US20070073043A1

Movatterモバイル変換

Info

Publication number: US20070073043A1
Application number: US11/228,206
Authority: US
Inventors: Kenya Shitara; Nobuo Hanai; Mamoru Hasegawa; Hiromasa Miyaji; Yoshihisa Kuwana
Original assignee: Kyowa Hakko Kogyo Co Ltd
Current assignee: KH Neochem Co Ltd
Priority date: 1991-09-18
Filing date: 2005-09-19
Publication date: 2007-03-29
Also published as: US20060057139A1; JP3440104B2; ES2195995T3; DE69233027T2; ATE239078T1; EP1013761B1; US5750078A; US20060058512A1; US7045129B2; US20020026036A1; EP0533199B1; DE69233027D1; US20030166876A1; US6437098B1; PT533199E; US20030095964A1; EP0533199A2; JPH05304989A; DE69233706T2; AU2458192A

Abstract

A humanized chimera antibody, a pharmaceutical composition comprising a humanized chimera antibody and a pharmaceutically acceptable carrier, and a method of treating cancer which comprises administering to a patient a pharmaceutically acceptable amount of the humanized chimera antibody, are disclosed.

Description

The present application is a divisional of application Ser. No. 10/166,626, filed Jun. 12, 2002 (allowed), which was a divisional of application Ser. No. 09/225,322, filed Jan. 5, 1999 (now U.S. Pat. No. 6,437,098), which was a divisional of application Ser. No. 08/454,680, filed May 31, 1995 (now U.S. Pat. No. 5,866,692), which was a divisional of application Ser. No. 08/408,133, filed Mar. 21, 1995 (now U.S. Pat. No. 5,750,078), which was a continuation of application Ser. No. 08/292,178, filed Aug. 17, 1994 (abandoned), which was a continuation of application Ser. No. 07/947,674, filed Sep. 17, 1992 (abandoned), which claims benefit of Japan Hei 3-238375, filed Sep. 18, 1991, the entire contents of each of which is hereby incorporated by reference.

FIELD OF THE INVENTION

This invention relates to a process for the production of humanized chimera antibody. In contrast to mouse monoclonal antibody, humanized chimera antibody does not cause formation of anti-mouse immunoglobulin antibody in the body of a patient. Thus, side effects are reduced or eliminated and half life in blood increases when the chimera antibody is used. Therapeutic effects which are superior to those obtained in the case of using mouse monoclonal antibody can be obtained in the treatment of human cancers and the like.

BACKGROUND OF THE INVENTION

It is known that, when mouse antibodies are administered to humans, they are recognized as foreign substances and cause formation of anti-mouse immunoglobulin antibodies in the human body, and the thus formed antibodies react with the administered mouse antibodies. As a result, side effects occur (J. Clin. Oncol., 2, 881 (1984); Blood, 65, 1349-1363 (1985); J. Natl. Cancer Inst., 80, 932 (1988); Proc. Natl. Acad. Sci. U.S.A., 82, 1242 (1985)), the antibodies are cleared away quickly (J. Nucl. Med., 26, 1011 (1985); Blood, 65, 1349-1363 (1985); J. Natl. Cancer Inst., 80, 937 (1988)) and effects of the antibodies are reduced (J. Immunol., 135, 1530 (1985); Cancer Res., 46, 6489 (1986)). When mouse monoclonal antibody is converted into humanized chimera antibody, human anti-mouse, immunoglobulin antibody form in minimal amounts if at all, and the half life of the chimera antibody in human blood is six times as long as that of mouse monoclonal antibody (Proc. Natl. Acad. Sci. U.S.A., 86, 4220 (1989)). In addition, it is probable that the Fc region of mouse antibody does not fully activate human complement and human effector cells, in comparison with the Fc region of human antibody. For example, the antitumor activity of mouse monoclonal antibody to ganglioside GD₂, which is effected via human effector cells, is improved when the monoclonal antibody is converted into chimera antibody that has the human antibody Fc region (J. Immunol., 144, 1382-1386 (1990)).

Ganglioside is one of the animal cell membrane-constituting constituting glycolipids and is composed of a sugar chain as a hydrophilic side chain, sphingosine as a hydrophobic side chain and fatty acids. It is known that expression of ganglioside varies depending on the type of cells, organs and animal species. In addition, it has been revealed that quantity and quality of the expressed ganglioside change during the canceration process of cells (Cancer Res., 45, 2405 (1985)). For example, it has been reported that gangliosides GD₂, GD₃, GM₂and the like which hardly exist in normal cells were found in the cells of neuroblastoma, lung small cell carcinoma and melanoma belonging to neuroectodermal-origin tumor which is said to be highly malignant (J. Exp. Med., 155, 1133 (1982); J. Biol. Chem., 257, 12752 (1982); Cancer Res., 47, 225 (1987); ibid., 47, 1098 (1987); ibid., 45, 2642 (1985); Proc. Natl. Acad. Sci. U.S.A., 80, 5392 (1983)).

Ganglioside GD₃has been found most frequently in melanoma cells among the neuroectodermal-origin tumors, and anti-ganglioside GD₃monoclonal antibodies (to be referred to as “anti-GD₃monoclonal antibody” hereinafter) belonging to the mouse IgM class and IgG class have been reported (Int. J. Cancer, 29, 269 (1982); J. Biol. Chem., 257, 12752 (1982); Cancer Res., 47, 225 (1987); Acta Neuropathol., 79, 317 (1989); Proc. Natl. Acad. Sci. U.S.A., 77, 6114 (1980); J. Exp. Med., 155, 1133 (1982); Proc. Natl. Acad. Sci. U.S.A., 81, 5767 (1984)).

KM-641 (EEPROM BP-3116) disclosed in EP-A-0 493 686 is an anti-GD₃monoclonal antibody belonging to the mouse IgG3 class, which reacts not only with ganglioside GD₃but also withganglioside 3′,8′-LD1 and is possessed of a broad range of antitumor spectrum. In addition, KY-5641 has stronger binding activities to antigens than anti-GD₃monoclonal antibody R24 which has been disclosed in J. Exp. Med., 155, 1133 (1982) and it shows strong antitumor activities.

The mouse monoclonal antibody R24 to the ganglioside GD₃was once used for the treatment of melanoma, but the administered mouse monoclonal antibody R24 did not fully exert its effect due to the formation of anti-mouse immunoglobulin antibody in the patient's body (Eur. J. Cancer Clin. Oncol., 24,suppl 2, s 65 (1988)).

Consequently, the use of chimera antibody for anti-GD₃monoclonal antibody would be advantageous in that anti-mouse immunoglobulin antibody does not form in the body, side effects are reduced or eliminated, its half life in blood is prolonged and its antitumor effector effect increases, and thus therapeutic effects of the chimera antibody which are superior to those of mouse monoclonal antibody can be obtained in the treatment of human cancers and the like.

Several processes for the production of humanized chimera antibodies are known. Humanized chimera antibody, in which constant regions of the heavy chain (to be referred to as “H chain” hereinafter) and the light chain (to be referred to as “L chain” hereinafter) of mouse monoclonal antibody are converted into human constant regions, is produced in animal cells making use of recombinant DNA techniques. Examples of such processes include a process in which humanized chimera antibody is produced using chromosomal DNA as a gene which encodes mouse H chain variable region (to be referred to as “V_H” hereinafter) and L chain variable region (to be referred to as “V_L” hereinafter) (Morrison et al., Proc. Natl. Acad. Sci. U.S.A., 81, 6851 (1984); Neuberger et al., Nature, 314, 268 (1985); Nishimura et al., Cancer Res., 47, 999 (1987); Dorai et al., J. Immunol., 139, 4232 (1987); Kameyama et al., FEBS letter, 244, 301 (1989)) and another process in which humanized chimera antibody is produced using cDNA (Gillies et al., J. Immunol. Methods, 125, 191 (1989); Liu et al., published International Application in Japan No. 2-501886). Cloning and base sequence determination of hybridoma cell chromosomal DNA which encodes mouse V_Hand V_Lrequire much time and labor in comparison with those of cDNA that encodes mouse V_Hand V_L. Consequently, the process in which cDNA is used for the production of humanized chimera antibody is more desirable than the chromosomal DNA process.

Gillies et al. have succeeded in expressing humanized chimera antibody in animal cells, making use of an expression vector for animal cells having inserted therein a humanized chimera H chain gene obtained by linking mouse V_H-encoding cDNA with human C_H-encoding chromosoal DNA, and a humanized chimera L chain gene obtained by linking mouse V_L-encoding cDNA with human C_L-encoding chromosomal DNA (J. Immunol. Methods, 125, 191 (1989)). However, when an attempt was made to prepare chimera antibodies from several tones of antibodies, a problem was found that there were certain chimera antibodies whose L chains could not be expressed without converting leader sequences. In addition, humanized chimera antibody can be produced more simply when cDNA which encodes human C_Hand C_Lis used instead of the human C_H- and C_L-encoding chromosomal DNA.

In published International Application in Japan No. 2-501886, Liu et al. discloses a process for the expression of humanized chimera antibody in animal cells, which comprises using an expression vector for animal cells having inserted therein a chimera H chain cDNA obtained by linking mouse V_H-encoding cDNA with human C_H-encoding cDNA and a chimera L chain cDNA obtained by linking mouse V_L-encoding cDNA with human C_L-encoding cDNA. According to this process, however, it is necessary to alter the J_Hportion of the V_H-encoding cDNA and the J_Lportion of the V_L-encoding cDNA by means of mutation, because the cDNA which encodes mouse V_Hor V_Lis linked with the human C_H- or C_L-encoding cDNA at the J region in the mouse variable region. In addition, with regard to the chimera L chain prepared using mouse Jk5, leucine which is one of the amino acids of theframework 4 is changed to isoleucine when made into humanized chimera antibody. Although amino acid sequence of complementarity-determining region (to be referred to as “CDR” hereinafter) is especially important for antigen-antibody binding, the amino acid sequence of the framework is also an important factor. For example, Riechmann et al. have prepared CDR graft antibody by grafting a rat antibody CDR into a human antibody framework and reported that binding activity of the antibody was reduced by the framework conversion and the antibody activity increased when amino acid sequence of the framework was partially changed (Nature, 332, 323 (1988)). Consequently, there is a possibility that the binding activity of humanized chimera antibody is undesirably reduced when the antibody is produced by the mouse Jk5-aided process disclosed by Liu et al.

In view of the above, when any mouse antibody is converted into humanized chimera antibody, it has been desired to simply and easily produce humanized chimera antibody in which amino acids of the mouse antibody variable region remain completely unchanged.

OBJECTS OF THE INVENTION

An object of the present invention is to provide a process for the production of humanized chimera antibody by which the chimera antibody is produced easily without changing any of the amino acids of its mouse antibody variable region. Another object of the present invention is to provide a humanized chimera antibody to ganglioside GD₃and a process for the production of such antibody.

SUMMARY OF THE INVENTION

The present invention relates to a process for producing humanized chimera antibody which comprises the steps of:

(1) constructing a cassette vector by inserting a cDNA coding for human antibody C_Hinto an expression vector for animal cell use and establishing a cloning site in the upstream region C_Hof said cassette vector for inserting a cDNA which encodes nonhuman animal V_H;
(2) digesting a cDNA coding for nonhuman animal antibody V_Hwith restriction enzymes;
(3) inserting said cDNA coding for nonhuman animal antibody V_Hinto the cassette vector, using a synthetic DNA which comprises a base sequence corresponding to the 5′-end side of said human antibody C_Hand a base sequence corresponding to the 3′-end side of said nonhuman animal antibody V_Hand is possessed of restriction enzyme recognition sites on both of its ends, thereby constructing a humanized chimera antibody H chain expression vector in which said cDNA coding for human antibody C_Hand said cDNA coding for nonhuman animal antibody V_Hare linked together through said synthetic DNA;
(4) constructing a cassette vector by inserting a cDNA coding for human antibody C_Linto an expression vector for animal cell use and establishing a cloning site in the upstream region of the C_Lof said cassette vector for inserting a cDNA which encodes nonhuman animal antibody V_L;
(5) digesting a cDNA coding for nonhuman animal antibody V_Lwith restriction enzymes;
(6) inserting said cDNA coding for nonhuman animal antibody V_Linto the cassette vector, using a synthetic DNA which comprises a base sequence corresponding to the 5-end side of said human antibody C_Land a base sequence corresponding to the 3′-end side of said nonhuman animal antibody V_Land is possessed of restriction enzyme recognition sites on both of its ends, thereby constructing a humanized chimera antibody L chain expression vector in which said cDNA coding for human antibody C_Land said cDNA coding for nonhuman animal antibody V_Lare linked together through said synthetic DNA;
(7) introducing these expression vectors into host cells to obtain a transformant; and
(8) culturing said transformant in an appropriate culture medium, thereby allowing the transformant to produce and accumulate a humanized chimera antibody, and collecting said humanized chimera antibody from the resulting culture broth.

