variable region of mouse GC 33. The nucleotide sequences of humanized GC33H variable regions ver.a, ver.c, ver.f, ver.h, ver.i, ver.j and ver.k are shown in SEQ ID NO: 74. 75, 76, 77, 78, 79, 80, the amino acid sequence of which is shown in SEQ ID NO: 81. 82, 83, 84, 85, 86, 87. The nucleotide sequence of humanized GC33L variable region ver.a is set forth in SEQ ID NO: 88, the amino acid sequence of which is set forth in SEQ ID NO: 89, respectively. In the humanized GC33H variable regions ver.i, ver.j and ver.k, the 6 th glutamic acid was replaced by glutamine, and the thermal stability of the antibody was significantly increased.

Reference example 2

Changes in the humanized GC33L chain

Regarding deamidation of a protein, it is known that the rate constant of deamidation reaction depends on the primary sequence. Asn-Gly is also known to be particularly susceptible to deamidation (Rocinson et al, Proc. Natl. Acad. Sci. USA 2001; 98; 944-949). SEQ ID NO: asn33 in CDR1 of the humanized GC33L chain ver.a variable region shown at 88 contains the primary sequence Asn-Gly, which is presumed to be susceptible to deamidation.

To evaluate the effect of deamidation of Asn33 on the binding activity of the antibody, an altered antibody was prepared by replacing Asn33 with Asp. Point mutations were introduced using the Quick Change site-directed mutagenesis kit (Stratagene). Specifically, a 50. mu.L reaction solution containing 125ng of sense primer (CTTGTA CAC AGT GAC GGA AAC ACC TAT: SEQ ID NO: 124), 125ng of antisense primer (ATA GGT GTT TCC GTC ACT GTG TAC AAG: SEQ ID NO: 125), 5. mu.L of 10 Xreaction buffer, 1. mu.L of dNTP mix, 10ng of HEFg κ with humanized GC33L strand ver.a cloned, and 1. mu.L of Pfu Turbo DNA polymerase was subjected to 12 cycles consisting of 95 ℃ for 30 seconds, 55 ℃ for 1 minute, and 68 ℃ for 9 minutes. The reaction product was digested with restriction enzyme DpnI at 37 ℃ for 2 hours, and introduced into XL1-Blue competent cells to obtain transformants. The variable region was excised from the clone containing the correct mutation and cloned again into the expression vector HEFg κ. Expression vector HEFg γ 1 containing humanized GC33H chain ver.k was introduced into COS7 cells using Fugene 6 (Roche). The culture supernatant of the cells transiently expressing the modified antibody is collected. Antibody concentration was quantified by sandwich ELISA using anti-human IgG antibody, and the binding activity of the altered antibody was evaluated by ELISA using plates coated with soluble GPC3 core protein. The binding activity of the altered antibody (N33D) in which Asn33 was replaced with Asp disappeared, suggesting that deamidation at Asn33 has a large effect on the binding activity.

Substitution of Gly34 for other amino acid residues was reported to inhibit deamidation of Asn33 (WO 03057881a 1). According to the above method, a series of altered antibodies G34A, G34D, G34E, G34F, G34H, G34N, G34P, G34Q, G34I, G34K, G34L, G34V, G34W, G34Y, G34R, G34S, G34T were prepared by substituting Gly34 with 17 amino acids other than Cys and Met using Quick Change site-directed mutagenesis kit. The binding activity of the antibody was evaluated using the culture supernatant of COS7 cells transiently expressing the antibody. As a result, it was found that the binding activity was maintained even when G34 was substituted with another amino acid residue other than Pro (G34P) and Val (G34V).

