KR20000010771A - Monoclonal antibody to cea, conjugates comprising said antibody, and theirtherapeutic use in an adept system - Google Patents

Abstract

PURPOSE: A method is provided to target an operating factor part to a cell representing an antibody CEA being existed in a mammal needing targeting the cell representing the antibody CEA. CONSTITUTION: The invention relates to an anti-CEA monoclonal antibody (named "806.077 antibody") of murine origin and useful for the diagnosis and therapy of cancer. The antibody complementarity determining regions (CDRs) have the following sequences: heavy chain; CDR1 DNYMH, CDR2 WIDPENGDTE YAPKFRG, CDR3 LIYAGYLAMD Y; and light chain CDR1 SASSSVTYMH, CDR2 STSNLAS, CDR3 QQRSTYPLT. Humanized antibodies are described. The antibody is preferably in the form of a conjugate with either an enzyme suitable for use in an ADEPT system, especially a carboxypeptidase, or with a co-stimulatory molecule such as the extracellular domain of human B7.1.

Description

Monoclonal Antibodies to CEA, Conjugates Containing the Antibodies and Therapeutic Uses in the ADP-based System

Transformation from normal tissue cells to tumor cells can be caused by altering the structure on the cell surface. The altered cell surface structure can act as an antigen, and the tumor modified structure represents one so-called tumor associated antigen [see, eg, Altered Glycosylation in Tumor Cells, Eds. Reading, Hakamori and Marcus 1988, Arthur R. Liss publ. .) Reference]. Such antigens can be used to generate and use single specific antibodies, for example for the first time by hybridoma techniques, as is now well known after initiation by Kohller and Milstein, Nature, 256, 495-497, 1975. have.

One tumor associated antigen is CEA (cancerous fetal antigen) as first disclosed in Gold and Freedman, J Exp Med, 121, 439, 1965. This antigen is present on the tumor cell surface and has also been demonstrated in serum.

The idea of using antibodies targeting tumor associated antigens as a preferred method for treating cancer has been suggested (Herlyn et al. (1980) Cancer Research 40, 717). Antibodies can be used to target a variety of chemicals and biologic agents to tumors, and the conjugates described above as the primary factor in various methods of in vitro and in vivo diagnostics have been particularly successful. In addition, the use of immunoconjugates in the treatment of cancer is also promising [Lord et al (1985) Trends in Biotechnology 3, 175; Vitetta et al (1987) Science 238, 1098. Such therapeutic uses have a greater technical need than diagnostic uses, and the tumor associated antigens targeted in these immunotherapeutic applications need to be highly tumor specific and not be expressed in significant amounts in human living tissue. While not intending to be limited to theoretical matters as well as specific tumor-associated tissue distribution properties, it is desirable in some applications that the antibody remain on the cell surface after antigen binding rather than rapidly internalizing. For example, in ADEPT (antibody directed enzyme prodrug therapy, see US Pat. Nos. 4757278 and 5,5959,90), it is preferred that the antibody remain on the cell surface to enable prodrug activation by antibody-enzyme conjugates.

Antibody conjugates are also used in tumor immunotherapy applications. The following describes the scientific background of the present application. To respond to immune stimulation, T-cells require two signals. The first signal is provided through the recognition of MHC expressing peptides by the T-cell receptor (TCR). However, TCR stimulation alone causes T-cell non-response or anergy, and a second or co-stimulation signal requires stimulation of specific T-cell activation and proliferation (Schwartz RHJExp.Med). , 1996, 184, 1-8). Upon receiving these two signals, the resulting cytotoxic T-cells kill the target cells to mediate the immune response. Many potential costimulatory molecules have been identified (eg, B7, ICAM, LFA-1 and LFA-3, CD40, CD70 and CD24, Galea-Lauri J. et al. Cancer Gene Therapy, 1996, 3, 202-213). The main costimulatory functions are two receptors CD28 and CTLA-4 (Hellstrom KE et al Immunol. Rev., 1995, 145, 123-145 and Lenschow DJ et al. Ann. Rev. Immunol., 1996, 14,233-258) It has been found to be provided by related molecules B7.1 (also known as CD80) and B7.2 (also known as CD86) that can interact. B7.1 and B7.2 are expressed on antigen expressing cells (APCs), such as dendritic cells, while CD28 and CTLA-4 are expressed on T-cells. B7.2 appears to be expressed constitutively on the surface of APC, but expression of B7.1 is upregulated after contact with T-cells. Similarly, CD28 is expressed on T-cells but alternately expresses CTLA-4 with downregulation after activation. Stimulation of CD28 and CTLA-4 by B7.1 and B7.2 represents a complex signaling pattern that regulates immune responses by controlling not only T-cell activation but also subsequent regulation of proliferation (Greene J. et al J. Biol. Chem., 1996, 271, 26762-26771). Sometimes the data reported by researchers studying these costimulatory molecules is due to this phenomenon.

In cancer, tumor-infiltrating lymphocytes have been identified, but the lack of an immune response to the tumor is due to T-cell anergy. Tumor cells may display specific or selective antigens on their surface, but lack B7.1 / B7.2 to avoid immune surveillance. This has been demonstrated in vivo experiments, where B7.1 / B7.2 transfected tumor cells are less tumorigenic than untransfected cells from the same cell line, and the transfected cells are preventive immunity when the parent cell is re-antigenated. Inducible (Townsend SE and Allison JP, 1993, Science, 259, 368-370). These results indicate that once stimulated, the immune response can be B7.1 / B7.2 independent. Hellstrom suggests that gene therapy results in the expression of B7.1 / B7.2 in tumor cells that could stimulate host responses that can reduce or eliminate disease. Gajusski (J. Immunol., 1996, 156, 465-472) and Matulonis et al., J. Immunol., 1996, 156, 1126-1131 are responsible for T-cell activation. B7.1 reported better than B7.2. It has been reported that the use of B7.1 as a solution (as a fusion with antibody constants) can only adequately provide costimulation to T-cells that accept TCR stimulation through independent sources (Linsley PS et al. J. Exp. Med., 1991, 173, 721-730).

Thus, there is a need for another improved anti-CEA antibody useful for the diagnosis and treatment of cancer.

The present invention relates to novel anti-CEA monoclonal antibodies (referred herein as "806.077 antibodies" or "806.077 Abs") useful for diagnosing and treating cancer.

1 shows antitumor activity of 806.077 antibody-CPG2 conjugate in the ADEPT model.

2 shows a plasmid map of pCF009.

FIG. 3 shows BIAcore data showing antibody-B7.1 fusion protein binding to immobilized CTLA4-Ig, with solid lines indicating test binding and dashed lines showing control.

The present invention is based on the discovery of a novel anti-CEA antibody referred to herein as an 806.077 antibody.

As a first aspect, the present invention provides an anti-CEA antibody comprising a complementarity determining region (CDR) in which the CDRs exhibit the following sequence:

a) heavy chain

CDR1 DNYMH (SEQ ID NO: 29)

CDR2 WIDPENGDTE YAPKFRG (SEQ ID NO: 31)

CDR3 LIYAGYLAMD Y (SEQ ID NO: 32)

b) light chain

CDR1 SASSSVTYMH (SEQ ID NO: 26)

CDR2 STSNLAS (SEQ ID NO: 27)

CDR3 QQRSTYPLT (SEQ ID NO: 28)

CDR or complementarity determining regions are sequences present in the hypervariable loops of antibody variable domains that are believed to be important for determining the specificity of antigen-antibody interactions [Kabat, EA, Lu, TT, Reid-Miller, M.Perry, HM & Gottesman, KS (1987), Sequences of Proteins of Immnunological Interest. 4th ed. Washington D.C.:United States Dept. of Health and Human Services; The reader is referred to this document for details of Kabat antibody residue number). However, CDRs described herein include framework residues involved in binding. For 806.077 antibodies, CDRs were determined through homology with the hypervariable sequences of other murine antibodies. As used herein, the terms "VK" and "VH" refer to variable regions contained in the light and heavy chains of an antibody, respectively. For structural analysis of antibody molecules, see Padlan (1994), Molecular Immunology 31, 169-217.

The light chain CDRs are as follows.

VK CDR1 Kabat residues 24-34, SASSSVTYMH (SEQ ID NO: 26);

VK CDR2 Kabat residues 50-56, STSNLAS (SEQ ID NO: 27);

VK CDR3 Kabat residues 89-97, QQRSTYPLT (SEQ ID NO: 28).

The heavy chain CDRs are as follows.

VH CDR1 Kabat residues 31 to 35B, DNYMH (SEQ ID NO: 29);

Preferred VH CDR1 Kabat residues 27 to 35B, FNIKDNYMH (SEQ ID NO: 30);

VH CDR2 Kabat residues 50-65, WIDPENGDTE YAPKFRG (SEQ ID NO: 31);

VH CDR3 Kabat residues 95-102, LIYAGYLAMD Y (SEQ ID NO: 32); And

Preferred VH CDR3 Kabat residues 93-102, including HVLIYAGYLA MDY (SEQ ID NO: 33).

Preferably the binding affinity of the CEA antigen is at least 10E-5M, more preferably at least 10E-6M, even more preferably at least 10E-7M, even more preferably at least 10E-8M, even more preferably At least 10E-9M, more preferably at least 10E-10M, in particular at least 10E-11M.

The term antibody, as used herein, generally refers to an immunoglobulin molecule (or a fragment thereof or a modified antibody construct such as a scFv that retains a specific CEA antigen binding). CDRs play a major role mainly in antigen binding, and non-CDR protein sequences are generally derived from immunoglobulins but may also be derived from the immunoglobulin domains of the immunoglobulin superfamily.

As a second aspect, the present invention provides a CEA antibody which optionally contains a humanized structure as follows.

Heavy chain variable region sequence (SEQ ID NO: 11)

Light chain variable region sequence (SEQ ID NO: 9)

Or a structure thereof, ie, any one of F (ab ') ₂ , F (ab'), Fab, Fv, single-stranded Fv, and V-min.

Among these, an F (ab ') ₂ fragment structure is preferable. Any suitable antibody fragment that retains 806.077 antibody binding properties may be used. For example, the recently reported antibody fragment "LF (ab) ₂ " is also preferred (Zapata (1995) in Protein Engineering, 8, 1057-1062). Disulfide bonded Fv may also be used. Alternatively, the antibody may form part of a conjugate as described below.

Preferred humanized antibodies contain any one or more of the following sequences, which may optionally be in the form of F (ab ') ₂ fragments.

Heavy chain variable region sequence, VH1 (SEQ ID NO: 55);

Light chain variable region sequence, VK4 (SEQ ID NO: 71);

Human CH1 heavy chain IgG3 constant region;

Human kappa light chain CL region; And

Human IgG3 hinge region.

As a third aspect, the present invention provides a polynucleotide sequence capable of encoding the heavy or light chain variable region of the CEA antibody of the present invention. Preferably the heavy or light chain variable region is fused (optionally via some binding sequence) to a gene encoding a protein effector moiety (part of the conjugate, see below), preferably through the antibody heavy chain. Generally, the fusion can be at the N terminus or C terminus of the antibody chain. In the case of B7 conjugates, the fusion is preferably at the N terminus of the antibody chain.

CPB has an N-terminal pro domain, which is thought to assist in the correct folding of the protein before it is removed to dissociate the active enzyme. proCPB removes the pro domain (eg, via trypsin treatment) from the N terminus of the fusion construct when its C terminus is fused to the N terminus of the antibody chain. Alternatively, if proCPB is bound to the C terminus of the antibody chain, a problem arises that the pro domain must be removed from the "middle" of the construct without destroying the fusion protein. The solution is to coexpress each of the pro domains (trans). The advantage of this method is that once the cell line is constructed, there is no need to trypsin activate the expressed fusion protein to remove the CPB pro domain. Constructs containing proCPB fused at the C-terminus to the N-terminus of the antibody chain provide the advantage of not having to construct co-expressing cell lines expressing large amounts of the pro domain while expressing large amounts of other proteins.

Conservative amino acid analogs of certain amino acid sequences herein, ie those that retain the binding properties of the CEA antibodies of the present invention, but that one or more conservative amino acids are substituted, deleted or added to result in different sequences. However, amino acid sequences specifically described herein are preferred. Common conservative amino acid substituents are listed in the following table.

Conservative substitutions Original amino acids Substitution example Preferred Substituents Ala (A) Val; Leu; Ile Val Arg (R) Lys; Gln; Asn Lys Asn (N) Gln; His; Lys; Arg Gln Asp (D) Glu Glu Cys (C) Ser Ser Gln (Q) Asn Asn Glu (E) Asp Asp Gly (G) Pro Pro His (H) Asn; Gln; Lys; Arg Arg Ile (I) Leu; Val; Met; Ala; Phe; Norleucine Leu Leu (L) Norleucine; Ile; Val; Met; Ala; Phe Ile Lys (K) Arg; Gln; Asn Arg Met (M) Leu; Phe; Ile Leu Phe (F) Leu; Val; Ile; Ala Leu Pro (P) Gly Gly Ser (S) Thr Thr Thr (T) Ser Ser Tyr (Y) Trp; Phe; Thr; Ser Phe Val (V) Ile; Leu; Met; Phe; Ala; Norleucine Leu

As used herein, variants that possess the property of hybridizing to a particular sequence of interest under stringent conditions as nucleic acid variants (deletions, substitutions and additions) of a particular nucleic acid sequence are also included in the present invention. Hybridization tests are described in Example 9 below. However, specifically described nucleic acid sequences are preferred. Also contemplated are peptide nucleic acids that are acceptable equivalents of polynucleotide sequences to ensure that there is at least no need for translation into proteins (Wittung (1994) Nature 368, 561).

In a fourth aspect, the present invention provides an antibody or antibody fragment as described herein characterized in that it is humanized.

Humanized antibodies, associated fragments or antibody binding constructs consist primarily of the structural backbone of human-derived immunoglobulin sequences that support nonhuman-derived amino acid sequences that exist around and within the antigen binding site (complementarity determining regions or CDRs). Polypeptides (this technique is known as CDR transplantation, which often includes some skeletal changes, see Examples below). Suitable methods are described, for example, in WO 91/09967, EP 0328404 and Queen et al. Proc Natl Acad Sci 86, 10029, Mountain and Adair (1989) Biotechnology and Genetic Engineering Reviews 10, 1 (1992), and as an alternative method of humanization, antibody coating of surface residues (EP 519596, Merck / NIH, Padlan et al. May also be considered. Preferred humanized 806.077 antibodies are any one of Examples 11-47 or 107-122. Preferred humanized heavy chain variable region is VH1 (see Examples). Preferred light chain variable regions are VK4 which optionally contain any of the additional changes described in Examples 107-109. Preferred human heavy chain constant region is IgG3.

As a fifth aspect, the present invention provides a chimeric humanized antibody. Antibody 806.077 The preparation of chimeric humanized antibody fragments of antibodies is described in Example 8. Chimeric antibodies are generally made by combining the variable regions from one species and the constant regions of another antibody from another species.

As used herein, the term "humanized" refers to any method of humanization, such as, for example, CDR transplantation or chimeric antibody preparation, or any hybridization method, such as a combination of chimeric light and CDR-grafted heavy chains. (See Example 110 for suitable embodiments).

Specifically, repeated in vivo administration of rodent antibodies, either alone or as conjugates, results in an immune response response to rodent antibodies in the receptor, a so-called HAMA (human anti mouse antibody) response. The HAMA response limits the effectiveness of the medication when repeated doses are required. Immunogenicity of the antibody can be reduced by chemical modification of the antibody with a hydrophilic polymer such as polyethylene glycol or by genetic engineering to make the antibody binding structure more human-like. For example, a recombinant chimeric antibody can be obtained by replacing the gene sequence for the variable domain of a rodent antibody that binds to CEA with the variable region domain of a human myeloma protein. The procedure of this method is described in detail in EP 194276, EP 0120694, EP 0125023, EP 0171496, EP 0173494 and WO 86/01533. Alternatively, a human antibody having the specificity of the original rodent antibody can be obtained by isolating or synthesizing the gene sequence encoding the CDRs in the CEA binding rodent antibody and substituting the corresponding sequence region in the homologous human antibody gene. . The procedure of this method is described in EP 023940, WO 90/07861 and WO91 / 09967. Alternatively, changing the surface residues of many of the variable domains in rodent antibodies to residues commonly found in human homologous antibodies yields rodent antibodies with surface “veneers” of the residues, which are present in the human body. Will be perceived as self. This method is described by Padlan et al (1991), Mol. Immunol. 28, 489.

As a sixth aspect, the invention provides a host cell transformed with a polynucleotide sequence encoding at least a variable region in a heavy or light chain of a CEA antibody of the invention, optionally in the form of a conjugate, or a mutant non-human animal generated from the host cell or Provide mutant plants.

In a seventh aspect, the present invention provides hybridoma 806.077 and modified cell lines thereof deposited as ECACC Accession No. 96022936.

The hybridoma 806.077 antibody was submitted to the European Animal Cell Culture Collection (ECACC) at the PHLS Center for Applied Microbiology and Research, Wiltshire SP4 0JG Salisbury Photon, UK, under the Budapest Treaty, February 29, 1996. It is deposited as 96022936.

In an eighth aspect, the present invention provides plasmid pNG3-Vkss-HuCk deposited as NCIMB Accession No. 40798.

Plasmid pNG3-Vkss-HuCk was deposited on April 11, 1996, at the National Industrial Marine Bacteria Collection Center (NCIMB) in St. Masha Drive 23, Aberdeen, AB2, Scotland, UK, under the Treaty of Budapest. Deposited.

In a ninth aspect, the present invention provides plasmid pNG4-VHss-HuIgG2CH1 'deposited as NCIMB Accession No. 40797.

Plasmid pNG4-VHss-HuIgG2CH1 'was deposited on April 11, 1996, at the National Industrial Marine Bacteria Collection Center (NCIMB) in Aberdeen Abbey 2 Al Wye St. Masha Drive 23, Aberdeen, AB, Scotland. Has been deposited as.

In a tenth aspect, the present invention provides plasmid pNG3-Vkss-HuCk-NEO deposited as NCIMB Accession No. 40799.

Plasmid pNG3-Vkss-HuCk-NEO was deposited on April 11, 1996, at the National Industrial Marine Bacteria Collection Center (NCIMB) in Aberdeen Abbey 2, Al Wye St. Marcia Drive 23, Aberdeen, Scotland, UK, under the Treaty of Budapest. Has been deposited as.

As an eleventh aspect, the present invention provides a method of preparing at least the variable region of the heavy or light chain of an anti-CEA antibody as described herein, the method comprising the following steps.

a) transforming the host cell with a polynucleotide sequence encoding at least the variable region of the heavy or light chain of the anti CEA antibody and optionally growing the transformed host cell into a mutant non-human mammal or mutant plant;

b) treating said host cell, mutant non-human mammal or mutant plant with conditions conducive to expression and selective secretion of at least variable regions; And

c) optionally at least partially purifying the variable region.

As a twelfth aspect, the present invention provides a method of preparing an antibody or conjugate as described herein, the method comprising the following steps.

a) treating a host cell, mutant non-human mammal or mutant plant, or 806.077 hybridoma, as described herein under conditions conducive to expression and optionally secretion of the antibody or conjugate; And optionally

b) at least partially purifying the antibody or conjugate.

The heavy chain variable region and the light chain variable region are preferably expressed in the same cell and then combined to form an anti CEA antibody. The heavy chain variable region or the light chain variable region is preferably fused (optionally via some binding sequence) to a gene encoding a protein effector moiety (part of the conjugate, see below), which fusion is via the antibody heavy chain. desirable. Generally, the fusion can be at the N terminus or C terminus of the antibody chain. In the case of B7 conjugates, fusion is preferred at the N-terminus of the antibody chain. CPB has an N-terminal pro domain that is thought to assist in the correct folding of the protein before removing the pro domain to dissociate the active enzyme. proCPB, when its C terminus is fused with the N terminus of the antibody chain, removes the pro domain from the N terminus of the fusion construct (eg, trypsin treatment). Alternatively, binding proCPB to the C terminus of the antibody chain raises the problem of removing the pro domain from the "middle" of the construct without destroying the fusion protein. The solution is to co-express the pro domain separately (in trans form). This approach has the advantage that once the cell line is constructed, there is no need for trypsin activation of the expressed fusion protein to remove the CPB pro domain. Constructs with proCPB, with the C terminus fused to the N terminus of the antibody chain, have the advantage of not having to construct a coexpressed cell line expressing a large amount of the pro domain while expressing a large amount of other proteins.

As a thirteenth aspect, the present invention provides a

a) culturing the hybridoma 806.077 antibody deposited as ECACC Accession No. 96022936 in medium under conditions conducive to expressing the antibody;

b) obtaining antibody 806.077 antibody from this culture medium; And

c) Optionally, providing a method of preparing monoclonal antibody 806.077, comprising preparing the F (ab ') ₂ fragment of antibody 806.077 antibody by enzymatic cleavage.

In a fourteenth aspect, the invention provides a conjugate containing an anti-CEA 806.077 antibody of the invention and an effector moiety as described herein. An effector moiety is an entity that has the effect of providing another activity (eg, an enzyme, toxin or radioactive ligand) to an 806.077 antibody in forming a conjugate.

In one embodiment, the effector moiety is preferably an enzyme suitable for use in the ADEPT system. In the international patent application WO 96/20011 published on July 4, 1996, the inventors have proposed a "reverse polarity" ADEPT system based on mutant human enzymes which have the advantage of low immunogenicity over eg bacterial enzymes. . Specific host enzymes were human pancreatic CPB (see, eg, [D253K] human CPB of Example 15 and [D253R] human CPB of Example 16) and prodrugs thereof (see Examples 18 and 19). Mutation of this host enzyme results in a change in the way the enzyme interacts with the prodrug in terms of substrate recognition compared to the native host enzyme. As a follow-up study by the inventors, the international application PCT / GB96 / 01975 (published March 6, 1997 as WO97 / 07796) discloses a study of mutant CPB enzyme / prodrug combinations for ADEPT. Preferred enzymes for ADEPT are any one of CPG2 or a reverse polarity CPB enzyme, such as [D253K] HCPB, [G251T, D253K] HCPB or [A248S, G251T, D25K] HCPB.

In addition, the 806.077 antibody conjugates can be used for tumor immunotherapy. Thus, in another preferred embodiment, the conjugate effector moiety is a costimulatory molecule, which is preferably B7, more preferably human B7.1 or B7.2, in particular human B7.1. Preferably, the conjugate is in the form of a fusion protein, which is preferably formed by binding the C-terminus of the co-stimulatory molecule to the N-terminus of the 806.077 antibody chain, preferably to the heavy-chain N-terminus of the antibody chain, The 806.077 antibody preferably contains an Fc antibody region, more preferably an F (ab ') ₂ antibody fragment, and more preferably the antibody is a humanized or human antibody. Particularly preferred conjugates are those described in Example 104 below.

The method of using antibodies to target costimulatory molecules to tumor cells provides antigen expression functions to tumor cells such that T-cells receive specific TCR stimulation from the tumor cells themselves and receive costimulatory signals from antibody targeted molecules. It is thought to give. The use of human or humanized antibodies is preferred for treating human tumors because the use of mouse antibodies in humans can stimulate an immune response that reduces the effects of repeated treatment. The use of fusion proteins in which the tumor antigen binding region is bound to the extracellular portion of the costimulatory molecule is novel. Hayden et al, Tissue Antigens, 1996, 48, 242-254 have reported on the use of bispecific antibody molecules that bind anti-CD28 binding domains and anti-tumor antigen binding domains. The molecule can interact with CD28 on T-cells, but the signal it can deliver is qualitatively different from the signal provided by the native CD28 ligand, such as binding affinity between B7.1 and CD28. Much greater than binding affinity. Interspecies homology between B7.1, B7.2 and CD28 indicates the evolutionary conservation of binding region sequences. In conclusion, it is contemplated that, for example, B7.1 from humans can interact with CD28 from mouse, providing similar costimulatory signals. Thus, human or humanized proteins are preferred for treating human diseases.

In addition, the use of human or humanized proteins in animal models will have similar effects to those expected in humans, through which human B7.1 / B7.2 and human / humanization can be used to treat human diseases. Relevant data on the efficacy of the antibody fusion proteins can be obtained.

Conjugation of the effector moiety to the antibody can be carried out by any suitable method, such as chemical bonding via a heterobifunctional linker or recombinant gene fusion techniques. In general, the fusion protein is a preferred conjugate, specifically a conjugate with HCPB or B7.

Preferred conjugates are molecules in which the effector moiety is selected from the following components, optionally in the form of fusion proteins.

a) enzymes suitable for use in the ADEPT system,

b) CPG2;

c) [G251T, D253K] HCPB;

d) [A248S, G251T, D253K] HCPB;

e) costimulatory molecules;

f) the extracellular domain of B7;

g) the extracellular domain of human B7.1; And

h) extracellular domain of human B7.2.

The conjugates of the present invention need not necessarily consist of one effector molecule and one antibody molecule. For example, the conjugate may comprise one or more effector molecules per antibody molecule. In general, the F (ab ′) ₂ antibody conjugate, which is a fusion between the antibody and the enzyme or extracellular domain of B7, may have 2 moles of enzyme or B7 per mole of antibody.

Particularly preferred conjugates are fusion proteins selected from any one of the following conjugates (sequences are described from the N terminus to the C terminus).

a) an antibody Fd'chain having a CH1 constant region derived from structure VH1 (SEQ ID NO: 55) / IgG3 / a hinge region derived from IgG3 having a C terminus bonded to an N terminus of [A248S, G251T, D253K] HCPB; And

Humanized 806.077 F (ab ') ₂ -{[A248S, G251T, D253K] HCPB} ₂ fusions containing an antibody light chain having a CL region derived from the structure VK4 (SEQ ID NO: 71) / kappa light chain;

b) [A248S, G251T, D253K, wherein the C terminus is linked via a (GGGS) ₃ linker to the N terminus of the antibody Fd 'chain having a CH1 constant region derived from structure VH1 (SEQ ID NO: 55) / IgG3 / a hinge region derived from IgG3 ] HCPB; And

{[A248S, G251T, D253K] HCPB} ₂ -humanized 806.077F (ab ′) ₂ fusions containing an antibody light chain having a CL region derived from the structure VK4 (SEQ ID NO: 71) / kappa light chain; And

c) a human B7.1 extracellular domain in which the C terminus is fused to the N terminus of the antibody Fd 'chain having a CH1 constant region derived from structure VH1 (SEQ ID NO: 55) / IgG3 / a hinge region derived from IgG3; And

(Human B7.1 extracellular domain) ₂ -humanized 806.077F (ab ') ₂ fusion containing an antibody light chain with a CL region derived from the structural VK4 (SEQ ID NO: 71) / kappa light chain.