The cassette vector to be used in the present invention is a vector which is obtained by inserting a cDNA that encodes a constant region of human antibody into an expression vector for animal cell use, in which a cloning site is located in the upstream region of the constant region for inserting a cDNA that encodes a variable region of nonhuman animal antibody. An expression vector for humanized chimera antibody can be constructed easily by inserting a variable region of nonhuman animal antibody into the cloning site of the cassette vector, using a synthetic DNA which comprises a base sequence corresponding to the 5′-end side of a constant region of human antibody and a base sequence corresponding to the 3′-end side of a variable region of nonhuman animal antibody and is possessed of restriction enzyme recognition sites on its both ends.

The present invention also relates to a humanized chimera antibody obtainable by the above-described process, a pharmaceutical composition comprising the humanized chimera antibody and a pharmaceutically acceptable carrier, and a method of treating cancer which comprises administering to a patient a pharmaceutically acceptable amount of said humanized chimera antibody.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 shows a restriction enzyme cleavage map of a 9.3 kb XbaI fragment of KM50 cell chromosomal DNA.

FIG. 2 shows a construction scheme for plasmid pKMB11.

FIG. 3 shows a construction scheme for plasmid pKMD6.

FIG. 4 shows a construction scheme for plasmid pEPKMA1.

FIG. 5 shows a construction scheme for plasmid pEPKMB1.

FIG. 6 shows a construction scheme for plasmid pAGE501.

FIG. 7 shows a construction scheme for plasmid pAGE109.

FIG. 8 shows a construction scheme for plasmid pAGE502.

FIG. 9 shows a construction scheme for plasmid pAGE503.

FIG. 10 shows a construction scheme for plasmid pSEd1.

FIG. 11 shows a construction scheme for plasmid pSE1D2.

FIG. 12 shows a construction scheme for plasmid pIG1SE1d2.

FIG. 13 shows a construction scheme for plasmid pIG1SE1d3.

FIG. 14 shows a construction scheme for plasmid pIG1SE1d4.

FIG. 15 shows a construction scheme for plasmid pPMOL2.

FIG. 16 shows a construction scheme for plasmid pPMOL3.

FIG. 17 shows a construction scheme for plasmid pchCKA7.

FIG. 18 shows a construction scheme for plasmid pchCKB1 (also SEQ ID NO:19).

FIG. 19 shows a construction scheme for plasmid pckCKC1.

FIG. 20 shows a construction scheme for plasmid pChiIgHB2.

FIG. 21 shows a construction scheme for plasmid pChiIgA1.

FIG. 22 shows plasmids pKM641HA3 and pKM641LA2.

FIG. 23 shows plasmid pChi641HA1.

FIG. 24 shows a construction scheme for plasmid pKM641HE1.

FIG. 25 shows a construction scheme for plasmid pKM641HF1.

FIG. 26 shows a construction scheme for plasmid pChi641HA1.

FIG. 27 shows a construction scheme for plasmid pChi641HAM1.

FIG. 28 shows plasmid DChi641LG11.

FIG. 29 shows a construction scheme for plasmid pChi641LG11.

FIG. 30 shows a construction scheme for plasmid pChi641LGM11.

FIG. 31 shows a pattern of SDS-PAGE (4 to 15% gradient gel) of purified anti-GD₃chimera antibody KM-871 (about 5 μg/lane) carried out under reductive condition (A) or non-reductive condition (B), where the lanes starting from the left respectively indicate electrophoretic patterns of molecular weight markers, human IgG standard, mouse anti-GD₃antibody KM-641 and anti-GD₃chimera antibody KM-871.

FIG. 32 is a graph showing reactivity of anti-GD3 chimera antibody KM-871 with ganglioside GD₃-positive G361 and SK-MEL-28 cells measured by fluorescent antibody technique with the cell number on the ordinate and the fluorescence intensity on the abscissa. A dotted line shows reactivity in the absence of the antibody, while a solid line shows reactivity in the presence of M4-871.

FIG. 33 its a graph showing complement-dependent cytotoxicity (CDC) of anti-GD₃chimera antibody KM-871 and anti-GD₃mouse antibody EM-641 against ganglioside GD3-positive G361 and SK-MEL-28 cells with cytotoxicity on the ordinate and an antibody concentration on the abscissa. A blackened bar shows CDC activity of NM-871, while a striped bar shows that of KM-641.

FIG. 34 is a graph showing antibody-dependent cell-mediated cytotoxicity (ADCC) of EM-871 and KM-641 against ganglioside GD₃-positive cell G361 with a ratio of effector cells to target cells on the ordinate and ADCC activity on the abscissa. A blackened bar shows ADCC activity of KM-871, a dotted bar shows ADCC activity of KM-641 and a striped bar shows control (in the absence of the antibody). PEN means polymorphonuclear leukocyte.

FIG. 35 is a graph showing therapeutic effect of KM-871 on transplanted tumors with the tumor size on the ordinate and days after transplantation of tumors on the abscissa, in which closed circle shows effect of anti-Sialyl Le^amonoclonal antibody AMC-462, open square shows that of KM-641 and open triangle shows that of KM-871.

DETAILED DESCRIPTION OF THE INVENTION

1. Construction of Cassette Vector

The cassette vector to be used in the present invention is constructed by inserting a cDNA which encodes a human antibody constant region into an expression vector for animal cell use. Essential components in the expression vector for animal cell use include promoter, enhancer, polyA signal, splicing signal, drug resistance gene as a selection marker (e.g., ampicillin resistance gene, etc.) and the like. Any expression vector for animal cell use may be used for this purpose, as long as it can contain and express the cDNA molecule which encodes a human antibody constant region. For example, pAGE107 (Cytotechnology, 3, 133 (1990)) is useful as such an expression vector. Examples of the promoter and enhancer for use in the expression vector for animal cell use include: SV40 early promoter and enhancer (J. Biochem., 101, 1307 (71987)); LTR promoter and enhancer of Moloney mouse leukemia virus (Biochem. Biophys. Res. Comun., 149, 960 (1987)); and immunoglobulin H chain promoter (Cell, 41, 479 (1985)) and enhancer (Cell, 33, 717 (1983)). The immunoglobulin H chain promoter and enhancer can be prepared using appropriate antibody-producing hybridoma cells, such as rat hybridoma KM50 cells which produce anti-human serum albumin antibody as disclosed in JP-A-60-258128 (the term “JP-A” as used herein means an “unexamined published Japanese patent application”). The following describes processes for the preparation of the immunoglobulin H chain promoter and enhancer making use of KM50 cells.

Each chromosomal DNA is obtained from cultured KY50 cells, and P3X63Ag8U.1 (to be referred to as “P3U1” hereinafter) cells (ATCC CRL1597) which are to be fused with KM50 and rat kidney cells in accordance with the procedure disclosed in Molecular Cloning (2nd. ed., Cold Spring Harbor Laboratory Press, 1989, p9.14). Next, a DNA fragment containing immunoglobulin promoter and enhancer and a gene of the variable region of activated immunoglobulin H chain, in which DNA rearrangement has been induced, is isolated from the chromosomal DNA extracted from KT450 cells, in accordance with the procedure discloses in FEBS letter 244, 301 (1989). The immunoglobulin promoter and enhancer are cut out from the thus isolated DNA fragment and inserted into the aforementioned expression vector for animal cell use. Plasmid pIg1SE1d-d4 is an illustrative example of the animal cell expression vector which contains the immunoglobulin H chain promoter and enhancer.

Next, a cloning site is established in the upstream region of a human constant region of a cassette vector, for inserting a cDNA which encodes a variable region of nonhuman animal antibody. Into the thus established cloning site is inserted a cDNA which encodes a variable region of nonhuman animal antibody, using a synthetic DNA which comprises a base sequence corresponding to the 5′-end side of a constant region of human antibody and a base sequence corresponding to the 3′-end side of a variable region of nonhuman animal antibody and is possessed of restriction enzyme recognition sites on both of its ends. In this way, a humanized chimera antibody expression vector is constructed in which the cDNA coding for human antibody constant region and the cDNA coding for the variable region of nonhuman animal antibody are linked together through the synthetic DNA. The synthetic DNA to be used may be prepared using a DNA synthesizer, based on the base sequence which corresponds to the 5′-end side of a constant region of human antibody and the base sequence that corresponds to the 3′-end side of a variable region of nonhuman animal antibody. Illustrative examples of cloning site-containing cassette vectors include a cassette vector pChiIaHB2 which is used for the construction of an expression vector for the expression of humanized chimera antibody H chain and a cassette vector pChiIgLA2 which is used for the construction of an expression vector for the expression of humanized chimera antibody L chain.

A cassette vector for use in the construction of an expression vector for the expression of humanized chimera antibody H chain is constructed, for example, by cutting out a human C_H-encoding cDNA-containing fragment, from an ApaI site in the vicinity of the 5′-end of the cDNA to its 3′-end, and inserting the fragment into an appropriate expression vector for animal cell use such as plasmid pIg1SE1d4 or the like. Then, a cloning site is established in the thus constructed cassette vector for inserting a cDNA which encodes a V_Hof nonhuman animal antibody. Into the thus established cloning site is then inserted a cDNA fragment encoding a nonhuman animal antibody V_H, which is obtained by digesting a V_H-encoding cDNA with an appropriate restriction enzyme, using a synthetic DNA molecule which comprises a base sequence corresponding to the 5′-end side (5′-end to ApaI site) of a human antibody C_Hand a base sequence corresponding to the 3′-end side of a nonhuman animal antibody V_Hand is possessed of restriction enzyme recognition sites on both of its ends. In this way, an expression vector for use in the expression of humanized chimera antibody H chain is easily obtained without altering amino acid sequence of the expressed V_H.

A cassette vector for constructing of an expression vector for the expression of humanized chimera antibody L chain may be constructed for example by introducing an EcoRV site into the vicinity of 5′-end side of a human C_L-encoding cDNA by means of mutation, cutting out a fragment from the resulting human cDNA from the EcoRV site to the 3′-end and inserting the fragment into an appropriate expression vector such as plasmid pIg1SE1d4 or the like. Then, a cloning site is established in the thus constructed cassette vector for inserting a cDNA-which encodes a nonhuman animal antibody V_L. Into the thus established cloning site is then inserted a cDNA fragment encoding a nonhuman animal antibody V_L, which is obtained by digesting a V_L-encoding cDNA with an appropriate restriction enzyme, using a synthetic DNA which comprises a base sequence corresponding to the 5′-end side (5′-end to EcoRV site) of a human antibody C_Land a base sequence corresponding to the 3′-end side of a nonhuman animal antibody V_Land is possessed of restriction enzyme recognition sites on both of its ends. In this way, an expression vector for use in the expression of humanized chimera antibody L chain is easily obtained without altering amino acid sequence of the expressed V_L.

Examples of the cDNAs which encode the human C_Hand human C_Ldescribed above are disclosed, for instance, in Cell 22, 197 (1982). Such cDNAs can be prepared from human antibody-producing myeloma cells, humanized monoclonal antibody-producing hybridoma cells, humanized chimera antibody-producing cells (SP2-PC chimera; FEBS Letters, 244, 301 (1-89)) and the like, in accordance with known procedures disclosed for instance in Proc. Natl. Acad. Sci. U.S.A. 62, 7025 (1985) and ibid., 79 7025 (1985). That is, cDNA is synthesized using mRNA extracted from the above-described cells, in accordance with the procedure disclosed in Molecular Cloning 2nd. ed.; 1989, p8.1. A library is prepared from the thus synthesized cDNA using a phage vector or a plasmid vector, in accordance with the procedure disclosed in Molecular Cloning 2nd. ed.; 1989, p8.1, 1.53. Next, a recombinant phage or a recombinant plasmid which contains human C_H-encoding cDNA or human C_L-encoding cDNA is obtained from the thus prepared library using a human antibody constant region or a human antibody variable region as a probe, in accordance with the procedure disclosed in Molecular Cloning 2nd. ed.; 1989, p8.1, 1.53. Base sequences of the human C_H-encoding cDNA and the human C_L-encoding cDNA are determined in accordance with the procedure disclosed in Molecular Cloning, 2nd. ed.; 1989, p13.1. Introduction of an appropriate restriction enzyme recognition site into the human C_L-encoding cDNA, for example insertion of an EcoRV recognition site into a region in the vicinity of the 5′-end of the cDNA, may be effected in accordance with the procedure disclosed in Molecular Cloning, 2nd. ed.; 1989, p15.1.

2. Production of Humanized Chimera Antibody

Firstly, cDNAs which encode V_Hand V_Lof nonhuman animal antibody, such as mouse anti-GD₃monoclonal antibody, are prepared in the following manner.

That is, cDNA is synthesized using mRNA extracted from appropriate hybridoma cells which produce mouse anti-GD₃monoclonal antibody, such as mouse anti-GD₃monoclonal antibody KM-641 (FERM BP-3116). A library is prepared from the thus synthesized cDNA using a phage vector or a plasmid vector. Next, a recombinant phage or a recombinant plasmid which contains V_H-encoding cDNA or V_L-encoding cDNA is obtained from the thus prepared library using a constant region or a variable region of nonhuman antibody, such as mouse antibody, as a probe. Base sequences of the V_H-encoding cDNA and the V_L-encoding cDNA are determined in accordance with the aforementioned procedure.