The amino acid sequences of the light chain CDR1 of the above-described altered antibody are set forth in SEQ id nos: 90(G34A), SEQ ID NO: 91(G34D), SEQ ID NO: 92(G34E), SEQ ID NO: 93(G34F), SEQ ID NO: 94(G34H), SEQ ID NO: 95(G34N), SEQ ID NO: 96(G34T), SEQ ID NO: 97(G34Q), SEQ ID NO: 98(G34I), SEQ ID NO: 99(G34K), SEQ ID NO: 100(G34L), SEQ ID NO: 101(G34S), SEQ id no: 102(G34W), SEQ ID NO: 103(G34Y), SEQ ID NO: 104(G34R), SEQ ID NO: 105(G34V), SEQ ID NO: 106 (G34P). The amino acid sequences of the above-described altered antibody light chain variable regions are set forth in SEQ ID NOs: 107(G34A), SEQ ID NO: 108(G34D), SEQ ID NO: 109(G34E), SEQ ID NO: 110(G34F), SEQ ID NO: 111(G34H), SEQ id no: 112(G34N), SEQ ID NO: 113(G34T), SEQ ID NO: 114(G34Q), SEQ ID NO: 115(G34I), SEQ ID NO: 116(G34K), SEQ ID NO: 117(G34L), SEQ ID NO: 118(G34S), SEQ id no: 119(G34W), SEQ ID NO: 120(G34Y), SEQ ID NO: 121(G34R), SEQ ID NO: 122(G34V), SEQ ID NO: 123 (G34P).

Reference example 3

Disruption of fucose transporter genes in CHO cells

1. Construction of targeting vectors

(1) Preparation of KO1 vector

PCR was performed from pcDNA3.1/Hygro (Invitrogen) using Hyg5-BH and Hyg3-NT primers to construct a hygromycin resistance gene (Hygr) having the same sequence as the 5 ' portion of the initiation codon of the fucose transporter gene, to which a BamH I site and a TGCGC sequence were added in the 5 ' portion, and a Not I site was added in the 3 ' portion of the region containing the signal to SV40polyA addition, and the Hygr fragment was excised.

Forward primer

Hyg5-BH 5’-GGA TCC TGC GCA TGA AAA AGC CTG AAC TCACC-3’(SEQ ID NO：128)

Reverse primer

Hyg3-NT 5’-GCG GCC GCC TAT TCC TTT GCC CTC GGA CG-3’(SEQ ID NO：129)

The fucose transporter targeting vector ver.1 (hereinafter referred to as KO1 vector) is a vector constructed by inserting the 5 'portion (SmaI at base 2,780 to BamHI at base 4,232 of the nucleotide sequence shown in SEQ ID NO: 126), the 3' portion (SacI at base 4,284 to base 10,934), and the Hygr fragment of the fucose transporter into pMC1DT-A vector (Yagi T, Proc. Natl. Acad. Sci. USA vol.87, 9918-9922, 1990). The KO1 vector is characterized in that since Hygr is not linked to a promoter, Hygr is expressed by the promoter of fucose transporter when homologous recombination occurs. However, if only 1 copy of the vector is introduced into the cell by homologous recombination, Hygr expression does not necessarily achieve resistance to hygromycin B. The KO1 vector was cleaved with NotI and introduced into cells. It is expected that the fucose transporter will delete 41 base pairs of exon 1 including the start codon due to the introduction of the KO1 vector, thereby losing function.

(2) preparation of pBSK-pgk-1-Hygr

The pBSK-pgk-1 vector was prepared by excising the promoter of the mouse pgk-1 gene from pKJ2 vector (Popo H, Biochemical Genetics vol.28, p299-308, 1990) using EcoRI-PstI and cloning it into the EcoRI-Pst I site of pBluescript (Stratagene). PCR was performed from pcDNA3.1/Hygro by using Hyg5-AV and Hyg3-BH primers, adding an EcoT22I site and a Kozak sequence to the 5 'portion of Hygr, adding a BamHI site to the 3' portion of a region containing a signal to SV40polyA addition, and then excising Hygr.

Forward primer

Hyg5-AV 5’-ATG CAT GCC ACC ATG AAA AAG CCT GAA CTCACC-3’(SEQ ID NO：130)

Reverse primer

Hyg3-BH 5’-GGA TCC CAG GCT TTA CAC TTT ATG CTT C-3’(SEQ ID NO：131)

The Hygr (EcoT22I-BamH I) fragment was inserted into the PstI-BamH I site of pBSK-pgk-1 to prepare pBSK-pgk-1-Hygr vector.