In this specification, the antibody hinge regions described in connection with the conjugates are shown according to the principles defined by Padlan in Molecular Immunology 31, 169-217, 1994, specifically see Table 2. Such particularly preferred conjugates are those in which 2 moles of enzyme or B7.1 are bound per mole of F (ab ') ₂ . The forward slash ("/") is a simple separate notation representing separate structural components joined by peptide bonds that form part of the conjugate.

The humanized sequences of the VH1 and / or VK4 variable regions have the advantage of preserving good binding with minimal additional changes required in the human backbone. The IgG3 hinge region has the advantage of providing good homogeneity of F (ab ') ₂ production and product.

In another preferred embodiment associated with the particularly preferred conjugate described above, the fusion is via an antibody light chain. In another preferred embodiment of the particularly preferred conjugate described above, the hinge region derived from the CH1 constant region / IgG structural factor derived from IgG3 may be replaced with the corresponding IgG2 factor.

If the effector is a toxin, this toxin moiety is generally cytotoxic and contains components that can kill cells after internalization.

The toxin moiety and the anti-CEA antibody may directly or indirectly bind to each other. The toxin moiety and the anti-CEA antibody are generally bound so that the anti-CEA antibody can bind to its target cells due to the geometry of the conjugate. Advantageously, the toxin moiety and the anti-CEA antibody are extracellularly stable and intracellularly unstable, so that the toxin moiety and the anti-CEA antibody are bound outside the target cell, but after the internalization the toxin moiety is bound. Thus, the conjugate has a site that is cleavable intracellularly and / or is stable extracellularly.

Examples of the conjugate in which the toxin moiety is directly bonded to the binding moiety of the target cell include that the toxin moiety and the anti-CEA antibody are bound by a disulfide bridge formed between a thiol group on the toxin moiety and a thiol group on the antiCEA antibody. Details regarding the preparation and properties of immunotoxins and other conjugates are given in European Patent Application EP 528 527 (Publication Number), which is incorporated herein by reference.

In a fifteenth aspect, the present invention provides a polynucleotide sequence capable of encoding a polypeptide of an antibody or conjugate as described above. The term "encryptable" is meant to include polynucleotide sequences that show degeneracy in the genetic code in that some amino acids are encoded by one or more codons.

As a sixteenth aspect, the present invention provides a vector containing the aforementioned polynucleotide.

As a seventeenth aspect, the present invention provides a host cell transformed with the above-described polynucleotide sequence or a mutant non-human animal or mutant plant generated from the host cell.

In an eighteenth aspect, the present invention provides a pharmaceutical composition comprising a pharmaceutically acceptable diluent or carrier in combination with a conjugate of the invention described herein, and optionally in a form suitable for intravenous administration.

As a nineteenth aspect, the present invention provides a conjugate as described herein for use as a medicament.

In a twentieth aspect, the present invention provides the use of the conjugate as described herein in the manufacture of a medicament for treating a tumor disease.

Dosage and dosage regimen will depend on the particular effector portion used, the population of target cells, and the history of the patient. Dosages of the conjugates generally range from 0.1 to 1 mg / kg body weight.

Conjugates of the invention are generally administered in the form of a pharmaceutical composition. Accordingly, according to the present invention, there is provided a pharmaceutical composition containing a pharmaceutically acceptable diluent or carrier with a conjugate (as described herein). Examples of such compositions are described in Example 10.

The pharmaceutical compositions of the present invention can be formulated in a variety of dosage forms. Generally, the conjugates of the present invention are preferably parenterally, preferably intravenously. Specific parenteral pharmaceutical compositions are formulated in unit dosage forms suitable for injection administration. Thus, examples of particularly suitable compositions include solutions, emulsions or suspensions containing immunotoxins together with pharmaceutically acceptable parenteral carriers or diluents. Suitable carriers or diluents include aqueous excipients such as water or saline, and nonaqueous excipients such as fixative oils or liposomes. This composition can contain an agent which improves the stability of the conjugate in this composition. For example, the composition may comprise a buffer. The concentration of the conjugate may vary, but is generally formulated at a concentration of about 1-10 mg / ml.

In a twenty-first aspect, the present invention provides an expression vector encoding an anti-CEA antibody of the present invention as described herein.

In a twenty second aspect, the present invention provides an expression vector encoding at least the variable region of the heavy or light chain of an anti-CEA antibody as described herein.

In a twenty third aspect, the present invention provides host cells transformed with the vectors as described herein for expression herein.

In a twenty fourth aspect, the present invention provides host cells transformed with the polynucleotide sequence as described herein.

Mammalian cells (CHO, COS, myeloma) have been used as hosts for the co-expression of cDNAs and fragments of antibodies H and L chains and have been used to generate antibodies with specified binding activity.Bebbington, C., 1991 , Methods, vol 2, p136-145 and Adair, J., 1992, Immunological Reviews, vol 130). For expression, expression of a construct that directly expresses an active CPB, COS or CHO cell expression system is preferred. The cDNA could be introduced through a plasmid and then integrated into the chromosomal DNA of CHO cells, in particular, or replicated with a very high copy number, especially in COS cells. In general, plasmids require a selection marker for continually persisting in the transfected host, a eukaryotic promoter effective for high efficiency transcription from cDNA, restriction enzyme sites convenient for cloning and polyadenylation, and transcription termination signals for message stability. Shall be. These various vectors are described in Bebbington, C. et al., 1992, Bio / Technology, vol 10, p. 169-175 and Wright, A., 1991, Methods, vol 2, p125-135, Suitable vectors are commercially available (eg pRc / CMV, Invitrogen Corp.).

The expression of certain antibody fragments in E. coli has been well reported (Pluckthun, A., Immunological Reviews, 1992, vol 130, p151-188 and Skerra, A., Current Opinion in Immunology, 1993, vol 5, p256-). 262). Intracellular expression of Fd and L chains has also been disclosed (Cabilly, S., 1989, Gene, vol 85, p553-557), but in vitro refolding and recombination of these chains is required to demonstrate binding activity. (Buchner, J and Rudolph, R., 1991, Bio / Technology, vol 9, p157-162). A more effective route for obtaining active antibody fragments is through Periplasm secretion (Better, M. et al., 1988, Science, vol 240, p1041-1043). The components of the H chain and L chain of the antibody fragment are coexpressed from a single plasmid. Each antibody chain has a leader peptide of bacteria that leads the chain to E. coli periflorum, where the leader is cleaved and the free chains bind to produce soluble and active antibody fragments. . This process is thought to be similar to the natural process in which antibody chains expressed in eukaryotic cells pass into the lumen of the endoplasmic reticulum before binding to complete antibodies. This process often indicates the presence of binding activity in the culture supernatant.

Some expression systems involve transforming host cells with vectors, such expression systems being known, for example, E. coli, yeast and mammalian hosts (Methods in Enzymology 185, Academic Press 1990). As other expression systems, for example, a target gene, preferably a vector, is cleaved from and then introduced into the pronucleus of the mammalian conjugate, preferably the target gene, which is preferably bound to the mammary gland promoter for inducing the expressed protein into the milk of the animal (usually Also included are one of the two nuclei present in the pronucleus (usually a male nucleus), followed by a mutant non-human mammal that is then implanted into wool. A percentage of the animals produced by the wool will carry and express the transgene integrated into the chromosome. In general, the integrated genes are passaged to offspring through normal breeding and are readily distributed to livestock. The desired protein is preferred because it can be obtained simply from the milk of mutant females. See, for example, the following publications: Simons et al. (1988), Bio / Technology 6: 179-183; Wright et al. (1991) Bio / Technology 9: 830-834; US 4,873,191 and US 5,322,775]. Manipulation of mouse embryos is described in Hogan et al. "Manipulating the Mouse Embryo; A Laboratory Manual", Cold Spring Harbor Laboratory 1986.

Mutant plant techniques are also described, for example, in Swain W.F. (1991) TIBTECH 9: 107-109; Ma J.K.C. et al (1994) Eur. J. Immunology 24: 131-138; Hiatt A. et al (1992) FEBS Letters 307; 71-75; Hein M.B. et al (1991) Biotechnology Progress 7: 455-461; Duering K. (1990) Plant Molecular Biology 15: 281-294.

If desired, host genes can be inactivated or modified via standard procedures as outlined below, which methods are described, for example, in Gene Targeting; A Practical Approach, IRL Press 1993. Cloning a gene or part of it into a vector and inserting a selection marker (eg Neo) into the gene so that its function is disrupted After linearizing this vector, embryonic liver (ES) cells (eg mouse) Transformation from the 129 / Ola state, usually by electropenetration, then homologous recombination occurs in a proportion of hepatocytes, resulting in the propagation of hepatocytes containing gene breakdown and blasts (eg from C57BL / 6J mice). After injection into the mother, it is transplanted into wool and developed.The chimeric progeny can be identified by the outer marker markers. To identify, breed the chimera by conjugating mice with genetic markers that differentiate between ES-derived and host subcellular-derived ones, and half of ES-derived conjugates will contain genetic modifications. Identify (50% of the offspring) that contain gene breakdown (eg by Southern blotting) The selected offspring are heterozygotes, can be propagated to another heterozygote, and then homozygous offspring Screening (approximately 25% of offspring): microinjection of DNA into the pronucleus, spheroplast fusion (Jakobovits et al. (1993) Nature 362: 255-258) or lipid-mediated transfection of ES cells (Lamb et (1993) Hybridization of mutant animals obtained by known techniques such as Nature Genetics 5 22-29) and mutant animals with gene knockout resulted in endogenous gene knockout and heterologous inheritance. Mutant animals with child substitutions are obtained.

ES cells containing target gene disruption can be further changed by transforming with a target gene sequence containing a specific change, preferably by cloning into a vector and then linearizing and transforming. After homologous recombination, the altered gene is introduced into the genome. Such embryonic stem cells can be used to obtain mutants as described above.

The term "host cell" includes all prokaryotic or eukaryotic cells suitable for expression techniques, such as bacteria, yeast, plant cells and non-human mammalian conjugates, oocytes, blast cells, embryonic liver cells and other cells suitable for mutation techniques. In addition, where allowed throughout this specification, the term "host cell" includes mutations originating from transformed non-human mammalian conjugates, oocytes, blasts, embryonic liver cells, plant cells and any other cells suitable for mutation techniques. Plants or non-human mammals.

In a twenty-fifth aspect, the present invention provides a method of treating a human or animal in need thereof, comprising administering to a human or animal a pharmaceutically effective amount of a conjugate as described herein.

In a twentysixth aspect, the invention expresses antigen CEA present in a mammal in need of targeting a cell expressing antigen CEA, comprising administering a pharmaceutically effective amount of a conjugate of the invention as described herein. A method of targeting an effector moiety to a cell is provided.

In a twenty seventh aspect, the present invention provides the use of an antibody as described above in a diagnostic method.

One of the diagnostic methods is immunoassay. In vitro assays for immunoassays based on novel antibodies according to the present invention may be performed using CEA present in a sample to be tested by using the antibody of the present invention in a labeled or unlabelled form according to conventional immunological techniques in the art. Measuring complex formation of the antibody. As one example, an antibody may be labeled with a detectable label, such as radiolabel, chemiluminescent, fluorescent or enzyme label. Alternatively, antibodies are detected through complexes formed with labeled substances or using unlabeled techniques such as biosensor methods based on surface plasmon resonance. The sample is in the form of a bodily fluid such as serum or a tissue preparation (histochemical assay).

For in vivo diagnostic purposes, the antibodies according to the invention are labeled with a suitable externally detectable label such as, for example, radiolabels or heavy metal atoms, and then administered to a subject for use in determining possible local accumulation in the body.

In in vitro diagnosis of cancer, fluorescent markers can be directly quantified after binding or detection of anti-CEA antibodies to enzymes such as horseradish peroxidase and bacterial luciferase that can produce measurable signals, or Can bind to radioisotopes. In standard immunoassays, such conjugates provide a means of measuring the presence or absence of CEA in body tissues and, consequently, provide a quick and convenient assay for diagnosing tumor disease. For general information on methods related to enzyme immunoassays, see the following references. ETMaggio, CRC Press and US Patent 3690 8334, US Patent 3791 932, US Patent 3817 837, US Patent 3850 578, US Patent 3853 987, US Patent 3867 517, US Patent 3901 654, US Patent 3935 074, US Patent 3984 533 , US Patent 3996 345 and US Patent 4098 876.

For in vivo diagnosis of cancer, anti-CEA antibodies can be bound to isotopes of elements such as yttrium, technetium or indium, or to heavy metal isotopes that can be measured with systemic imaging cameras (Larson, SM, 1987, Radiology, 165, 297-304).

In the treatment of cancer, preferred embodiments include anti-CEA antibodies capable of directly killing cancer cells or binding to effector moieties capable of killing and killing suitable prodrugs, particularly in the ADEPT system. Tumor cells in ADEPT can be selectively killed by binding anti-CEA antibodies to enzymes that can catalyze the conversion of non-toxic doses of prodrugs into potent toxic drug compounds. Administration of the conjugate localizes the enzyme activity at the tumor site. Continuous prodrug administration produces toxic drugs locally, resulting in selective killing of the tumor site. This method is described in WO88 / 07378, US Patent 4,975,278, US Patent 5,405,990 and WO89 / 10140. In addition, antibody 806.077 can also be used in tumor immunotherapy as described above by binding to a costimulatory molecule.

Tumor cells can be selectively killed by direct or chemical derivatization of anti-CEA antibodies and binding to macrocycle chelators containing high energy radioisotopes such as ⁹⁰ Y, ¹³¹ I and ¹¹¹ In. Anti-CEA antibodies act to localize isotopes to tumors, and the radiation emitted by these isotopes destroys the DNA of surrounding cells, killing tumors.

Tumor cells can also selectively kill anti-CEA antibodies by binding to cytotoxic drugs and cytostatic drugs such as methotrexate, chlorambucil, adriamycin, daunorubicin and vincristine. The drug has been used in hospitals for many years, but therapies that can be used have been limited because of nonspecific toxicity. The binding of the drug to the CEA antibody localizes the drug only at the tumor site, and thus the dose of drug delivered only to the tumor without causing the unacceptable side effects of the drug acting on the bone marrow or other tissues such as the nervous system. Can be increased.

Antibodies of the invention are effective for many applications that have been improved by reducing the size of the antibody binding structure thereby improving tissue permeability and other pharmacokinetic properties of the pharmaceutical composition. This effect can be obtained by removing the Fc region of the antibody molecule by enzymatic or genetic engineering to generate recombinant Fab 'or F (ab') ₂ fragments.

In addition, genetic engineering can be used to further reduce the size of anti-CEA antibodies. That is, the Fv containing CDRs can be manipulated and expressed in a separated state, and then chemically crosslinked using, for example, disulfide bridges. Alternatively, the light and heavy chain domains that make up the Fv structure can be generated as a single polypeptide chain (SCFv) by fusing a Fv domain with a linker peptide sequence consisting of from the native C terminus of one domain to the N terminus of the other domain ( PCT / US87 / 02208 and US Patent 4704692). Alternatively, a single Fv domain may be expressed in isolation to form a single domain antibody or dAb as described by Ward et al Nature (1989) 341,544. Another form of anti-CEA antibody encompassed by the present invention is a V-min construct as disclosed in International Patent Application WO94 / 12625 (Inventor Slater and Tims). Abbreviations used herein are as follows.

ADEPT Antibody-Directed Enzyme Prodrug Therapy APC Antigen expressing cells CDR Complementarity Determination Zone CEA Carcinoma Fetal Antigen CL Constant domains of antibody light chains CPB Carboxypeptidase B CPG2 Carboxypeptidase G2 DAB Substrate 3,3'-diaminobenzidine tetrahydrochloride DEPC Diethylpyrocarbonate DMEM Dulbecco's Modified Eagle Badge ECACC European Animal Cell Culture Collection Center EIA Enzyme Immunoassay ELISA Enzyme-Coupled Immunosorbent Assays

FCS Fetal bovine serum Fd Heavy chain of Fab, Fab 'or F (ab') ₂ optionally containing a hinge HAMA Human anti mouse antibody HCPB Human carboxypeptidase B, preferably of the pancreatic carboxypeptidase B Hinge (IgG hinge) Proline-rich short peptide containing cysteine that bridges two heavy chains HRPO Horseradish Peroxidase NCA Nonspecific crosslinking reaction antigens NCIMB National Industrial Marine Bacteria Collection PBS Phosphate buffered saline PCR Polymerase chain reaction preproCPB ProCPB with an N-terminal leader sequence proCPB CPB with N-terminal pro domain SDS-PAGE Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis TBS Tris buffered saline VH Variable region of heavy chain antibody chains VK Variable region of the light chain antibody chain

The present invention is described in detail through the following examples, but is not limited thereto (supported by reference examples after the examples).

Unless otherwise indicated, DNA is recovered and purified using a GENECLEAN ^™ II kit (Stratech Scientific Ltd. or Bio 101 Inc.). The kit comprises 1) 6M sodium iodide; 2) concentrated solution of sodium chloride, tris and EDTA to prepare sodium chloride / ethanol / water rinse; 3) Glassmilk-contains 1.5 ml vials containing 1.25 ml of an aqueous suspension of a specially prepared silica matrix. DNA purification techniques are based on the Vogelstein and Gillespie methods known from Proceedings of the National Academy of Sciences USA (1979) Vol 76, p. 615. Briefly, the kit procedure is as follows. Three volumes of sodium iodide solution contained in the kit are added to one volume of gel fragment. The mixture is heated at 55 ° C. for 10 minutes to melt agarose, then glassmilk (5 to 10 ml) is added and mixed well, then left at room temperature for 10 minutes. Spin down the glass milk and wash three times with NEW WASH (0.5 mL), a component of the kit. Air dry to remove wash buffer from this glassmilk. The dried glass milk was incubated with water (5-10 mL) at 55 ° C. for 5-10 minutes to elute DNA. The aqueous supernatant containing the eluted DNA is recovered by centrifugation. The elution step is repeated to collect the supernatant.

Competent E. coli DH5α cells were obtained from Life Technologies Limited (DH5α competent cells of maximum potency).

Trace preparation of the double-stranded plasmid DNA is made using the RPM ^™ DNA preparation kit or similar product from Bio101 Incorporated (List # 2070-400). The kit contains an alkaline lysate solution that dissociates plasmid DNA from bacterial cells and a glass milk in a spin filter that adsorbs dissociated DNA, which is then sterile in water or 10 mM Tris-HCl, 1 mM EDTA, pH7.5 Elute.

Serum-free medium is OPTIMEM ^™ I Reduced Serum Medium with GibcoBRL Product No. 31985.

LIPOFECTIN ^™ Reagent (GibcoBRL Product No. 18292-011) is a cationic lipid N- [1- (2,3-dioleyloxy) propyl] -n, n, n-trimethylammonium chloride (DOTMA) contained in membrane filtered water. ) And dioleyl phosphatidylethanolamine (DOPE) 1: 1 (w / w) liposome preparation. This reagent spontaneously binds to DNA to form a lipid-DNA complex [Felgner et al. Proc. Natl. Acad. Sci. USA (1987) 84, 7431].

G418 (sulphate) is GENETICIN ^™ (GibcoBRL Product No. 11811), a gentamicin-based aminoglycoside antibiotic used as a screening agent in molecular genetic experiments.

AMPLITAQ ^™ (available from Perkin Elmer Setus) was used as a thermostable DNA polymerase.

General molecular biology procedures follow the method described in "Molecular Cloning-A Laboratory Manual" Second Edition, Sambrook, Fritsch and Maniatis (Cold Spring Harbor Laboratory, 1989).

Example 1

Discovery and Preparation of Hybridoma Cell Line 806.077

BALB / C mice from 8 to 10 weeks of age were immunized subcutaneously with a CEA first dose (10 μg) contained in phosphate buffered saline (0.1 mL) and Freund's complete adjuvant (0.1 mL). Two weeks later and another two weeks later, further challenge was made with an additional dose of CEA (10 μg) in phosphate buffered saline (0.1 mL) mixed with Freund's incomplete adjuvant (0.1 mL). After 32 weeks, the animals were finally administered CEA (10 μg) in phosphate buffered saline at an intravenous immunization dose and then killed 3 days later. The spleens were removed and prepared and fused with NS0 cells (obtained from the European Animal Cell Culture Collection under Accession No. 85110503) according to standard methods (Kohler and Milstein, Nature (1975) 256, 495). The resulting cells were dispersed in 96 well culture dishes and then incubated for 2 weeks. Supernatants of the resulting hybridomas were selected by EIA (enzyme immunoassay). 102 wells were positive for native CEA from a total of 1,824 wells generated in five fusions. Seventeen wells were positive in fusion 806. Cells contained in these wells were cloned by limiting dilution and the resulting clones were tested by EIA. 17 Cell lines from 10 wells of the original wells were successfully cloned. One cell line, called 806.077, was deposited in European Animal Cell Culture Collection Accession No. 96022936. The table below outlines the antibody production programs that led to the discovery of 806.077 antibody hybridomas.

antigen Acceptance Rest period (weeks) Number of fusions Number of test wells CEA ^* positive number by EIA The last selected number Untreated CEA normal 8 to 20 28 13,920 99 0 Desialized CEA normal 8 to 12 14 5,568 12 ^* 0 Bonded CEA normal 10 to 12 8 3,168 One 0 Untreated CEA normal > 30 5 1,824 102 3 ^* Desalialized immunization tested against untreated CEA produced many positive groups when tested against immunogens.

Example 2

Preparation of 806.077 Antibody from Deposited Hybridoma Cell Line ECACC 96022936

2.1 Preparation of Serum-Containing Medium

1 ml of the cryopreserved ampoule stored in liquid nitrogen was taken out and thawed rapidly in a 37 ° C. water bath. The contents were transferred aseptically to sterile 15 ml centrifuge tubes. The cells were resuspended by dropwise addition of 10 ml of Dulbecco's Modified Eagle's Medium (DMEM) containing 10% (v / v) fetal bovine serum (FCS). The suspension was centrifuged at 50 xg for 10 minutes, then the supernatant was aseptically removed and the pellet was pre-injected with 95% air 5% carbon dioxide gas in 5 ml DMEM 5 ml, 10% FCS and 1% Resuspend in L-glutamine. This flask was incubated at 36.5 ° C. in the dark.

After 3 days, the entire contents of the flask were transferred to a larger 75 mL flask and passaged by dilution with DMEM, 10% FCS and 1% L-glutamine (final viable cell concentration = 2-3 × 10 ⁵ cells / mL). In a similar manner it was further extended to a 162 ml flask.

The culture supernatant for purification was prepared by 500 ml spin culture in a 850 ml spin bottle. The culture was inoculated in a rotary bottle previously gas-injected with a water of 2 × 10 ⁵ viable cells / ml, rotated at 3 rpm, and incubated at 36.5 ° C. After full growth of the cultures, harvest was generally taken 500-800 hours after inoculation, when cell viability was below 10% and IgG concentration reached a maximum.

2.2 Treatment of culture harvest

After harvesting, the culture supernatant of the rotating bottle was clarified by centrifugation at 60 × g for 30 minutes. Sodium azide (0.02% w / v) was added to the clarified supernatant as a preservative and stored at 4 ° C. in the dark until purification.

2.3 Purification of 806.077 Antibodies

806.077 antibody hybridoma supernatant (3 L) was adjusted to pH 7.5 with dilute aqueous sodium hydroxide solution and then filtered through 0.45 μιη (Millipore MILLIDISK ^™ ). The filtered antibody supernatant was purified from phosphate buffered saline ("PBS"; 8 mM Na ₂ HPO ₄ , 1.5 mM KH ₂ PO ₄ , 150 mM NaCl, 2.5 mM KCl, pH7.3, such as Oxoids, in the form of reconstitution tablets). Obtained on the affinity column of Protein G (e.g., Protein G Fast Flow SEPHAROSE ^™ , Pharmacia Product No. 17.0618.03; inner diameter 5 cm x 6.5 cm = 130 ml) equilibrated at 4 ° C at a flow rate of 4 ml / min. Charged in. The column was washed with PBS (260 mL) under the same flow rate, the antibody was eluted with 100 mM sodium citrate (pH 2.6) to collect fractions and the eluate was irradiated with UV absorbance (280 nm). The UV absorbance fraction containing the antibody was increased and immediately adjusted to pH 7 and concentrated to about 2 mg / ml by ultrafiltration using a 30 kDa cutoff membrane (eg Amicon YM30). Dialysis in 50 mM Tris-HCl (pH 7.0) buffer using a 6-8 kDa porosity cutoff membrane (eg SPECTRAPOR ^™ 1) yielded 110 mg of 806.077 antibody with purity> 95% on SDS-PAGE.

Example 3

Selectivity of 806.077 Antibodies

To assess the selectivity, a number of human normal tissues and tumor tissues were screened for reactivity with antibody 806.077 using a sensitive three-step indirect immunohistochemistry for acetone-fixed cryostat fragments.

Immunohistochemistry was performed on sections of human tissue obtained after resection surgery or post mortem. In order to preserve the proper morphology and antigenicity, the tissues were obtained as fresh as possible, cut into small pieces (about 0.5 cm 3) and then frozen in liquid nitrogen and stored at -80 ° C. Tissue fragments (6μ) are cut on a cryostat, mounted on polylysine coated slides (e.g. Blue TECHMATE ^™ slides, Dako), fixed in ice cold acetone for 2 minutes, wrapped in foil and stored at -80 ° C. It was.

The slides were thawed at room temperature and then the foil was stripped just before use. Each fragment was cut with a diamond marker and 100 μl of 806.077 antibody diluted to 2 μg / ml with Tris buffered saline (TBS), or 100 μl of CEA / NCA reaction control (A5B7 antibody) at 2 μg / ml in TBS. , Or 100 μl of the MOPC isotype control (Sigma Chemical Company, St. Louis, List No. M9269, at 2 μg / ml) in TBS, or a related positive control such as LP34 (Dako). All subsequent incubations were carried out in a humidified room at room temperature for 30 minutes. All washing steps were performed twice with TBS. After incubation, slides were washed and 100 μl of a second antibody reagent containing 1/50 rabbit anti-mouse immunoglobulin (Dako Patts) bound to horseradish peroxidase and 1/5 of normal serum (Sigma) in TBS Was added to each fragment.