A fragment of the V_H-encoding cDNA, ranging from the 5′-end to an appropriate restriction enzyme site near the 3′-end (to be referred to as “site A” hereinafter), is cut out and inserted into the cloning site of the aforementioned cassette vector, using a synthetic DNA which comprises a base sequence corresponding to the 5′-end side of a human antibody C_Hand a base sequence corresponding to the 3′-end side (from 3′-end to site A) of a nonhuman animal antibody V_Hand is possessed of restriction enzyme recognition sites on both of its ends. In this way, an expression vector for use in the expression of humanized chimera antibody H chain is constructed by linking the human antibody C_H-encoding cDNA with the nonhuman antibody V_H-encoding cDNA through the synthetic DNA. In the same way, a Fragment of the V_L-encoding cDNA, ranging from the 5′-end to an appropriate restriction enzyme site near the 3′-end (to be referred to as “site B” hereinafter), is cut out and inserted into the cloning site of the aforementioned cassette vector, using a synthetic DNA molecule which comprises a base sequence corresponding to the 5′-end side of a human antibody C_Land a base sequence corresponding to the 3′-end side (from 3′-end to site B) of a nonhuman animal antibody V_Land is possessed of restriction enzyme recognition sites on both of its ends. In this way, an expression vector for use in the expression of humanized chimera antibody L chain is constructed by linking the human antibody C_L-encoding cDNA with the nonhuman antibody V_L-encoding cDNA through the synthetic DNA.

A transformant which is capable of producing humanized chimera antibody is obtained by transforming appropriate host cells with the thus prepared expression vectors for use in the expression of the H chain and L chain of humanized chimera antibody.

Any type of cells may be used as host cells for use in the introduction of the humanized chimera antibody expression vectors, as long as these cells are capable of expressing the humanized chimera antibody. Illustrative examples of such host cells include mouse SP2/0-Ag14 cells (ATCC CRL1581; to be referred to as “SP2/0 cells” hereinafter), mouse P3X63-Ag8.653 (ATCC CRL1580) and CHO cells which are deficient in dihydrofolate reductase gene (to be referred to as “dhfr” hereinafter) (Urlaub et al., Proc. Natl. Acad. Sci. U.S.A., 77, 4216 (1980)).

Introduction of the expression vectors for use in the expression of the H chain and L chain of humanized chimera antibody into host cells may be effected for example by the electroporation technique disclosed in JP-A-2-257891. A transformant capable of producing the humanized chimera antibody may be selected using RPMI1640 medium supplemented with G418 and fetal calf serum, in accordance with the procedure disclosed in JP-A-2-257891. A transformant, KM-871, which produces humanized chimera antibody that reacts with ganglioside GD₃is an illustrative example of the transformant capable of producing humanized chimera antibody. KM-871 has been deposited on Aug. 13, 1991, with Fermentation Research Institute, Agency of Industrial Science and Technology of 1-3, Higashi 1-chome, Tsukuba-shi, Ibaraki, Japan under the Budapest Treaty, and has been assigned the accession number FERM BP-3512.

For the cultivation of the thus-obtained transformants, any medium can be used as long as the desired antibody can be produced and accumulated in the medium. An example of such medium is RPM!1640 medium supplemented with G418 and fetal calf serum. The transformants may be inoculated into 200 μl to 100 ml of the above-mentioned medium to give a cell concentration of 1×10⁵to 1×10⁷cells/ml and cultivated at 37° C. in a 5% CO₂incubator for 1 to 7 days. The desired chimera antibody is produced and accumulated in the culture medium.

Activity of the humanized chimera antibody in the culture broth is measured by enzyme-linked immunosorbent assay (ELISA method; E. Harlow et al., Manual of Antibody Experiments, Cold Spring Harbor Laboratory Press, 1988). Productivity of the humanized chimera antibody in the transformant can be improved making use of a dhfr amplification system in accordance with the procedure disclosed in JP-A-2-257891.

The humanized chimera antibody thus produced can be purified from supernatant fluid of the aforementioned cultured mixture making use of a protein A column (E. Harlow et al., Manual of Antibody Experiments, Cold Spring Harbor Laboratory Press, 1988). Illustrative examples of humanized chimera antibodies obtained in this way include those which react with ganglioside GD₃, such as humanized chimera antibody EM-871 and the like.

Reactivity of humanized chimera antibody is measured by ELISA method. The molecular weight of the H chain, the L chain or the entire molecule of purified humanized chimera antibody is measured by means of polyacrylamide gel electrophoresis (SDS-PAGE), Western blotting method (E. Harlow et al., Manual of Antibody Experiments, Cold Spring Harbor Laboratory Press, 1988) or the like.

Binding activity, or avidity, of the humanized chimera antibody to ganglioside GD₃to a cultured cancer cell line is measured by means of the fluorescent antibody technique, the ELISA method or the like. Complement-dependent cytotoxicity (CDC activity) and antibody-dependent cell-mediated cytotoxicity (ADCC activity) of humanized chimera antibody to a cultured cancer cell line are measured in accordance with the procedures disclosed in Menekigaku Jikken Nyumon, (Manual of Immunological Experiments) Matsuhashi et al., Gakkai Shuppan Center, Japan, 1981).

The humanized chimera antibodies according to the present invention can be used alone as an anticancer agent. They may be formulated into an anticancer composition together with at least one pharmaceutically acceptable carrier. For instance, the humanized chimera antibodies are dissolved in physiological saline, an aqueous solution of glucose, lactose or mannitol and the like. The powder of the humanized chimera antibodies for injection can be prepared by lyophilizing the humanized chimera antibodies in accordance with the conventional method and mixing the lyophilized products with sodium chloride. The anticancer composition may further contain additives conventionally used well known in the art of medical preparation, for example, pharmaceutically acceptable salts.

The humanized chimera antibodies according to the present invention can be administered in the form of the above-described anticancer composition to mammals including human in a dose of 0.2 to 20 mg/kg/day. The dose may vary depending on the age, condition, etc. of patients. The administration of the anticancer composition can be effected by intraveous infection once a day (single administration or consecutive administration) or intermittently one to three times a week or once every two to three weeks.

The antincancer composition is expected to be useful for treating cancer such as melanoma, neuroblastoma and glioma.

The following examples and reference examples are provided to further illustrate the present invention. It is to be understood, however, that the examples are for purpose of illustration only and are not to be construed to limit the invention.

EXAMPLE 1

Construction of Cassette Vector

1. Isolation of Promoter and Enhancer Genes of K-50 Cell-Derived Immunoglobulin H Chain

(1) Preparation of Chromosomal DNA from KM50 Cells, P3U1 Cells and Rat Kidney

Chromosomal DNA was prepared in the following manner in accordance with the procedure disclosed in Molecular Cloning, Maniatis et al., 1989, p9.16.

1.2×10⁸KM50 cells, 2×10⁸P3U1 cells (ATCC CRL1597) and 1.6 g of rat kidney (a kidney sample frozen at −80° C. was smashed thoroughly using a mallet) were each suspended in 2 ml of a buffer solution (pH 7.5) containing 10 mM Tris-HCl, 150 mM sodium chloride and 10 mM sodium ethylenediaminetetraacetate (to be referred to as “EDTA” hereinafter). To this suspension were added 0.8 mg of Proteinase K (Sigma Chemical Co.) and 10 mg of sodium lauryl sulfate (to be referred to as “SDS” hereinafter). After incubation at 37° C. for 10 hours, the resulting mixture was extracted with the same volume of phenol (once), chloroform (twice) and ether (once) in this order, and the extract was dialyzed for 10 hours against a buffer solution (pH 7.5) containing 10 mM Tris-HCl and 1 mM EDTA. A DNA solution was recovered from the dialysis tube and Ribonuclease A (Sigma Chemical Co.) was added thereto to give a final concentration of 20 μg/ml. After incubating at 37° C. for 6 hours to decompose RNA completely, the resulting solution was mixed with 15 mg of SDS and 1 mg of Proteinase K and incubated at 37° C. for 10 hours. The thus treated solution was extracted with the same volume of phenol, chloroform and ether (twice for each) in this order, and the extract was dialyzed for 10 hours against a buffer solution (pH 7.5) containing of 10 mM Tris-HCl and 1 mM EDTA. The DNA solution was recovered from the dialysis tube and used as a chromosomal DNA sample. A DNA concentration of each sample was determined by measuring the absorbance at 260 nm and, as a result, it was found that 1.6 mg, 1.5 mg and 1.9 mg of chromosomal DNA was obtained from 1.2×10³KM50 cells, 2×10⁸P3U1 cells and 1.6 g of rat kidney, respectively.

(2) Identification of Activated Immunoglobulin H Chain Gene in KM50 Cells by Southern Blotting

A 3 μg portion of each of the chromosomal DNA samples obtained in the above step (1) from KM50 cells, P3U1 cells and rat kidney was dissolved in 25 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. Each of the thus prepared solution was mixed with 15 units of XbaI (Takara Shuzo Co., Ltd.; all restriction enzymes used in the following experiments were purchased from the same company) and incubated at 37° C. for 2 hours to cleave the chromosomal DNA at the XbaI site. The reaction mixture was subjected to agarose gel electrophores is, resulting DNA fragments were transferred to a nitrocellulose filter in accordance with the method of Southern et al. (J. Mol. Biol., 98, 503, (1975)) and then subjected to hybridization in the known method (Rameyama et al., FEBS Letters, 244, 301-306 (1989)) using a mouse JH probe which is disclosed in the FEBS Letters article. A band equivalent to about 9.3 kb was observed only in the DNA sample of KM50 cells. In consequence, it was considered that the XbaI fragment of immunoglobulin DNA found in this band contained the activated immunoglobulin H chain gene derived from KM50 cells.

(3) Preparation of KM50 Cell Chromosomal DNA Library

A 60 μg portion of the K-50 cell chromosomal DNA obtained in the above step (2) was dissolved in 250 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 150 units of XbaI and incubated at 37° C. for 2 hours to cleave the chromosomal DNA at the XbaI site. The reaction mixture was subjected to agarose gel electrophoresis and a 9.3 kb-equivalent fraction was recovered as about 2 μg of 9.3 kb DNA sample of KY-50 cells, making use of the DEAE paper method (Maniatis et al., Molecular Cloning, 1989, p6.24). Separately, a 3 μg portion of lambda-ZAP (Stratagene Cloning Systems) to be used as a vector was dissolved in 200 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 50 units of XbaI and incubated at 37° C. for 2 hours to cleave the DNA at the XbaI site. The resulting reaction mixture was extracted with phenol-chloroform and then treated with ethanol to precipitate and recover about 3 μg of DNA. The thus recovered DNA sample was dissolved in a 100 μl of 100 mM Tris-HCl buffer (pH 7.5), and the resulting solution was mixed with 1 unit of alkaline phosphatase (Takara Shuzo Co., Ltd.) to effect dephosphorylation of restriction enzyme cleavage ends of the vector DNA. The resulting reaction mixture was extracted with phenol-chloroform and then treated with ethanol to precipitate and recover 2 μg of DNA. The thus recovered DNA sample was dissolved in 10 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5) and 1 mM EDTA to serve as a vector DNA sample. Next, 0.2 μg of the thus prepared vector DNA sample and 0.2 μg of the KM50 cell-derived 9.3 kb DNA sample were dissolved in 5 μl of a buffer solution containing 66 mM Tris-HCl (pH 7.5), 6.6 mM magnesium chlorite, 10 mM dithiothreitol (to be referred to as “DTT” hereinafter) and 0.1 mM adenosine triphosphate (to be referred to as “ATP” hereinafter)(to be referred to as “T4 ligase buffer” hereinafter). The resulting solution was mixed with 175 units of T4 DNA ligase (Takara Shuzo Co., Ltd.) and incubated at 4° C. for 3 days. A 2 μl portion of the resulting reaction mixture was subjected to lambda phage packaging in the known method (Maniatis et al., Molecular Cloning, 1989, p2.95) using GigaPack Gold purchased from Stratagene Cloning Systems.E. coliBE4 cells were infected with this phage to obtain 200,000 phage clones. 100,000 out of these phage clones were fixed on nitrocellulose filters in the known method (Maniatis et al., Molecular Cloning, 1989, p2.112).

(4) Selection of Recombinant DNA Containing a Gene of the Activated (Anti-Human Serum Albumin) Immunoglobulin H Chain Variable Region in KM50 Cells

Two clones showing strong reaction with the³²P-labeled mouse JH probe at 65° C. were isolated from the 100,000 phage clones prepared in the above step (3) in accordance with the procedure of Rameyama et al. (FEBS Letters, 44, 301-306, 1989). When the phage DNA was recovered in the conventional manner (Maniatis et al., Molecular Cloning, 1989, p2.118-2.169), it was found that the 9.3 kb XbaI fragment of the KM50 cell-derived chromosomal DNA was incorporated into the phage DNA.