(3) Preparation of KO2 vector

The fucose transporter targeting vector ver.2 (hereinafter referred to as KO2 vector) was constructed by inserting the 5 'portion of the fucose transporter (SmaI at base 2,780 to BamHI at base 4,232 of the nucleotide sequence shown in SEQ ID NO: 126), the 3' portion (SacI at base 4,284 to base 10,934) and the pgk-1-Hygr fragment into the pMC1DT-A vector. Unlike the KO1 vector, the pgk-1 gene promoter in the KO2 vector is linked to Hygr, so that resistance to hygromycin B can be obtained even when only 1 copy of the vector is introduced into cells by homologous recombination. The KO2 vector was cleaved with NotI and introduced into cells. It is expected that the fucose transporter will delete 46 base pairs of exon 1 including the start codon due to the introduction of the KO2 vector, thereby losing function.

(4) preparation of pBSK-pgk-1-Puror

The pPUR vector (BD Biosciences) was cleaved with PstI and BamHI, and the digested fragment (Puror) was inserted into the PstI-BamHI site of pBSK-pgk-1, thereby preparing pBSK-pgk-1-Puror.

(5) Preparation of KO3 vector

The fucose transporter targeting vector ver.3 (hereinafter referred to as KO3 vector) was constructed by inserting the 5 'portion of the fucose transporter (SmaI at base 2,780 to BamHI at base 4,232 of the nucleotide sequence shown in SEQ ID NO: 126), the 3' portion (SacI at base 4,284 to base 10,934) and the pgk-1-Puror fragment into the pMC1DT-A vector. Further, the 3' -end of pgk-1-Puror was ligated with a sequence that binds to a primer used for screening as shown below. The KO3 vector was cleaved with NotI and introduced into cells. It is expected that the fucose transporter will delete 46 base pairs of exon 1 including the start codon due to the introduction of the KO3 vector, thereby losing function.

Reverse primer

RSGR-A 5’-GCT GTC TGG AGT ACT GTG CAT CTG C-3’(SEQID NO：132)

The fucose transporter gene is knocked out by using the 3 targeting vectors.

2. Introduction of vector into CHO cell

HT supplements (100X) (Invitrogen, catalog No. 11067-030) and penicillin-streptomycin (Invitrogen, catalog No. 15140-122) were added to CHO-S-SFMII HT (Invitrogen, catalog No. 12052-098) in amounts of 1/100, respectively, relative to the volume of CHO-S-SFMII HT. CHO DXB11 cells were subcultured in this medium (hereinafter referred to as SFMII (+)) which was also used for culturing cells after gene transfer. At 8X 10⁶Individual cells/0.8 mL of the total amount of the cellsCHO cells were suspended in Dulbecco's phosphate buffer (hereinafter referred to as PBS. Invitrogen, Cat. No. 14190-144). To the cell suspension, 30. mu.g of the targeting vector was added, and the cell suspension was transferred into a Gene Pulser cup (4mm) (Bio-Rad, Cat. No. 1652088). After the cup was left on ice for 10 minutes, the vector was introduced into the cells by electroporation using GENE-PULSER II (Bio-Rad, No. 340BR) under the conditions of 1.5kV and 25. mu. FD. After introduction of the vector, the cells were suspended in 200ml of SFMII (+) medium and transferred to 20 96-well flat-bottomed plates (Iwaki, catalog No. 1860-096) at 100. mu.l/well. Placing the plate on CO₂After incubation at 37 ℃ for 24 hours in an incubator, reagents were added.

3. First stage of knock-out

KO1 vector or KO2 vector was introduced into CHO cells, and 24 hours later, the cells were selected using hygromycin B (Invitrogen, Cat. No. 10687-010). Hygromycin B was dissolved in SFMII (+) at a concentration of up to 0.3mg/ml and added at 100. mu.l/well.