The slides were incubated again and washed with TBS. 100 μl of porcine anti-rabbit immunoglobulin (1/50 dilution with 1/5 normal serum in TBS) bound to horseradish peroxidase, the final detection antibody, was added to each fragment and incubated. After washing thoroughly. DAB substrate (3,3'-diaminobenzidine tetrahydrochloride) was prepared by dissolving 1 tablet of DAB (Sigma) in TBS (17 ml) with hydrogen peroxide (17 μl), followed by Paste Paper (eg Whatman) Dropping through 4). After incubation for 3 minutes, the slides were knocked off to remove excess DAB, and the slides were washed with TBS. After counter staining with hematoxylin (eg, Meyer hematoxylin, Shanden), fragments were dehydrated with alcohol and xylene and mounted with a non-aqueous synthetic mounting agent (eg, E-Z mounting agent, Shanden product) prior to microscopy.

Antibody binding areas were observed with brown staining on the fragments. In order to evaluate the binding degree of the 806.077 antibody to the tissue was graded as follows.

+++ (strong) => antibody binding to 75% of tumor cells

++ (medium) = antibody binding to tumor cells from 50% to 75%

Antibody binding to tumor cells of + (about) = 25% -50%

+/- (min) = unfocused antibody binding to a small area of tumor cells

-= Not dyed.

The cancerous fetal antigen (CEA) is a member of an immunoglobulin gene phase with one putative variable region-like domain region (N domain; 108 amino acids) and three constant domain similar regions A1B1, A2B2 and A3B3. Is 92 amino acids, B domain consists of 86 amino acids (Hefta, 1992, Cancer Research 52: 5647-5655). CEA also contains two signal peptides at the amino and carboxy termini, respectively. These signal peptides are all removed during post-translational processing, where the signal peptide at the carboxy terminus is replaced with glycosylphosphatidylinositol (GPI) moieties.

A number of CEA related proteins have been reported with various homology with CEA (Thompson, 1991, J. of Clinical Laboratory Analysis, 5: 344-366). These proteins include NCA1 and NCA2, which are nonspecific cross-reactive antigens. These related proteins are expressed in certain normal tissues, including granulocytes and normal lung epithelium. The anti-CEA monoclonal antibodies found so far mostly cross-react with proteins of any of the aforementioned related proteins and react with certain normal tissues, particularly with granules or lung epithelium.

The anti-CEA antibody, 806.077, was found to be CEA selective, not cross-reactive with granulocytes, and exhibited minimal staining for 4/14 of normal lung tissue tested.

806.077 antibody was first screened for tumor and NCA selectivity as tissue culture supernatant. This screening was performed using the supernatant pure water and the supernatant diluted 1:10. As a result, the binding of the antibody to colon tumors was the same, but the binding to normal lung and spleen tissues was the same when compared to A5B7. The figure was found to be very reduced. Antibodies were affinity purified (described in Example 2) and screening was repeated for another tumor and tissue type.

The antibody was also tested against a panel of colorectal tumor fragments, which showed that the optimal screening concentration of 806.077 antibody was 2 μg / ml. Therefore, all screening tests were subsequently performed using this concentration of antibody. The results of the screening test are as follows. Reactivity of the 806.077 antibody was compared to A5B7 (selective test using concentration 2 μg / ml) for the following tumor / normal tissues.

806.077 Antibody Tumor Reactivity

Colon Cancer (n = 17)

All 17 tumors tested showed moderate to strong reactivity (equivalent to A5B7 ++ / +++).

Breast cancer (n = 6)

Medium / weak staining (+ / ++) in 2/6 tumors, minimal staining (+/-) in 2/6 tumors.

NSCLC arm (n = 6)

Strong staining (+++) in 2/6 tumors, moderate staining (++) in 1/6 tumors, weak staining (+) in 2/6 tumors.

Stomach cancer (n = 2)

Strong staining (+++) in 1/2 tumor, weak staining (+) in 1/2 tumor.

Ovarian cancer (n = 3) and prostate cancer (n = 3)

No staining observed in all tumors tested.

In all cases equivalent reactivity was observed for A5B7.

Normal tissue reactivity

Lung (NCA reactivity) (n = 14)

Weak staining (+) in 4/14 lung tissue, not staining in 10/14 tissue (-).

A5B7 is moderately bound to 1/14 lung tissue (++), weakly bound to 10/14 tissue (+), and minimally bound to 1/14 tissue (+/-).

Spleen (Granulocyte / NCA Reactivity) (n = 6)

No staining was observed in all spleen tissues tested.

A5B7 binds moderately (++) to 1/6 tissue and shows weak binding (+) to 5/6 tissue.

Post normal tissue (n = 13)

Moderate / weak reactivity (++ / +) was observed only in the esophagus, skin, colon and pancreatic tissue (normal tissue expressing CEA). Similar binding was observed for A5B7. In addition to benign tissues such as colon, skin, esophagus and pancreas, negative tissues such as cerebellum, midbrain, cerebrum, smooth muscle, liver, kidney, aorta, stomach and heart were used.

Example 4

Generation of 806.077 Antibody F (ab ') Fragments

Pisin (10 mg) was suspended in 50 mM cysteine (3 mL; BDH 37218) and 50 mM Tris-HCl (pH 7.0) solution and incubated at 37 ° C. for 30 minutes. Excess cysteine was removed via size exclusion chromatography (Sephadex ^™ G-25 column, 1.5 cm × 25 cm; Pharmacia) using 50 mM Tris-HCl, pH 7.0 buffer. The decrease in the concentration of pisin was measured by ultraviolet absorbance at A280 nm (the absorbance of 1 mg / ml of solution was read as 2 in 1 cm cell) and found to be 1.65 mg / ml.

A solution of 806.077 antibody (100 mg) dissolved in 50 mM Tris-HCl buffer (pH 7.0) (50 mL) and freshly prepared reduced pisin (5 mg; added in 3 mL of this solution) were digested at 37 ° C. for 20 hours. It was. The digest was diluted with the same amount of PBS, followed by the protein G affinity column (Pharmacia, SEPHAROSE ^™ Fast Flow, inner diameter 5.0 cm × 6.5 cm = 125 ml; at 50 ° C. using 50 mM Tris-HCl pH 7.0 buffer). Pre-equilibration) and run at a constant flow rate of 3 ml / min. The column was washed with 50 mM sodium acetate (pH 4.0) (250 mL) to remove low molecular weight fragments, followed by elution of F (ab ') ₂ with 50 mM sodium citrate (pH 2.8) and the elution of ultraviolet light. Absorbance was monitored at A280 nm. The eluate containing F (ab ') ₂ was adjusted to pH 7, the dialysis buffer was exchanged for 100 mM sodium phosphate / 100 mM sodium chloride / 1 mM EDTA, pH 7.2, with a cutoff of 10 kDa. Concentrated to 8 mg / ml using membrane filtration (eg, Amicon ^™ YM10) (1 mg / ml solution reads absorbance at 280 nm in 1.4 cm cell at 1.4). 42 mg of 95% pure F (ab ') ₂ was obtained in 65% yield.

Example 5

Preparation of 806.077 Antibody F (ab ') ₂ -carboxypeptidase G2 Conjugate

As a linker for derivatization of 806.077 antibody F (ab ') ₂ , SATA (S-acetyl thioglycolic acid N-hydroxysuccinimide ester, Sigma Product No. A9043) was used. SMPB [4- (p-maleimidophenyl) butyric acid N-hydroxysuccinimide ester, Sigma Product No. M6139] was used as a linker for derivatization of carboxypeptidase G2 (CPG2).

5.1 F (ab ') ₂ Derivatization

DMSO in a solution of F (ab ') ₂ fragment (40 mg, prepared as described in Example 4) dissolved in 100 mM phosphate / 100 mM NaCl / 1 mM EDTA (pH 7.2) (buffer A; 5 mL) SATA (0.28 mg) in (28 μl) was mixed. After 40 minutes at room temperature, the resulting solution was charged to a desalting column (SEPHADEX ^™ G-25, inner diameter 1.5 cm × 50 cm = 100 ml; equilibrated with buffer A at 4 ° C.) and run at a flow rate of 1.2 ml / min to give an excess of The reagent was removed. The eluate was monitored with ultraviolet absorbance under A280 nm. SATA derivatized F (ab ') ₂ was collected and derivatized F (ab') by mixing with 500% hydroxylamine HCl / 500 mM sodium phosphate / 30 mM EDTA (pH 8.0) for 10 minutes at room temperature. ) ₂ was deacetylated. Protein concentration was determined by ultraviolet absorbance at 280 nm, with 1 mg / ml solution showing an absorbance value of 1.4 in 1 cm cells. This solution was diluted with buffer A to a concentration of about 1 mg / ml. The linker loading was determined by Elman-SH analysis and was found to be 1.8 to 2.0 linker / F (ab ') ₂ moles.

5.2 CPG2 Derivatization

Mass purification of CPG2 from Pseudomonas RS-16 was performed as described in Sherwood et al., (1985), Eur. J. Biochem., 148, 447-453. The preparation of F (ab ') ₂ and IgG antibodies bound to CPG enzymes was carried out by known methods, for example disclosed in PCT WO89 / 10140. CPG is available from Center for Applied Microbiology and Research, Wilshire SSP4 Zero Jay Salisbury Potton Down. CPG2 can also be obtained through recombinant techniques. Nucleotide coding sequences of CPG2 are described in Minton, NP et al., Gene (1984) 31, 31-38. This coding sequence is described by Chambers, SP et al., Appl. Microbiol., Biotechol. (1988), 29, 572-578 and Saccharomyces cerevisiae [Clarke, LE et al., J. Gen Microbiol., (1985) 131, 897-904. Total gene synthesis has been described in M. Edwards in Am. Biotech. Lab (1987), 5, 38-44. Expression of heterologous proteins in E. coli has been reviewed by FAO Marston in DNA Cloning Vol. III, Practical Approach Series, IRL Press (Editor DM Glover), 1987, 59-88. Protein expression in yeast has been reviewed by Methods in Enzymology Volume 194, Academic Press 1991, Edited by C. Guthrie and GR Fink.

CPG enzymes are available from Sigma Chemical Company, Dorset Wool Fancy Road, UK. For CPG enzymes see Goldman, P. and Levy, CC, PNAS USA, 58: 1299-1306 (1967) and Levy, CC and Goldman P., J. Biol. Chem., 242: 2933-2938 (1967) ]. Carboxypeptidase G3 enzymes are disclosed in Yasuda, N. et al., Biosci. Biotech. Biochem., 56: 1536-1540 (1992). Carboxypeptidase G2 enzymes are described in European Patent 121 352.

CPG2 (50 mg; recombinant enzyme from E. coli) was dialyzed in 100 mM sodium phosphate / 100 mM sodium chloride (pH 7.2) (buffer B) and diluted to 8 mg / ml (1 mg / ml solution was 1 280 nm absorbance value is read as 0.6 in cm cells).

SMPB (Sigma) was dissolved in DMSO at a concentration of 10 mg / ml. CPG2 (50 mg in Buffer B at a concentration of 8 mg / ml) was mixed with SMPB solution (0.108 ml; 1.08 mg) and reacted at room temperature for 120 minutes. Excess reagent was removed at a rate of 1.2 ml / min on a desalting column (Sephadex G-25, inner diameter 1.5 cm × 50 cm = 100 ml; equilibrated with Buffer B at 4 ° C.). Derivatized CPG2 was collected and its concentration measured at UV A280 nm (1 mg / ml solution read 280 nm absorbance at 0.6 in 1 cm cell). This solution was diluted to a CPG2 concentration of about 1 mg / ml. Linker loading was determined by adding a known amount of 2-mercaptoethanol to maleimido derivatized CPG2 via 'reverse' Elman assay and then analyzing unreacted -SH. As a result, the linker loading was found to be 2.0 to 2.4 linker / CPG2 moles.

5.3 splicing

Deacetylated derivatized F (ab ') ₂ and derivatized CPG2 were mixed in equal amounts under nitrogen and the mixture (about 80 mL, total protein concentration about 1 mg / mL) was left at room temperature for 20 hours. The reaction was terminated by the addition of 10% v / v of 100 mM aqueous glycine to the reaction. The resulting crude conjugate mixture was buffer exchanged by dialysis with low salt buffer (50 mM sodium acetate, pH 6.0) and then dye-ligand affinity column equilibrated at 4 ° C. with the same buffer (column where dye binds to CPG2, For example, loaded into ACL Mimetic Green 1, 2.5 cm × 10 cm = 50 ml), the unreacted derivatized F (ab ′) ₂ was removed. The conjugate and derivatized CPG2 were eluted with 50 mM acetate / 500 mM NaCl, pH 6.0, at a flow rate of 2.0 ml / min, and the eluate was monitored by UV (A280 nm).

The crude conjugate still containing derivatized CPG2 was concentrated to about 12 ml using an ultrafiltration device (eg, Amicon YM10 ^™ ) with a cutoff of 10 kDa to bring the total protein concentration to 5 mg / ml 10% v / v 10 mM zinc sulfate (Sigma Z 0251) in water was added to compensate for the loss of CPG2 in the process described above. Then, 50 mM sodium acetate / 150 mM sodium chloride (pH 6.0) was introduced at a flow rate of 1 ml / min to size exclusion chromatography at 4 ° C. (e.g. SEPHACRYL S-300 HR ^TM Pharmacia, inner diameter 2.5 cm × 25 cm). = 500 mL) to collect the fractions, monitored at UV A280 nm to obtain the conjugate fractions, and then the column fractions were SDS-PAGE to separate unreacted derivatized CPG2 and conjugates.

The peaks containing the conjugate [F (ab ') ₂ : CPG2 1: 2, 1: 1 and 2: 1] were then collected and concentrated by ultrafiltration to 1.3 mg / ml, where the protein concentration was at A280 nm. UV absorbance was measured (1 mg / ml was 1.0). The purity of the conjugate was determined by SDS PAGE and found to contain 12 mg of the total conjugate, the composition of which was 65% in the 1: 1 ratio of the conjugate, 20% in the 1: 2 or 2: 1 ratio of the conjugate, and free. The derivatized F (ab ') ₂ of <5% and the free form of derivatized CPG2 were <5%.

Example 6

Tumor Suppressive Activity of the 806.077 Antibody F (ab ') ₂ -CPG2 Conjugates Mixed with a Prodrug

Tumor inhibitory activity of the 806.077 antibody F (ab ′) ₂ -CPG2 conjugate prepared as described in Example 5 was determined by prodrug N (4- [N, N-bis (2-chloroethyl) amino] -phenoxycarbo Nil) -L-glutamic acid (called "PGP" in this example, disclosed in US Pat. No. 5,405,990 and Blakey et al., Br. J. Cancer 72, 1083-88, 1995) The colorectal tumor xenograft model was evaluated.

A group of 8 to 10 athymic female nude mice were injected subcutaneously with 1 × 10 ⁷ LoVo colorectal tumor cells (ECACC 87060101). 806.077 antibody F (ab ') ₂ -CPG2 conjugate (250 U CPG2 enzyme activity Kg ^-1 ) or phosphate buffered saline (170 mM NaCl, 3.4 mM KCl, 12 mM Na ₂ HPO when tumors were 4-5 mm in diameter) ₄ , 1.8 mM KH ₂ PO ₄ , pH 7.2) was injected intravenously. 72 hours later PGP prodrug was injected intraperitoneally (40 mg / kg three times at 1 hour intervals). Then, the length of the tumor was measured in two directions three times a week, the volume of the tumor was calculated according to the following equation.

Volume = ∏ / 6 × D ² × d

Where d is the large diameter of the tumor and d is the small diameter. Tumor volume is expressed in size relative to the tumor volume present at the start of the prodrug portion during treatment. Tumor inhibitory activity was compared to the control group that gave PBS instead of the conjugate or prodrug. Tumor inhibitory activity is the growth retardation rate, defined as the time taken for the volume of treated tumor to increase fourfold—the time for the volume of control tumor to increase fourfold, and the volume of treated tumor 14 days after prodrug administration. T / C values, defined as the volume of control tumors. Statistical effectiveness of the tumor suppressor effect was determined using a variance (one-way) test assay.

Tumor inhibitory activity of the 806.077 antibody F (ab ′) ₂ -CPG2 conjugate mixed with PGP prodrug is shown in FIG. 1. Summary of tumor inhibition data is as follows.

Tumor Inhibitory Activity of 806.077 Antibody F (ab ′) ₂ -CPG2 Conjugates Mixed with PGP Prodrugs for LoVo Tumor Xenografts

concrete Dose (U / kg) T / C (%) Growth delay (days) Validity (p) 806.077F (ab ') ₂ -CPG2 250 16.5 14 <0.01 500 4.7 22 <0.01

The results showed that the 806.077 antibody F (ab ') ₂ -CPG2 conjugate mixed with PGP prodrug showed statistically significant tumor decay and prolonged growth retardation compared to the control group.

Example 7

Cloning and Sequencing of Variable Regions of 806.077 Antibody Heavy and Light Chain Genes

7.1 Preparation of Cellular RNA

There are several ways to isolate polyA + mRNA from eukaryotic cells (Sambrook J., Fritsch EF, Maniatis T., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Second Edition, 1989, Chapter 8, p3, Referred to herein as "maniatis"). Among them, cytoplasmic RNA was prepared using frozen hybridoma cell pellets containing 1 × 10 ⁹ cells stored at −80 ° C. as described in Favoloro et al., Methods in Enzymology 65, 718-749. It was.

The cells were treated with ice cold lysis buffer containing 400 u of ribonuclease inhibitor (RNAguard; Pharmacia Product No. 27-0815-01) [140 mM NaCl, 1.5 mM MgCl ₂ , 10 mM Tris-HCl pH 8.6 and 0.5% NP40 ( Polyglycol ether nonionic surfactant; nonylphenoxy polyethoxy ethanol, Sigma Product No. 127087-87-0)] and resuspended for 10 seconds. The solution was layered on the same amount of ice-cold lysis buffer containing 24% sucrose (w / v) and 1% NP-40 and left on ice for 5 minutes. This preparation was then centrifuged at 4000 rpm for 30 minutes at 4 ° C. in a bench top centrifuge (Sorval RT6000B), and then the cytoplasmic phase was removed in the same amount of 2 × PK buffer [200 mM Tris (pH 7.5), 25 mM EDTA. , 300 mM NaCl and 2% (w / v) SDS]. Protease K (Sigma Product No. P2308) was added to a final concentration of 200 μg / ml and the mixture was incubated at 37 ° C. for 30 minutes.

This preparation was extracted with the same amount of phenol / chloroform, the aqueous phase was separated, and 2.5 volumes of ethanol were added and mixed. This solution was stored at -20 ° C overnight. After centrifugation (4000 rpm, 30 minutes at 4 ° C. with a bench top centrifuge Sorval RT6000B), the supernatant was discarded and the pellet dried in a vacuum desiccator followed by diethylpyrocarbonate (DEPC) treated water (mania) Prepared as described in TIS, cited above) and dissolved in 250 [mu] l and collected. RNA content was measured by spectrophotometer and concentration was calculated by considering absorbance 1 at 260 nm as 40 μg / ml.

7.2 Preparation of the first strand variable region cDNA

Various methods of synthesizing cDNA are described in Maniatis (chapter 8). Oligonucleotide primers used are described in Marks et al. J. Mol. Biol (1991) 222, 581-597. cDNA was prepared as follows. RNA (5 mg) was added to 10 μl of 5 × reverse transcriptase buffer [250 mM Tris (pH 8.3), 40 mM MgCl ₂ and 50 mM DTT] in a microcentrifuge tube, 1 μl of forward primer (25 pM), 1.25 mM dNTP 10 Mix with 5 μl, 10 mM DTT, 0.5 μl RNAguard (Pharmacia) and add DEPC treated water to make 50 μl final volume. The reaction mixture was heated to 70 ° C. for 10 minutes, then slowly cooled to 37 ° C. and then 100 u (0.5 μl) of M-MLV reverse transcriptase (Pharmacia Product No. 27-0925-01) were added and the reaction was stirred at 37 ° C. Incubated for 1 hour at. The forward primer used to make light chain cDNA is oligonucleotide CK2FOR (SEQ ID NO: 1) designed to hybridize to the CK constant region of the murine kappa light chain gene. To prepare heavy chain cDNA, forward primer CG1FOR (SEQ ID NO: 2) was used, which hybridizes to the CH1 constant domain of murine IgG1.

7.3 Amino Acid Sequencing

The heavy and light chains of the 806.077 antibody were separated by SDS-PAGE and Western blotting, and the N terminus of the isolate was analyzed for amino acid sequence. As a result, the N terminus of the light chain was chemically blocked and the heavy chain obtained the sequence of the first 34 N term residues (SEQ ID NO: 3). Based on this amino acid sequence, specific DNA reverse primers were designed for use in 806.077 heavy chain variable region PCR. This primer is called SP1back (SEQ ID NO: 7).

7.4 Isolation of Antibody Gene Fragments by PCR

Genes of 806.077 heavy and light chain variable regions were isolated using cDNA prepared as described above as a template. General reaction conditions are as follows.

5 µl of cDNA reaction solution, 5 µl of dNTP (2.5 mM), 5 µl of 10x enzyme buffer (500 mM KCl, 100 mM Tris (pH 8.3), 15 mM MgCl ₂ and 0.1% gelatin), 1 µl of 25 pM / µl reverse primer. , 1 μl 25 pM / μl forward primer, 0.5 μl thermostable DNA polymerase and DEPC treated water to make a final volume of 50 μl were added. PCR conditions were 90 seconds at 94 ° C .; 60 seconds at 55 ° C .; 25 cycles of 120 seconds at 72 ° C. were performed and finally incubated at 72 ° C. for 10 minutes.

Under general reaction conditions, oligonucleotide CK2FOR (SEQ ID NO: 1) was used as the forward primer for light chain cDNA preparation, and CG1FOR (SEQ ID NO: 2) was used for the preparation of heavy chain cDNA oligonucleotide. In order to obtain a desired PCR product, various reactions using various reverse primers were performed on each of the heavy and light chains.

In the case of the 806.077 light chain, a specific PCR product was obtained by using reverse primers VK1back (SEQ ID NO: 4) and VK4back (SEQ ID NO: 5) in the analysis. Similarly, for heavy chains, specific PCR products were obtained using VH1back (SEQ ID NO: 6) and SP1back primers (SEQ ID NO: 7). The reaction product was analyzed on 2% agarose gel. DNA was purified by cleaving the product of the desired size.

7.5 Cloning PCR Products into Bluescript KS + Vectors

One restriction site was introduced into each 5 'region (reverse primer) oligonucleotide and 3' region (forward primer) oligonucleotide for each antibody fragment. Thus, separate PCR products for the VH and VK PCR reactions were cloned into the appropriate enzyme restriction sites of the Bluescript vector KS + (stratazine cloning system) using standard DNA manipulation methods (eg, PCR product VH1back / CG1For was converted to PstI / HindIII). Cloned VK4back / CK2For into SacI / HindIII). DNA was prepared from the resulting clones, and each construct was sequenced accurately using an automated fluorescence sequencing device (Applied Biosystems) using at least 12 clones. Analyzed sequences were reviewed, compared and sorted using appropriate computer software. Consensus sequences for the VH and VK genes were obtained and translated into the corresponding amino acid sequences.

The DNA sequence and amino acid sequence obtained for the 806.077 light chain variable (VK) region were described in SEQ ID NO: 8 and SEQ ID NO: 9, respectively. The DNA sequences and amino acid sequences obtained for the 806.077 heavy chain variable (VH) region are set forth in SEQ ID NO: 10 and SEQ ID NO: 11, respectively. Clones containing light chains were designated VK4 and clones containing heavy chain sequences were designated VH14A.

Example 8

Construction of Chimeric Light and Heavy Chain Fd Genes

The heavy and light chain genes cloned into Bluescript (VK4 and VH14A described in Example 7) were isolated by PCR using primers to introduce new unique enzyme restriction sites while specifically amplifying only the appropriate genes of the variable region. This unique restriction site allowed the gene fragments of the variable region to be cloned, forming skeletal structures in DNA fragments encoding both appropriate antibody signal sequences and human constant regions. The sequences of the signal and constant regions of the light chain and heavy chain Fd were previously cloned into pNG3 and pNG4, which are derivatives of the pSG5 eukaryotic plasmid expression vector, respectively.

Vector pNG3 was prepared as follows. Plasmid pSG5 (stratazine Product No. 216201) was digested with SalI and XbaI to remove the existing SV40 promoter and polylinker sequences. Plasmid pNG1 was obtained by introducing new polylinkers using oligonucleotide sequences 34 and 35 that hybridized with each other and cloned into SalI and XbaI cleaved pSG5 plasmids. This pNG1 plasmid was digested with BglII and HindIII and cloned into this site was a BglII-HindIII CMV promoter fragment derived from pcDNA3 (Invitrogen Product No. V790-20) to obtain plasmid pNG2. Finally, the polyA region derived from pSG5 was isolated by PCR as described in section 7.4 of Example 7, except that oligonucleotide sequences 36 and 37 were used with plasmid pSG5. This PCR product was cleaved with XmaI and BamHI, purified by electrophoresis on 2% agarose gel (using GENECLEAN, see Example 7) and ligated into XmaI-BamHI cleaved pNG2 plasmid to obtain pNG3.

pNG4 vectors were prepared as follows. The pNG3 vector was further modified such that the SacI restriction enzyme recognition site present in the cloned CMV promoter fragment collapsed with changes in its DNA sequence. Such modifications were carried out using a two-step PCR mutagenesis reaction using a pNG3 vector as a template. The PCR used two complementary oligonucleotide primers (SEQ ID NOs: 38 and 39) to mutate the SacI recognition sequence and two contiguous primers (SEQ ID NOs: 40 and 41) for product amplification. These two primer pairs (SEQ ID NOS: 38 and 41) and (SEQ ID NOs. 39 and 40) were used in a standard PCR reaction (described in Section 7.4 of Example 7) to obtain two PCR products for the first time, which were 2% agarose. It was separated by electrophoresis on gel. Each product was mixed in equimolar amounts and reamplified using adjacent primers (SEQ ID NOs: 40 and 41) under standard PCR reaction conditions to amplify the final PCR product obtained by grafting together. This product was serially digested with restriction enzymes NcoI and HindIII and cloned into pNG3 vectors prepared by appropriate restriction digestion to replace the mutated (SacI ⁻ ) fragments with the original pNG3 NcoI-HindIII (SacI ⁺ ) fragments. The new vector thus obtained was called pNG4.

Clones (VK4) containing 806.077 murine light chains in Bluescript KS + vectors were selected and amplified using oligonucleotide primers 077VK-UP (SEQ ID NO: 12) and 077VK-DOWN (SEQ ID NO: 13). In a similar manner, 806.077 heavy chain clones (VH14A) were amplified using 077VH-UP (SEQ ID NO: 14) and 077VH-DOWN (SEQ ID NO: 15). PCR was carried out as follows. 5 μl of dNTP (2.5 mM) in 100 ng of plasmid DNA, 5 μl of 10 × enzyme buffer (see above), 1 μl of 25 pM / μl reverse primer, 1 μl of 25 pM / μl forward primer, 0.5 μl of thermostable DNA polymerase and 50 Appropriate amount of DEPC treated water was added to make a final volume of μL. PCR conditions consisted of 15 cycles of 90 seconds at 94 ° C, 60 seconds at 55 ° C, 120 seconds at 72 ° C, and finally incubation at 72 ° C for 10 minutes. The product was analyzed on 2% agarose gel. This DNA was purified and DNA fragments digested with relevant restriction enzymes for subsequent vector cloning.