(5) Base Sequence of the Gene of the Activated (Anti-Human Serum Albumin) Immunoglobulin H Chain Variable Region in KM50 Cells

Restriction enzyme cleavage mans of the two clones obtained in the above step (4) was prepared by digesting them with various restrict on enzymes and it was found that completely the same DNA fragment (9.3 kb) has been inserted into these clones (FIG. 1). Next, base sequence of a part of the 9.3 kb DNA fragment, which was considered to contain the promoter and variable regions of the rat immunoglobulin H chain, was determined in accordance with the Sanger method (Sanger et al., Proc. Natl. Acad. Sci. U.S.A., 74, 5463 (1977); M13 Cloning and Sequencing Handbook, Amersham). In SEQ ID NO: 1, a region containing octamer sequences such as ATGCAAAT and TATA box sequences such as TTGAAAA and the like can be regarded as the immunoglobulin promoter region.

2. Construction of Heterologous Protein Expression Vector Using Promoter and Enhancer of the Activated (Anti-Human Serum Albumin) Immunoglobulin H Chain Variable Region in KM50 Cells

(1) Construction of pKMB11

A 1 μg portion of the 9.3 kb fragment of the immunoglobulin H chain variable region gene obtained in 1-(5) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of BglII and 10 units of HindIII and incubated at 37° C. for 2 hours to cleave the DNA fragment at the BalII and HindIII sites. The resulting reaction mixture was subjected to agarose gel electrophoresis and 0.01 μg of a DNA fragment containing 0.8 kb immunoglobulin promoter was recovered. Separately, a 1 μg portion of a plasmid pBR322-BglII (Kuwana et al., FEBS Letters, 219, 360 (1987)) was dissolved in 30 μl of a buffer solution containing 10 MM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of BglII and 10 units of HindIII and incubated at 37° C. for 2 hours to cleave the plasmid at the BglII and HindIII sites. The resulting reaction mixture was subjected to agarose gel electrophoresis, a DNA fragment of about 4.2 kb was recovered. A 0.1 μg portion of the thus obtained pBR322-BglII derived DNA fragment of about 4.2 kb and 0.01 μg of the immunoglobulin promoter-containing DNA fragment were dissolved in 20 μl of a T4 ligase buffer, and the resulting solution was mixed with 175 units of T4 DNA ligase (Takara Shuzo Co., Ltd.) and incubated at 4° C. for 24 hours. Using the resulting reaction mixture, transformation ofE. coliHB101 (J. Mol. Biol., 41, 459 (1969)) was carried out in accordance with the method of Scott et al. (M. Shigesada, Saibo Kogaku, 2, 616 (1983)) to isolate a colony having ampicillin resistance (to be referred to as “ApR” hereinafter). Plasmid DNA was recovered from the colony to obtain pKMB11 as shown inFIG. 2.

(2) Construction of pKMD6

In order to establish an appropriate restriction enzyme recognition site in downstream region of the immunoglobulin promoter, the plasmid pKMB311 constructed in the above step (1) was digested with nuclease BAL31 from the NcoI site. A 10 μg portion of the plasmid pKMB11 was dissolved in 100 ∥l of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM potassium chloride. The thus prepared solution was mixed with 30 units of NcoI and incubated at 37° C. for 2 hours to cleave the plasmid at the NcoI site. The resulting reaction mixture was extracted with phenol and chloroform and treated with ethanol. The thus precipitated DNA fragments were dissolved in 100 μl of BAL31 buffer which contained 20 mM Tris-HCl (pH 8.0), 600 mM sodium chloride, 12 mM calcium chloride, 12 mM magnesium chloride and 1 mM EDTA, and the resulting solution was mixed with 0.25 unit of BAL31 (Bethesda Research Laboratories, Inc. (BRL)) and incubated at 37° C. for 5 seconds. The reaction was stopped by extracting the reaction mixture with phenol. After extraction with chloroform and precipitation with ethanol, 1 μg of DNA was recovered. A 0.1 μg portion of the thus obtained DNA sample and 0.01 μg of a synthetic DNA linker SalI were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 175 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the resulting reaction mixture, transformation ofE. coliHB101 was carried out in accordance with the method of Scott et al. to isolate an ApR colony. Plasmid DNA was recovered from the colony to obtain pKMD6 as shown inFIG. 3. The base sequence of the BAL31-digested portion of this plasmid was determined in accordance with the Sanger method and it was found that bases up to the third base (the 303 position base in the SEQ ID NO: 1) upstream from the initiation codon ATG of the immunoglobulin gene.

(3) Construction of pEPKMA1, pEPKMB1 and pAGE501

Since original promoter and enhancer of the immunoglobulin gene are separated from each other, it is necessary to construct a vector in which the promoter and enhancer are connected together so that it can be used as a vector for the expression of a heterologous protein. The following manipulation was carried out to construct such vectors.

A 1 μg portion of the 9.3 kb fragment of the immunoglobulin H chain variable region gene obtained in 1-(5) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of EcoRV and 10 units of XbaI and incubated at 37° C. for 2 hours to cleave the DNA fragment at the EcoRV and XbaI sites. The resulting reaction mixture was subjected to agarose gel electrophoresis and 0.1 μg of a DNA fragment of about 1 kb containing the immunoglobulin enhancer region was recovered. Separately, a 1 μg portion of the plasmid pKMD6 obtained in the above step (2) was dissolved in 100 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of BalI and incubated at 37° C. for 2 hours to cleave the plasmid at the BglII site. The resulting reaction mixture was extracted with phenol and chloroform and precipitated with ethanol. The thus precipitated DNA fragments were dissolved in 40 μl of DNA polymerase I buffer containing 50 mM Tris-HCl (pH 7.5), 10 nM magnesium chloride, 0.1 mM dATP (deoxyadenosine triphosphate), 0.1 mM dCTP (deoxycytidine triphosphate), 0.1 mM dGTP (deoxyguanosine triphosphate) and 0.1 mM. dTTP (deoxythymidine triphosphate). The resulting solution was mixed with 6 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 90 minutes to convert the adhesive 5′-end formed by the BglII digestion into blunt end. The reaction was stopped by extracting the reaction mixture with phenol. After extraction with chloroform and precipitation with ethanol, the resulting DNA fragments were dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of HindIII and incubated at 37° C. for 2 hours to cleave the DNA fragment at the HindIII site. The resulting reaction mixture was subjected to agarose gel electrophoresis, 0.1 μg of a DNA fragment of about 0.8 kb containing the immunoglobulin promoter region was recovered. Next, a 0.2 μg portion of plasmid pUC18 (Messing, Methods in Enzymology, 101, 20 (1983)) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mm sodium chloride. The thus prepared solution was mixed with 10 units of HindIII and 10 units of XbaI and incubated at 37° C. for 2 hours to cleave the plasmid at the HindIII and XbaI sites. The resulting reaction mixture was subjected to agarose gel electrophoresis, 0.1 μg of a DNA fragment of about 2.7 kb was recovered. A 0.1 μg portion of the thus obtained pKMD6-derived 0.8 kb DNA fragment, 0.02 μg of the DNA fragment containing the immunoglobulin enhancer region and 0.1 μg of the pUC18 fragment were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 175 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the resulting reaction mixture, transformation ofE. coliHB101 was carried out to isolate an ApR colony. Plasmid DNA was recovered from the colony to obtain pEPKMA1 as shown inFIG. 4.

Next, a 1 μg portion of the plasmid pEPKMA1 was dissolved in 100 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of XbaI and incubated at 37° C. for 2 hours to cleave the plasmid at the XbaI site. The resulting reaction mixture was extracted with phenol and chloroform and precipitated with ethanol. The thus precipitated DNA fragments were dissolved in 40 μl of the aforemetioned DNA polymerase I buffer solution, and the resulting solution was mixed with 6 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 90 minutes to convert the cohesive 5′-end formed by the XbaI digestion into blunt end. The reaction was stopped by extracting the reaction mixture with phenol. After extraction with chloroform and precipitation with ethanol, DNA fragments was recovered. The thus obtained DNA sample and 0.01 μg of a synthetic DNA XhoI linker (Takara Shuzo Co., Ltd.) were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 175 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the resulting reaction mixture, transformation ofE. coliHB101 was carried out to isolate an ApR colony. Plasmid DNA was recovered from the colony to obtain pEPKMB1 as shown inFIG. 5.

Next, SV40 early gene promoter and enhancer regions (to be referred to as “P_SE” hereinafter) of an expression vector pAGE107 for use in the expression of heterologous genes in animal cells (Miyaji et al., Cytotechnology, 3, 133 - 140 (1990)) were converted into KM50-derived immunoglobulin H chain promoter and enhancer (to be referred to as “P_IH” hereinafter) of pEPKMB1 in the following manner.

A 1 μg portion of the plasmid pAGE107 was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 150 mM sodium chloride. The thus prepared solution was mixed with 10 units of SalI and 10 units of XhoI and incubated at 37° C. for 2 hours to cleave the plasmid at the SalI and XhoI sites. The resulting reaction mixture was subjected to agarose gel electrophoresis and 0.5 μg of a DNA fragment of about 5.95 kb containing G418 resistance gene was recovered. Next, a 1 μg portion of the plasmid pEPKMB1 was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 150 mM sodium chloride. The thus prepared solution was mixed with 10 units of SalI and 10 units of XhoI and incubated at 37° C. for 2 hours to cleave the plasmid at the SalI and XhoI sites. The resulting reaction mixture was subjected to agarose gel electrophoresis and 0.1 μg of a DNA fragment of about 1.7 kb containing immunoglobulin promoter and enhancer regions was recovered. A 0.1 μg portion of the thus obtained pAGE107-derived 5.95 kb DNA fragment and 0.02 μg of the DNA fragment containing immunoglobulin promoter and enhancer regions were dissolved in 20 ∥l of the T4 ligase buffer, and the resulting solution was mixed with 175 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the resulting reaction mixture, transformation ofE. coliH3101 was carried out to isolate an ApR colony. Plasmid DNA was recovered from the colony to obtain pAGE501 as shown inFIG. 6.

(4) Construction of pAGE109

One of the two EcoRI cleavage sites in plasmid pAGE106 was deleted in the following manner to construct pAGE109.

A 2 μg portion of the expression vector pAGE106 for use in the expression of heterologous genes in animal cells (JA-A 3-22979 or EP-A-0 405 28.5) was dissolved in 100 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of SacI and incubated at 370C for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 1.5 μg of a pAGE106 DNA fragment (4.3 kb) was recovered which contained the SV40 early gene promoter and G418 resistance gene cleaved with EcoRI and SacI. The thus recovered DNA fragment was dissolved in 40 μl of the DNA polymerase I buffer solution, and the resulting solution was mixed with 5 units ofE. coliDNA polymerase I large fragment and incubated at 16° C. for 2 hours to convert the cohesive 3′-end formed by the SacI digestion and the cohesive 5′-end formed by the EcoRI digestion into blunt ends. The resulting reaction mixture was extracted with phenol and chloroform and then treated with ethanol. The thus precipitated sample was dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 4 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain plasmid pAGE109 as shown inFIG. 7.

(5) Construction of pAGE502

Plasmid pAGE502 was constructed in the following manner in order to convert the SV40 promoter and enhancer of pAGE107 into immunoglobulin H chain promoter and enhancer.

A 2 μg portion of the plasmid pAGE107 disclosed in JP-A-3-22979 or EP-A-O 405 285 was dissolved in 100 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of HindIII and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to phenol-chloroform extraction and ethanol precipitation and the thus recovered sample was dissolved in 40 μl of the DNA polymerase I buffer solution. The resulting solution was mixed with 5 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 2 hours to convert the cohesive 5′-end formed by the HindIII digestion into blunt end. The resulting reaction mixture was extracted with phenol and chloroform and then treated with ethanol. The thus precipitated sample was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 nM sodium chloride. The thus prepared solution was mixed with 10 units of XhoI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 1.5 μg of a pAGE107 DNA fragment of about 5.95 kb was obtained which contained G418 resistance gene and ApR gene cleaved with XhoI and HindIII.

Next, a 2 μg portion of the plasmid pAGE501 obtained in the above step (3) was dissolved in 100 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 175 mM sodium chloride. The thus prepared solution was mixed with 10 units of SalI and incubated at 37° C. for 4 hours. After subjecting the resulting reaction mixture to phenol-chloroform extraction and ethanol precipitation, the thus recovered sample was dissolved in 40 μl of the DNA polymerase I buffer solution. The resulting solution was mixed with 5 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 2 hours to convert the cohesive 5′-end formed by the SalI digestion into blunt end. The resulting reaction mixture was extracted with phenol and chloroform and then treated with ethanol. The thus precipitated sample was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of XhoI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.2 μg of a pAGE501 DNA fragment of about 1.8 kb was obtained which contained KM50 immunoglobulin H chain promoter and enhancer genes cleaved with XhoI and SalI.