4. Screening of homologous recombinants by PCR

(1) Preparation of samples for PCR

Homologous recombinants were screened by PCR. CHO cells used in the screening were cultured in 96-well plates. After removing the culture supernatant, 50. mu.l/well of a cell lysis buffer was added, and the cells were first heated at 55 ℃ for 2 hours and then at 95 ℃ for 15 minutes to inactivate proteinase K, thereby preparing a template for PCR. The buffer for cell lysis in each well included 5. mu.l of 10 × LA buffer II (Takara Bio Inc., with LA Taq added), 2.5. mu.l of 10% NP-40(Roche, Cat. No. 1332473), 4. mu.l of proteinase K (20mg/ml, Takara BioInc., Cat. No. 9033), and 38.5. mu.l of distilled water (Nacalai Tesque, Cat. No. 36421-35).

(2) Conditions for PCR

The PCR reaction mixture contained 1. mu.l of the above PCR sample, 5. mu.l of 10 × LA buffer II, and MgCl₂5. mu.l (2.5mM), 5. mu.l dNTP (2.5mM), and primers (10. mu.l each)M) 2. mu.l, LATaq (5 IU/. mu.l cat. No. RR002B) 0.5. mu.l, and distilled water 29.5. mu.l (50. mu.l total). TP-F4 and THygro-R1 were used as PCR primers in the screening of cells into which the KO1 vector had been introduced, and TP-F4 and THygro-F1 were used as PCR primers in the screening of cells into which the KO2 vector had been introduced.

The PCR conditions of the KO1 vector-introduced cells were such that preheating was performed at 95 ℃ for 1 minute, amplification cycles of 95 ℃ for 30 seconds, 60 ℃ for 30 seconds, and 60 ℃ for 2 minutes were performed 40 cycles, and reheating was performed at 72 ℃ for 7 minutes. The PCR conditions for the KO2 vector-introduced cells included: preheating was carried out at 95 ℃ for 1 minute, amplification cycles of 95 ℃ for 30 seconds and 70 ℃ for 3 minutes were carried out 40 cycles, and reheating was carried out at 70 ℃ for 7 minutes.

The primers used are shown below. In a cell sample in which homologous recombination with the KO1 vector or the KO2 vector has occurred, about 1.6kb or 2.0kb of DNA, respectively, will be amplified. The primer TP-F4 was designed for the 5' genomic region of fucose transporter outside the vector, and the THygro-F1 and THygro-R1 were designed for the Hygr gene in the vector.

Forward primer (KO1, KO2)

TP-F45’-GGA ATG CAG CTT CCT CAA GGG ACTCGC-3’(SEQ ID NO：133)

Reverse primer (KO1)

THygro-R1 5’-TGC ATC AGG TCG GAG ACG CTG TCG AAC-3’(SEQ ID NO：134)

Reverse primer (KO2)

THygro-F 15’-GCA CTC GTC CGA GGG CAA AGGAAT AGC-3’(SEQ ID NO：135)

PCR screening results

A total of 918 cells into which the KO1 vector had been introduced were analyzed, and 1 cell was considered to be a homologous recombinant (homologous recombination rate: about 0.1%). A total of 537 cells introduced with the KO2 vector were analyzed, and 17 cells were considered to be homologous recombinants (homologous recombination rate: about 3.2%).

Southern blot analysis

Further, homologous recombination was confirmed by Southern blotting. A total of 10. mu.g of genomic DNA was prepared from the cultured cells according to standard methods for analysis in Southern blots. PCR was performed using the following two primers to make a 387bp probe corresponding to SEQ ID NO: 126 to 2,500 of base 2,113 of the nucleotide sequence shown in seq id no, for use in Southern blotting. Genomic DNA was cut with BglII.

Forward primer

Bgl-F：5’-TGT GCT GGG AAT TGA ACC CAG GAC-3’(SEQ IDNO：136)

Reverse primer

Bgl-R：5’-CTA CTT GTC TGT GCT TTC TTC C-3’(SEQ ID NO：137)

By cleavage with Bgl II, the blot showed a band of about 3.0kb from the fucose transporter chromosome, a band of about 4.6kb from the chromosome homologously recombined with the KO1 vector, and a band of about 5.0kb from the chromosome homologously recombined with the KO2 vector. The experiment included 1 cell that homologously recombined with the KO1 vector and 7 cells that homologously recombined with the KO2 vector. The only cell obtained by homologous recombination with the KO1 vector was designated 5C1, and later analysis indicated that the cell comprised multiple cell populations. The cells were therefore cloned by limiting dilution before use in the experiment. One of the cells obtained by homologous recombination with the KO2 vector was named 6E 2.