To secrete antibody light chains, a double-stranded DNA cassette was designed that contains both information about Kozak recognition sequences and light chain signal sequences. These cassettes were hybridized with each other into two oligonucleotides (SEQ ID NOS: 42 and 43) that were subsequently cloned between the HindIII and SacII restriction sites of the pNG3 plasmid (suitably restriction digested and isolated using standard methods). It is composed. The DNA sequence of SEQ ID NO: 46 containing the sequence of the human light chain kappa constant region was then digested with XmaI and XhoI, inserted between the pNG-Vkss plasmid digested with XhoI and XmaI to insert vector pNG3-Vkss-HuCk (NCIMB 40798). Got it. In addition, pNG3-Vkss-HuCk (derived from the pSG5-Neo vector, a pSG5 plasmid variant supplied by S. Green from Genca Pharmaceuticals; other sources include vectors such as pMC1neo (stratazine product number 213201)). The neomycin resistance gene expression cassette was cloned. The neomycin resistance gene expression cassette was cloned into the XbaI site of pNG3-Vkss-HuCk as an XbaI fragment and then checked for orientation via restriction enzyme digestion. Thus, the plasmid pNG3-Vkss-HuCk-Neo (NCIMB 40799) was obtained. The light chain gene sequence described above was directly cloned between the SacII and XhoI sites of the pNG3-Vkss-HuCk-Neo vector and inserted into the frame in frame conformation. PCR fragments obtained from the light chain genes were digested with SacII and XhoI restriction enzymes and similarly cloned into expression vectors containing restriction digested VK signals and HuCK constant region coding sequences. The resulting chimeric 806.077 light chain sequences are set forth in SEQ ID NOs: 16 and 17.

Similarly, for the secretion of antibody heavy chains, a double-stranded DNA cassette containing both Kozak recognition sequences and information on heavy chain signal sequences was designed. This cassette was hybridized with each other into two oligonucleotides (SEQ ID NOS: 44 and 45) that were subsequently cloned between the HindIII and EcoRI restriction sites of the pNG4 plasmid (suitably restriction digested and isolated using standard methods). It is composed. Thus, the heavy chain gene sequence can be directly cloned between the EcoRI and SacI sites of the pNG4-VHss vector and inserted into the backbone conformation format. The DNA sequence of SEQ ID NO: 47 containing the coding sequence of the human heavy chain IgG2CH1 'constant region (SEQ ID NOS: 22 and 23) was then digested with SacI and XmaI, cloned between pNG4-VHss digested with SacI and XmaI to vector pNG4-VHss -HuIgG2CH1 '(NCIMB 40797) was obtained. PCR fragments obtained as heavy chain genes were digested with EcoRI and SacI restriction enzymes and similarly cloned into expression vector pNG4-VHss-HuIgG2CH1 'restriction digested and containing the VH signal and the HuIgG2 CH1' constant region coding sequence. The chimeric 806.077 HuIgG2 Fd chain sequences thus obtained are set forth in SEQ ID NOs: 18 and 19.

In some cases it is desirable to use other types of chimeric heavy chain Fd constructs. For this purpose, heavy chain vector variants containing HuIgG1CH1 '(SEQ ID NOS: 20 and 21) or HuIgG3CH1' (SEQ ID NOS: 24 and 25) are prepared in place of the HuIgG2CH1 '(SEQ ID NOS: 22 and 23) gene.

The sequences set forth in SEQ ID NOs: 46 and 47 are prepared by several methods. Examples are described in Edwards (1987) Am. Biotech. Lab. 5, 38-44, Jayaraman et al. (1991) Proc. Natl. Acad. Sci. USA 88, 4084-4088; Foguet and Lubbert (1992) Biotechniques 13, 674-675 and Pierce (1994) Biotechniques 16, 708. The sequences set forth in SEQ ID NOs: 46 and 47 are described in Jayaraman et al. (1991) Proc. Natl. Acad. Sci. Preferred is a PCR method similar to that described in USA 88, 4084-4088.

Individual heavy and light chain sequences are constructed once with a heavy chain Fd gene expression cassette (including a coexpression vector construct where the promoter and gene are cleaved into BglII / SalI fragments and then cloned into the BamHI / SalI site of the light chain vector). This construct is transfected with NS0 myeloma cells (ECACC 85110503) by standard electropenetration and the transfectants are selected using G418 resistance, which is a property carried on the expression plasmid construct as a marker for selection.

Alternatively, the complete heavy chain Fd and light chain genes, as HindIII / XmaI fragments, from each vector can be simply cleaved and cloned into another expression vector system of choice.

Example 9

Hybridization test using nucleic acid variants of specific nucleic acid sequences

9.1 Hybridization Test

Methods for detecting nucleic acid variants containing sequences associated with specific 806.077 antibody sequences are as follows. This nucleic acid variant was isolated by DNA / RNA or gel electrophoresis from eukaryotic cells immobilized on the membrane as in bacterial colony-derived DNA or cDNA library selection described above, followed by Northern hybridization (Maniatis et al., Chapter 7, p39). Or purified nucleic acid fragments that have been transferred to suitable membranes in connection with Southern hybridization (Maniatis, Chapter 9, p31) techniques.

9.2 Hybridization Probes

Hybridization probes can be obtained from all fragments of DNA or RNA encoding a particular 806.077 antibody nucleic acid sequence of interest, more particularly from variable regions, particularly regions encoding CDR3. Synthetic oligonucleotides or their complementary sequences can be used as specific probes for this CDR3 coding region.

Hybridization probes can be prepared by adding a radioactive 5 ′ phosphate group derived from [γ- ³² P] ATP to a synthetic oligonucleotide through the action of a T4 polynucleotide kinase. 20 pmol of this oligonucleotide was added to 100 mM Tris, pH7.5, 10 mM MgCl ₂ , 0.1 mM spermidine, 20 mM dithiothreitol (DTT), 7.55 μM ATP, 55 μCi [γ- ³² P] ATP and 2.5 uL 4 polynucleotide kinase (Pharmacia Biotechnology Limited, Uppsala, Sweden) is added to 20 μl of reaction solution. The reaction solution was incubated at 37 ° C. for 30 minutes and then incubated at 70 ° C. for 10 minutes before use in hybridization. Methods of preparing hybridization probes from oligonucleotides (Chapter 11) or methods of preparing hybridization probes from DNA and RNA fragments (Chapter 10) are described in Maniatis. In addition, many proprietary kits are available on the market.

9.3 Hybridization Conditions

The margin containing nucleic acid was prehybridized in a suitable closed container at 65 ° C. for at least 1 hour in 100 ml of solution containing 6 × SSC, 0.1% SDS and 0.25% skim milk powder (Marvel ^™ ). Proprietary hybridization devices such as Model HB-1 (Techne Ltd) provide the reproducible conditions of this experiment.

The prehybridization solution was then replaced with 10 ml of probe solution containing 6 × SSC, 0.1% SDS, 0.25% skim milk powder (Marvel ^™ ) and oligonucleotide probes as described above. The solution was incubated for 5 minutes at 65 ° C. in the solution, and then the temperature was gradually lowered to 30 ° C. or lower. The probe solution was then discarded and washed with 100 ml of a solution containing 6 × SSC, 0.1% SDS for 5 minutes at room temperature. The new batch containing the same solution was washed again at 30 ° C. and increased to 10 ° C. for 5 minutes to 60 ° C. each wash.

After washing, the room was dried and exposed to an X-ray film such as Hyperfilm ^™ MP (Amersham International) at −70 ° C. in a dense film cassette using a paste tungstate condensing screen to enhance the image of the photograph. The film was exposed for a suitable period (usually overnight) and then developed to observe a photographic image of the radioactive area present on the margin. The relevant nucleic acid sequence can be identified because it represents a photographic image, unlike a completely unrelated sequence that does not represent an image. Generally, the relevant sequences are positive at the highest wash temperature (60 ° C.). However, related sequences may also be positive only at low wash temperatures (50 ° C., 40 ° C. or 30 ° C.).

This result also depends on the nature of the probe used. Long nucleic acid fragment probes need to be hybridized for a long time at high temperature but can remain bound to the relevant sequences at high wash temperatures and / or low salt concentrations. Shorter mixed or degenerate oligonucleotide probes may require low stringency wash conditions such as low temperature and / or high Na ⁺ concentrations. Considerations for hybridization protocols are given in Maniatis (Chapter 11).

Example 10

drugs

Representative pharmaceutical dosage forms containing 806.077 antibodies that can be used for treatment in combination with a suitable prodrug are described below.

ADEPT Injection Solution

Sterile aqueous solutions for injection contain the following ingredients per ml of solution.

806.077 Antibody-CPG2 Conjugate 1.0 mg Sodium Acetate Trihydrate 6.8 mg Sodium chloride 7.2 mg Tween 20 0.05 mg

The typical dose of conjugate to adults is 30 mg, followed by three doses of 1 g prodrug every three hours after 3 days. Suitable CPG2 conjugates are any one of the conjugates described in Examples 105 and 106. A combination with HCPB may be used instead of the CPG2 conjugate shown in the table. Suitable HCPB conjugates are any one of the conjugates described in Examples 48-101.

Injection solution for tumor immunotherapy

Sterile aqueous solutions for injection contain the following components per ml of solution.

806.077 Antibody-B7 Conjugate 1.0 mg Sodium Acetate Trihydrate 6.8 mg Sodium chloride 7.2 mg Tween 20 0.05 mg

The typical dose of conjugate to adults is 30 mg. Suitable conjugates are those described in Example 104.

Example 11

Construction of Early 806.077 Humanized Antibody Heavy and Light Chain Variable Region Genes

First, the outline of the humanization method is as follows. The purpose of antibody humanization is to bind the binding site of a non-human antibody to the support skeleton of a human antibody while maintaining the specific antigen binding affinity and specificity of the parent antibody. The possibility of such antibody manipulation is due to its close sequence and structural homology with immunoglobulins from other mammalian species.

This method, in its most basic form, described six hypervariable regions or complementarity determining regions (CDRs) derived from the Fv region of an antibody, as first described in Johns et al., Nature (1986) 321, 522-525. Transition to other areas. Experience has shown, however, that in addition to the CDRs, the amino acids in the antibody backbone must be transferred before the process can be successfully performed because these residues sometimes contact and affect the shape of the CDR loop.

In the case of the 806.077 antibody, the "early" humanized antibody form was made to contain the six CDRs of the mouse and substituents of many backbone residues. Another variant (Examples 12-47) was made by using this construct again as a template and replacing it with another “mouse” residue. Other details of this embodiment are as described in the earlier humanized constructs.

Human antibody heavy chain variable region NEWM (Poljak, R.J. et al (1974) PNAS 71 3440-3444) and light chain kappa variable region REI [Palm, W and Hilschmann, N.Z. (1975) Physiol. Chem. 356 167-191] to create a human antibody framework of the receptor. Many examples of successful humanization using this Fv skeleton have been reported in the literature, and the domains of the two proteins have been elucidated in three dimensions. Based on the comparison of the 806.077 heavy and light chain variable region protein sequences of the mouse with the consensus sequence (and each sequence composition) of the most closely related mouse subgroup Kabat and the human NEWM and REI protein sequences, the mouse CDRs and other additional Initial DNA sequences were designed that encode early humanized antibodies incorporating important skeletal substituents.

The 806.077 CDR sequences of the injected mice were set forth in SEQ ID NOs: 26, 27 and 28 for the light chain variable regions and were found at positions 24 to 34 (CDR1), 50 to 56 (CDR2) and 89 to 97 (CDR3), respectively. CDRs introduced into the heavy chain variable region are set forth in SEQ ID NOs: 29, 31 and 32 located at positions 31-35 (CDR1), 50-65 (CDR2), 95-102 (CDR3) (using Kabat nomenclature), respectively. In the heavy chain variable region, additional change V24A to the NEWM backbone; S27F; T28N; F29I; S30K; V71A; A92H; R93V (using Kabat nomenclature) was done but there were no additional backbone changes in the light chain variable region.

Individual synthetic DNA sequences were designed that encode the initial forms of the 806.077 humanized antibody heavy chain (806.077HuVH1) and light chain (806.077HuVK1) variable regions incorporating changes in the CDRs and any other backbone residues. The antibody variable gene sequences set forth in SEQ ID NOs: 48 and 53 can be prepared by a variety of methods, see for example Edwards (1987) Am. Biotech. Lab. 5, 38-44, Jayaraman et al. (1991) Proc. Natl. Acad. Sci. USA 88, 4084-4088, Foguet and Lubbert (1992) Biotechniques 13, 674-675 and Pierce (1994) Biotechniques 16, 708. The DNA sequences set forth in SEQ ID NOs: 48 and 53 are preferably prepared by PCR methods similar to those described in Jayaraman et al. (1991) Proc. Natl. Acad. Sci. USA 88, 4084-4088.

The humanized 806.077 antibody variable light chain gene sequences (SEQ ID NOs: 49 and 50) were cloned into the pNG3-Vkss-HuCK-Neo (NCIMB 40799) expression vector by skeletal registration. For this purpose, a synthetic PCR DNA fragment encoding the humanized variable light chain gene (SEQ ID NO: 48) is digested with SacII and XhoI restriction enzymes, and similarly pNG3 containing the digested VK signal sequence and HuCK constant region coding sequence. Cloned into-Vkss-HuCK-Neo vector. The DNA and protein sequences containing its signal sequence together with the fully humanized 806.077 light chain sequence (806.077HuVK1-HuCK) thus obtained are set forth in SEQ ID NOs: 51 and 52, respectively, and the vector is described as pNG3-Vkss-806.077HuVK1-. Called HuCK-Neo.

Similarly, in the case of humanized antibody heavy chains, the humanized variable heavy chain gene sequences (SEQ ID NOs: 54 and 55) were directly cloned into pNG4-VHss-HuIgG2CH1 '(NCIMB 40797) expression vector and inserted in skeletal matching. For this purpose, synthetic PCR fragments obtained for the humanized heavy chain gene (SEQ ID NO: 53) were digested with EcoRI and SacI restriction enzymes, similarly restriction digested and containing the VH signal sequence and the HuIgG2 CH1 'constant region coding sequence. It was cloned into pNG4-VHss-HuIgG2CH1 'vector. The DNA and protein sequences containing their signal sequences together with the fully humanized 806.077 Fd heavy chain sequence (806.077 HuVH1-HuIgG2 Fd) thus obtained are described in SEQ ID NOs: 56 and 57, respectively, and the vector is pNG4-VHss-806.077 HuVH1-HuIgG2 was called CH'1.

Initial humanized antibody constructs were prepared by constructing coexpression plasmids containing 806.077 HuVK1 light chain and 806.077 HuVH1 heavy chain variable region antibody genes. The plasmid pNG3-Vkss-807.077HuVK1-HuCK-Neo vector containing the humanized light chain variable region HuVK1 (SEQ ID NOs: 49 and 50) was digested with restriction enzymes BamHI and SalI and the resulting vector was run on a 1% agarose gel. After vector band was cut and purified. Plasmid pNG4-VHss-806.077HuVH1-HuIgG2 CH1 '(containing the humanized 806.077 HuVH1 heavy chain variable regions (SEQ ID NOs: 56 and 57)) was digested with restriction enzymes BglII and SalI and the reaction proceeded onto 2% agarose. The fragment bands were then cut out and purified. The recovered DNA fragment was then ligated in the pNG3-Vkss-806.077HuVK1-HuCK-Neo vector prepared above to obtain the desired HuVH1 / HuVK1 coexpression vector clone.

This construct was used to transfect NS0 myeloma cells (ECACC 85110503) via standard electropenetration techniques and the transfectants were selected for G418 antibiotic resistance. The clones obtained were tested for antibody expression via the antihuman antibody Fd ELISA and CEA binding ELISA assays described below.

In the CEA ELISA, each well of 96 well immunoplate (NUNC MAXISORB ^™ ) was coated with 50 ng of CEA in 50 mM carbonate / bicarbonate coating buffer, pH 9.6 (buffer capsule-Sigma C3041) and incubated overnight at 4 ° C. Plates were washed three times with PBS + 0.05% Tween 20 and then blocked by reacting for 1 hour at room temperature with 150 μl of 1% BSA in PBS + 0.05% Tween 20 per well. This plate was washed as described above, 100 μl of sample per well was added and incubated for 2 hours at room temperature. Again wash the plate three times with PBS + 0.05% Tween 20, add 1/500 dilution of HRPO labeled goat antihuman kappa antibody (Sigma A7164) to 1% BSA in PBS-Tween20 at 100 μl per well and shake Incubated at room temperature for at least 1 hour on a substrate. This plate was washed as described above and then again once more with PBS. To detect the degree of binding, a color solution (dissolved in 1 capsule of phosphate-citrate buffer (Sigma P4922), 100 ml of water, to which one 30 mg tablet of ο-phenylenediamine dihydrochloride (Sigma P8412) was added) was added. Add 100 μl per well and incubate for up to 15 minutes. 75 µl of 2M H ₂ SO ₄ is added to the reaction solution to stop the reaction, and the absorbance is read at 490 nm.

In the antihuman antibody Fd ELISA, each well of a 96 well immunoplate was coated with 1.2 μg of antihuman Fd antibody (binding site PC075) in an amount added in 50 mM carbonate / bicarbonate coating buffer, pH 9.6 (buffer capsule-Sigma C3041). And incubated at 4 ° C. overnight. Plates were washed three times with PBS + 0.05% Tween 20 and then blocked for 1 hour at room temperature using 150 μl of 1% BSA in PBS + 0.05% Tween 20 per well. This plate was washed as described above, 100 μl of sample per well was added and incubated for 2 hours at room temperature. Again wash the plate three times with PBS + 0.05% Tween 20 and add HRPO labeled goat antihuman kappa antibody (Sigma A7164) to 1% BSA in PBS-Tween20 at 100 μl per well and for 1 hour at room temperature on shake substrate. It was incubated for a longer time. This plate was washed as described above and then again once more with PBS. To detect the degree of binding, a coloring solution or the like was added as described in the CEA binding assay.

Clones found to have the highest expression and CEA binding rates were selected and further propagated in 24 well plates and retested. According to the criteria of this assay, the best clones were selected and propagated for use in 1 liter of production, and then the whole propagated cultures were seeded using 1:10 dilution (ie, 100 ml in 900 ml of fresh culture medium). Added), and further propagation for 14 days. Human F (ab ') ₂ antibody fragments were then purified from the culture supernatants as described in Example 102.

Examples 12-38

Other combinations of humanized heavy and light chain variable region genetic variants: 806.077 Construction of humanized heavy and light chain variable region variants.

Initial humanized 806.077 variable region genes were used to construct other gene constructs containing skeletal residues from other mice. Modification of the gene sequence was carried out by cassette mutagenesis (in most cases). In this technique, a portion of the original gene is removed from the complete plasmid vector by restriction digestion using two suitable native enzymes, and then together to form a double-stranded DNA cassette (a DNA fragment with appropriate aggregate ends) in the thus prepared plasmid. Direct ligation and replacement of the two complementary oligonucleotides hybridized to contain the desired DNA changes while reconstituting the gene. Another combination of mutations in the heavy or light chain can be obtained by simple exchange of DNA fragments between suitable variants using useful intrinsic restriction enzyme sites.

In addition to the original sequences HuVK1 (SEQ ID NOs: 49 and 50), three different variants for the humanized light chain variable regions were prepared, which were called HuVK2, HuVK3 and HuVK4, respectively. The light chain variable region variant HuVK2 is a change in the original HuVK1 coding sequence to obtain the amino acid change M4L (Kabat nomenclature), and the mutation of this gene (SEQ ID NO: 49) is carried out by cassette mutagenesis. Plasmid pNG3-Vkss-806.077HuVK1-HuCK-Neo (containing fully humanized light chains (SEQ ID NOs: 49 and 50)) was digested with restriction enzymes SacII and NheI. This digest was then loaded onto a 2% agarose gel and the cut fragments were separated from the rest of the vector. Vector DNA was cut out of the gel and recovered and stored at −20 ° C. until needed. Two oligonucleotides (containing the desired base changes) were designed and synthesized (SEQ ID NOs: 58 and 59). These two oligonucleotides were hybridized by adding 200 pmol of each oligonucleotide to a total of 30 μl of water, heating to 95 ° C. and then slowly cooling the solution to 30 ° C. 100 pmol of the annealed DNA product was then directly ligated in the aforementioned vector. This DNA "cassette" exchange allowed for direct placement of the desired HuVK2 DNA and protein sequences (SEQ ID NOs: 60 and 61) in the expression vector pNG3-Vkss-806.077HuVK2-HuCK-Neo.

In a similar manner, amino acid change D1Q using synthetic oligonucleotides (SEQ ID NOs: 62 and 63); Q3V; HuVK3 with M4L (Kabat nomenclature) was constructed to obtain the DNA and protein sequences (SEQ ID NOS 64 and 65) of the desired HuVK3 directly positioned in the expression vector pNG3-Vkss-806.077HuVK3-HuCK-Neo.

The light chain variable region variant HuVK4 did not have a unique restriction enzyme site available close to the mutation site and was prepared using other techniques. HuVK4 with amino acid change L47W was prepared using PCR mutation technique. Vector pNG3-806.077HuVK1-HuVK-Neo was used as a template for two PCR reactions (94 ° C, 90 seconds; 55 ° C, 60 seconds; 72 ° C, 120 seconds for 15 times, all buffers and the like as described above) It was. Reaction A was used for synthetic oligonucleotide sequence primers (SEQ ID NOs: 66 and 67) and reaction B was used for synthetic oligonucleotide sequence primers (SEQ ID NOs: 68 and 69). The products of these PCR reactions (A and B) were fragments of 535 base pairs and 205 base pairs in length, respectively. The reaction product was run on 2% agarose gel and separated from other background product. The gel was cut out of the expected size band and recovered. A mixture of products A and B in various amounts was made and subjected to PCR reaction using synthetic oligonucleotide sequences 66 and 68. The resulting product (about 700 base pairs) was digested with restriction enzymes SacII and XhoI and the degradation product was separated on a 2% agarose gel. The expected 310 base pair band was cut out from this gel and recovered. This fragment was ligated into a vector pNG3-806.077HuVK1-HuVK-Neo vector (preliminarily cleaved with the restriction enzyme SacII / XhoI) and the desired HuVK4 DNA and protein in the expression vector pNG3-Vkss-806.077HuVK4-HuCK-Neo. Sequences (SEQ ID NOs: 70 and 71) could be made.

In addition to the original HuVH1 sequences (SEQ ID NOs: 54 and 55), six variants of the humanized heavy chain variable region were made, which were called HuVH2 to HuVK7, respectively. The heavy chain variable region variant HuVH2 is a variant of the original HuVH1 coding sequence which obtained a amino acid change G49A (Kabat nomenclature) by mutating the gene (SEQ ID NO: 54) by cassette mutagenesis. Plasmid pNG4-VHss-806.077HuVH1-HuIgG2 CH1 '(containing humanized fully IgG2 heavy chain Fd (SEQ ID NOs: 56 and 57)) was digested with restriction enzymes StuI and NotI. This digest was then loaded onto a 2% agarose gel and the cut fragments were separated from the rest of the vector. Vector DNA was cut out of the gel and recovered and stored at −20 ° C. until needed. Two oligonucleotides were designed and synthesized (SEQ ID NOs: 72 and 73), and then the product obtained by hybridization was directly ligated to the vector prepared as described above. This DNA “cassette” exchange yielded the desired HuVH2 DNA and protein sequences (SEQ ID NOS: 74 and 75) that were positioned directly in the expression vector pNG4-VHss-806.077HuVH2-HuIgG2 CH1 '.

Similarly, amino acid change T73S using synthetic oligonucleotides (SEQ ID NOs: 76 and 77); HuVH3 with F78A (Kabat nomenclature) was constructed, in which case the vector pNG4-VHss-806.077HuVH1-HuIgG2 CH1 'was digested using restriction enzymes NotI and SacII. The synthetic DNA cassette was directly ligated into the vector thus prepared, thereby inserting the DNA and protein sequences (SEQ ID NOS: 78 and 79) of the desired HuVH3 in the expression vector pNG4-VHss-806.077HuVH3-HuIgG2 CH1 '.

Amino acid change G49A; T73S; And HuVH4 consisting of F78A (Kabat nomenclature) is a combination of HuVH2 (SEQ ID NOS: 74 and 75) and HuVH3 (SEQ ID NOS: 78 and 79). This digested the vector pNG4-VHss-806.077HuVH3-HuIgG2 CH1 'with the enzymes NotI and NheI and isolated about 200 base pairs of the NotI / NheI restriction fragment obtained by separating on a 2% agarose gel. The fragment was recovered and ligated to the vector pNG4-VHss-806.077HuVH2-HuIgG2 CH1 '(decomposed by restriction enzymes with the same NotI and NheI before purification of the vector fragment). As a result, a clone containing the DNA and protein sequences (SEQ ID NOs: 80 and 81) of the desired HuVH4 was obtained in the expression vector pNG4-VHss-806.077HuVH4-HuIgG2 CH1 '.

HuVH5 with amino acid change V67A (Kabat nomenclature) was constructed using synthetic oligonucleotides (SEQ ID NOs: 82 and 83). In addition, the vector pNG4-VHss-806.077HuVH1-HuIgG2 CH1 'was digested with restriction enzymes NotI and SacI. The synthetic DNA cassette was directly ligated to the vector thus prepared to obtain DNA and protein sequences (SEQ ID NOs 84 and 85) of the desired HuVH5 inserted in the expression vector pNG4-VHss-806.077HuVH5-HuIgG2 CH1 '.

HuVH6 with amino acid changes V67A, T73S and F78A (Kabat nomenclature) was constructed using synthetic oligonucleotides (SEQ ID NOs: 86 and 87), and for this mutant the vector pNG4-VHss-806.077HuVH1-HuIgG2 CH1 'was restriction enzyme NotI. And SacII. The synthetic DNA cassette was directly ligated to the thus prepared vector to obtain DNA and protein sequences (SEQ ID NOS 88 and 89) of the desired HuVH6 inserted in the expression vector pNG4-VHss-806.077HuVH6-HuIgG2 CH1 '.