Next, 0.1 μg of the thus obtained pAGE107 HindIII-XhoI fragment (about 5.95 kb) and 0.1 μg of the pAGE501 SalI-XhoI fragment (about 1.8 kb) were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain plasmid PAGE502 as shown inFIG. 8.

(6) Construction of pAGE503

One of the two EcoRI cleavage sites in plasmid pAGE502 was deleted in the following manner to construct pAGE503.

A 2 μg portion of the plasmid pAGE109 obtained in the above step (4) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of HindIII and 10 units of ClaI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.2 μg of a pAGE109 DNA fragment of about 1 kb was recovered which contained the poly(A) signal gene of beta-globin and SV40 early genes cleaved with ClaI and HindIII.

Next, a 2 μg portion of the plasmid pAGE502 obtained in the above step (5) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of HindIII and 10 units of ClaI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and then to the aforementioned DEAE paper method to recover about 1 μg of a pAGE502 DNA fragment of about 6.1 kb which contained KM50 immunoglobulin H chain promoter and enhancer genes, ApR gene and G418 resistance gene cleaved with HindIII and ClaI. Next, 0.1 μg of the thus obtained pAGE109 HindIII-ClaI fragment (about 1 kb) and 0.1 μg of the pAGE502 HindIII-ClaI fragment (about 6.1 kb) were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain plasmid pAGE503 as shown inFIG. 9.

(7) Construction of pSE1d1

A dhfr gene was introduced into plasmid pAGE107 in the following manner to construct plasmid pSE1d1.

A 2 μg portion of the plasmid pAGE107 disclosed in JP-A 3-22979 or EP-A-0 405 285 was dissolved in 100 μl of a buffer solution containing 100 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of EcoRI and incubated at 37° C. for 4 hours. After subjecting the resulting reaction mixture to phenol-chloroform extraction and ethanol precipitation, the thus recovered sample was dissolved in 40 μl of the DNA polymerase I buffer solution. The resulting solution was mixed with 5 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 2 hours to convert the cohesive 5′-end formed by the EcoRI digestion into blunt end. The resulting reaction mixture was extracted with phenol and chloroform and then treated with ethanol. The thus precipitated sample was dissolved in 30 μl of a buffer solution which was composed of 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of HindIII and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 1.5 μg of a pAGE107 DNA fragment of about 5.6 kb was recovered which contained G418 resistance gene and ApR gene cleaved with EcoRI and HindIII.

Next, a 2 μg portion of a plasmid pSV2-dhfr (Subramani et al., Mol. Cell. Biology, 1, 854 (1981)) was dissolved in 100 μl of a buffer solution containing 10 MM. Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of BglII and incubated at 37° C. for 4 hours. After subjecting the resulting reaction mixture to phenol-chloroform extraction and ethanol precipitation, the thus recovered sample was dissolved in 40 μl of the DNA polymerase I buffer solution. The resulting solution was mixed with 5 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 2 hours to convert the cohesive 5′-end formed by the BglII digestion into blunt end. The resulting reaction mixture was extracted with phenol and chloroform and then treated with ethanol. The thus precipitated sample was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of HindIII and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis, about 0.2 μg of a pSV2-dhfr DNA fragment of about 0.76 kb was recovered which contained dhfr gene cleaved with BglII and HindIII.

Next, 0.1 μg of the thus obtained pAGE107 HindIII-EcoRI fragment (about 5.6 kb) and 0.1 μg of the pSV2-dhfr BglII-HindIII fragment (about 0.76 kb) were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain plasmid pSE1d1 as shown inFIG. 10.

(8) Construction of pSE1d2

The HindIII cleavage site was removed from the plasmid pSE1d1 in the following manner to construct plasmid pSE1d2.

A 2 μg portion of the plasmid pSE1d1 obtained in the above step (7) was dissolved in 100 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 nm sodium chloride. The thus prepared solution was mixed with 10 units of HindIII and incubated at 37° C. for 4 hours. After subjecting the resulting reaction mixture to phenol-chloroform extraction and ethanol precipitation, the thus recovered sample was dissolved in 40 μl of the DNA polymerase I buffer solution. The resulting solution was mixed with 5 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 2 hours to convert the cohesive 5′-end formed by the HindIII digestion into blunt end. The resulting reaction mixture was extracted with phenol and chloroform and then treated with ethanol. The thus precipitated sample was dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain plasmid pSE1d2 as shown inFIG. 11.

(9) Construction of pIg1SE1d2

The dhfr gene was introduced into plasmid pAGE503 in the following manner to construct plasmid pIg1SE1d2.

A 2 μg portion of the plasmid pAGE503 obtained in the above step (6) was dissolved in 100 μl of a buffer solution containing 100 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mm sodium chloride. The thus prepared solution was mixed with 10 units of ClaI and incubated at 37° C. for 4 hours. After subjecting the resulting reaction mixture to phenol-chloroform extraction and ethanol precipitation, the thus recovered sample was dissolved in 40 μl of the DNA polymerase I buffer solution. The resulting solution was mixed with 5 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 2 hours to convert the cohesive 5′-end formed by the ClaI digestion into blunt end. The resulting reaction mixture was extracted with phenol and chloroform and then treated with ethanol. The thus precipitated sample was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of MluI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 1 μg of a pDAG-503 DNA fragment of about 5.4 kb was recovered which contained the KM50 immunoglobulin H chain promoter and enhancer genes cleaved with ClaI and MluI.

Next, a 2 μg portion of the plasmid pSE1d2 obtained in the above step (8) was dissolved in 100 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of XhoI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to phenol-chloroform extraction and ethanol precipitation and the thus recovered sample was dissolved in 40 μl of the DNA polymerase I buffer solution. The resulting solution was mixed with 5 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 2 hours to convert the cohesive 5′-end formed by the XhoI digestion into blunt end. The resulting reaction mixture was extracted with phenol and chloroform and then treated with ethanol. The thus precipitated sample was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of MluI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 1 μg of a pSE1d2 DNA fragment of about 3.8 kb was recovered which contained. dhfr gene cleaved with XhoI and MluI.

Next, 1 μg of the thus obtained pAGE503 ClaI-MluI fragment (about 5.4 kb) and 1 μg of the pSE1d2 XhoI-MluI fragment (about 3.8 kb) were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain plasmid pIg1SE1d2 as shown inFIG. 12.

(10) Construction of pIg1SE1d3

The ApaI cleavage site was removed from the plasmid pIg1SE1d2 in the following manner to construct plasmid pIg1SE1d3.

A 2 μg portion of the plasmid pIg1SE1d2 obtained in the above step (9) was dissolved in 100 μl of a buffer solution containing 10 mm Tris-HCl (pH 7.5) and 6 mM magnesium chloride. The thus prepared solution was mixed with 10 units of ApaI and incubated at 37° C. for 4 hours. After subjecting the resulting reaction mixture to phenol-chloroform extraction and ethanol precipitation, the thus recovered sample was dissolved in 40 μl of the DNA polymerase I buffer solution. The resulting solution was mixed with 5 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 2 hours to convert the cohesive 3′-end formed by the ApaI digestion into blunt end. The resulting reaction mixture was extracted with phenol and chloroform and then treated with ethanol. The thus precipitated sample was dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain plasmid pIg1SE1d3 as shown inFIG. 13.

(11) Construction of pIg1S1d4

In order to establish a cloning site between HindIII cleavage site and EcoRI cleavage site of the plasmid pIg1SE1d3, plasmid pIg1SE1d4 was constructed by inserting the synthetic DNA shown in SEQ ID NO: 5 into the plasmid pIg1SE1d3 in the following manner.

A 2 μg portion of the plasmid pIg1SE1d3 obtained in the above step (10) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of HindIII and 10 units of EcoRI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 1 μg of a pIg1SE1d3 DNA fragment of about 9.2 kb was recovered which contained the KM50 immunoglobulin H chain promoter and enhancer genes, ApR gene, G418 resistance gene and dhfr gene cleaved with HindIII and EcoRI.

Next, 0.1 μg of the thus obtained pIg1SE1d3 HindIII-EcoRI fragment (about 9.2 kb) and 10 ng of the synthetic DNA (SEQ ID NO: 2) were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain plasmid pIg1SE1d4 as shown inFIG. 14.

3. Preparation of Moloney Mouse Leukemia Virus Long Terminal Repeat (to be Referred to as “MoLTR” Hereinafter)

Since MoLTR is known to have promoter and enhancer activities (Kuwana et al., Biochem. Biophys. Res. Comun., 149, 960 (1987)), a plasmid pPMOL3 containing MoLTR was prepared in the following manner in order to use MoLTR as cassette vector promoter and enhancer.

A 3 μg portion of the plasmid pPMOL1 disclosed in JP-A 1-63394 was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 7 mM magnesium chloride and 6 mM 2-mercaptoethanol. The thus prepared solution was mixed with 10 units of ClaI and incubated at 37° C. for 4 hours. After subjecting the resulting reaction mixture to phenol-chloroform extraction and ethanol precipitation, the thus recovered sample was dissolved in 40 μl of the DNA polymerase I buffer solution. The resulting solution was mixed with 5 units ofE. coliDNA polymerase I Klenow fragment and incubated at 16° C. for 2 hours to convert the cohesive 5′-end formed by the ClaI digestion into blunt end. The reaction was stopped by phenol extraction, followed by chloroform extraction and ethanol precipitation to recover 2 μg of DNA fragments. The thus precipitated DNA sample and 0.01 μg of a synthetic DNA XhoI linker (Takara Shuzo Co., Ltd.) were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 175 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the resulting reaction mixture, transformation ofE. coliHB101 was carried out to obtain plasmid pPMOL2 as shown inFIG. 15. Next, a 3 μg portion of the thus obtained plasmid pPMOL2 was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 7 mM. magnesium chloride, 10 mM sodium chloride and 6 mM 2-mercaptoethanol. The thus prepared solution was mixed with 10 units of SmaI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to phenol-chloroform extraction and ethanol precipitation and 2 μg of DNA fragments were recovered. The thus recovered DNA sample and 0.01 μg of a synthetic DNA EcoRI linker (Takara Shuzo Co., Ltd.) were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 175 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the resulting reaction mixture, transformation ofE. coliHB101 was carried out to obtain plasmid pPMOL3 as shown inFIG. 16.

4. Cloning of H Chain Constant Region (Cg1) cDNA and L Chain Constant Region (Ck) cDNA of Human Immunoglobulin IgG1

(1) Preparation of mRNA from Chimera Antibody-Producing SP2-PC Chimera-1 Cells

Using a mRNA extraction kit, Fast Track (No. K1593-02, available from Invitrogen), 6.2 μg of mRNA was obtained from 1×10⁸cells of chimera antibody-producing SP2-PC Chimera-1 which has anti-phosphorylcholine activity and is disclosed in FEBS Letters (244, 301 - 306 (1989)).

(2) Preparation of SP2-PC Chimera-1 cDNA Library and Cloning of Human Immunoglobulin H Chain Constant Region (Cg1) cDNA and L Chain Constant Region (Ck) cDNA

A 2 μg portion of the mRNA obtained in the above step (1) was subjected to EcoRI adaptor addition using cDNA Synthesis Kit (No. 27-9260-01, available from Pharmacia) followed by kination. The resulting cDNA solution was subjected to phenol-chloroform extraction and ethanol precipitation to recover 4 μg of cDNA. The thus recovered cDNA was dissolved in 20 μl of sterile water, and the resulting solution was subjected to agarose gel electrophoresis to recover about 0.3 μg of a DNA fragment of about 1.8 kb and about 0.3 μg of a DNA fragment of about 1.0 kb.

Next, a 5 μg portion of the vector PUC18 was dissolved in 100 μl of a buffer solution containing 100 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 50 units of EcoRI and incubated at 37° C. for 4 hours to cleave the pUC18 DNA at its EcoRI cleavage site. The resulting reaction mixture was subjected to phenol-chloroform extraction and ethanol precipitation to recover about 3 μg of a pUC18 DNA fragment cleaved with EcoRI.

Next, 0.1 μg of the thus obtained pUC18 EcoRI fragment (about 2.7 kb) and the 1.8 kb and 1.0 kb cDNA fragments (0.1 μg for each) prepared from the SP2-PC Chimera-1 cells were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 50 units of T4 DNA ligase and incubated at 4° C. for 24 hours.

Using the thus obtained recombinant plasmid DNA, transformation ofE. coliL7392 was carried out. About 3,000 colonies thus obtained were fixed on nitrocellulose filters. Two³²P-labeled probes were prepared from human immunoglobulin constant region chromosomal genes (Cg1 as an IgG1 H chain constant region and Ck as an IgG1 L chain constant region) which have been isolated by Kameyama et al. (FEBS Letters, 244, 301 (1989)). From colonies which showed strong reactions at 65° C. with these probes, one showing strong reaction with Cg1 (pPCVHhCGI1) and the other showing strong reaction with Ck (pPCVLhCK1) were obtained.