7. Second stage of knockout

Cell lines completely deficient in the fucose transporter gene were established from cells that had undergone homologous recombination with the KO1 vector and the KO2 vector using 3 vectors. The combination of vector and cells was performed as follows. Method 1 combined KO2 vector and 5C1 cells (KO1), method 2 combined KO2 vector and 6E2 cells (KO2), and method 3 combined KO3 vector and 6E2 cells (KO 2). Each vector was introduced into appropriate cells and 24 hours later screening was initiated using hygromycin B, puromycin (Nacalaitesque, cat. No. 29455-12). The final concentration of hygromycin B was 1mg/ml in method 1 and 7mg/ml in method 2. In method 3, the final concentration of hygromycin B is 0.15mg/ml and the final concentration of puromycin is 8. mu.g/ml.

8. Screening of homologous recombinants by PCR

To select cells from method 1, PCR was performed to detect cells that had undergone homologous recombination with both the KO1 vector and the KO2 vector. To screen for cells from method 2, the following PCR primers were designed: TPS-F1 is designed as set forth in SEQ ID NO: 126 in the region of bases 3,924-3,950, the SHygro-RI was designed in the region of bases 4,248-4,274. The PCR primers will amplify 350bp of the fucose transporter gene region with the deletion due to the KO2 vector. Therefore, in the PCR screening in method 2, cells which did not produce an amplification product of 350bp were considered as cells in which the fucose transporter gene was completely deleted. PCR conditions were pre-heating at 95 ℃ for 1 minute, 35 cycles of amplification at 95 ℃ for 30 seconds and 70 ℃ for 1 minute, and re-heating at 70 ℃ for 7 minutes.

Forward primer

TPS-F1：5’-CTC GAC TCG TCC CTA TTA GGC AAC AGC-3’(SEQ ID NO：138)

Reverse primer

SHygro-RI：5’-TCA GAG GCA GTG GAG CCT CCA GTC AGC-3’(SEQ ID NO：139)

In method 3, TP-F4 was used as the forward primer, and RSGR-A was used as the reverse primer. PCR conditions were pre-heating at 95 ℃ for 1 minute, 35 cycles of amplification at 95 ℃ for 30 seconds, 60 ℃ for 30 seconds, and 72 ℃ for 2 minutes, and heating at 72 ℃ for 7 minutes. In a sample of cells in which homologous recombination with the KO3 vector had occurred, about 1.6kb of DNA was amplified. Cells that had undergone homologous recombination with the KO3 vector and cells that had undergone homologous recombination with the KO2 vector were detected by this PCR procedure.

PCR screening results

In method 1, a total of 616 cells were analyzed, 18 of which were homologously recombinant (homologously recombination rate of 2.9%). In method 2, a total of 524 cells were analyzed, of which there were 2 homologously recombinant cells (the rate of homologous recombination was about 0.4%). In method 3, a total of 382 cells were analyzed, of which 7 cells were homologously recombinant (the rate of homologous recombination was about 1.8%).

Southern blot analysis

Is carried out according to the methodSouthern blottingAnd (6) analyzing. As a result, 1 cell completely lacking the fucose transporter gene was found in the analyzed cells. In the first stage of knock-out, PCR andSouthernthe analysis of the blots was consistent, but in the second stage of knockdown, the results were inconsistent. The possible reasons for this are: 1. cells that have undergone homologous recombination independently with KO1 or KO2 in method 1 are mixed together; 2. fucose transporter genes are not a pair (2 genes), but a plurality of pairs (or more than 3 genes); 3. during the culture of the knockout first-stage established cell line, the copy number of the fucose transporter gene remaining in the subcultured cells increases.