Amino acid change G49A; V69A; HuVH7 with T73S and F78A (Kabat nomenclature) combines HuVH2 (SEQ ID NOs: 74 and 75) and HuVH6 (SEQ ID NOs: 88 and 89) variants. This conjugate was obtained by digesting the vector pNG4-VHss-806.077HuVH6-HuIgG2 CH1 'vector with the enzymes NotI and NheI and isolating about 200 base pairs of NotI / NheI restriction fragment via separation on a 2% agarose gel. This fragment was recovered and ligated to the vector pNG4-VHss-806.077HuVH2-HuIgG2 CH1 'vector (prepared with the same NotI and NheI restriction enzymes and purified to the vector fragment). As a result, a clone containing the desired HuVH7 DNA and protein sequences (SEQ ID NOs: 90 and 91) in the expression vector pNG4-VHss-806.077HuVH7-HuIgG2 CH1 'was obtained.

This combination of humanized heavy and light chain variable gene variants cleaves the heavy chain Fd gene variant expression cassette (containing the gene cleaved with the promoter and the BglII / SalI fragment) and clones this fragment into the BamHI / SalI site of the light chain variant vector. Co-expression vector constructs were obtained. Examples of possible variant combinations based on the aforementioned humanized heavy and light chain variants are shown in Table 1 below.

Combination of Humanized Heavy and Light Chain Variable Region Variants Example number Heavy chain variable region Sequence number Light chain variable region Sequence number Coexpressed Plasmid Vectors 11 HuVH1 54 and 55 HuVK1 49 and 50 pNG 806HuVH1 / HuVK1 / HuIgG2 12 HuVH1 54 and 55 HuVK2 60 and 61 pNG 806HuVH1 / HuVK2 / HuIgG2 13 HuVH1 54 and 55 HuVK3 64 and 65 pNG 806HuVH1 / HuVK3 / HuIgG2 14 HuVH1 54 and 55 HuVK4 70 and 71 pNG 806HuVH1 / HuVK4 / HuIgG2 15 HuVH2 74 and 75 HuVK1 49 and 50 pNG 806HuVH2 / HuVK1 / HuIgG2 16 HuVH2 74 and 75 HuVK2 60 and 61 pNG 806HuVH2 / HuVK2 / HuIgG2 17 HuVH2 74 and 75 HuVK3 64 and 65 pNG 806HuVH2 / HuVK3 / HuIgG2 18 HuVH2 74 and 75 HuVK4 70 and 71 pNG 806HuVH2 / HuVK4 / HuIgG2 19 HuVH3 78 and 79 HuVK1 49 and 50 pNG 806HuVH3 / HuVK1 / HuIgG2 20 HuVH3 78 and 79 HuVK2 60 and 61 pNG 806HuVH3 / HuVK2 / HuIgG2 21 HuVH3 78 and 79 HuVK3 64 and 65 pNG 806HuVH3 / HuVK3 / HuIgG2 22 HuVH3 78 and 79 HuVK4 70 and 71 pNG 806HuVH3 / HuVK4 / HuIgG2 23 HuVH4 80 and 81 HuVK1 49 and 50 pNG 806HuVH4 / HuVK1 / HuIgG2 24 HuVH4 80 and 81 HuVK2 60 and 61 pNG 806HuVH4 / HuVK2 / HuIgG2 25 HuVH4 80 and 81 HuVK3 64 and 65 pNG 806HuVH4 / HuVK3 / HuIgG2 26 HuVH4 80 and 81 HuVK4 70 and 71 pNG 806HuVH4 / HuVK4 / HuIgG2 27 HuVH5 84 and 85 HuVK1 49 and 50 pNG 806HuVH5 / HuVK1 / HuIgG2 28 HuVH5 84 and 85 HuVK2 60 and 61 pNG 806HuVH5 / HuVK2 / HuIgG2 29 HuVH5 84 and 85 HuVK3 64 and 65 pNG 806HuVH5 / HuVK3 / HuIgG2 30 HuVH5 84 and 85 HuVK4 70 and 71 pNG 806HuVH5 / HuVK4 / HuIgG2 31 HuVH6 88 and 89 HuVK1 49 and 50 pNG 806HuVH6 / HuVK1 / HuIgG2 32 HuVH6 88 and 89 HuVK2 60 and 61 pNG 806HuVH6 / HuVK2 / HuIgG2 33 HuVH6 88 and 89 HuVK3 64 and 65 pNG 806HuVH6 / HuVK3 / HuIgG2 34 HuVH6 88 and 89 HuVK4 70 and 71 pNG 806HuVH6 / HuVK4 / HuIgG2 35 HuVH7 90 and 91 HuVK1 49 and 50 pNG 806HuVH7 / HuVK1 / HuIgG2 36 HuVH7 90 and 91 HuVK2 60 and 61 pNG 806HuVH7 / HuVK2 / HuIgG2 37 HuVH7 90 and 91 HuVK3 64 and 65 pNG 806HuVH7 / HuVK3 / HuIgG2 38 HuVH7 90 and 91 HuVK4 70 and 71 pNG 806HuVH7 / HuVK4 / HuIgG2

Similar to Example 11, Example 14 was prepared by constructing a coexpression plasmid containing 806.077 HuVK4 light chain and 806.077 HuVH1 heavy chain variable region antibody genes. To this end, the plasmid pNG3-Vkss-806.077HuVK4-HuCK-Neo vector containing the humanized light chain variable region HuVK1 (SEQ ID NOs: 70 and 71) is digested with restriction enzymes BamHI and SalI, and the digested vector is depleted with 1% agarose. The gel was run and the bands of the vector were purified. Plasmid pNG4-VHss-806.077HuVH1-HuIgG2 CH1 '(containing the humanized 806.077 HuVH1 heavy chain variable regions (SEQ ID NOs: 56 and 57)) was digested with restriction enzymes BglII and SalI, and the reaction was isolated by running on 2% agarose. The fragment bands were then cut and purified. The recovered DNA fragment was ligated to the prepared pNG3-Vkss-806.077HuVK4-HuCK-Neo vector to obtain a clone of the desired HuVH1 / HuVK4 coexpression vector.

This construct was used to transfect NS0 myeloma cells (ECACC 85110503) via standard electropenetration techniques as described in Example 11, and the resulting transfectants were selected for G418 resistance. The resulting clones were tested for antibody expression by anti-human antibody Fd ELISA and CEA binding ELISA assay. Clones found to exhibit the best expression rate and CEA binding rate were selected and propagated to express the product. Human F (ab ') ₂ antibody fragments were then purified from the culture supernatants as described in Example 102.

Examples 39-47

Expression of Humanized F (ab ') ₂ Fragments with Different Kinds of Human Heavy Chain Constants

Other kinds of chimeric heavy chain Fd constructs may be used. Therefore, another heavy chain vector variant was prepared containing either one of HuIgG1 CH1 '(SEQ ID NOS: 20 and 21) or HuIgG3CH1' (SEQ ID NOS: 24 and 115) instead of the HuIgG2CH1 'gene (SEQ ID NOS: 22 and 23) as a constant region. The produced vectors were pNG4-VHss-HuIgG1 CH1 'and pNG4-VHss-HuIgG3 CH1', respectively. The heavy chain antibody variable region of interest is obtained by cleaving the appropriate pNG4-VHss- "VH variable region" -HuIgG2 CH1 'plasmid with EcoRI and SacI restriction enzymes, which are similarly restricted digested pNG4-VHss-HuIgG1CH1' or pNG4-VHss. Cloning into the -HuIgG3 CH1 'vector yielded a complete heavy chain Fd sequence. As described above, once the individual heavy and light chain sequences are constructed, the heavy chain Fd gene expression cassette (containing both the promoter and gene) is digested by restriction digestion, and the resulting fragment is cloned into the appropriate site of the light chain vector to achieve final concurrency. Expression vectors can be made. Table 2 below describes Examples 39-47 in which the various heavy and light chain variable regions were combined with many kinds of human heavy chain constant regions.

As Example 44, vector pNG4-VHss-HuIgG3 CH1 'was digested with restriction enzymes EcoRI and SacI and fragments of the vector were isolated as described above. HuVH1 heavy chain antibody variable regions (SEQ ID NOs: 54 and 55) are digested by digesting the plasmid pNG4-VHss-806.077HuVH1-HuIgG2 CH1 'with EcoRI and SacI restriction enzymes and fragments similarly digested pNG4-VHss-HuIgG3 CH1'. Cloning into the vector yielded the complete vector pNG4-VHss-806.077HuVH1-HuIgG3 CH1 'containing the humanized fully IgG3 heavy chain Fd sequences (SEQ ID NOs: 94 and 95). Light chain vector pNG3-Vkss-806.077 obtained by digesting the heavy chain Fd gene expression cassette (containing both the promoter and the gene) with a BglII / SalI fragment and digesting the vector band previously digested with restriction enzymes BamHI and SalI and isolated on 1% agarose The BamHI / SalI site of HuVK1-HuCK-Neo (including HuVK1-HuCK humanized light chain sequences 51 and 52) was cloned. As a result, a co-expression vector (pNG 806HuVH1 / HuVK3 / HuIgG3) capable of expressing a humanized 806.077 HuVH1 / HuVK1-HuIgG3 / kappa Fd antibody fragment was obtained.

Example number Humanized heavy chain Sequence number Humanized light chain Sequence number Coexpressed Plasmid Vectors 39 HuVH1-HuIgG1 92 and 93 HuVK1-HuCK 51 and 52 pNG 806HuVH1 / HuVK1 / HuIgG1 40 HuVH1-HuIgG2 56 and 57 HuVK1-HuCK 51 and 52 pNG 806HuVH1 / HuVK1 / HuIgG2 41 HuVH1-HuIgG3 94 and 95 HuVK1-HuCK 51 and 52 pNG 806HuVH1 / HuVK1 / HuIgG3 42 HuVH1-HuIgG1 92 and 93 HuVK3-HuCK 96 and 97 pNG 806HuVH1 / HuVK3 / HuIgG1 43 HuVH1-HuIgG2 56 and 57 HuVK3-HuCK 96 and 97 pNG 806HuVH1 / HuVK3 / HuIgG2 44 HuVH1-HuIgG3 94 and 95 HuVK3-HuCK 96 and 97 pNG 806HuVH1 / HuVK3 / HuIgG3 45 HuVH1-HuIgG1 92 and 93 HuVK4-HuCK 98 and 99 pNG 806HuVH1 / HuVK4 / HuIgG1 46 HuVH1-HuIgG2 56 and 57 HuVK4-HuCK 98 and 99 pNG 806HuVH1 / HuVK4 / HuIgG2 47 HuVH1-HuIgG3 94 and 95 HuVK4-HuCK 98 and 99 pNG 806HuVH1 / HuVK4 / HuIgG3

As Example 47, vector pNG4-VHss-HuIgG3 CH1 'was digested with EcoRI and SacI restriction enzymes and the resulting vector fragment was isolated. HuVH1 heavy chain antibody variable regions (SEQ ID NOs: 54 and 55) were digested by digesting the pNG4-VHss-HuVH1-HuIgG2 CH1 'plasmid with EcoRI and SacI restriction enzymes and fragments similarly digested pNG4-VHss-HuIgG3 CH1' vector Cloned into. As a result, a complete vector pNG4-VHss-806.077HuVH1-HuIgG3 CH1 'containing a humanized fully IgG3 heavy chain Fd sequence (SEQ ID NOs: 94 and 95) was obtained. The heavy chain Fd gene expression cassette (both promoter and gene) was digested with BglII / SalI fragments, light chain vector pNG3-Vkss- previously digested with restriction enzymes BamHI and SalI, isolated on 1% agarose and purified vector bands. It was cloned into the BamHI / SalI site of the 806.077 HuVK4-HuCK-Neo vector (containing HuVK4-HuCK humanized light chain sequences 98 and 99). As a result, a coexpression vector construct pNG 806HuVH1 / HuVK4 / HuIgG3 capable of expressing a humanized HuVH1 / HuVK1-HuIgG3 / kappa Fd antibody fragment was obtained.

All other examples presented in Table 2 above were also prepared in a manner similar to that described in Examples 44-47. However, for constructs containing human IgG1, heavy chain Fd gene expression cassettes (including promoters and genes) digested with BglII / BamHI fragments (because of the SalI restriction site inherent in the HuIgG1 CH1 'constant region genes are present) Cloning was made by cloning to the BamHI site of the suitably prepared light chain vector. In such cases, the orientation of the heavy chain cassette should be checked. This confirmation is observed by comparison of fragments from similarly degraded HuIgG2 forms (Examples 11 to 38), which can be compared to fragment size produced by restriction digestion (eg digestion with restriction enzyme HindIII) followed by gel electrophoresis analysis. do. If the pattern of fragments for these two constructs match, then the heavy chain cassette is considered inserted in the correct orientation.

As described above in Example 11, these constructs were used to transfect NS0 myeloma cells (ECACC 85110503) via standard electropenetration and the transfectants were selected for resistance to G418. The resulting clones were tested for antibody expression through anti-human antibody Fd ELISA and CEA binding ELISA assays, and genes were expressed by selecting, extending and propagating clones identified as having the best expression and CEA binding rates. As before, human F (ab ′) ₂ antibody fragments were purified from the culture supernatants as described in Example 102.

Example 48

Preparation of Humanized 806.077F (ab ') _2- [A248S, G251T, D253K] HCPB Fusion Proteins

This example shows a method for preparing a gene encoding 806.077 humanized Fd heavy chain fragment bound to [A248S, G251T, D253K] HCPB, and the gene encoding the humanized light chain of 806.077 and human carboxypeptidase. Co-expression of the pro domain of B together with a gene encoding a method for obtaining an F (ab ') ₂ protein having a [A248S, G251T, D253K] HCPB molecule at the C terminus of each heavy chain fragment. The constant and hinge regions of the humanized Fd heavy chain fragments were obtained from human IgG3 antibody isotypes. The protein expressed as such is called an antibody-enzyme fusion protein.

(a) Preparation of a gene encoding 806.077 humanized Fd heavy chain fragment bound to [A248S, G251T, D253K] HCPB and cloning to pEE6

The gene encoding humanized 806.077 Fd bound to [A248S, G251T, D253K] HCPB was prepared by PCR from pZEN 1921 (Reference Example 2). Primary PCR was performed using 100 mM oligonucleotide sequence 100 in buffer containing 100 mM Tris-HCl (pH8.3), 50 mM KCl, 1.5 mM MgCl ₂ , 0.125 mM of each dATP, dCTP, dGTP and dTTP. And SEQ ID NO: 101 (100 pM each) and template pZEN1921 (2 ng). Incubated at 94 ° C. for 5 minutes and then thermostable DNA polymerase (2.5 u, 0.5 μl) was added and the mixture was layered with mineral oil (100 μl) and the reaction mixture was stirred at 94 ° C. for 1 minute at 53 ° C. 25 cycles of incubation at 1 min and 2.5 min at 72 ° C. followed by 10 min at 72 ° C. The resulting 536 base pair PCR product was electrophoretically separated on a 1% agarose (Agarose type I, Sigma A-6013) gel, the separated bands were cut from the gel and the DNA fragments isolated.

Secondary PCR was performed using template IgG3-pBSIIKS + (8.7 ng, Reference Example 4), oligonucleotide sequences 102 and 103 and 954 base pair fragments isolated as described above. Two PCR products were mixed in PCR buffer as described above (0.2 ng / μl, 1.0 ng / μl or 5.0 ng / μl concentrations). The mixture was incubated at 94 ° C. for 5 minutes and then circulated 10 times for 1 minute at 94 ° C. and 4 minutes at 63 ° C. Oligo sequences 101 and 102 (100 pM each) in PCR buffer (50 μl) were added. After incubation at 94 ° C. for 3 minutes, the mixture was cycled 25 times at 94 ° C. for 1.5 minutes, at 53 ° C. for 2 minutes, and at 72 ° C. for 2 minutes and finally at 72 ° C. for 10 minutes. In the process, base G at position 508 in SEQ ID NO: 115 was changed to base A.

PCR products of 1434 base pairs were isolated by electrophoresis on 1% agarose gel and purified, followed by 10 mM Tris HCl, pH 7.9, 50 mM NaCl, 10 mM MgCl ₂ , 1 mM DTT, and BSA (100 μg / ml). NheI (20u) and XbaI (80u) (New England BioLabs, Inc.) in 100 μl total volume containing) were digested at 37 ° C. for 4 hours. The resulting fragments were again separated by electrophoresis on 1% agarose gel and purified. In a similar digestion manner, the vector pNG4-VHss-806.077huVH1-HuIgG2CH1 '(10 μg; Example 11) was cleaved with NheI and XbaI and digested plasmids with small intestinal alkaline phosphatase (1 μl; New England Biolabs, 10 u / [Mu] l) was added to remove the 5 'phosphate group and then further incubated at 37 [deg.] C. for 30 minutes. The phosphatase activity was destroyed by incubation at 70 ° C. for 10 minutes. NheI-XbaI cleaved plasmids were purified from agarose gels. The NheI-XbaI digested PCR product thus obtained (about 500 ng) was purified by 50 mM Tris-HCl (pH 7.8), 10 mM MgCl ₂ , 10 mM DTT, 1 mM ATP, 50 μg / ml BSA and 400 u T4 DNA ligase (New England). Ligation of the plasmid DNA (about 200 ng) with the truncated plasmid DNA (about 200 ng) in 20 μl of the solution containing BioLabs, Inc. for 4 hours. 1 µl of this reaction solution was used to transform 20 µl of competent E. coli DH5α cells. Transformed cells were plated on L agar + 100 μg / ml ampicillin. Potential clones containing the gene for humanized 806.077 Fd- [A248S, G251T, D253K] HCPB were identified by PCR. Each clone was PCR treated as described above using oligonucleotide sequences 104 and 105. PCR reaction samples (10 μl) were analyzed by electrophoresis on 1% agarose gel. Clones containing the genes of interest were identified through the presence of a 512 base pair PCR product. Clones producing 512 base pair bands were used for DNA microfabrication. This DNA sample was digested with HindIII and XbaI to examine for the presence of 3751 base pair and 1862 base pair fragments. This DNA was used for mass production of plasmid DNA using clones containing fragments obtained upon cleavage with HindIII and XbaI, and the sequence of the insert was confirmed by DNA sequencing. The sequence of the expected insert is set forth in SEQ ID NO: 112. Of the clones examined, two clones contained the expected sequence but one base was mutated. Clone 54 (also referred to as pMF195) had base T instead of base A at position 605 in SEQ ID NO: 112, and clone 68 (also referred to as pMF198) had base C instead of expected base T at position 1825. The sequence set forth in SEQ ID NO: 112 is a buffer containing pMF195 and pMF198 (10 μg each) containing 20 mM tris acetate (pH 7.9), 50 mM potassium acetate, 10 mM magnesium acetate, 1 mM DTT and BSA (100 μg / ml). Prepared by digestion with XmaI (10u) and XbaI (100u) (New England Biolabs) in 100 μl). 215 base pair fragments derived from pMF195 and vector fragments derived from pMF198 (after treatment with alkaline phosphatase) were separated from 1% agarose gels and ligated together as described above. The ligation mixture was used to transform competent DH5α cells. The transformed cells were plated on L agar + ampicillin and the resulting colony's DNA was XmaI and XbaI digested to select colonies for the presence of 5400 base pairs and 215 base pair fragments. Positive clones were used to mass produce plasmid DNA and the sequence of the insert was confirmed by DNA sequencing. The plasmid containing the 806.077Fd- [A248S, G251T, D253K] HCPB gene obtained from clone 102 was called pMF213. HindIII-XbaI fragments derived from pMF213 were selected by pEE6 in DH5α (PCR using oligonucleotide sequences 106 and 107 for 2228 base pair inserts). This plasmid was converted to pEE6.hCMV by converting HindIII site upstream of the hCMV promoter into Is a derivative of Stephens and Cockett (1989) Nucleic Acids Research 17, 7110).

(b) Preparation of co-expression vector for expression of antibody-enzyme fusion protein

To prepare vectors capable of expressing antibody-enzyme fusion proteins in eukaryotic cells, GS-System ^™ (Celltech Biologics) was used (WO 87/04462, WO 89/01036, WO 86/05807 and WO 89/10404). ). This procedure involves cloning the humanized antibody light chain gene into the HindIII-XmaI region of vector pEE14. This vector is described in Bebbington in METHODS: A Companion to methods in Enzymology (1991) 2, 136-145. To construct the expression vector, the plasmids pEE14 and pNG3-VKss-806.077HuVK4-HuCK-Neo (Example 14) were digested with HindIII and XmaI as described above. From each digest, the appropriate vector (from pEE14) and insert (pNG3-VKss-806.077 HuVK4-HuCK-Neo derived 732 base pairs) were isolated from 1% agarose gel and ligated together to transform competent DH5α cells. . Transformed cells were plated onto L agar + ampicillin (100 μg / ml). Colonies were selected via restriction enzyme analysis of DNA isolates, ie DNA was digested with HindIII and XmaI to detect the presence of 732 base pair fragments. Colonies producing 732 base pair restriction fragments were used for mass production of plasmid DNA and the sequence of the insert was confirmed by DNA sequence analysis. The plasmid containing the humanized light chain sequence set forth in SEQ ID NO: 70 in pEE14 was called pEE14-806.077HuVK4-HuCK.

To prepare the coexpression vector, pMF221 (10 μg) was added to buffer (100 μg / ml) containing 10 mM Tris-HCl (pH 7.9), 150 mM NaCl, 10 mM MgCl ₂ , 1 mM DTT and BSA (100 μg / ml). Μl) was digested with BglII (20u) and SalI (40U), and the 4560 base pair fragment was isolated and purified by agarose gel electrophoresis. Similarly, pEE14-806.077HuVK4-HuCK was digested with BamHI (40u) and SalI (40u), 9.95 kb vector fragments were isolated and ligated to BglII-SalI fragments from pMF221 and cloned into DH5α. The resulting colonies were selected by PCR using two groups of oligonucleotides (SEQ ID NOS: 104 and 105, and SEQ ID NOs: 108 and 109). DNA sequences were analyzed using clones providing 185 base pairs and 525 base pairs, respectively, of the PCR product. The clone with the correct sequence was called pMF228-light chain / Fd-mutant HCPB coexpression vector in DH5α. Humanized Fd-mutant HCPB sequences are set forth in SEQ ID NO: 113. Residues 1 to 19 are signal sequences, residues 20 to 242 are humanized variable regions and IgG3 CH1 regions, residues 243 to 306 are IgG3 hinge regions, and residues 307 to 613 are residues 248, 251 and 253 in the human HCPB sequence. This is another mutant HCPB sequence. Changes in the HCPB sequence are shown at positions 554 (Ser), 557 (Thr) and 559 (Lys) in SEQ ID NO: 113, respectively.

(c) preparation of a vector for expressing the pro domain of proHCPB

As a second eukaryotic expression plasmid, pEE12 containing the gene for the prepro sequence of preproHCPB to secrete the pro domain (called pro-L) with additional C-terminal leucine residues is disclosed in International Patent Application No. WO96 / 20011. Prepared as described in Reference Example 17. Plasmid pMF161 was prepared by PCR from pMF18 as described for the unmodified prepro sequence, except oligonucleotide sequences 110 and 111 were used. The 359 base pair fragment was cloned into pBluescript to obtain pMF141, which was then cloned into pEE12 to give pMF161. The protein sequence of pro-L is set forth in SEQ ID NO: 114.

(d) Expression of antibody enzyme fusion proteins in eukaryotic cells

For expression in eukaryotic cells, COS-7 cells were cotransfected using a vector containing a gene capable of expressing an antibody enzyme fusion protein (pMF228) and a vector containing a pro-L sequence (pMF161). COS cells are transformed with the origin defective SV40 virus and have been widely used to transiently express various proteins in a short period of time because of the ability to replicate circular plasmids containing the SV40 replication origin in very high copy numbers. CV-1. Two widely used COS cell clones are COS-1 and COS-7. Basic methods for transfecting COS cells are described in Bebbington in Methods: A Companion to Methods in Enzymology (1991) 2, p. 141. 10% heat inactivation via a method known as lipofection (Felgner et al. In Methods: A Companion to Methods in Enzymology (1993) 5, 67-75) to express HCPB COS-7 cells (2 × 10 ⁵ ) using plasmid vectors pMF48 and pMF67 (2 μg each) in a 6-well culture plate added with 2 ml of Dulbecco's Modified Eagle's Medium (DMEM) containing purified fetal bovine serum (FCS). ) Was transfected. The cells were incubated for 20 hours at 37 ° C. in a CO ₂ incubator. The mixture of plasmid DNA mixed in serum-free medium (200 μl) was slowly mixed with LIPOFECTIN ^™ reagent (12 μl) and incubated for 15 minutes at room temperature. The cells were washed with serum free medium (2 mL). Serum-free medium (600 μl) was added to DNA / LIPOFECTIN ^™ and the mixture was layered on the cells and incubated for 6 hours at 37 ° C. in a CO ₂ incubator. Medium containing DNA was replaced with normal DMEM containing 10% FCS and cells were incubated for 72 hours as described above. Cell supernatant (diluted 1:10 with 0.025 M Tris-HCl, pH 7.5; 125 μl) activity against Hipp-Glu (5 hours assay in 250 μl total volume) as substantially described in Example 103 Was analyzed. Supernatant dilutions hydrolyzed Hipp-Glu substrate by 18.4%.

Alternatively, an unmodified pro domain (derived from plasmid pMF67 described in Reference Example 17 of International Application No. WO 96/20011) may be used in place of the pro-L expression plasmid in this experiment.

Mass expression of proteins derived from COS cells is described in Ridder et al. (1995) in GENE 166, 273-276 and Blasey et al. (1996) in CRYOTECHNOLOGY 18, 183-192.

To investigate stable expression in CHO cells, the method described in Bebbington in METHODS: A Companion to Methods in Enzymology (1991) 2, 136-145 was used for GS selection using MSX 25 μM and 50 μM. . Alternatively, CHO cells may be transfected using a lipofection method substantially as described above to transfect COS cells. Cells are transfected using plasmids pMF228 and pMF161 or a mixture of pMF228 and pMF67. Supernatants of viable colonies are selected by CEA ELISA (described in Example 11) and Western analysis (described below) for the presence of a 170 kDa band corresponding to a given antibody enzyme fusion protein. In addition, the supernatant is appropriately diluted to select for enzymatic activity as described in Example 103. Colonies expressing the desired antibody enzyme fusion protein were incubated in the required amount (see ME Reff (1993) in Current Opinion in Biotechnology 4, 573-576 and references cited therein) and fused from this cell culture supernatant. Protein was purified using one or more of the methods described in Example 102.