(3) Introduction of EcoRV Size into Human Ig k Chain Constant Region

An EcoRV site was introduced into 5 -end side of the human Ig k chain constant region by means of site-specific mutagenesis using a kit purchased from Promega (Catalogue No. Q6210). A 2 μg portion of the plasmid pPCVLhCK1 was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of KpnI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.2 μg of a pPCVLhCK1 DNA fragment of about 0.8 kb was recovered which contained the human immunoglobulin L chain constant region cleaved with EcoRI and KpnI.

Next, a 2 μg portion of pSELECT1 (a kit available from Promega, Catalogue No. Q6210) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 50 mM sodium chloride. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of KpnI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 1 μg of a pSELECT1 DNA fragment of about 5.7 kb cleaved with EcoRI and KpnI was recovered.

Next, 0.1 μg of the pPCVLhCK1 EcoRI-KpnI fragment (about 0.8 kb) and 0.1 pg of the pSELECT1 EcoRI-KpnI fragment (about 5.7 kb) obtained above were dissolved in 20 μl of the T4 ligase buffer, and the resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliJM109 was carried out to obtain plasmid pchCKA7 as shown inFIG. 17.

Next, using the plasmid pchCKA7 thus obtained and the synthetic DNA of SEQ ID NO: 6 as mutagenesis primer, the ACC sequence of the human immunoglobulin L chain constant region (12 to 14 position bases from the N-terminal) was converted into GAT in order to construct a plasmid pchCKB1 (FIG. 18) in which an EcoRV site was introduced into the converted site.

Next, the EcoRV site of the plasmid pchCKB1 was converted into HindIII cleavage site in the following manner. A 2 μg portion of the plasmid pchCKB1 obtained above was dissolved in 10 μl of a buffer solution containing 100 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of EcoRI and incubated at 37° C. for 4 hours. After subjecting the resulting reaction mixture to phenol-chloroform extraction and ethanol precipitation, the thus recovered sample was dissolved in 40 μl of the DNA polymerase I buffer solution. The resulting solution was mixed with 5 units ofE. coliDNA polymerase I Klenow fragment and incubated at 37° C. for 30 minutes to convert the cohesive 5′-end formed by the EcoRI digestion into blunt end. The resulting reaction mixture was extracted with phenol and chloroform and then treated with ethanol. The thus precipitated sample was dissolved in 20 μl of the T4 ligase buffer containing 0.1 μg of HindIII linker (Takara Shuzo Co., Ltd.), and the resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain plasmid pchCKC1 as shown inFIG. 19.

5. Construction of Cassette Vector

(1) Construction of a Cassette Vector for Use in the Construction of Humanized Chimera Antibody H Chain Expression Vector

A 2 μg portion of the plasmid pIg1SE1d4 obtained in the aforementioned step 2-(11) was dissolved in 30 μl of a buffer solution containing 10 mM Tri-s-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of EcoRV and 10 units of ApaI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarase gel electrophoresis and about 1.5 μg of a pIg1SE1d4 DNA fragment of about 9.2 kb cleaved with EcoRV and ApaI was recovered.

Next, a 2 μg portion of the plasmid pPCVHhCGI1 obtained in the aforementioned step 4-(2) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5) and 6 mM magnesium chloride. The thus prepared solution was mixed with 10 units of ApaI and 10 units of SmaI and incubated at 37° C. for 1 hour. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.2 μg of a pPCVHhCGI1 DNA fragment of about 1 kb was recovered which contained the human immunoglobulin H chain constant region gene cleaved with ApaI and SmaI.

Next, 0.1 μg of the pIg1SE1d4 EcoRV-ApaI fragment (about 9.2 kb) and 0.1 μg of the pPCVHhCGI1 ApaI-SmaI fragment (about 1 kb) prepared above were dissolved in 20 μl of the T4 ligase buffer. The resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain a plasmid pChiIgHB2 (FIG. 20) as a cassette vector for use in the construction of a humanized chimera antibody H chain expression vector.

(2) Construction of a Cassette Vector for Use in the Construction of Humanized Chimera Antibody L Chain Expression Vector

A 2 μg portion of the plasmid pIg1SE1d4 obtained in the aforementioned step 2-(10) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of EcoRV and 10 units of HindIII and incubated a 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 1.5 μg of a pIg1SE1d4 DNA fragment of about 9.2 kb cleaved with EcoRV and HindIII was recovered.

Next, a 2 μg portion of the plasmid pchCKC1 obtained in the aforementioned step 4-(3) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 6 mM magnesium chloride and 100 mM sodium chloride. The thus prepared solution was mixed with 10 units of EcoRV and 10 units of HindIII and incubated at 37° C. for 1 hour. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.2 μg of a pPCVLhCK1 DNA fragment of about 0.6 kb was recovered which contained the human immunoglobulin L chain constant region gene cleaved with EcoRV and HindIII.

Next, 0.1 μg of the pIg1SE1d4 EcoRV-HindIII fragment (about 9.2 kb) and 0.1 μg of the pchCKC1 EcoRV-HindIII fragment (about 0.6 kb) prepared above were dissolved in 20 μl of the T4 ligase buffer. The resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain a plasmid pChiIgLA1 (FIG. 21) as a cassette vector for use in the construction of a humanized chimera antibody L chain expression vector.

EXAMPLE 2

Anti-GD₃Chimera Antibody

1. Preparation of mRNA from a Hybridoma Cell Line Capable of Producing Mouse Anti-GD₃Monoclonal Antibody KM-641

Using a mRNA extraction kit, Fast Track (No. X1593-02, available from Invitrogen), 34 μg of mRNA was prepared from 1×10⁸cells of a hybridoma cell line which is capable of producing mouse anti-GD₃monoclonal antibody KM-641 prepared by the method described later in Reference Example shown below.

2. Preparation of cDNA Libraries of KM-641 H Chain and L Chain Genes

Using ZAP-cDNA Synthesis Kit (No. sc200400), a cDNA synthesis kit available from Stratagene Cloning Systems, cDNA having EcoRI adaptor on its 5′-end and XhoI adaptor on its 3′-end was prepared from 3 μg of the mRNA obtained in theabove procedure 1. About 6 μg of the cDNA was dissolved in 10 μl of sterile water and subjected to agarose gel electrophoresis to recover 0.1 μg of an H chain-corresponding cDNA fragment of about 1.8 kb and 0.1 μg of an L chain-corresponding cDNA fragment of about 1.0 kb. Next, 0.1 μg of the 1.8 kb cDNA fragment, 0.1 μg the 1.0 kb cDNA fragment and 1 μg of Uni-ZAP XR (available from Stratagene Cloning Systems; a preparation obtained by existing Lambda ZAPII vector with EcoRI and XhoI, followed by treatment with calf intestine alkaline phosphatase) to be used as a vector were dissolved in 11 μl of the 175 ligase buffer, and the resulting solution was mixed with 175 units of T4 DNA ligase and incubated at 12° C. for 10 hours and then at room temperature for 2 hours. A 4 μl portion of the resulting reaction mixture was subjected to lambda phage packaging using Giga Pack Gold (Stratagene Cloning Systems) in accordance with the conventional method (Maniatis et al. , Molecular Cloning, 1989, p2.95). AnE. colistrain PLK-F was infected with the thus packaged product in accordance with the conventional method (Maniatis et al., Molecular Cloning, 1989, p2.95-107) to obtain an H chain cDNA library and an L chain cDNA library, each containing about 10,000 phage clones. Next, these phage particles were fixed on nitrocellulose filters in accordance with the conventional method (Maniatis et al., Molecular Cloning, 1989, p2.112).

3. Cloning of Monoclonal Antibody KM-641 H Chain and L Chain cDNA

Two³²P-labeled probes were prepared from an EcoRI fragment of about 6.8 kb containing a mouse immunoglobulin constant region chromosomal gene Cgl (Roeder et al., Proc. Natl. Acad. Sci. U.S.A., 78, 474 (1981)) and a mouse Ck gene-containing HindIII-BamHI fragment of about 3 kb (Sakano et al., Nature, 280, 288 (1979)).: A phage clone which showed strong reaction at 65° C. with one of these two probes were obtained from each of the H chain cDNA library and the L chain cDNA library prepared in theabove procedure 2 in accordance with the conventional method (Maniatis et al., Molecular Cloning, 1989, p2.108). Next, using ZAP-cDNA Synthesis Kit (No. sc200400), a cDNA synthesis kit of Stratagene Cloning Systems each of the thus obtained phage clones was introduced into plasmid pBluescript to isolate a recombinant plasmid pKM641HA3 containing the KM-641 H chain cDNA and a recombinant plasmid pKM641LA2 containing the KM-641 L chain cDNA. When each of the plasmids pK-641HA3 and pKM641LA2 was digested with EcoRI and XhoI, it was found that cDNA of about 1.6 kb had been inserted into the former plasmid, and cDNA of about 0.9 kb into the latter (FIG. 22).

4. Immunoglobulin Variable Region Base Sequences of FM-641 H Chain cDNA (pKM641HA3) and KM-641 L Chain cDNA (pKM641LA2)

Immunoglobulin variable region base sequences of the plasmids pKM641HA3 and pKM641LA2 obtained in theabove procedure 3 were determined by the dideoxy method (Maniatis et al., Molecular Cloning, 1989, p13.42) using Sequenase Version 2.0 DNA Sequencing Kit (United States Biochemical Corporation). The results are shown in SEQ ID NO: 7 and SEQ ID NO: 9. The plasmid pKM641LA2 was a complete cDNA containing a leader sequence and having a methionine-corresponding sequence which was assumed to be the initiation codon ATG located close to the 5′-end. The plasmid pKM641HA3, on the other hand, did not have such a methionine-corresponding initiation codon-like sequence on its 5′-end side, and its leader sequence was partially deficient.

5. Construction of KM-641 Chimera H Chain Expression Vector

H chain variable region gene obtained by cleaving the plasmid pKM641HA3 variable region at the 5′-end AluI site and 3′-end StyI site was ligated with the cassette vector for use in the construction of the humanized chimera antibody H chain obtained in Example 1 using the synthetic DNA sequences shown in SEQ ID NO: 11 and SEQ ID NO: 13, thereby constructing a humanized chimera antibody H chain expression vector pchi641HA1 (FIG. 23).

Firstly, the DNA shown in SEQ ID NO: 13 (seeFIG. 23) was synthesized using a DNA synthesizer. This synthetic DNA comprises a base sequence derived from plasmid pKM641HA3 ranging from the 3′-end of its immunoglobulin H chain variable region to a StyI cleavage site in the vicinity of the 3′-end and a base sequence derived from plasmid pAGE28 ranging from the 5′-end of its immunoglobulin H chain constant region to an ApaI cleavage site in the vicinity of the 5′-end. Thus, the synthetic DNA has a StyI cleavage site and an ApaI cleavage site on both of its end. Next, the thus synthesized DNA was introduced into the plasmid pKM641HA3 in the following manner.

A 3 μg portion of the plasmid pKM641HA3 was dissolved in 30 μl of a buffer solution containing 50 mM Tris-HCl (pH 7.5), 10 mM magnesium chloride, 50 mM sodium chloride and 1 mM DTT. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of StyI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.3 μg of a DNA fragment of about 0.41 kb was recovered. Next, a 3 μg portion of pAGE28 (Mizukami et al., J. Biochem., 101, 1307-1310 (1987)) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 7 mM magnesium chloride and 6 mM 2-mercaptoethanol. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of ApaI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 2 μg of a DNA fragment of about 2.45 kb was recovered. Next, 0.1 μg of the pKM641HA3 EcoRI-StyI fragment (about 0.41 kb) and 0.1 μg of the pAGE28 EcoRI-ApaI fragment (about 2.45 kb) prepared above and 0.3 μg of the synthetic DNA of SEQ ID NO: 13 were dissolved in 20 μl of the T4 ligase buffer solution. The resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain a plasmid pKM641HE1 as shown inFIG. 24.

Since the thus constructed plasmid pKM641HE1 lacks a leader sequence, the following attempt was made to supplement the plasmid with the leader sequence using the synthetic DNA of SEQ ID NO: 11.

A 3 μg portion of the plasmid pKM641HE1 was dissolved in 30 μl of a buffer solution containing 10 m: Tris-HCl (pH 7.5), 7 mM magnesium chloride and 6 mM 2-mercaptoethanol. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of ApaI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.4 μg of a DNA fragment of about 0.42 kb was recovered. Next, a 0.4 μg portion of the thus prepared pKM641HE1 EcoRI-ApaI fragment (about 9.42 kb) was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 7 mM magnesium chloride, 50 mM sodium chloride and 6 mM 2-mercaptoethanol. The thus prepared solution was mixed with 10 units of AluI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to phenol-chloroform extraction and ethanol precipitation and about 0.3 μg of a DNA fragment of about 0.4 kb was recovered.

Next, 0.1 μg of the pKM641HE1 AluI-ApaI fragment (about 0.4 kb) and 0.1 μg of the pAGE28 EcoRI-ApaI fragment (about 2.45 kb) prepared above and 0.3 μg of the synthetic DNA of SEQ ID NO: 11 were dissolved in 20 μl of the T4 ligase buffer solution. The resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain a plasmid pKM641HF1 as shown inFIG. 25.