11. Analysis of fucose expression

Fucose expression was analyzed by PCR in 26 cells judged to be homologous recombinants. 100ul of a medium containing 5 ug/g/mlmL of lentil lectin,FITC conjugates (vector laboratories, Cat. No. FL-1041), 2.5% FBS, 0.02% sodium azide in PBS (hereinafter referred to as FACS solution), 1X 10 in total will be in an ice bath⁶Individual cells were stained for 1 hour. The cells were then washed 3 times with FACS solution and analyzed with facscalibur (becton dickinson). The results show that onlySouthernFucose transporter gene judged in blot analysisThe expression of fucose was decreased in the completely deleted cells.

From the above results, the following judgment can be made.

Based on the fact that only 1 fucose transporter gene among 616 cells was completely deleted, the frequency of homologous recombination was about 0.16%, which is very low. As described above, PCR andSouthernthere are several possible reasons for the inconsistency of the results of the blot analysis. However, the cell line obtained in method 3 may not contain a mixture of cells that undergo homologous recombination independently with the KO2 vector and the KO3 vector because 2 reagents are used for the selection. In addition, it was not possible to determine by PCR that all other cell lines in which homologous recombination had occurred included multiple cell populations. As described above, if 3 or more than 3 fucose transporter genes are present, the targeting of the genes in the cell will be very difficult. Only when a vector such as KO1 vector, which is difficult to express Hygr, is used and a large number of cells are selected, homologous recombinants can be obtained.

Claims (11)

1. An anti-glypican 3 antibody composition comprising an antibody having a heavy chain variable region comprising a light chain variable region whose amino acid sequence is as set forth in SEQ ID NO: 25, and the amino acid sequence thereof is shown in SEQ ID NO: 26 and the amino acid sequence thereof is shown in SEQ ID NO: 27 and a CDR3 of light chain variable region comprising the amino acid sequence set forth in SEQ ID NO: 45 and the amino acid sequence thereof is shown in SEQ ID NO: 46 and the amino acid sequence thereof is shown in SEQ ID NO: 60, wherein the sugar chain component of the CDR3 is changed so that the percentage of fucose-deficient antibodies is 20% or more.

2. An anti-glypican 3 antibody composition comprising a heavy chain variable region having any one of the following (1) to (7) and an amino acid sequence thereof as set forth in SEQ ID NO: 89 light chain variable region antibody:

(1) the amino acid sequence is shown as SEQ ID NO: 81;

(2) the amino acid sequence is shown as SEQ ID NO: 82;

(3) the amino acid sequence is shown as SEQ ID NO: 83, the heavy chain variable region;

(4) the amino acid sequence is shown as SEQ ID NO: a heavy chain variable region shown at 84;

(5) the amino acid sequence is shown as SEQ ID NO: 85;

(6) the amino acid sequence is shown as SEQ ID NO: 86; or

(7) The amino acid sequence is shown as SEQ ID NO: 87, and a heavy chain variable region as shown in,

characterized in that the sugar chain composition is changed so that the proportion of fucose-deficient antibodies is 20% or more.

3. An anti-glypican 3 antibody composition comprising an antibody having a heavy chain variable region comprising a light chain variable region whose amino acid sequence is as set forth in SEQ ID NO: 25, and the amino acid sequence thereof is shown in SEQ ID NO: 26 and the amino acid sequence thereof is shown in SEQ ID NO: 27, and a CDR3 of the light chain variable region comprising CDR1, CDR2 and CDR3 of any one of (1) to (15):