(e) Western analysis

Western blot analysis was performed as follows. An amount (20 μl) of each supernatant sample was mixed with the same amount of sample buffer (62.5 mM Tris, pH 6.8, 1% SDS, 10% sucrose and 0.05% bromophenol blue) with or without reducing agent. Samples were incubated at 65 ° C. for 10 minutes and then electrophoresed on an 8-18% acrylamide gradient gel (EXCEL ^™ gel system from Pharmacia Biotechnology Products) according to the manufacturer's instructions in a MULTIPHOR ^™ II apparatus (LKB Produkter AB). . After electrophoresis, the separated proteins were transferred to membranes (HYBOND ^™ C-Super, Amersham International) using a NOVABLOT ^™ device (LKB Produkter AB). After blotting, the membrane was air dried.

The presence of antibody fragments was detected using antihuman kappa antibody (Sigma A7164, goat antihuman kappa light chain peroxidase conjugate) used at a 1: 2500 dilution. The presence of human antibody fragments was visualized using a chemiluminescence system (ECL ^™ detection system, Amersham International).

Examples 49-74

Preparation of other humanized 806.077F (ab ') ₂ -mutant HCPB fusion proteins

This example shows a method for producing a gene encoding 806.077 humanized Fd heavy chain fragment bound to mutant HCPB (D253K; G251T, D253K; A248S, G251T, D253K) and the gene encoding 806.077 humanized light chain. And co-expression with a gene encoding the pro domain of human carboxypeptidase B to produce an F (ab ′) ₂ protein having a mutant HCPB molecule at the C terminus of each heavy chain fragment. The constant and hinge regions of the humanized Fd heavy chain fragments were obtained from human IgG1 or IgG2 or IgG3 antibody isotypes. The expressed protein was called antibody enzyme fusion protein.

The procedure described in Example 48 was repeated using the appropriate sequence described in Table 3 below. Oligonucleotides used for PCR construction and clone selection can be easily obtained from appropriate sequences.

To change the mutant HCPB sequence, instead of the plasmid pZEN1921, the PCR template described in paragraph (a) of Example 48, pZEN1860 (Reference Example 1) for [G251T, D253K] HCPB or pICI1713 for [D253K] HCPB (International Application). No. WO96 / 20011).

In order to change the antibody heavy chain constant region and the hinge region, pNG4-VHss-HuIgG1CH1 '(described in Examples 39 to 47) or pNG4-VHss- in place of the vector IgG3-pBSIIKS +, a PCR template described in paragraph (a) of Example 48, was used. HuIgG2CH1 '(NCIMB 40797) was used.

To change the humanized antibody light chain sequence, pEE14-806.077HuVK1-HuCK or pEE14-806.077HuVK3-HuCK was used in place of the vector pEE14-806.077HuVK4-HuCK described in paragraph (b) of Example 48. The vectors pEE14-806.077HuVK1-HuCK and pEE14-806.077HuVK3-HuCK are derived from pNG-VHss-806.077HuVK1-Neo and pNG-VHss-806.077NuV instead of HindIII-XmaI fragments derived from pNG-VHss-806.077HuVK4-Neo, respectively. Prepared as described for pEE14-806.077HuVK4-HuCK in paragraph (b) of Example 48, except that the 732 base pair of HindIII-XmaI fragments (described in Examples 12-38) were used.

The antibody-enzyme fusion protein variants used in each example are as described in Table 3 below.

Example number Humanized heavy chain Humanized light chain Mutant HCPB Enzyme 49 HuVH1-HuIgG3 HuVK4-HuCK [D253K] HCPB 50 HuVH1-HuIgG3 HuVK4-HuCK [G251T, D253K] HCPB 51 HuVH1-HuIgG3 HuVK1-HuCK [A248S, G251T, D253K] HCPB 52 HuVH1-HuIgG3 HuVK1-HuCK [D253K] HCPB 53 HuVH1-HuIgG3 HuVK1-HuCK [G251T, D253K] HCPB 54 HuVH1-HuIgG3 HuVK3-HuCK [A248S, G251T, D253K] HCPB 55 HuVH1-HuIgG3 HuVK3-HuCK [D253K] HCPB 56 HuVH1-HuIgG3 HuVK3-HuCK [G251T, D253K] HCPB 57 HuVH1-HuIgG1 HuVK4-HuCK [A248S, G251T, D253K] HCPB 58 HuVH1-HuIgG1 HuVK4-HuCK [D253K] HCPB 59 HuVH1-HuIgG1 HuVK4-HuCK [G251T, D253K] HCPB 60 HuVH1-HuIgG1 HuVK1-HuCK [A248S, G251T, D253K] HCPB 61 HuVH1-HuIgG1 HuVK1-HuCK [D253K] HCPB 62 HuVH1-HuIgG1 HuVK1-HuCK [G251T, D253K] HCPB 63 HuVH1-HuIgG1 HuVK3-HuCK [A248S, G251T, D253K] HCPB 64 HuVH1-HuIgG1 HuVK3-HuCK [D253K] HCPB 65 HuVH1-HuIgG1 HuVK3-HuCK [G251T, D253K] HCPB 66 HuVH1-HuIgG2 HuVK4-HuCK [A248S, G251T, D253K] HCPB 67 HuVH1-HuIgG2 HuVK4-HuCK [D253K] HCPB 68 HuVH1-HuIgG2 HuVK4-HuCK [G251T, D253K] HCPB 69 HuVH1-HuIgG2 HuVK1-HuCK [A248S, G251T, D253K] HCPB 70 HuVH1-HuIgG2 HuVK1-HuCK [D253K] HCPB 71 HuVH1-HuIgG2 HuVK1-HuCK [G251T, D253K] HCPB 72 HuVH1-HuIgG2 HuVK3-HuCK [A248S, G251T, D253K] HCPB 73 HuVH1-HuIgG2 HuVK3-HuCK [D253K] HCPB 74 HuVH1-HuIgG2 HuVK3-HuCK [G251T, D253K] HCPB

Example 75

Preparation of the [A248S, G251T, D253K] HCPB- (humanized 806.077) F (ab ') ₂ fusion protein

This example shows a method for producing a gene encoding pro- [A248S, G251T, D253K] HCPB bound to a humanized (form 1 VH with human IgG3) Fd heavy chain fragment of antibody 806.077 and to a human body of the 806.077 antibody. The present invention relates to a method of coexpressing a light chain (form 4 VK having CK) with a gene encoding the light chain. This example provides a F (ab ′) ₂ protein with pro- [A248S, G251T, D253K] HCPB molecules at each N terminus of the heavy chain fragment. This enzyme is activated by enzymatic removal of the pro domain using trypsin.

Standard molecular biology techniques such as restriction enzyme digestion, ligation, kinase reaction, dephosphorylation, polymerase chain reaction (PCR), bacterial transformation, gel electrophoresis, buffer preparation and DNA production, purification and isolation are described in Maniatis et. al., (1989) Molecular Cloning, A Laboratory Manual; Second Edition: as described in the Cold Spring Harbor Laboratory, Cold Spring Harbor, New York, or following procedures recommended by the manufacturer of the particular product. In most cases the enzyme was purchased from New England Biolabs, but procedures equivalent to other products may be used. Oligonucleotide sequences were subjected to controlled pores of 5'dimethoxytrityl base-blocked nucleoside-2-cyanoethyl-N, N'-diisopropyl-phosphoamidite in an Applied Biosystems 380A DNA synthesizer. Blocked nucleosides bound to the glass support were prepared according to the protocol provided by Applied Biosystems Inc. using about 0.2 μmol.

The HCPB mutants, native HCPB and HCPB fusion proteins can be used in the HPLC described in Example 103 or Example 20 of International Application WO96 / 20011 for the properties of converting hippuryl-L-glutamic acid or hipfuryl-L-arginic acid to hippuric acid. The analysis was carried out using a system analysis method.

Immunoassay techniques are based on the procedures described in Tijssen (1985) Practice and Theory of Enzyme Immunoassays, Laboratory Techniques in Biochemistry and Molecular Biology Volume 15, Elsevier Science Publishers, Amsterdam, or procedures recommended by the manufacturer of a particular product. It was carried out using.

In order to make plasmids capable of expressing antibody enzyme fusion proteins in eukaryotic cells, two plasmids pEE6 (a derivative of pEE6.hCMV in which the HindIII restriction enzyme site upstream of the hCMV promoter was converted to the BglII site {Stephens and Cockett, 1989, Nucleic Acids Research, 17, 7110}) and pEE12 (a derivative of pSV2.GS with a number of restriction sites removed (Bebbington et al, 1992, Bio / Technology, 10.169)) were used GS-System (Celltech Biologics). (See international applications WO 87/04462, WO 89/01036, WO 86/05807 and WO 89/10404 for specific details).

a) Cloning of pre-pro-HCPB to the restriction enzyme XmaI cleavage site (position 1048 in SEQ ID NO: 124)

The double-stranded DNA of plasmid pMF18 (disclosed in Reference Example 19 of international application WO 96/20011), a construct containing pre-pro-HCPB cloned in vector pBluescript II KS + (starratazine), was prepared using standard DNA techniques (Qiagen plasmid Kit or equivalent) and limited digestion with care to ensure complete digestion with HindIII and XmaI enzymes. The restriction enzyme HindIII cleaves the pMF18 plasmid immediately before the pre-sequence initiation of the HCPB gene and XmaI cleaves at the codon for amino acid 240 (proline) of the mature protein. The resulting HindIII to XmaI DNA fragments are called pre-pro-HCPB fragments. DNA of the correct size containing the pre-pro-HCPB fragment (about 1061 base pairs) was purified.

Double-stranded DNA of the plasmid vector pUC19 (New England Biolabs) was prepared, restriction digested with HindIII and XmaI, and purified in a similar manner as in the pre-pro-HCPB fragment (about 2651 base pairs). To clone the HCPB gene fragment into the pUC19 vector, the molar ratio of vector to insert was used at about 1: 2.5 in the presence of T4 DNA ligase, 1 mM ATP and enzyme buffer and the final DNA concentration contained at about 2.5 ng / μl. One ligation mixture was prepared. After the ligation reaction, the E. coli strain DH5α was transformed using the DNA mixture. A certain amount of cells were plated on L-agar nutrient medium containing 100 μg / ml of ampicillin for plasmid vector selection and incubated overnight at 37 ° C. Several colonies were taken and used for the preparation of double stranded plasmid DNA traces. This DNA sample was analyzed by restriction enzyme digestion and constructs of the correct orientation were identified. The plasmid containing the pre-pro-HCPB fragment up to the XmaI site present in the mature gene was called pCF003.

b) Cloning of five amino acids of [A248S, G251T, D253K] HCPB + linker and VH from position G241

To isolate HCPB from the Fd sequence, a neutral peptide linker containing (glycine-glycine-glycine-serine) ₃ was introduced into the sequence during PCR. To create a mutant [A248S, G251T, D253K] HCPB sequence fragment (described in Reference Example 2) and add the peptide linker and the first five amino acids of humanized 806.077VH, primer CME 00971 and in the presence of about 5 ng of pZen1921 DNA; PCR was initiated using 100 pMol of CME 00972 (SEQ ID NOS: 122 and 123) and a final concentration of 200 μM of dNTP, Taq polymerase reaction buffer and 2.5 U of Taq polymerase in a final volume of 100 μL. The mixture was heated at 94 ° C. for 10 minutes prior to addition of the Taq enzyme, and the PCR incubation was circulated for 1.5 minutes at 94 ° C., 2 minutes at 55 ° C., 2 minutes at 72 ° C. and at 72 ° C. at the end of the reaction. Incubate once for 10 minutes. PCR products containing the resulting [A248S, G251T, D253K] HCPB fragment (about 298 base pairs) were subjected to agarose gel electrophoresis to analyze the presence of the correct size of DNA and found that the correct size of bands was mainly present. Purification of the remaining product from the reaction mixture, separation of excess reagents using a microconcentrator column (Centricon ^™ 100, Amicon), followed by ethanol / sodium acetate precipitation, centrifugation, vacuum drying and resuspension in distilled water DNA was isolated. The isolated DNA was restriction digested with XmaI and EcoRI and the bands of the correct size (about 271 base pairs) were purified.

The double-stranded DNA (Qiagen Plasmid Kit or equivalent) of plasmid pCF003 (as described above) prepared using standard DNA techniques was digested with XmaI and EcoRI enzymes to purify bands of the correct size (about 2696 base pairs).

To clone the mutant HCPB gene fragment into the vector, the molar ratio of vector to insert in the presence of T4 DNA ligase, 1 mM ATP and enzyme buffer was adjusted to about 1: 2.5 (pCF003 1: [A248S, G251T, D253K] HCPB fragment PCR product). Ligation mixture was prepared, which was used as 2.5) and containing a final DNA concentration of about 2.5 ng / μl. After the ligation reaction, the E. coli strain DH5α was transformed using the DNA mixture. A certain amount of cells were plated on L-agar nutrient medium containing 100 μg / ml of ampicillin for plasmid vector selection and incubated overnight at 37 ° C. About 200 colonies were taken and plated in 2 replicates on sterile Nitrocellulose filtrate (Schleicher and Schull) prewet on an L-agar nutrient medium plate containing 100 μg / ml of ampicillin for plasmid vector selection and plated at 37 ° C. Incubate overnight. One of the two repeating plates was stored at 4 ° C. to be used as a living cell source of colonies, and the other one was denatured to fix DNA obtained from individual colonies on nitrocellulose. The nitrocellulose filter was separated from the agar plate, followed by 1 minute in 10% SDS, 2 minutes in 0.5M NaOH, 1.5M NaCl, 4 minutes in 0.5M NaOH, 1.5M NaCl, 0.5M NaOH, Subsequently, they were placed continuously on immersion (Watman) immersed in 1.5M NaCl for 2 minutes, ⑤ 0.5M Tris pH 7.4, 1.5M NaCl for 2 minutes, and ⑥ 2x SSC (standard citric acid saline) for 2 minutes. This site was then placed in a room (Watman) soaked in 10 × SSC and subjected to UV treatment (Spectrolinker XL-1500 UV crosslinker) to crosslink the denatured DNA to nitrocellulose. The filter was air dried at room temperature and then prehybridized at 60 ° C. for 1 hour with slow stirring in 6 × SSC solution (eg using a Techne HB-1D hybridizer). Prehybridization blocks any nonspecific DNA binding site.

To determine colonies containing the DNA insert of interest, the crosslinked DNA in nitrocellulose margin was radiolabeled ³² P- prepared from [A248S, G251T, D253K] HCPB purified PCR DNA fragment (see above). Hybridization with DNA probes. Approximately 50 ng of DNA was labeled with ³² P-dCTP 50 μCi (> 3000 Ci / mMol) in a total volume of 50 μl (Pharmacia T7 Quick Prime Kit) using T7 DNA polymerase, and the reaction was carried out at 37 ° C. for 15 minutes. Proceeded. The labeled probe was heated at 95 ° C. for 2 minutes to denature double-stranded DNA, and then immediately 10 ml of 6 × SSC at 60 ° C. was added and this solution was used to replace the preliminary hybridization solution. Incubate with slow stirring at 60 ° C. for about 3 hours. This hybridization solution was then discarded and the filtrate was washed twice at 60 ° C. twice each with 2 × SSC for 15 minutes. The room was slowly blotted dry and covered with an attachment film (Saran ^™ wrap or similar) and then exposed to X-ray film (eg Kodak X-OMAT-ARS ^™ ) overnight at room temperature. After the film was developed, it was identified as a colony containing the desired insert that exhibited the strongest exposure (darkest spot) on the X-ray film. In this series of experiments, about 15% of colonies showed positive hybridization. Of these, 12 colonies were selected and used for further screening tests. These colonies were taken from the margin of replication, streaked and grown on L agar nutrient medium containing 100 μg / ml of ampicillin, and then grown in L broth nutrient medium containing 100 μg / ml of ampicillin.

The colonies thus selected were used to prepare double stranded plasmid DNA traces. This DNA sample was analyzed by restriction enzyme digestion and constructs of the correct orientation were identified. To determine what changes were made to the DNA sequence during PCR, DNA was prepared by standard techniques (quiagen plasmid kits or the like) by taking multiple clones with exact restriction maps and inserting the insert into different oligonucleotide primers. Was sequenced using. Identify the construct of the correct sequence and place the plasmid containing the pre-pro- [A248S, G251T, D253K] HCPB-linker-humanized 806.077VH gene up to the PstI site (5th amino acid site) (position 1301 of SEQ ID NO: 124). Was called pCF004.

c) Cloning of humanized 806.077 Fd

Double-stranded DNA of plasmid pNG4-VHss-HuVH1-806.077-IgG3CH1 ', a construct containing human IgG3 CH1 cloned in vector pNG4 (see Example 44) and a humanized 806.077 Form 1 VH with hinge region, was used as standard DNA Prepared using the technique (Qiagen plasmid kit or the like) and restriction digestion with PstI and XmaI enzymes. DNA of the correct size containing the humanized 806.077 Fd fragment (about 854 base pairs) was purified. Double-stranded DNA of the plasmid vector pUC19 (New England Biolabs) was prepared and subjected to restriction digestion with PstI and XmaI, and purified in a similar manner to the humanized 806.077 Fd fragment (about 2659 base pairs).

After restriction digestion, a certain amount of purified DNA sample was taken, its purity was investigated, and the concentration was estimated by agarose gel electrophoresis compared with a known standard. From this estimate, the ligation mixture for cloning the humanized 806.077 Fd gene fragment into the pUC19 vector was used with a molar ratio of vector to insert of about 1 to 2.5 and a final DNA concentration of about 2.5 ng / μl, and T4 DNA ligase , 1 mM ATP and enzyme buffer were used together.

After the ligation reaction, the E. coli strain DH5α was transformed using the DNA mixture. A predetermined amount of cells were plated on L agar nutrient medium containing 100 μg / ml of ampicillin for plasmid vector selection and incubated overnight at 37 ° C. Several colonies were taken and used for the preparation of double stranded plasmid DNA traces. This DNA sample was analyzed by restriction enzyme digestion and constructs of the correct orientation were identified. The plasmid containing the 806.077 Fd fragment humanized from the PstI site to the XmaI site was called pCF005.

d) Cloning of humanized 806.077 Fd into pre-pro- [A248S, G251T, D253K] HCPB-linker constructs

Double-stranded DNA of plasmid pCF005 (described above) was prepared using standard DNA techniques (Qiagen plasmid kit or the like) and restriction digested with PstI and EcoRI enzymes. DNA of the correct size containing the humanized 806.077 Fd fragment (about 870 base pairs) was purified. Double-stranded DNA of the plasmid vector pCF004 (described above) was prepared and subjected to restriction digestion with PstI and EcoRI, and then purified by a method similar to that for the humanized 806.077 Fd fragment (about 3950 base pairs). To clone the humanized 806.077 Fd gene fragment into the pCF004 vector, the molar ratio of vector to insert was about 1 to 2.5 and the final DNA concentration was about 2.5 ng / in the presence of T4 DNA ligase, 1 mM ATP and enzyme buffer. The ligation mixture was made to a μl.

After the ligation reaction, the DNA mixture was used to transform E. coli strain DH5α. A certain amount of cells were plated onto L agar nutrient medium containing 100 μg / ml of ampicillin for plasmid vector selection and incubated overnight at 37 ° C. Several colonies were taken and used for micropreparation of double stranded plasmid DNA. This DNA sample was analyzed by restriction enzyme digestion and constructs of the correct orientation were identified. The plasmid containing pre-pro- [A248S, G251T, D253K] HCPB-Linker-Fd (humanized 806.077) in pUC19 was called pCF006.

e) Cloning of pre-pro- [A248S, G251T, D253K] HCPB-Linker- (humanized 806.077) Fd into pEE6 hCMV vector

Double-stranded DNA of plasmid pCF006 (described above) was prepared using standard DNA techniques (Qiagen plasmid kit or the like) and restriction digested with HindIII and EcoRI enzymes. DNA of the correct size containing the fusion protein (about 2185 base pairs) was purified.

Two-stranded DNA of the plasmid vector pEE6 (described above) was prepared, subjected to restriction digestion with HindIII and EcoRI, and purified in a similar manner as the fusion protein (about 4775 base pairs). To clone the humanized 806.077 Fd fusion protein into the pEE6 vector, the molar ratio of vector to insert was about 1 to 2.5 and the final DNA concentration was about 2.5 ng / in the presence of T4 DNA ligase, 1 mM ATP and enzyme buffer. The ligation mixture was made to a μl. After the ligation reaction, the DNA mixture was used to transform E. coli strain DH5α. A certain amount of cells were plated onto L agar nutrient medium containing 100 μg / ml of ampicillin for plasmid vector selection and incubated overnight at 37 ° C. Several colonies were taken and used for micropreparation of double stranded plasmid DNA. This DNA sample was analyzed by restriction enzyme digestion and constructs of the correct orientation were identified. The plasmid containing pre-pro- [A248S, G251T, D253K] HCPB-Linker-Fd (humanized 806.077) in pEE6 was called pCF007.

f) Cloning of 806.077 light chain form 4 humanized into pEE12 vector

Double-stranded DNA of plasmid pNG3-VKss-806.077-HuVK4- HuCK-Neo, a construct containing the humanized 806.077 form HuVK4 with human CK cloned in vector pNG3 (see Examples 12-38), was prepared using standard DNA techniques (see Qiagen plasmid kit or the like) and restriction digestion with HindIII and EcoRI enzymes. DNA of the correct size containing the humanized 806.077 light chain (about 2022 base pairs) was purified. Double-stranded DNA of the plasmid vector pEE12 was prepared, subjected to restriction digestion with HindIII and EcoRI, and purified in a similar manner to the humanized 806.077 light chain (about 7085 base pairs). To clone the humanized 806.077 light chain into the pEE12 vector, the molar ratio of vector to insert was about 1 to 2.5 and the final DNA concentration was about 2.5 ng / μl in the presence of T4 DNA ligase, 1 mM ATP and enzyme buffer. The ligation mixture was prepared. After the ligation reaction, the DNA mixture was used to transform E. coli strain DH5α. A certain amount of cells were plated onto L agar nutrient medium containing 100 μg / ml of ampicillin for plasmid vector selection and incubated overnight at 37 ° C. Several colonies were taken and used for micropreparation of double stranded plasmid DNA. This DNA sample was analyzed by restriction enzyme digestion and constructs of the correct orientation were identified. The plasmid containing the humanized 806.077 light chain form 4 was called pCF008 / 4.

g) Cloning of CMVp-pre-pro- [A248S, G251T, D253K] HCPB-Linker- (humanized 806.077) Fd in pCF008 / 4

Double-stranded DNA of plasmid pCF007 (described above) was prepared using standard DNA techniques (Qiagen plasmid kit or the like) and restriction digested with BglII and SalI enzymes. Restriction enzyme BglII cleaves the pCF007 plasmid before the beginning of the CMV MIE leader, promoter and fusion protein gene. The restriction enzyme SalI cleaves about 520 base pairs after the stop codon of the mature protein. DNA of the correct size containing the fusion protein (about 4844 base pairs) was purified. Double-stranded DNA of the plasmid vector pCF008 / 4 was prepared and subjected to restriction digestion with BamHI and SalI, and purified by a similar method as the fusion protein (about 7436 base pairs). To clone the [A248S, G251T, D253K] HCPB-linker- (humanized 806.077) Fd fusion gene into the pCF008 / 4 vector, the mole ratio of vector to insert was about 1 to 2.5 and T4 DNA ligase, 1 The ligation mixture was prepared so that the final DNA concentration was about 2.5 ng / μL in the presence of mM ATP and enzyme buffer. After the ligation reaction, the DNA mixture was used to transform E. coli strain DH5α. A certain amount of cells were plated onto L agar nutrient medium containing 100 μg / ml of ampicillin for plasmid vector selection and incubated overnight at 37 ° C. Several colonies were taken and used for micropreparation of double stranded plasmid DNA. This DNA sample was analyzed by restriction enzyme digestion and constructs of the correct orientation were identified. The plasmid containing the gene for pro- [A248S, G251T, D253K] HCPB-linker-F (ab ') ₂ (humanized 806.077 antibody) in GS expression vector pEE12 is called pCF009 and the map of this plasmid is shown in FIG. Shown in The DNA and amino acid sequences of the light chain HuVK4 are set forth in SEQ ID NOs: 70 and 71. The DNA sequence of the pre-pro- [A248S, G251T, D253K] HCPB-linker-Fd (humanized 806.077) is set forth in SEQ ID NO: 124 and the corresponding amino acid sequence is set forth in SEQ ID NO: 125.

h) Expression of pro- [mutant] HCPB-linker-F (ab ') ₂ (humanized 806.077) from mouse myeloma cells

The following method was used to express all (D253K; G251T, D253K; and A248S, G251T, D253K) mutant pro-HCPB enzyme fusion proteins in myeloma. Preferred mouse myeloma cell line is NS0 (Galfre and Milstein, 1981, Methods in Enzymol., 73, 3-46), European Animal Cell Culture Collection, PHLS CAMR (Esp4 0 Jay Wiltshire Salisbury Photon Down, UK). (ECACC List No. 85110503). The cells were grown in Dulbecco's Modified Eagle's Medium (DMEM; Gibco / BRL) containing 10% heat inactivated fetal bovine serum (FCS).

Two plasmids pCF009 containing neomycin resistance genes for selecting stable cell lines resistant to G418 to express pro- [A248S, G251T, D253K] HCPB-Linker-F (ab ') ₂ (humanized 806.077) (Described above) and pRc / RSV (Invitrogen List No. V780-20) were used. Each plasmid (0.5 to 10 [mu] g) via lipofection (Felgner et al., In Methods: A Companion to Methods in Enzymology, 1993, 5, 67-75), which delivers polynucleotides via cationic lipids into eukaryotic cells. ) About 8 × 10 ⁶ NS0 cells were transfected at about 5 μg. The cells were harvested by centrifugation, washed with serum free medium (30 mL), then resuspended in 800 μl medium and kept at 37 ° C. in tissue culture flasks until the addition of DNA. Serum-free medium (450 μl) was slowly mixed with LIPOFECTIN ^™ reagent (50 μl) and incubated for 30-45 minutes at room temperature. This mixture was added to 500 μl of medium containing plasmid DNA mixture (less than 100 μl) and left at room temperature for 15 minutes. To this plasmid DNA-LIPOFECTIN ^™ mixture was added serum-free medium (600 μl) and the mixture was added to cells incubated for about 5 hours at 37 ° C. in a CO ₂ incubator. The medium containing this DNA was exchanged with normal DMEM medium (8 mL) containing 10% FCS and the cells were incubated overnight. The medium was again exchanged with normal DMEM medium (8 mL) containing 10% FCS and the cells were incubated for 24 hours without selection as described above. At the end of the culture, the medium was exchanged with DMEM containing 10% FCS and G418 selector (1.5 mg / ml), and the cells were diluted with this medium (1: 4 to 1: 20) (about 0.5 to 1.5 cells per plate). x 10 ⁶ ) Microtiter wells (150 μl per well; two or more plates per dilution). This microtiter plate was periodically examined for incubation for at least 2 weeks at 37 ° C. in a CO ₂ incubator and for the formation of viable clones.

Medium of wells containing one viable clone was taken for use in the test and exchanged with fresh medium (containing G418). The harvested medium was tested for antibody binding to CEA in ELISA [International Application WO96 except for replacing the second antibody solution with anti-human (goat anti-human kappa light chain peroxidase conjugate, Sigma A7164) instead of anti-mouse. / 20011 Test by the method as described in Reference Example 5, paragraph 1)]. Using positive samples in the CEA ELISA, the reaction was carried out at 700 占 폚 / ml in trypsin (50 mM Tris-HCl and 150 mM NaCl, pH 7.6) at 4 ° C for 1 hour, followed by addition of 5 times or more soybean trypsin inhibitor. [A248S, G251T, D253K] HCPB enzyme activity (described above) was tested after activation (removing the pro domain from the fusion protein). As a result, several clones were identified that produced a medium positive for 806.077 antibody binding to CEA and [A248S, G251T, D253K] HCPB enzyme activity. This clone was used in International Application WO96 / 20011 Reference Example 5 paragraph j, except that the non-reducing Western blot analysis [antibody solution was replaced with anti-human (goat anti-human kappa light chain peroxidase conjugate, Sigma A7164) instead of anti mouse. Tests as described were further identified to clones that produced predominantly F (ab ') ₂ (806.077) fusion proteins. The clones were grown and tested for stable production of fusion proteins during passage, and the clones with the highest production rates were cultured in mass and frozen in liquid nitrogen using standard techniques.

Amplification, high concentration expression and fermentation of fusion proteins from NS0 myeloma cells were performed in a similar manner as described in Bebbington et al. (1992) in Bio / Technology 10, 169-175. The fusion protein was purified and the pro-sequences isolated as described in Example 102.

Examples 76-101

Cloning and expression of other variants of the pro-HCPB-linker- (humanized 806.077) Fd + (humanized 806.077) light chain

The method of generating a fusion protein with other variants of HCPB is similar to that described in detail in Example 75 (described above), except that instead of the [A248S, G251T, D253K] HCPB used in paragraph b) of Example 75, [ D253K] HCPB or [G251T, D253K] HCPB were used and the plasmid DNA used in the PCR reaction was either pICI1713 (international application WO96 / 20011, described in Example 15) or pZEN1860 (Reference Example 1). After cloning, identification and sequencing, the humanized up to the PstI site (amino acid 5) present in the pre-pro- [D253K] HCPB-linker or pre-pro- [G251T, D253K] HCPB-linker obtained and the base vector pUC19 Plasmids containing the 806.077VH gene were used in place of pCF004 in subsequent cloning reactions.

The method for generating fusion proteins with other CH1 domains is similar to the method described in detail in Example 75 (described above), except that human IgG1 or IgG2 CH1 instead of 806.077-HuVH1-IgG3CH1 'in Example 75c). A plasmid containing a humanized 806.077 VH form 1 and hinge region was used (SEQ ID NOS: 92 and 56, respectively). After cloning, identification and sequence identification, the plasmids containing the obtained IgG1 or IgG2 sequences were used in place of pCF005 in subsequent cloning reactions.

The method of producing a fusion protein with other variants of the humanized 806.077 light chain is similar to the method described in detail in Example 75 (described above), except that in humans instead of 806.077-HuVK4-HuCK in paragraph f) of Example 75 Plasmids containing the converted 806.077 Lc Form 1 or Form 3 were used (SEQ ID NOs: 51 and 96, respectively). After cloning, identification and sequence identification, the plasmids containing the alternative light chain sequences obtained were used in place of pCF008 / 4 for subsequent cloning reactions. Fusion protein variants (76-101) of each example are listed in Table 4 below.

Example number Humanized heavy chain Humanized light chain Mutant HCPB Enzyme 76 HuVH1-HuIgG3 HuVK4-HuCK [D253K] HCPB 77 HuVH1-HuIgG3 HuVK4-HuCK [G251T, D253K] HCPB 78 HuVH1-HuIgG3 HuVK1-HuCK [A248S, G251T, D253K] HCPB 79 HuVH1-HuIgG3 HuVK1-HuCK [D253K] HCPB 80 HuVH1-HuIgG3 HuVK1-HuCK [G251T, D253K] HCPB 81 HuVH1-HuIgG3 HuVK3-HuCK [A248S, G251T, D253K] HCPB 82 HuVH1-HuIgG3 HuVK3-HuCK [D253K] HCPB 83 HuVH1-HuIgG3 HuVK3-HuCK [G251T, D253K] HCPB 84 HuVH1-HuIgG1 HuVK4-HuCK [A248S, G251T, D253K] HCPB 85 HuVH1-HuIgG1 HuVK4-HuCK [D253K] HCPB 86 HuVH1-HuIgG1 HuVK4-HuCK [G251T, D253K] HCPB 87 HuVH1-HuIgG1 HuVK1-HuCK [A248S, G251T, D253K] HCPB 88 HuVH1-HuIgG1 HuVK1-HuCK [D253K] HCPB 89 HuVH1-HuIgG1 HuVK1-HuCK [G251T, D253K] HCPB 90 HuVH1-HuIgG1 HuVK3-HuCK [A248S, G251T, D253K] HCPB 91 HuVH1-HuIgG1 HuVK3-HuCK [D253K] HCPB 92 HuVH1-HuIgG1 HuVK3-HuCK [G251T, D253K] HCPB 93 HuVH1-HuIgG2 HuVK4-HuCK [A248S, G251T, D253K] HCPB 94 HuVH1-HuIgG2 HuVK4-HuCK [D253K] HCPB 95 HuVH1-HuIgG2 HuVK4-HuCK [G251T, D253K] HCPB 96 HuVH1-HuIgG2 HuVK1-HuCK [A248S, G251T, D253K] HCPB 97 HuVH1-HuIgG2 HuVK1-HuCK [D253K] HCPB 98 HuVH1-HuIgG2 HuVK1-HuCK [G251T, D253K] HCPB 99 HuVH1-HuIgG2 HuVK3-HuCK [A248S, G251T, D253K] HCPB 100 HuVH1-HuIgG2 HuVK3-HuCK [D253K] HCPB 101 HuVH1-HuIgG2 HuVK3-HuCK [G251T, D253K] HCPB

Example 102

Purification of Proteins Containing 806.077 Antibody Sequences

Several methods were used alone or in combination to purify or propagate recombinant F (ab ') ₂ or antibody-enzyme fusion proteins from myeloma cells, CHO cells or COS cell supernatants. Murine 806.077F (ab ') ₂ , chimeric 806.077F (ab') ₂ constructs and fully humanized 806.077F (ab ') ₂ constructs, and antibody-enzymes incorporating these F (ab') ₂ constructs Fusion protein constructs were purified using one or more different methods, such as affinity chromatography, anion exchange chromatography, or protein A / protein G chromatography. These techniques can also be used to purify the 806.077 antibody-B7 fusion (see Example 104).

a) antigen affinity chromatography

The cancerous fetal antigen (CEA), causing parental murine 806.077 antibodies, was immobilized on a column (using Pharmacia). In brief, immobilization is achieved through stable ester bonds to NHS-activated Sepharose ^™ high performance media pre-filled in a column (HiTrap ^™ ); The binding of CEA to the activated substrate was performed according to the standard guidelines provided with the product.

Preparation of 1 ml Affinity Column

The CEA stock solution (8 mg / ml) was first diluted with coupling buffer (0.2 M sodium hydrogen carbonate, 0.5 M sodium chloride; pH 8.3) to a final concentration of 0.5 mg / ml. The new column was washed with 6 ml of ice cold 1 mM HCl at a flow rate not exceeding a flow rate of 1 ml / min. Immediately, CEA ligand (1 ml at 0.5 mg / ml) was injected onto the column. Both ends of this column were sealed and left at room temperature for 30 minutes. Nonspecifically bound ligands were washed out of the column by inactivating the remaining activator without binding to the ligand and alternating high and low pH washes three times. The buffers used were 0.5M ethanolamine, 0.5M sodium chloride (pH 8.3) and 0.1M sodium acetate and 0.5M sodium chloride (pH 4.0). 6 ml of each buffer was introduced onto the column substrate for each wash procedure. Finally, the column was washed with stock buffer (0.05 M Na ₂ HPO ₄ , 0.1% NaN _3, pH 7.0).

Purification Procedure

Cell culture supernatants containing the desired F (ab ') ₂ or fusion constructs, such as chimeric 806.077 F (ab') ₂ , humanized 806.077F (ab ') _2, or antibody-enzyme fusion proteins, were treated with phosphate buffered saline ( pH 1: 1) and passed through a 1 ml affinity column at a flow rate of 1 ml / min. This column was previously equilibrated with phosphate buffered saline (pH 7.2; 50 mM sodium phosphate, 150 mM sodium chloride). The cell supernatant was passed through this column and washed with 10 column volumes of phosphate buffered saline. The eluate was collected in 1 ml fractions, while bound F (ab ') ₂ was eluted with 5 column volumes of 100 mM sodium citrate (pH 3.0). Eluted F (ab ') ₂ is Western blot using a suitable antibody peroxidase conjugate (antihuman kappa light chain-peroxidase conjugate, when fully humanized F (ab') ₂ , using Sigma A-7164) It was analyzed by color development with hydrogen peroxide and 4-chloro-1-naphthol. Appropriate fractions were collected and concentrated if necessary using a centrifugal concentrator (Centricon ^™ 30).

b) anion exchange chromatography

The cell culture supernatant containing the desired F (ab ') or fusion construct, such as chimeric 806.077F (ab') ₂ , humanized 806.077F (ab ') _2, or antibody-enzyme fusion protein, was added to the ionic strength of the solution. Diafiltration was performed with 50 mM Tris (using stirred cells with a membrane with a cut off of 10,000 molecular weight) until equivalent to column equilibration buffer. An aliquot of 40 ml of the superfiltered supernatant was loaded on a suitable column (Pharmacia Mono Q ^™ 10/10 HR) at a rate of 2 ml / min. This column was previously equilibrated with 50 mM Tris, pH 8.0. The supernatant was passed through this column and the column was again washed to baseline with equilibration buffer. The material bound on the column was then eluted using 15 column volumes of 0-50% buffer B (50 mM Tris, 1M sodium chloride pH 8.0). Elution fractions were collected (4 ml per fraction), using a suitable antibody peroxidase conjugate (antihuman kappa light chain-peroxidase conjugate, Sigma A-7164 in a fully humanized F (ab ') ₂ case, with hydrogen peroxide and Western blot analysis, developing with 4-chloro-1-naphthol, identified the fractions containing F (ab ') ₂ . Appropriate fractions were collected and concentrated if necessary in a centrifugal concentrator (Centricon ^™ 30).

c) Protein A and Protein G Purification

Desired F (ab ') ₂ or fusion constructs (eg, 806.077F (ab') ₂ , chimeric 806.077 IgG ₁ or IgG ₂ or IgG ₃ ; pro-HCPB-linker-806.077F (ab ') ₂ , 806.077F Cell culture supernatants containing (ab ′) ₂ -HCPB) were diluted 1: 1 with phosphate buffered saline and then loaded onto a column previously equilibrated with phosphate buffered saline (pH 7.2). The column was washed again with phosphate buffered saline to baseline, then 100 mM sodium citrate (pH 3.0) for F (ab ') ₂ and 50 mM glycine and 100 mM sodium chloride (pH 10.8) for fusion proteins. To elute bound F (ab ') ₂ or fusion proteins. The elution fractions were combined and neutralized by adding 125 μl of 2M Tris per 1 mL of elution volume. Fractions containing F (ab ') ₂ were combined and concentrated if necessary using a centrifugal concentrator.

d) pro-sequence cutting

In the case of fusion proteins containing covalently linked pro-sequences (eg pro-HCPB-linker-806.077F (ab ') ₂ ), the pro sequence is cleaved by incubating the fusion with trypsin. This procedure on the milligram (fusion) scale was carried out as follows. Trypsin and fusion proteins were mixed at a ratio of 1: 1000 (trypsin: fusion). This mixture was incubated at room temperature (about 22 ° C.) for 24 hours to completely cleave the pro sequence. This mixture was recycled using any of the usual chromatographic purification or concentration protocols to separate the fusion protein from the pro sequence.

Example 103

Activity Analysis of Antibody-Enzyme Fusion Proteins Containing Mutant Human CPB Against Hipp-Glu Prodrug Analogues

Hippuryl-L-glutamic acid (Hipp-Glu; with cell culture supernatants or purified antibody enzyme fusion proteins containing mutants of human CPB (D253K; G252T, D253K; A248S, G251T, D253K: Examples 48-101)); The conversion of Reference Example 9) of international application WO96 / 20011 to hipfuric acid was analyzed using HPLC-based analysis.

The reaction mixture (250 μl) was dissolved in 4 μg of purified fusion protein or cell culture supernatant (pure water or diluted to 0.025M Tris-HCl pH7.5; 125 μl) and 0.025M Tris-HCl (pH 7.5). It contains 0.5 mM Hipp-Glu. This sample was incubated at 37 ° C. for 5 hours. The reaction was terminated by addition of 250 μl of a mixture of 30% methanol, 70% phosphate buffer (50 mM; pH 6.5), 0.2% trifluoroacetic acid, and the amount of hippuric acid produced was determined using HPLC (Hewlett Packard using a diode array). 1090 Series 11).

Sample (50 μl) was injected onto a column (25 cm; HICHROM ^™ Hi-RPB) and separated using a mobile phase of 15% methanol, 85% phosphate buffer (50 mM; pH 6.5) at a flow rate of 1 ml / min. . The amount of product obtained (hipfuric acid) was determined from a calibration curve made of known amounts of hipfuric acid (Sigma-H6375). The results are expressed as the rate of conversion of the substrate to product at 37 ° C. for 30 minutes to 24 hours depending on the conversion rate.

In the case of an antibody enzyme fusion protein with N-terminal proCPB, the pro domain is first removed by treatment with trypsin (700 μg / ml) in 50 mM Tris-HCl (pH 7.6), 150 mM NaCl for 1 hour at 4 ° C.

Example 104

Preparation of human B7.1-humanized 806.077F (ab ') ₂ fusion protein (hB7-806)

As in Reference Example 3, a fusion protein containing the signal sequence and extracellular domain of human B7.1 fused directly to the 5 'coding region of the humanized 806.077 antibody Fd chain was constructed using PCR technique. The HindIII-NheI fragment was then made that contains the extracellular domain and the natural signal sequence of human B7.1 fused to the V _H region of the humanized 806.077 antibody heavy chain. This is cloned into a suitable vector, such as pNG4-V _H ss-HuIgG2CH1 'or pNG4-V _H ss-HuIgG3CH1' (see Examples 39 to 47) (replacement of bases 1 to 423 in SEQ ID NO: 18) to human B7.1-humanization 806.077 Fd fusion gene was obtained. A suitable vector containing the fused and humanized 806.077 L chain (VK4 form of humanized 806.077 light chain is pCF008 / 4 after constructing a co-expression vector using an expression system such as that described herein; Example 75 Co-expression). These vectors were used to transfect NS0 myeloma cells and select colonies through CEA binding activity measured in culture supernatants. Other humanized sequences are described in Examples 39-47.

The hB7-806 fusion protein was expressed from a suitable cell line and purified using a Protein A column as described in Reference Example 3 or using any one of the methods described in Example 102. Purification methods other than the Protein A column are also preferred for preparing humanized 806.077 antibody fragments and fusion proteins thereof. Fusion proteins can be tested using the assays described in Reference Example 3 for antigen-receptor binding and T-cell costimulatory activity when bound to LS 174T cells.

Example 105

Preparation of Chimeric and Humanized 806.077 F (ab ') ₂ -CPG2 Conjugates

The procedure described in Example 5 was repeated using any one of the chimeric forms described in Example 8 or one of the humanized forms described in Examples 39-47, instead of the rat F (ab ') ₂ protein.

Example 106

Preparation of Humanized 806.077 Fab-CPG2 Enzyme Fusion Protein

Humanized 806.077 antibody and bacterial CPG2 enzyme fusion protein constructs were constructed using a PCR method similar to that described for the construction of HuVK4 in Examples 12-38, wherein the specifically designed primers were used for PCR reactions. To amplify the antibody and enzyme gene components (so that the resulting DNA product contains overlapping complementary sequences) and then bind via another “graft / bind” PCR reaction to produce a complete antibody enzyme fusion gene. The fusion protein was obtained by binding the 3 'end of the Fd humanized 806.077 antibody heavy chain gene to the 5' end of the CPG2 structural coding gene to prepare a Fab-CPG2 fusion protein coding gene. In this construct, the humanized 806.077 antibody heavy chain gene component terminates after residue K236 for HuVH1-HuIgG1 Fd heavy chain (SEQ ID NO: 93) or after residue Val 237 for HuVH1-HuIgG2 Fd heavy chain (SEQ ID NO: 57). Or after residue Val 237 for HuVH1-HuIgG3 Fd heavy chain (SEQ ID NO: 95) (hence in each case all sequences belonging to the hinge region are excluded), the first CPG2 located at the C terminus for the signal sequence cleavage site. To residues (Minton et al (1984) Gene 31, 31-38). However, in order to obtain adequate antibody binding and enzymatic properties, it may be desirable to incorporate additional residues at the linkages between the two components.

The fusion protein was then digested and isolated and cloned into a suitable vector, such as pNG4-VHss-HuIgG2CH1 '(NCIMB 40797), where the fusion gene DNA fragments cloned the original antibody gene into the gene of the fusion protein. Made by replacing A coexpression vector was constructed in a similar manner as described in Example 11 followed by coexpression of the humanized 806.077 light chain and fusion. Colonies were selected by transfecting NS0 myeloma cells using the co-expression vector and detecting the presence of CEA and Fd binding activity for each culture supernatant as described above. The fusion protein was purified through a Protein A column and found to have both antigenic and enzymatic properties using standard test methods.

Example 107

Other combinations of humanized heavy and light chain variable regions based on light chain sequence VK4

The procedure described in Examples 12-38 was repeated using a modified sequence in which the tyrosine residue (Tyr) at position 35 of SEQ ID NO: 71 was substituted with the phenylalanine residue (Phe) instead of the humanized light chain variable sequence VK4 (SEQ ID NO: 71). .

Example 108

Other combinations of humanized heavy and light chain variable regions based on light chain sequence VK4

The procedure described in Examples 12-38 was repeated using a modified sequence in which the phenylalanine residue (Phe) at position 72 of SEQ ID NO: 71 was replaced with a leucine residue (Leu) instead of the humanized light chain variable sequence VK4 (SEQ ID NO: 71). .

Example 109

Other combinations of humanized heavy and light chain variable regions based on light chain sequence VK4

Instead of the humanized light chain variable sequence VK4 (SEQ ID NO: 71), the tyrosine residue (Tyr) at position 35 of SEQ ID NO: 71 is replaced with a phenylalanine residue (Phe) and the phenylalanine residue (Phe) at position 72 of SEQ ID NO: 71 is a leucine residue ( The procedure described in Examples 12-38 was repeated using a modified sequence substituted with Leu).

Example 110

Combination of humanized heavy chain variable region and chimeric light chain sequence

The procedure described in Examples 12-38 was repeated with the chimeric sequence of SEQ ID NO: 17 described in Example 8 instead of the humanized light chain variable sequence.

Examples 111-113

Expression of humanized F (ab ') ₂ fragments with modified light chain VK4 variable sequences

Example using the variable light sequence set forth in Example 107 used to prepare a replacement for the humanized light chain sequence of SEQ ID NO: 99 wherein the tyrosine residue (Tyr) at position 57 of SEQ ID NO: 99 was replaced with a phenylalanine residue (Phe) The procedure described in 39-47 was repeated.

Example 111 is a combination of HuVH1-HuIgG1 with SEQ ID NO: 99 modified as described above.

Example 112 is a combination of HuVH1-HuIgG2 with SEQ ID NO: 99 modified as described above.

Example 113 is a combination of HuVH1-HuIgG3 with SEQ ID NO: 99 modified as described above.

Examples 114-116

Expression of humanized F (ab ') ₂ fragments with modified light chain VK4 variable sequences

The variable light chain sequence described in Example 108 was used to prepare a replacement for the humanized light chain sequence of SEQ ID NO: 99 wherein the phenylalanine residue (Phe) at position 94 of SEQ ID NO: 99 was replaced with a leucine residue (Leu). The procedure described in Examples 39-47 was repeated.

Example 114 is a combination of HuVH1-HuIgG1 with SEQ ID NO: 99 modified as described above.

Example 115 is a combination of HuVH1-HuIgG2 with SEQ ID NO: 99 modified as described above.

Example 116 is the combination of HuVH1-HuIgG3 with SEQ ID NO: 99 modified as described above.

Examples 117-119

Expression of humanized F (ab ') ₂ fragments with modified light chain VK4 variable sequences

In the humanized light chain sequence of SEQ ID NO: 99, wherein the tyrosine residue (Tyr) at position 57 of SEQ ID NO: 99 and the phenylalanine residue (Phe) at position 94 of SEQ ID NO: 99 are replaced with phenylalanine residues (Phe) and leucine residues (Leu), respectively The procedure described in Examples 39-47 was repeated using the variable light chain sequence described in Example 109 used to make a replacement for.

Example 117 is a combination of HuVH1-HuIgG1 with SEQ ID NO: 99 modified as described above.

Example 118 is a combination of HuVH1-HuIgG2 with SEQ ID NO: 99 modified as described above.

Example 119 is a combination of HuVH1-HuIgG3 with SEQ ID NO: 99 modified as described above.

Examples 120-122

Expression of Humanized F (ab ') ₂ Fragments with Chimeric Light Chain Sequences

The procedure of Examples 39-47 was repeated using the chimeric light chain sequence described in Example 110 instead of the humanized light chain sequences used in Examples 39-47.

Example 120 is a combination of HuVH1-HuIgG1 with a modified chimeric light chain sequence as described above.

Example 121 is a combination of HuVH1-HuIgG2 with a modified chimeric light chain sequence as described above.

Example 122 is a combination of HuVH1-HuIgG3 with a modified chimeric light chain sequence as described above.

Example 123

Preparation of Humanized Fusion Proteins Based on Modified Light Chain VK4 Sequences

The procedure described in Example 48 was repeated using a plasmid containing the modified VK4 sequences described in Examples 107 and 111-113 instead of the plasmid pEE14-806.077HuVK4-HuCK.

Example 124

Preparation of Humanized Fusion Proteins Based on Modified Light Chain VK4 Sequences

The procedure described in Example 48 was repeated using a plasmid containing the modified VK4 sequences described in Examples 108 and 114-116 instead of plasmid pEE14-806.077HuVK4-HuCK.

Example 125

Preparation of Humanized Fusion Proteins Based on Modified Light Chain VK4 Sequences

The procedure described in Example 48 was repeated using a plasmid containing the modified VK4 sequences described in Examples 109 and 117-119 instead of plasmid pEE14-806.077HuVK4-HuCK.

Example 126

Preparation of Humanized Fusion Proteins Based on Chimeric Light Chain VK4 Sequences

The procedure described in Example 48 was repeated using a plasmid containing the chimeric light chain sequences described in Examples 110 and 120-122 instead of plasmid pEE14-806.077HuVK4-HuCK.

Example 127

Preparation of Humanized Fusion Proteins Based on Modified Light Chain VK4 Sequences

The procedure described in Example 75 was repeated using a plasmid containing the modified VK4 sequences described in Examples 107 and 111-113 instead of plasmid pCF008 / 4.

Example 128

Preparation of Humanized Fusion Proteins Based on Modified Light Chain VK4 Sequences

The procedure described in Example 75 was repeated using a plasmid containing the modified VK4 sequences described in Examples 108 and 114-116 instead of plasmid pCF008 / 4.

Example 129

Preparation of Humanized Fusion Proteins Based on Modified Light Chain VK4 Sequences

The procedure described in Example 75 was repeated using a plasmid containing the modified VK4 sequences described in Examples 109 and 117-119 instead of plasmid pCF008 / 4.

Example 130

Preparation of Humanized Fusion Proteins Based on Chimeric Light Chain VK4 Sequences

The procedure described in Example 75 was repeated using a plasmid containing the chimeric light chain sequences described in Examples 110 and 120-122 instead of plasmid pCF008 / 4.

Reference Example 1

Preparation of Gene Sequences for [G251T, D253K] HCPB

The method of cloning [G251T, D253K] HCPB in E. coli is very similar to the method described in Example 15 of international application WO96 / 20011. PICI266 was used as the cloning vector, but [D253K} HCPB present in plasmid pICI1713 (Example 15 of International Application No. WO96 / 20011) was used as a starting material for PCR site directed mutagenesis. In this case, site directed mutagenesis was used during PCR amplification of the gene to change the codon at amino acid position 251 of the mature gene from glycine to threonine (GGC to ACT), ie G251T. In addition, many other mutations were generated at the same site (G251) using the oligonucleotide mixture having a codon change in G251. Each mutant gene was identified by sequencing the mutation site prior to complete gene sequencing after transformation and hybridization. In this reference only the oligonucleotides introducing the G251T mutation will be considered. Two PCR mixtures were prepared by the method as described in International Application No. WO96 / 20011 Example 15. Primers used in the first reaction are CAN 00402 (SEQ ID NO: 116) and CAN 00734 (SEQ ID NO: 117). Primers used in the second reaction are CAN 00284 (SEQ ID NO: 118) and CAN 01076 (SEQ ID NO: 119). In both reactions, the starting DNA is pICI1713.

Using a certain amount of two PCR reactions, DNA was analyzed for correct size (approximately 750 base pairs and 250 base pairs) and the concentration was confirmed by agarose gel electrophoresis. Another PCR was performed using the two PCR products together with two terminal primers (CAN 00402 (SEQ ID NO: 116) and CAN 00284 (SEQ ID NO: 118)). Using a certain amount of this PCR product, agarose gel electrophoresis was used to determine whether the DNA was the correct size (about 1000 base pairs). Purify the remaining product obtained in the reaction mixture, limit digestion of the isolated DNA with enzymes NcoI and EcoRI, and bands of the correct size (about 1000 base pairs) in a manner similar to that described in Example 16 of International Application No. WO 96/20011. Purified.

pICI266 double-stranded DNA was digested with NcoI and EcoRI enzymes and purified to the correct size of DNA (about 5600 base pairs). Purity was investigated by agarose gel electrophoresis using a limited amount of restriction digested and purified vector and insert DNA samples, and the concentration was estimated by comparison with known standards. Based on these estimates, the ligation mixture was prepared to clone the HCPB gene into the pICI266 vector in a similar manner as described in International Application No. WO96 / 20011 Example 16.

After the ligation reaction, the E. coli strain DH5α was transformed with the DNA mixture, and then colonies were taken and tested by hybridization. Multiple clones were taken for plasmid DNA preparation and the sequences on PCR mutation regions were analyzed to identify clones with G251T changes in a similar manner as described in International Application No. WO96 / 20011 Example 16. From this sequence analysis, a clone containing a plasmid having the desired [G251T: D253K] HCPB gene sequence was selected and called this plasmid pZEN1860.

Reference Example 2

Preparation of Gene Sequences for [A248S, G251T, D253K] HCPB

The method of cloning [A248S, G251T, D253K] HCPB in E. coli was very similar to the method described in Reference Example 1. [G251T, D253K] HCPB gene in plasmid pZEN1860 (Reference Example 1) was used instead of pICI1713 as a starting material for PCR site directed mutagenesis. In this case, site directed mutagenesis was used during PCR amplification of the gene to change the codon at amino acid position 248 in the mature gene from alanine to serine (GCT to TTC), ie A248S. Two PCR mixtures were prepared by the method as described in Reference Example 1. Primers used in the first reaction are CAN 00402 (SEQ ID NO: 116) and CAN 00720 (SEQ ID NO: 120). Primers used in the second reaction are CAN 00284 (SEQ ID NO: 118) and CAN 00726 (SEQ ID NO: 121). The starting DNA in these two reactions is pZEN1860.

PCR, cloning, expression and identification methods were carried out in the same manner as in Reference Example 1. From the sequencing results, clones containing the plasmid with the desired [A248S, G251T, D253K] HCPB gene sequence were selected and the plasmid was called pZEN1921.

Reference Example 3

Preparation and characterization of human B7.1-mouse A5B7 F (ab ') ₂ fusion protein (AB7)

Methods for preparing, purifying and characterizing recombinant murine A5B7 F (ab ') ₂ antibodies have been published (WO 96/20011, Reference Example 5). CDNA sequences for human B7.1 antigen (also called CD80) have been disclosed separately (Freeman GJ et al, Journal of Immunology, 1989, 143, 2714-2722). In this example, "AB7" means human B7.1-mouse A5B7 F (ab ') ₂ fusion protein, and "A5B7" means anti-CEA antibody called A5B7.

PCR-based assays were used to isolate the native signal sequence of human B7.1 and the extracellular domains (amino acids 1-242) from cDNAs derived from Raji culture cells (ATCC CCL86), which were then matured in mouse A5B7 Fd fragments. 'Fused directly upstream of the coding sequence. As a result, a B7.1 sequence having PCR primers 187/96 and 204/96 (SEQ ID NOs 126 and 127) and an A5B7 Fd subsequence having PCR primers 203/96 and 205/96 (SEQ ID NOs: 128 and 129) were isolated. After purifying the PCR product, it was mixed in approximately equivalent molar amount and fused by PCR with primers 187/96 and 205/96. The resulting PCR product was purified and digested with HindIII and BstEII (New England Biolabs, UK, Sti4 Tiwiczyn Willberry Way) and cloned into the HindIII-BstEII region of pAF1 using standard procedures. Thereby obtaining a full length human B7.1-mouse A5B7 Fd fusion. This fusion gene (SEQ ID NOS: 130-131) was transferred to EcoRI- on GS-system ^™ expression vector pEE6 (Celltech Biologics, SL1 4ene Slow Bath Road, UK) according to the protocol described in International Application WO 96/20011 Reference Example 5. Cloning into HindIII fragment yielded vector pAB7.1.

The BglII-SalI fragment containing the B7.1-A5B7 Fd expression cassette was cloned between the BglII and SalI sites of the vector pAF6 described above to obtain a vector (pAB7.2) capable of co-expressing the fusion protein and the A5B7 L chain. Vector pAB7.2 was then used to transform NS0 myeloma cells and select colonies with the ability to grow in the absence of glutamine. Cell lines expressing fusion proteins were identified by measuring CEA binding activity in culture supernatants using the ELISA described above. Cell lines expressing an appropriate amount of fusion protein (1D4) were selected to purify and characterize the AB7 fusion protein.

Purification and Characterization of AB7 Fusion Proteins

The secreted recombinant B7.1 (35-242) -A5B7F (ab ') ₂ , AB7 material, such as protein A Sepharose 4 fast flow (Pharmacia Biotech, HL1 3 UKH1 Street Albans Grosvenor Road 23) And purified from the culture supernatant using a Protein A agarose substrate. This substrate was washed using binding buffer (3M NaCl, 1.5M glycine, pH 8.9) 2 × 8 substrate volume. Culture supernatants containing AB7 were diluted 1: 1 in binding buffer. The washed substrate was added to the diluted culture supernatant (1 ml substrate fixed volume per 40 ml of diluted supernatant) and incubated for 2 hours at 4 ° C. under moderate shaking. The substrate was precipitated by centrifugation and ca. 75% of the supernatant was carefully separated. This substrate was resuspended in residual supernatant and the slurry was again charged to the column. This column was washed with 5-6 column volumes of 150 mM NaCl, 10 mM NaH ₂ PO ₄ (pH 7.4). The buffer was then changed to 100 mM sodium citrate (pH 2.8) and the elution fractions were collected. 2 M Tris buffer (pH 8.5) was added to this fraction and titrated to about pH 7.0. This eluted fraction was analyzed by non-reducible SDS-PAGE to retain the peak AB7 fraction (s) as product.

N-terminal sequencing

AB7 samples were analyzed by reducing SDS-PAGE and blotted onto polyvinylidene difluoride (PVDF) membranes (apparatus, gels, blotting membranes and methods were based in Boulevard 4202, Sorrento Valley, San Diego, 92121, USA NOVEX products). Protein bands were stained with Coomassie Blue and bands of about 70 kDa (ie, B7.1-Fd fusions) were N-terminal sequenced [Applied Biosystems, 494 Protein Sequencer (Perkin Elmer, ABI division, Kelvin close, Birchwood) Science Park North, Warrington, WA3 7PB). The analyzed sequence was consistent with the expected sequence of mature B7 (ie, after leader leader cleavage from amino acid 35 in SEQ ID NO: 131, Val Ile His Val et al.).

BIAcore analysis

AB7 was prepared using a BIAcore surface plasmon resonator made by BayerCore (A1 1 3 Hertz Street Albans Grosvenor Road 23, UK) [Greene JL, Leytze GM, Emswiler J, Peach R, Bajorath J, Cosand W, and Linsley PS. (1996) J. Biol. Chem. 271, 26762-26771, according to the BIAcore assay of the CD80 / CTLA-4 interaction described. Purified AB7 product samples were injected onto the CTLA4-Ig amine bound surface and the control amine bound surface. Binding could be clearly confirmed on the CTLA4-Ig surface compared to the control surface (see FIG. 3). In addition, when CTLA4-Ig was injected onto the amine-bound AB7 surface, binding between CTLA4-Ig and AB7 was confirmed.

The above-mentioned data together with the data of anti-CEA ELISA showed that the purified AB7 fusion protein had the biological properties of all the components, that is, the binding activity between the antigen and the receptor.

Costimulatory Activity of AB7 Fusion Proteins

The nature of the AB7 fusion protein, which can provide costimulatory signals to T cells when bound to CEA expressing tumor cells, was tested by adjusting the costimulatory assay described above (Jenkins et al. (1991) J. Immunol. 147). : 2461). LS174T colorectal tumor cells expressing CEA (fixed for 5 minutes at room temperature using 0.5% paraformaldehyde) were incubated with 10 μg / ml of AB7 fusion protein [0.5% human serum (Sigma AB, Dorsett, UK) RPMI 1640 medium (Sigma Chemical Company, Inc.) (Gibco, Life Technologies, Paisley, Scotland), stirred at 4 ° C. for 2 hours). Wash these cells twice before use and use fluorescein isothiocyanate (FITC) -bound goat antimouse Ig (Becton-Dickinson UK Ltd, Oxford) and flow cytometer (Facscan, Becton Dickinson) The binding of the fusion protein was confirmed. To use non-antigen human T cells in this assay, anti-T cell receptor antibodies (anti-CD3 antibody, OKT-3 Orthoclinical Diagnostics, Amersham, UK) previously coated on each well of a 96 well plate Stimulated T cell receptor (TCR). OKT-3 prepared purified antibody [2 μg / ml in bicarbonate coated buffer, pH 9.6 (preformed capsule, Sigma)] at 4 ° C. in a flat-bottomed 96-well microtiter plate (Costar Corporation, Cambridge, Mass., USA). Fixed by incubation overnight, then washed 3-4 times with PBS. 5% human purified peripheral T cells were negatively depleted from the donor human blood (ie, removing components other than T cells) using magnetic beads (Dynabeads, Dallas, Oslo, Norway). The wells were added at a concentration of 2 × 10 ⁵ / well in 50 μl of serum containing RPMI 1640 medium. LS174T cells fused to this well were added at a concentration of 5 × 10 ⁴ / well in 50 μl of a mixture of RPMI 1640 medium + 5% human serum. Finally the volume in all wells was charged to 200 μl using RPMI 1640 medium + 5% human serum. After 48 hours the cultures were pulsed with 1.25 μCi [ ³ H] thymidine (Amersham International) and harvested with a semi-automatic cell harvester (TomTec harvester, Wallak, UK) after 16 hours. The amount of [ ³ H] thymidine in the DNA was quantified using a liquid scintillation counter (Betaplate Scint and Betaplate counter, Wallak, UK). Typical costimulatory analysis data is shown in Table 5 below.

Concurrent stimulus data ΑCD3 coated on wells at 2 μg / ml (cpm) T cell alone 3582 T cell + αCD28 28178 T cell + LS174T 12303 T cell + LS174T + αCD28 25759 T cell + LS174T / fusion protein 41755 αCD3 = antiCD3 antibody; αCD28 = anti-CD28 antibody

T cells that do not receive first antigen stimulation require both T cell receptors and costimulatory signals. In this assay the T cell receptor signal is provided by the αCD3 antibody. Co-stimulation with αCD28 (Becton-Dickinson, 0.6 μg / ml) stimulated uptake of [ ³ H] thymidine by at least 8-fold compared to the use of αCD3 alone. The presence of tumor cells did not significantly affect this stimulus. [ ³ H] thymidine uptake when providing costimulatory signals through AB7 fusion proteins bound to tumor cells, at least three times higher than provided via tumor cell alone, or at least 11 times higher than observed without costimulation. Stimulated. The apparent stimulus provided by the tumor cell alone may be caused by residual accessory cells present in the purified T-cell population. A similar increase in T cell proliferation was consistently observed in wells containing tumor cell bound fusion proteins in each of the five analyzes performed compared to wells containing tumor cells unbound with T cells.

Reference Example 4

Preparation of IgG3-pBSIIKS +

This example relates to a method for producing a vector containing genes for human IgG3 heavy chain constant and hinge regions.

Genes containing the sequence set forth in SEQ ID NO: 115, the sequence of which is similar to SEQ ID NO: 25 but whose residues 312 and 501 in SEQ ID NO: 25 have been changed to C and G (residues 8-508), are described in Jayaraman et al. 1991) Proc. Natl. Acad. Sci USA 88, 4084-4088 prepared by PCR using a method similar to that described.

This gene was made up of two parts called IgG3A and IgG3B. This gene was cloned into the SacI and XmaI sites of pBluescript KS + (Stratazine Cloning Systems), respectively, to obtain vectors IgG3A-pBSIIKS + clone A7 and IgG3B-pBSIIKS + clone B17, respectively. IgG3A was prepared to stretch beyond the PmaCI restriction site (CACGTG at positions 334 to 339 in SEQ ID NO: 115). Similarly, IgG3B was prepared such that the 5 'end of this sequence was upstream of the PmaCI restriction site. To obtain the desired IgG3 gene sequence, intermediate IgG3A and IgG3B vectors were cleaved with AflIII and PmaCI. Vector fragments from IgG3A-pBSIIKS + clone A7 (2823 bp) and insert fragments from IgG3B-pBSIIKS + clone B17 (666 bp) were isolated and purified by electrophoresis on 1% agarose gel. This fragment was ligated and the E. coli strain DH5α was transformed using the ligation mixture. Clones containing the gene of interest were identified by digesting the DNA isolates with SacI and XmaI to obtain 520 bp fragments. The sequence of the insert was confirmed by DNA sequencing and clone F3 was called IgG3-pBSIIKS +.

Sequence list

(1) General Information:

(i) Applicant:

(A) Name: Geneca Limited

(B) Street: Stanhof Gate 15

(C) city: London

(E) Country: United Kingdom

(F) ZIP code: ZIP1 W 6 L

(G) Phone: 0171 304 5000

(H) Fax: 0171 304 5151

(I) Telex: 0171 834 2042

(ii) Name of invention: protein

(iii) Number of sequences: 131

(iv) form of computer reading:

(A) Medium type: floppy disk

(B) Computer: IBM PC compatible model

(C) operating system: PC-DOS / MS-DOS

(D) Software: PatentIn Release # 1.0, Version # 1.30 (EPO)

(2) Information about SEQ ID NO: 1

(i) Sequence Properties:

(A) Length: 32 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO:

(i) Sequence Properties:

(A) Length: 31 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO:

(i) Sequence Properties:

(A) Length: 34 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO:

(i) Sequence Properties:

(A) Length: 24 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO:

(i) Sequence Properties:

(A) Length: 24 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO:

(i) Sequence Properties:

(A) Length: 22 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO:

(i) Sequence Properties:

(A) Length: 41 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 8

(i) Sequence Properties:

(A) Length: 357 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO:

(i) Sequence Properties:

(A) Length: 108 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 10:

(i) Sequence Properties:

(A) Length: 360 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO:

(i) Sequence Properties:

(A) Length: 120 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 12:

(i) Sequence Properties:

(A) Length: 39 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO:

(i) Sequence Properties:

(A) Length: 30 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 14:

(i) Sequence Properties:

(A) Length: 36 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 15:

(i) Sequence Properties:

(A) Length: 33 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 16:

(i) Sequence Properties:

(A) Length: 705 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 17:

(i) Sequence Properties:

(A) Length: 235 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 18:

(i) Sequence Properties:

(A) Length: 765 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 19:

(i) Sequence Properties:

(A) Length: 255 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 20:

(i) Sequence Properties:

(A) Length: 121 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 21:

(i) Sequence Properties:

(A) Length: 369 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 22:

(i) Sequence Properties:

(A) Length: 116 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 23:

(i) Sequence Properties:

(A) Length: 348 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 24:

(i) Sequence Properties:

(A) Length: 167 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 25:

(i) Sequence Properties:

(A) Length: 501 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 26:

(i) Sequence Properties:

(A) Length: 10 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 27:

(i) Sequence Properties:

(A) Length: 7 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 28:

(i) Sequence Properties:

(A) Length: 9 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 29:

(i) Sequence Properties:

(A) Length: 5 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 30:

(i) Sequence Properties:

(A) Length: 9 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 31:

(i) Sequence Properties:

(A) Length: 17 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 32:

(i) Sequence Properties:

(A) Length: 11 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 33:

(i) Sequence Properties:

(A) Length: 13 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 34:

(i) Sequence Properties:

(A) Length: 60 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 35:

(i) Sequence Properties:

(A) Length: 60 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 36:

(i) Sequence Properties:

(A) Length: 29 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 37:

(i) Sequence Properties:

(A) Length: 29 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 38:

(i) Sequence Properties:

(A) Length: 39 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 39:

(i) Sequence Properties:

(A) Length: 39 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 40:

(i) Sequence Properties:

(A) Length: 19 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 41:

(i) Sequence Properties:

(A) Length: 20 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 42:

(i) Sequence Properties:

(A) Length: 77 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 43:

(i) Sequence Properties:

(A) Length: 71 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 44:

(i) Sequence Properties:

(A) Length: 61 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 45:

(i) Sequence Properties:

(A) Length: 61 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 46:

(i) Sequence Properties:

(A) Length: 357 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 47:

(i) Sequence Properties:

(A) Length: 381 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 48:

(i) Sequence Properties:

(A) Length: 342 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 49:

(i) Sequence Properties:

(A) Length: 321 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 50:

(i) Sequence Properties:

(A) Length: 107 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 51:

(i) Sequence Properties:

(A) Length: 705 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 52:

(i) Sequence Properties:

(A) Length: 235 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 53:

(i) Sequence Properties:

(A) Length: 385 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 54:

(i) Sequence Properties:

(A) Length: 360 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 55:

(i) Sequence Properties:

(A) Length: 120 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 56:

(i) Sequence Properties:

(A) Length: 765 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 57:

(i) Sequence Properties:

(A) Length: 255 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 58:

(i) Sequence Properties:

(A) Length: 40 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 59:

(i) Sequence Properties:

(A) Length: 46 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 60:

(i) Sequence Properties:

(A) Length: 321 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 61:

(i) Sequence Properties:

(A) Length: 107 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: peptide

(xi) sequence:

(2) Information about SEQ ID NO: 62:

(i) Sequence Properties:

(A) Length: 40 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 63:

(i) Sequence Properties:

(A) Length: 46 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 64:

(i) Sequence Properties:

(A) Length: 321 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 65:

(i) Sequence Properties:

(A) Length: 107 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 66:

(i) Sequence Properties:

(A) Length: 22 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 67:

(i) Sequence Properties:

(A) Length: 39 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 68:

(i) Sequence Properties:

(A) Length: 21 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 69:

(i) Sequence Properties:

(A) Length: 39 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 70:

(i) Sequence Properties:

(A) Length: 321 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 71:

(i) Sequence Properties:

(A) Length: 107 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 72:

(i) Sequence Properties:

(A) Length: 64 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 73:

(A) Length: 68 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 74:

(i) Sequence Properties:

(A) Length: 360 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 75:

(i) Sequence Properties:

(A) Length: 120 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 76:

(i) Sequence Properties:

(A) Length: 80 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 77:

(i) Sequence Properties:

(A) Length: 74 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 78:

(i) Sequence Properties:

(A) Length: 360 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 79:

(i) Sequence Properties:

(A) Length: 120 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 80:

(i) Sequence Properties:

(A) Length: 360 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 81:

(i) Sequence Properties:

(A) Length: 120 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 82:

(i) Sequence Properties:

(A) Length: 80 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 83:

(i) Sequence Properties:

(A) Length: 74 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 84:

(i) Sequence Properties:

(A) Length: 360 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 85:

(i) Sequence Properties:

(A) Length: 120 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 86:

(i) Sequence Properties:

(A) Length: 80 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 87:

(i) Sequence Properties:

(A) Length: 74 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 88:

(i) Sequence Properties:

(A) Length: 360 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 89:

(i) Sequence Properties:

(A) Length: 120 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 90:

(i) Sequence Properties:

(A) Length: 360 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 91:

(i) Sequence Properties:

(A) Length: 120 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 92:

(i) Sequence Properties:

(A) Length: 780 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 93:

(i) Sequence Properties:

(A) Length: 260 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 94:

(i) Sequence Properties:

(A) Length: 918 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 95:

(i) Sequence Properties:

(A) Length: 306 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 96:

(i) Sequence Properties:

(A) Length: 705 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 97:

(i) Sequence Properties:

(A) Length: 235 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 98:

(i) Sequence Properties:

(A) Length: 705 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 99:

(i) Sequence Properties:

(A) Length: 235 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 100:

(i) Sequence Properties:

(A) Length: 54 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 101:

(i) Sequence Properties:

(A) Length: 50 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 102:

(i) Sequence Properties:

(A) Length: 46 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 103:

(i) Sequence Properties:

(A) Length: 55 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 104:

(i) Sequence Properties:

(A) Length: 21 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 105:

(i) Sequence Properties:

(A) Length: 21 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 106:

(i) Sequence Properties:

(A) Length: 30 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 107:

(i) Sequence Properties:

(A) Length: 23 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 108:

(i) Sequence Properties:

(A) Length: 18 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 109:

(i) Sequence Properties:

(A) Length: 20 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 110:

(i) Sequence Properties:

(A) Length: 19 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 111:

(i) Sequence Properties:

(A) Length: 47 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 112:

(i) Sequence Properties:

(A) Length: 1870 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 113:

(i) Sequence Properties:

(A) Length: 613 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 114:

(i) Sequence Properties:

(A) Length: 96 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 115:

(i) Sequence Properties:

(A) Length: 520 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 116:

(i) Sequence Properties:

(A) Length: 31 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 117:

(i) Sequence Properties:

(A) Length: 23 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 118:

(i) Sequence Properties:

(A) Length: 88 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 119:

(i) Sequence Properties:

(A) Length: 30 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 120:

(i) Sequence Properties:

(A) Length: 20 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 121:

(i) Sequence Properties:

(A) Length: 30 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 122:

(i) Sequence Properties:

(A) Length: 22 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 123:

(i) Sequence Properties:

(A) Length: 90 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 124:

(i) Sequence Properties:

(A) Length: 2154 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 125:

(i) Sequence Properties:

(A) Length: 716 amino acids

(B) Sequence type: amino acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

(2) Information about SEQ ID NO: 126:

(i) Sequence Properties:

(A) Length: 42 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 127:

(i) Sequence Properties:

(A) Length: 45 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 128:

(i) Sequence Properties:

(A) Length: 45 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 129:

(i) Sequence Properties:

(A) Length: 40 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(xi) sequence:

(2) Information about SEQ ID NO: 130:

(i) Sequence Properties:

(A) Length: 1446 base pairs

(B) Sequence type: nucleic acid

(C) Number of chains: 1 chain

(D) Form: straight chain

(ii) Type of sequence: other nucleic acid

(ix) sequence characteristics

(A) Name / Key: CDS

(B) Presence location: 16..1435

(xi) sequence:

(2) Information about SEQ ID NO: 131:

(i) Sequence Properties:

(A) Length: 473 amino acids

(B) Sequence type: amino acid

(D) Form: straight chain

(ii) Type of sequence: protein

(xi) sequence:

Claims (15)

An anti-CEA antibody ("806.077 Ab") comprising a complementarity determining region (CDR) containing the following sequence: a) heavy chain CDR1: DNYMH (SEQ ID NO: 29) CDR2: WIDPENGDTE YAPKFRG (SEQ ID NO: 31) CDR3: LIYAGYLAMD Y (SEQ ID NO: 32) and b) light chain CDR1: SASSSVTYMH (SEQ ID NO: 26) CDR2: STSNLAS (SEQ ID NO: 27) CDR3: QQRSTYPLT (SEQ ID NO: 28). The antibody of claim 1, wherein the heavy chain CDR1 and CDR3 are as follows. CDR1: FNIKDNYMH (SEQ ID NO: 30), and CDR3: HVLIYAGYLA MDY (SEQ ID NO: 33). The antibody of claim 1, which contains the following structure, optionally humanized. Heavy chain variable region sequence (SEQ ID NO: 11) Light chain variable region sequence (SEQ ID NO: 9) The humanized antibody of claim 3, wherein the antibody comprises one or more of the following sequences and is optionally in the form of an F (ab ′) ₂ fragment. A heavy chain variable region sequence that is VH1 (SEQ ID NO: 55); Light chain variable region sequence that is VK4 (SEQ ID NO: 71); Human CH1 heavy chain IgG3 constant region; Human kappa light chain CL region; And Human IgG3 hinge region. A conjugate comprising the antibody according to any one of claims 1 to 4 and an effector moiety. 6. The conjugate of claim 5, wherein the effector addition is selected from the following molecules and is optionally in the form of a fusion protein. a) enzymes suitable for use in the ADEPT system; b) CPG2; c) [G251T, D253K] HCPB; d) [A248S, G251T, D253K] HCPB; e) costimulatory molecules; f) the extracellular domain of B7; g) the extracellular domain of human B7.1; And h) extracellular domain of human body B7.2. The method of claim 6, wherein the following fusions, namely a) the antibody Fd 'chain of the CH1 constant region derived from structure VH1 (SEQ ID NO: 55) / IgG3 / the hinge region derived from IgG3, wherein the C terminus is fused to the N terminus of [A248S, G251T, D253K] HCPB, and Humanized 806.077 F (ab ') ₂ -{[A248S, G251T, D253K] HCPB} ₂ fusions containing an antibody light chain having a CL region derived from the structure VK4 (SEQ ID NO: 71) / kappa light chain; b) [A248S, G251T, D253K, wherein the C terminus is fused via a (GGGS) ₃ linker to the N terminus of the antibody Fd 'chain having a CH1 constant region derived from structure VH1 (SEQ ID NO: 55) / IgG3 / a hinge region derived from IgG3 ] HCPB, and {[A248S, G251T, D253K] HCPB} ₂ -humanized 806.077F (ab ′) ₂ fusions containing an antibody light chain having a CL region derived from the structure VK4 (SEQ ID NO: 71) / kappa light chain; And c) a human B7.1 extracellular domain in which the C terminus is fused to the N terminus of an antibody Fd 'chain having a CH1 constant region derived from structure VH1 (SEQ ID NO: 55) / IgG3 / a hinge region derived from IgG3, and Characterized by being a fusion protein selected from the (human B7.1 extracellular domain) ₂ -humanized 806.077F (ab ') ₂ fusion containing the antibody light chain having a CL region derived from the structure VK4 (SEQ ID NO: 71) / kappa light chain. Conjugates (sequences are written from N terminus to C terminus). The polynucleotide sequence which can encode the antibody of any one of Claims 1-4, or the polypeptide of the conjugate of any one of Claims 5-7. The vector containing the polynucleotide of Claim 8. A mutant animal or a mutant plant other than a host cell transformed with the polynucleotide sequence according to claim 8 or a human grown from the host cell. Hybridoma 806.077 deposited with ECACC Accession No. 96022936. A pharmaceutical composition comprising a pharmaceutically acceptable diluent or carrier together with the conjugate of any one of claims 5 to 7 and optionally in a form suitable for intravenous administration. The conjugate of Claims 5-7 for use as a medicament. a) subject to treatment under conditions prone to expressing the host cell according to claim 10, the mutant mammal or mutant plant except humans, or the hybridoma according to claim 11, and optionally any one of claims 1 to 4. Secreting the antibody of claim 1 or the conjugate of any one of claims 5-7; And optionally b) at least partially purifying the antibody or conjugate Method for producing the above-described antibody or conjugate comprising a. A method of treating a human or animal in need thereof, comprising administering to the human or animal a pharmaceutically effective amount of the conjugate of any one of claims 5 to 7.