Next, immunoglobulin H chain variable region of the thus obtained plasmid pKM641HF1 was introduced into the aforementioned cassette vector pChiIgHB2 in the following manner.

A 3 ∥g portion of the plasmid pKM641HF1 was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 7 mM magnesium chloride and 6 mM 2-mercaptoethanol. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of ApaI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.5 μg of a DNA fragment of about 0.44 kb was recovered. Next, a 3 μg portion of the pChiIgHB2 was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 7 mM magnesium chloride and 6 mM 2-mercaptoethanol. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of ApaI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to phenol-chloroform extraction and ethanol precipitation and about 3 μg of DNA was recovered. Next, 0.1 μg of the pKM641HF1 EcoRI-ApaI fragment (about 0.44 kb) and 0.1 μg of the pChiIgHB2 EcoRI-ApaI fragment (about 10.1 kb) prepared above were dissolved in 20 μl of the T4 ligase buffer. The resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain a plasmid pChi641HA1 as shown inFIG. 26.

Next, KM50-derived immunoglobulin H chain promoter and enhancer regions of the thus obtained plasmid pChi641HA1 were converted into MoLTR in the following manner.

A 3 μg portion of the plasmid pChi641HA1 was dissolved in 30 μl of a buffer solution containing 50 mM Tris-HCl (pH 7.5), 10 mM magnesium chloride, 50 mM sodium chloride and 1 mM DTT. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of XhoI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.2 μg of a DNA fragment of about 8.8 kb was recovered. Next, a 3 μg portion of the pPMOL3 prepared inprocedure 2 of Example 1 was dissolved in 30 μl of a buffer solution containing 50 mM Tris-HCl (pH 7.5), 10 mM magnesium chloride, 50 mM sodium chloride and 1 mM DTT. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of XhoI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.3 μg of a DNA fragment of about 0.63 kb containing MoLTR was recovered. Next, 0.1 μg of the pChi641HA1 EcoRI-XhoI fragment and 0.1 μg of the pPMOL3 EcoRI-XhoI fragment prepared above were dissolved in 20 μl of the T4 ligase buffer solution. The resulting solution was mixed with 175 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the resulting reaction mixture, transformation ofE. coliHB101 was carried out to obtain a plasmid pChi641HAM1 (FIG. 27) as a KM-641 chimera H chain expression vector.

6. Construction of KY-641 Chimera L Chain Expression Vector

L chain variable region gene obtained by cleaving the plasmid pKM641LA2 variable region gene at its 5′-end EcoRI site and 3′-end HindIII site was ligated with the cassette vector for the expression of chimera L chain, using the synthetic DNA shown in SEQ ID NO: 15, thereby constructing an L chain expression vector pchi64!LG11 (FIG. 28).

Firstly, the DNA of SEQ ID NO: 15(seeFIG. 29) was synthesized using a DNA synthesizer. This synthetic DNA comprises a base sequence corresponding to a region of the plasmid pKM641LA2 ranging from the 3′-end of the immunoglobulin L chain variable region to a HindIII cleavage site in the vicinity of the 3′-end and a base sequence corresponding to a region of the plasmid pChiIgLA1 ranging from the 5′-end to an EcoRV cleavage site in the vicinity of the 5′-end. Thus, it has a HindIII cleavage site and an EcoRV cleavage site on both ends. Next, the thus synthesized DNA was introduced into the plasmid pKM641LA2 in the following manner.

A 3 μg portion of the plasmid pKM641LA2 was dissolved in 30 μl of a buffer solution containing 10 mM Tris-HCl (pH 7.5), 7 mM magnesium chloride, 50 mM sodium chloride and 6 mM 2-mercaptoethanol. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of HindIII and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.3 μg of a DNA fragment of about. 0.35 kb was recovered. Next, a 3 μg portion of pChiIgLA1 was dissolved in 30 μl of a buffer solution containing 50 mM Tris-HCl (pH 7.5), 10 mM magnesium chloride, 50 mM sodium chloride and 1 mM DTT. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of EcoRV and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to phenol-chloroform extraction and ethanol precipitation and about 3 μg of DNA was recovered and dissolved in 10 μl of the TE solution (a buffer solution containing 10 mM Tris-HCl and 1 mM EDTA (pH 7.5)). Next, 0.1 μg of the pKM641LA2 EcoRI-HindIII fragment (about 0.35 kb) and 0.1 μg of the pChiIgLA1 EcoRI-EcoRV fragment (about 9.7 kb) prepared above and 0.3 μg of the synthetic DNA of SEQ ID NO: 15 were dissolved in 20 μl of the T4 ligase buffer solution. The resulting solution was mixed with 350 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the thus obtained recombinant plasmid DNA, transformation ofE. coliHB101 was carried out to obtain a plasmid pChi641LG11 as shown inFIG. 29.

Next, KM50-derived immunoglobulin H chain promoter and enhancer regions of the thus obtained plasmid pChi641LG11 were converted into MoLTR in the following manner.

A 3 μg portion of the plasmid pChi641LG11 was dissolved in 30 μl of a buffer solution containing 50 mM Tris-HCl (pH 7.5), 10 mM magnesium chloride, 50 mM sodium chloride and 1 mM DTT. The thus prepared solution was mixed with 10 units of EcoRI and 10 units of XhoI and incubated at 37° C. for 4 hours. The resulting reaction mixture was subjected to agarose gel electrophoresis and about 0.2 μg of a DNA fragment of about 8.3 kb was recovered. Next, 0.1 Ag of the pChi641LG11 EcoRI-XhoI fragment and 0.1 μg of the pPMOL3 EcoRI-XhoI fragment prepared above were dissolved in 20 μl of the T4 ligase buffer. The resulting solution was mixed with 175 units of T4 DNA ligase and incubated at 4° C. for 24 hours. Using the resulting reaction mixture, transformation ofE. coliH3101 was carried out to obtain a plasmid pChi641LGM11 (FIG. 30) as a KM-641 chimera L chain expression vector.

7. Expression of Anti-GD₃Chimera Antibody in SP2/0 Cells

Introduction of plasmid into SP2/0 cells was carried out making use of the electroporation technique in accordance with the method of Miyaji et al. (Cytotechnology, 3, 133-140 (1990)).

The plasmids pChi641LG11 and pChi641HA1 (2 μg for each), or the plasmids pChi641LGY11 and pChi641HA11 (2 μg for each), were simultaneously introduced into 2×10⁶of SP2/0 cells, and the resulting cells were suspended in 40 ml of RPMI1640-FCS(10) which has been prepared by supplementing RPMI1640 medium (Nissui Pharmaceutical Co., Ltd.) with 10% of FCS, 1/40 volume of 7.5% NaHCO₃, 3% of 200 mM L-glutamine solution (available from GIBCO) and 0.5% of a penicillin-streptomycin solution (GIBCO, a solution containing 5,000 units/ml of penicillin and 5,000 units/ml of streptomycin). The thus prepared cell suspension was distributed in 200 μl-portions into wells of a 96-well microtiter plate (Flow Laboratories), and the cells were cultured at 37° C. in a CO₂incubator. After 24 hours of the culturing, G418 (GIBCO) was added to the cell suspension to a final concentration of 0.5 mg/ml, and the culturing was continued for 1 to 2 weeks. When transformant colonies were developed and grown into confluent stages, culture broths were recovered from the wells to measure anti-GD₃chimera antibody activities by ELISA method in the following manner.

A 2 ng portion of GD₃(available from Iatron) or other type of ganglioside was dissolved in 2 μl of ethanol solution containing 5 ng of phosphatidylcholine (Sigma Chemical Co.) and 2.5 ng of cholesterol (Sigma Chemical Co.). A 20 μl portion of the thus prepared solution or the same volume of its dilution solution was distributed into each well of a 96-well microtiter μlate (available from Greiner). After air-drying, blocking was effected with PBS containing 1% BSA. To each well was added 50 to 100 μl of a culture supernatant of a transformant, a purified mouse monoclonal antibody solution or a purified chimera antibody solution. After reaction at 4° C. for 10 hours, each well was washed with PBS and charged with 50 to 100 μl of peroxidase-labeled protein A (Funakoshi Pharmaceutical Co., Ltd.), followed by 1 to 2 hours of reaction at room temperature. After washing with PBS, 50 to 100 μl of ABTS substrate solution prepared by dissolving 550 mg of

diammonium

2,2′-azinobis(3-ethylbenzothiazoline-6-sulfonate) in 0.1 M citrate buffer (pH 4.2) and adding 1 μl/ml of hydrogen peroxide to the resulting solution just before its use was added to each well to develop color, and OD₄₁₅of the reaction mixture was measured.

Among clones thus obtained, culture broth of a clone having the highest activity measured by ELISA method contained anti-GD₃chimera antibody in an amount of about 0.1 μg/ml.

The clone having anti-GD₃chimera antibody activity was suspended in the aforementioned RPMI1640-FCS(10) medium supplemented with 0.5 mg/ml of G418 and 50 nM of methotrexate (to be referred to as “MTX” hereinafter) to a final cell density of 1-2×10⁵cells /ml. The thus prepared cell suspension was distributed in 2-ml portions into wells of a 24 well plate, and the cells were cultured at 37° C. for 2 to 3 weeks in a CO₂incubator to induce clones resistant to 50 nM MTX. When the thus induced clones were grown into confluent stages, anti-GD₃chimera antibody activities in the culture broths were measured by the ELISA method. Among clones thus obtained, culture broth of a 50 nM MTX-resistant clone having the highest activity measured by ELISA method contained anti-GD₃chimera antibody in an amount of about 0.3 μg/ml.

The 50 nM MTX-resistant clone was suspended in the RPMI1640-FCS(10) medium supplemented with 0.5 mg/ml of G418 and 200 nM of MTX to a final cell density of 1-2×10⁵cells /ml. The thus prepared cell suspension was distributed in 2-ml portions into wells of a 24 well plate, and the cells were cultured at 37° C. for 2 to 3 weeks in a CO₂incubator to induce clones resistant to 200 nM MTX. When the thus induced clones were grown into confluent stages, anti-GD₃chimera antibody activities in the culture broths were measured by the ELISA method. Among clones thus obtained, culture broth of a 200 nM MTX-resistant clone having the highest activity measured by ELISA method contained anti-GD₃chimera antibody in an amount of about 2 μg/ml. This 200 nM MTX-resistant clone was named transformant KM-871.

Expression of anti-GD₃chimera antibody protein in the transformant KM-871 was confirmed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) in the following manner.

The transformant KM-871 was suspended in GIT medium (Nihon Seiyaku Co., Ltd.) supplemented with 0.5 mg/ml of G418 and 200 ny of MTX to a final cell density of 1-2×10⁵cells/ml. The thus prepared cell suspension was distributed in 100-ml portions in 175 cm²flasks (available from Greiner), and the cells were cultured at 37° C. for 3 to 5 days in a CO₂incubator. When the cells were grown into confluent stage, the resulting culture broth (about 900 ml) was recovered and subjected to salting-out with 50% ammonium sulfate. Using Affigel Protein A MAPS-II Kit (Bio-Rad Laboratories), about 100 μg of purified anti-GD₃chimera antibody KM-871 was obtained. About 5 μg of the thus purified anti-GD₃chimera antibody EM-871 was subjected to electrophoresis in accordance with the conventional method (Laemmli, Nature, 227, 680 (1970)) to check its molecular weight. The results are shown inFIG. 31. As shown in the figure, under reductive conditions, molecular weights of the chimera H chain and the chimera L chain were found to be about 50 kilodaltons and about 25 kilodaltons, respectively, thus confirming expression of the H and L chains having correct molecular weights. Under non-reductive conditions, molecular weight of the chimera antibody was found to be about 150 kilodal-ons which confirmed expression of the correct size antibody consisting of two H chains and two L chains.

8. Reaction Specificity of Anti-GD₃Chimera Antibody KM-871

Reactivities of the anti-GD₃chimera antibody with ganglioside GM₁, N-acetyl GM₂(Boehringer-Mannheim Corp.), N-glycolyl GM₂, N-acetyl GM₃, N-glycolyl GM₃, GD_1a, GD_1b(Iatron), GD₂, GD₃(Iatron), GT_1b(Funakoshi Pharmaceutical Co., Ltd.) and GQ_1b(Iatron) were measured by the ELISA method. In this instance, GM₁and GD_1awere purified from bovine brain, N-glycolyl GM₂and N-glycolyl GM₃from mouse liver, N-acetyl GM₃from dog erythrocytes and GD₂from a cultured cell line IMR32 (ATCC CCL127), in accordance with the conventional method (J. Biol. Chem., 263, 10915 (1988)). The results are shown in Table 1.