(1) the amino acid sequence is shown as SEQ ID NO: 90, and the amino acid sequence of the CDR1 is shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(2) the amino acid sequence is shown as SEQ ID NO: 91, and the amino acid sequence of CDR1 shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(3) the amino acid sequence is shown as SEQ ID NO: 92, and the amino acid sequence of the CDR1 is shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(4) the amino acid sequence is shown as SEQ ID NO: 93, and the amino acid sequence of the CDR1 is shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(5) the amino acid sequence is shown as SEQ ID NO: 94, and the amino acid sequence thereof is shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(6) the amino acid sequence is shown as SEQ ID NO: 95, and the amino acid sequence of the CDR1 is as shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(7) the amino acid sequence is shown as SEQ ID NO: 96, and the amino acid sequence thereof is shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(8) the amino acid sequence is shown as SEQ ID NO: 97, and the amino acid sequence of the CDR1 is shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(9) the amino acid sequence is shown as SEQ ID NO: 98, and the amino acid sequence of CDR1 is set forth in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(10) the amino acid sequence is shown as SEQ ID NO: 99, and the amino acid sequence of the CDR1 is shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(11) the amino acid sequence is shown as SEQ ID NO: 100, and the amino acid sequence of the CDR1 is shown as SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(12) the amino acid sequence is shown as SEQ ID NO: 101, and the amino acid sequence of CDR1 is set forth in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(13) the amino acid sequence is shown as SEQ ID NO: 102, and the amino acid sequence thereof is shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(14) the amino acid sequence is shown as SEQ ID NO: 103, and the amino acid sequence of the CDR1 is shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60, CDR 3;

(15) the amino acid sequence is shown as SEQ ID NO: 104, and the amino acid sequence thereof is shown in SEQ ID NO: 46, and the amino acid sequence thereof is set forth in SEQ ID NO: 60 of the CDRs of CDR3 shown in figure 60,

characterized in that the sugar chain composition is changed so that the proportion of fucose-deficient antibodies is 20% or more.

4. An anti-glypican 3 antibody composition comprising a peptide having an amino acid sequence as set forth in SEQ ID NO: 87 and a light chain variable region selected from the group consisting of (1) to (15) below:

(1) the amino acid sequence is shown as SEQ ID NO: 107 or a light chain variable region;

(2) the amino acid sequence is shown as SEQ ID NO: 108;

(3) the amino acid sequence is shown as SEQ ID NO: 109;

(4) the amino acid sequence is shown as SEQ ID NO: 110;

(5) the amino acid sequence is shown as SEQ ID NO: 111;

(6) the amino acid sequence is shown as SEQ ID NO: 112, or a light chain variable region as shown;

(7) the amino acid sequence is shown as SEQ ID NO: 113;

(8) the amino acid sequence is shown as SEQ ID NO: 114;

(9) the amino acid sequence is shown as SEQ ID NO: 115, or a light chain variable region thereof;

(10) the amino acid sequence is shown as SEQ ID NO: 116, the light chain variable region;

(11) the amino acid sequence is shown as SEQ ID NO: 117;

(12) the amino acid sequence is shown as SEQ ID NO: 118;

(13) the amino acid sequence is shown as SEQ ID NO: 119;

(14) the amino acid sequence is shown as SEQ ID NO: 120, or a light chain variable region; and

(15) the amino acid sequence is shown as SEQ ID NO: 121, and a light chain variable region as shown in,

characterized in that the sugar chain composition is changed so that the proportion of fucose-deficient antibodies is 20% or more.

5. The antibody composition according to any one of claims 1 to 4, wherein the sugar chain composition thereof is changed so that the proportion of fucose-deficient antibodies is 50% or more.

6. The antibody composition according to any one of claims 1 to 4, wherein the sugar chain composition thereof is changed so that the proportion of fucose-deficient antibodies is 90% or more.

7. A method for producing an anti-glypican 3 antibody using a cell, characterized in that a gene encoding the anti-glypican 3 antibody as claimed in any one of claims 1 to 6 is introduced into the cell having a reduced ability to add fucose to a sugar chain.

8. The method for producing an anti-glypican 3 antibody as claimed in claim 7, wherein the cell having a decreased ability to add fucose to a sugar chain is a cell lacking in fucose transporter.

9. The method for producing an anti-glypican 3 antibody according to claim 7, wherein the cell has an ability to form a bisecting N-acetylglucosamine structure on a sugar chain.

10. The method for producing an anti-glypican 3 antibody according to claim 8, wherein the cell has an expression vector containing a DNA encoding N-acetylglucosaminyltransferase III.

11. A method for preparing an anti-glypican 3 antibody, the method comprising the steps of:

(a) introducing a gene encoding the anti-glypican 3 antibody as claimed in any one of claims 1 to 4 into a cell having a reduced ability to add fucose to a sugar chain, and

(b) the cells are cultured.