	TABLE 1


	Binding activity of
	antibody (OD₄₁₅)

	Anti-GD₃chimera	Mouse anti-GD₃
Ganglioside	antibody (0.3 μg/ml)	antibody (0.4 μg/ml)

N-acetyl GM₃	0.007	0.006
N-glycolyl GM₃	0	0
N-acetyl GM₂	0	0
N-glycolyl GM₂	0	0
GM₁	0	0
GD₂	0	0
GD₃	0.717	1.33
GD_1a	0	0
GD_1b	0	0
GT_1b	0	0
GQ_1b	0	0.16

As shown in Table 1, anti-GD₃chimera antibody KM-871 and mouse anti-GD₃antibody KM-641 reacted only with GD₃, thus showing no changes in the reaction specificity by the chimera formation.

9. Reactivity of Anti-GD₃Chimera Antibody KM-871 by fluorescent antibody technique

The cultured human malignant melanoma SK-MEL-28 (ATCC HTB72) and G361 cells (JCRB) both of which produced ganglioside GD3 were placed in a microtube (Treff) to give a cell number of 1×10⁶cells per tube and washed by centrifugation (1,200 rpm, 5 minutes) with PBS. 50 μl of anti-GD₃chimera antibody EM-871 (10 μg/ml) was added to the microtube and the mixture was allowed to react at 4° C. for 30 minutes. Thereafter, the cells were washed three times by centrifugation (1,200 rpm, 5 minutes) with PBS, then 20 μl of fluorescein isocyanate-labeled Protein A (Boehringer Mannheim-Yamanouchi, 30-fold diluted) was added and, after stirring, the mixture was allowed to react at 4° C. for 30 minutes. Thereafter, the cells were washed three times by centrifugation (1,200 rpm, 5 minutes) with PBS, further suspended in PBS, and submitted for analysis using FCS-1 flow cell sorter (Nippon Bunko).

Control tests without the addition of KM-871 were performed by the same analytical procedure as mentioned above.

The results are shown inFIG. 32. The fluorescence intensity peak for KM-871 showed shifting to the right (increased fluorescence intensity) as compared with the control, indicating that this antibody had reacted directly with ganglioside GD, on the surface of the SK-MEL-28 and G361 cells.

10. In Vitro Antitumor Effect of Anti-GD₃Chimera Antibody KM-871 (Complement-Dependent Cytotoxicity CDC)

(a) Preparation of Target Cells

Suspensions of the target cells, namely SK-MEL-28 cells and G361 cells, in RPMI-1640 medium supplemented with 10% FCS were respectively adjusted to a cell concentration of 1×10⁷cells/ml, Na₂⁵¹CrO₄was added to a concentration of 100 μCi/1×10⁷cells, reaction was performed at 37° C. for 1 hour and, thereafter, the cells were washed three times with the medium. The cells were allowed to stand in the medium at 4° C. for 30 minutes for spontaneous dissociation and then centrifuged (1,200 rpm, 5 minutes), and the medium was added to adjust the cell concentration to 4×10⁶cells/ml.

(b) Preparation of Complement

Serum from three healthy subjects were mixed to serve as a source of human complement.

(c) CDC Activity Measurement

To U-bottomed 96-well plates was added anti-GD₃chimera antibody E-871 or anti-GD₃mouse antibody KM-641 to final concentrations within the range of 0.05 μg/ml to 50 μg/ml. To each well were added 2×10⁵target cells. Reaction was performed at room temperature for 1 hour. The supernatants were removed by centrifugation (1,200 rpm, 5 minutes), the complement solution prepared as described above under (b) was added in 150-μl portions (final concentration of 15 v/v %), and reaction was performed at 37° C. for 1 hour. After centrifugation (1,200 rpm, 5 minutes), the amount of⁵¹Cr in each supernatant was determined using a γ-counter. The amount of⁵¹Cr resulting from spontaneous dissociation was determined by adding to target cells the medium alone in place of the antibody and complement solution and determining the amount of⁵¹Cr in the supernatant in the same manner as described above. The total amount of free⁵¹Cr was determined by adding 5 N sodium hydroxide in place of the antibody and complement solution, proceeding as described above, and determining the amount of⁵¹Cr in the supernatant.

The CDC activity was calculated as follows:

CDC activity (%) = \frac{\begin{matrix} Amount of^{51} Cr in \\ sample supernatant \end{matrix} - \begin{matrix} Amount of^{51} Cr resulting \\ from spontaneous dissociation \end{matrix}}{\begin{matrix} Total amont of \\ free^{51} Cr \end{matrix} - \begin{matrix} Amount of^{51} Cr resulting \\ from spontaneous dissociation \end{matrix}} \times 100

The results are shown inFIG. 33. It was found from the results that chimera antibody KM-871 showed strong cytotoxicity against the SK-MEL-28 and G361 cells as compared to mouse antibody KM-641, which indicates that chime-a antibody KM-871 would be clinically more useful than mouse antibody KM-641.

11. In Vitro Antitumor Effect of Anti-GD₃Chimera Antibody KM-871 (Antibody-Dependent Cell-Mediated Cytotoxicity:ADCC)

(a) Preparation of Target Cells

The target SK-MEL-28 and G361 cells were prepared in the same manner as described above under 10 (a).

(b) Preparation of Effector Cells

50 ml of human venous blood was collected, 0.5 ml of heparin sodium (Takeda Chemical Industries, 1,000 units/ml) was added, and the mixture was stirred gently and then centrifuged (1,500 to 1,800 g, 15 minutes) using Polymorphprep (Nycomed Pharma AS). The layer of lymphocytes and polymorphonuclear leukocytes was separated, and the cells were washed three times by centrifugation (1,500 to 1,800 g, 15 minutes) with PRMI-1640 medium and suspended in RPMI-1640 medium supplemented with 10% FCS (5×10⁶cells/ml) for use as effector cells.

(c) ADCC Activity Measurement

To U-bottomed 96-well plates was added anti-GD₃chimera antibody KM-871 or anti-GD₃mouse antibody KM-641 in 50-μl portions to final concentrations-of 10 μg/ml. To each well were-added 100 μl of target cells (2×10⁵cells) and 50 μl of effector cells (5×10⁵cells) so that the ratio of effector cells to target cells should be 50:1 or 100:1. Reaction was performed at 37° C. for 4 hours, followed by centrifugation (1,200 rpm, 5 minutes). The amount of⁵¹Cr in each supernatant was determined using a γ-counter. The amount of⁵¹Cr resulting from spontaneous dissociation was determined by adding to target cells the medium alone in place of the antibody and effector cells and measuring the amount of⁵¹Cr in the supernatant in the same manner as described above. The total amount of free⁵¹Cr was determined by adding 5 N sodium hydroxide in μlace of the antibody and effector cells, proceeding as described above, and determining the amount of⁵¹Cr in the supernatant.

The ADCC activity was calculated as follows:

ADCC activity (%) = \frac{\begin{matrix} Amount of^{51} Cr in \\ sample supernatant \end{matrix} - \begin{matrix} Amount of^{51} Cr resulting \\ from spontaneous dissociation \end{matrix}}{\begin{matrix} Total amont of \\ free^{51} Cr \end{matrix} - \begin{matrix} Amount of^{51} Cr resulting \\ from spontaneous dissociation \end{matrix}} \times 100

As a control, the medium was added in place of the antibodies, the procedure mentioned above was followed, and the amount of 51Cr in the control supernatant was determined for ADCC activity calculation.

The results are shown inFIG. 34. In both cases of using lymphocytes and polymorphonuclear leukocytes as effector cells, chimera antibody KM-871 showed strong antibody-dependent cell-mediated cytotoxicity against the G361 cells as compared to mouse antibody KM-871, which indicates that chimera antibody KM-871 would be clinically more useful than mouse antibody Re-641.

12. In Vivo Therapeutic Effect of Anti-GD₃Chimera Antibody KM-871 (Therapeutic Effect on Transplanted Tumors)

Human malignant melanoma G361 cells (1×10⁷cells) were intracutaneously transplanted to abdominal parts of Balb/c nu/nu mice (5 to 7 aminals/group). Anti-GD₃chimera antibody KM-871 (100 μg/animal) was intravenously administered into mice four times starting from the next day of the transplantation of the tumor cells. To the mice of the control group, 100 μg of anti-GD₃mouse antibody KM-641 or anti-Sialyl Lea monoclonal antibody AMC-462 (ECACC 86050801) was intravenously administered five times starting from the day of the transplantation. The therapeutic effect on transplanted tumor cells was determined in terms of tumor size (volume) calculated by the following equation.
Tumor size (mm³)=0.4×a×b²

- a: major axis
- b: minor axis

The results are shown inFIG. 35. As shown inFIG. 35, remarkable growth of tumors was observed in the control group to which AMC-462 was administered, while the growth of tumors was significantly suppressed in the group to which KM-641 was administered. KM-871 showed further stronger therapeutic effect so that the establishment of tumors was completely inhibited 65 days after the transplantation.

REFERENCE EXAMPLE 1

(1) Preparation of Antigen

In 30 ml of chloroform/methanol (2/1) solution were dissolved 5 μg of ganglioside GD₃having NeuAcα2→NeuAcα2→3Gal sugar chain on its non-reducing end (Iatron), 0.5 μmol of dipalmitoylphosphatidylcholine (Sigma Chemical Co.), 0.5 μmol of cholesterol (Nakalai Tesque), 0.05 μmol of dipalmitoylphosphatidylic acid (Sigma Chemical Co.) and 2.5 μg of Lipid A (Funakoshi Pharmaceutical Co., Ltd.). The thus prepared solution was warmed at 45° C. to remove solvents, thereby obtaining a uniform lipid thin film. After completely removing solvents by sacking the film for 1 hour using a vacuum pump, the resulting film was mixed with 0.5 ml of PBS solution and stirred at 45° C. to obtain an antigen solution.

(2) Preparation of Antibody-Producing Cells

Mice was immunized by administering 0.5 ml of the antigen solution obtained in the above step (1) into the caudal vein once every week for 7 weeks. For further immunization, ganglioside GD₃-positive SK-MEL-28 (ATCC HTB 72) cells (1×10⁷cells) were intraperitoneally administered once every week for three weeks. On the third day after the last administration, spleen cells were prepared from each mouse for use in the following cell fusion.

(3) Preparation of Mouse Myeloma Cells

A mouse myeloma cell line P3-U1 having 8-azaguanine resistance was cultured in normal medium (RPMI1640 medium containing 10% fetal calf serum (FCS)) to obtain 2×10⁷or more cells for use in the following cell fusion as parent cells.

(4) Preparation of Hybridoma

The spleen cells and myeloma cells obtained in the above steps (2) and (3), respectively, were used in ratio of 10:1 and subjected to cell fusion in accordance with the aforementioned procedure. After culturing at 37° C. for 14 days in HAT medium (prepared by supplementing normal medium with hypoxanthin (10⁻⁴M), thymidine (1.5×10⁻⁵M) and aminopterine (4×10⁻⁷M)) under an atmosphere of 5% CO₂, fused cells were selected and cultured in HT medium (HAT medium minus aminopterine). Then, active wells were selected by assaying the antibody titers against ganglioside GD₃, and after changing to normal medium, cloning was repeated twice. Thereafter, hybridomas which showed specific reaction with ganglioside GD₃were selected by enzyme immunoassay or immunohistological evaluation (ABC method). That is, 2 ng of ganglioside GM₃(purified from dog erythrocytes in accordance with the method of Nores et al., J. Immunol., 139, 3171 (1987)) and 2 ng of ganglioside GD3 (Iatron) were dissolved in 2 ml of ethanol solution containing 5 ng of phosphatidylcholine (Sigma Chemical Co.) and 2.5 ng of cholesterol (Sigma Chemical Co.). The thus prepared solution was distributed in 20-μl portions into wells of a 96 well microtiter μlate (Flow Laboratories), air-dried and then subjected to blocking using 1% BSA-PBS solution. Each of the resulting hybridoma culture supernatant was distributed in 50-μl portions into the plate wells carrying a ganglioside GD₃adsorbed and the plate carrying ganglioside GM₃adsorbed thereon, and the reaction was allowed to proceed at 4° C. for 18 hours.

After the reaction, a hybridoma strain capable of producing mouse monoclonal antibody specifically reactive with ganglioside GD.sub.3 but not with ganglioside GM.sub.3 were selected in accordance with the known method (Cancer Res., 46, 4438 (1986)). This mouse monoclonal antibody was named “mouse monoclonal antibody KM-641”, and the hybridoma which produces this antibody was named “hybridoma KM-641”. The hybridoma KM-641 has been deposited on Sep. 27, 1990, with Fermentation Research Institute, Agency of Industrial Science and Technology, 1-3,Higashi 1 chome, Tsukuba-shi, Ibaraki, JAPAN, under the Budapest Treaty and has been assigned the designation as FERM BP-3116.

The present invention provides a process for the production of humanized chimera antibody wherein the chimera antibody can be produced easily without changing any one of amino acids of its mouse antibody variable region, as well as a humanized chimera antibody specific for ganglioside GD.sub.3.

While the invention has been described in detail and with reference to specific embodiments thereof, it will be apparent to one skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention.