JPWO2012105699A1 - Production of antibodies with high complement-dependent biological activity - Google Patents

Abstract

The present invention provides a method for producing an antibody having a high complement-dependent biological activity when producing an antibody having a complement-dependent biological activity in various animal cells. A method for producing an antibody, characterized in that a host animal cell expressing a gene encoding a protein having β-1,4-galactosyltransferase (B4Gal-T) activity is used. And host cell lines that stably express such antibodies.

Description

  The present invention relates to a method for producing an antibody having a high complement-dependent biological activity. In particular, the present invention has β-1,4-galactosyltransferase (sometimes referred to as B4Gal-T or B1,4GT) activity when producing an antibody having high complement-dependent biological activity in animal cells. The present invention relates to a method for producing an antibody, an antibody produced by the method, and a host cell system that stably expresses such an antibody, characterized by using a host animal cell in which a gene encoding a protein is stably expressed.

  Because of their high specificity and high binding activity, antibodies have been widely used as research, various diagnostic agents, and as preventive and therapeutic agents for various diseases. In the United States, more than 20 antibody drugs have already been approved and used in the fields of cancer, autoimmune diseases, infectious diseases, allergic diseases, etc., and according to estimates by Data Monitor, the global market is 600 in 2015. It is expected to exceed $ 100 million. Furthermore, more than 140 types of human monoclonal antibodies are currently undergoing clinical trials (Non-patent Document 1).

  For example, in the field of cancer, in recent years, clinical application of cancer treatment with monoclonal antibody drugs has been actively progressed, and certain results have been achieved by the combined use with various chemotherapeutic agents (Non-patent Document 2, 3). These include anti-Her2 / neu antibodies (trastuzumab / Herceptin), anti-CD20 antibodies (rituximab / Rituxan), anti-EGFR antibodies (cetuximab / Erbitux), anti-VEGF antibodies (bevacizumab / Avastin), breast cancer, lymphoma, It is effective in some cancer patients such as colorectal cancer and lung cancer.

  In the field of autoimmune diseases, anti-CD20 antibodies (rituximab / Rituxan), anti-TNFα antibodies (remicade / Infliximab and humira / Adalimumab), and anti-IL-6R antibodies (actemura / Tocirizumab), etc. It is used clinically and forms a large market in areas such as systemic sclerosis, Crohn's disease, and systemic lupus erythematosus. Furthermore, in the field of allergic diseases, anti-IgE antibodies (xolair / Omalizumab) are used for bronchial asthma and allergic rhinitis.

  On the other hand, in the field of infectious diseases, anti-RSV antibodies (synagis / Palivizumab) are used as a preventive agent for RSV infection and form a large market. Furthermore, clinical development of antibody drugs against HIV, influenza virus, rabies virus, and hepatitis virus is now underway.

  However, the antibody drugs currently developed, including the antibody drugs listed above, have been improved compared to conventional drugs, but their drug efficacy is still limited and have reached a satisfactory level. It cannot be said that it is. That is, there is a strong demand for the development of a new antibody drug having superior biological activity or a method for enhancing the activity of an existing antibody drug.

  Under such circumstances, in recent years, search for antibodies with higher biological activity, research and development of methods for enhancing the activity of existing antibodies, research and development of methods for extending the half-life in blood, and the like have been made. Specifically, a method for enhancing antibody-dependent cellular cytotoxicity (hereinafter sometimes referred to as “ADCC activity”) by modifying the sugar chain of the antibody (Patent Documents 1, 2, 3, Non-Patent Documents 4, 5, and 6) and complement dependent cytotoxicity (hereinafter referred to as “CDC activity”) by the IgG1 / IgG3 recombination method of the antibody constant region (Fc region). A typical example is the enhancement method (Patent Document 4, Non-Patent Document 7). However, the present situation is that there is still a strong demand for antibodies having high complement-dependent biological activity or methods for their easy industrialization.

  By the way, the importance of the Fc region has been pointed out for the expression of the effector activity of the antibody, and ADCC activity and CDC activity via the Fc region are regarded as important. Regarding ADCC activity, as an approach for enhancing the activity, there have been reported cases in which a sugar chain bound to an Fc region of an antibody is modified and a peptide sequence in the Fc region is modified. In the former case, it has been reported that deletion of fucose bound to the reducing end of the N-glycoside-linked complex type sugar chain in the Fc region of the antibody results in high activity (Non-Patent Documents 4, 5, 8). In the latter case, it has been reported that high ADCC activity is also caused by amino acid substitution at a specific site in the peptide sequence of the Fc region of an antibody (Non-patent Document 10).

  On the other hand, the CDC activity is roughly classified into (i) a case where the peptide sequence of the Fc region or other region is modified, and (ii) a case where the sugar chain of the Fc region is modified. As an example of (i), the method for enhancing CDC activity by the recombination method of IgG1 / IgG3 described above (Patent Document 4, Non-Patent Document 7) is representative, but the problem inherent in IgG3 antibody (blood Short of half-life, difficulty of purification due to weak binding to protein A, and ease of aggregation, etc.), Fc region amino acid substitution (Patent Document 5), CDR ( Examples of amino acid substitution (complementarity determining region) (patent document 6) and further examples of amino acid substitution of hinge region (patent document 7) have been reported one after another. On the other hand, as an example from the viewpoint of (ii), an increase in the proportion of sugar chains in which galactose is bound to the non-reducing end of an N-glycoside-linked complex sugar chain affects the complement-dependent biological activity. However, there are reports that these increases enhance CDC activity (Non-patent Document 9) and reports that increases do not necessarily lead to enhancement of CDC activity (Non-patent Document 8). It is not clear that the change is less than twice. In addition, the sugar chain modification methods disclosed in Non-Patent Documents 8 and 9 are enzyme treatment methods in vitro and are not suitable for mass production.

Patent 4290423 Patent 4368530 JP2009-275049 WO2007 / 011041 WO2011 / 091078 WO2009 / 018411 WO2009 / 006520

Nature reviews, 9, p767-774 (2010) Lancet Oncol., 5, p.292-302 (2004) J. Clin. Oncology, 17, p268-276 (1999) J. Biol. Chem., 277, p26733-26740 (2002) BMC Biotechnology, 7:84, p1-13 (2007) Biotechnology and Bioengineering, 87, p614-622 (2004) Cancer Res., 68, p3863-3872 (2008) Biotechnol. Prog. 21, 1644-1652 (2005) Current Opinion in Immunology, 20, p471-478 (2008) Mol. Cancer Ther. 7, p2517-2527 (2008)

  In order to supply more effective and inexpensive antibody drugs to society in the background as described above, the present invention provides an antibody production method that facilitates mass production that enhances complement-dependent biological activity of antibodies produced in various animal cells. It is demanded. Furthermore, the shortage of manufacturing equipment is a major problem worldwide because antibody drugs are not only expensive, but also due to an increase in the required production volume and increase in types. In order to solve these problems, In addition, antibody drugs are required to have high biological activity.

  In order to achieve the above-mentioned problems, the complement-dependent biological activities of antibodies produced in various different host cells were investigated, and in particular, antibodies that were significantly different were bound to their Fc regions. The structure of the sugar chain was analyzed, the cause of the difference in its biological activity was investigated, and countermeasures were examined. As a result, in order to obtain an antibody having high complement-dependent biological activity, a cell in which a gene encoding a protein having β-1,4-galactosyltransferase (B4Gal-T) activity is stably expressed (preferably It is effective to use vertebrate cells, particularly mammalian cells), or animal cells into which a gene encoding a protein having B4Gal-T activity is introduced (preferably vertebrate cells, particularly mammalian cells) as host cells. As a result, the present invention was completed.

  That is, the present invention relates to an antibody production method described below, an antibody produced by the method, and a host cell system that stably expresses such an antibody.

[1] culturing a host animal cell expressing a gene encoding a protein having β-1,4-galactosyltransferase (B4Gal-T) activity and a gene encoding an antibody having complement-dependent biological activity; and Complement-dependent biological activity comprising the step of purifying the above-mentioned antibody in which the N-glycoside-linked complex sugar chain bound to the Fc region is modified by the expressed B4Gal-T activity from the cultured host animal cell. A method for producing an antibody comprising:
[2] The culturing step comprises:
(i) a gene encoding an antibody having complement-dependent biological activity in a host animal cell in which a gene encoding a protein having β-1,4-galactosyltransferase (B4Gal-T) activity is stably expressed. Culturing the cells introduced in an expressible manner, or
(ii) a host animal cell into which both a gene encoding a protein having β-1,4-galactosyltransferase (B4Gal-T) activity and a gene encoding an antibody having complement-dependent biological activity can be expressed. Culturing,
The method according to [1] above, comprising:
[3] The gene encoding the protein is
(i) the polynucleotide of SEQ ID NO: 7;
(ii) a polynucleotide that hybridizes with the polynucleotide of SEQ ID NO: 7 under a stringent condition and encodes a protein having B4Gal-T activity;
(iii) a polynucleotide encoding the amino acid sequence of SEQ ID NO: 8; or
(iv) The above-mentioned [1] or [2], which comprises an amino acid sequence having 80% or more identity with the amino acid sequence of SEQ ID NO: 8, and comprising a polynucleotide encoding a protein having B4Gal-T activity Method.
[4] The method according to any one of [1] to [3] above, wherein the host animal cell is a vertebrate cell, a mammalian cell, or a CHO cell.
[5] Furthermore,
(i) a step of isolating an antibody in which the proportion of G1 and G2 complex type sugar chains is 50% or more among N-glycoside-linked complex type sugar chains bound to the Fc region of the antibody; or
(ii) including the step of isolating an antibody having a G2 complex type sugar chain ratio of 50% or more among N-glycoside-linked complex type sugar chains bound to the Fc region of the antibody. ] The method as described in any one of [4].
[6] The complement-dependent biological activity is complement-dependent cytotoxic activity, complement-dependent virus neutralizing activity, complement-dependent CMV neutralizing activity, and complement-dependent anti-CD20 antibody. The method according to any one of [1] to [5] above, which is an activity selected from the group consisting of cytotoxic activity.
[7] Furthermore,
(i) isolating an antibody exhibiting complement-dependent biological activity that is at least 3 times higher than an antibody whose sugar chain is not modified; or
(ii) The method according to any one of the above [1] to [6], comprising a step of isolating an antibody exhibiting a complement-dependent biological activity that is 10 times or more higher than an antibody whose sugar chain is not modified. The method described.
[8] The method according to any one of [1] to [7], wherein the antibody is a monoclonal antibody.
[9] The method according to any one of [1] to [8] above, wherein the antibody is IgG or IgG1.
[10] An antibody produced by the method according to any one of [1] to [9] above.
[11] An antibody produced by the method according to [5] or [7] above.
[12] A host cell system that stably expresses the antibody according to [11] above.

  Since the antibody produced by the method of the present invention greatly enhances complement-dependent biological activity, it is expected to increase the therapeutic effect on various diseases not only for anti-tumor antibodies but also for antibodies against pathogenic microorganisms. The

  The method for producing an antibody with high complement-dependent biological activity according to the present invention is a particularly useful method when producing antibodies having various complement-dependent biological activities on an industrial scale. Further, in a preferred embodiment of the method of the present invention, the sugar chain bound to the Fc region of the produced antibody is human type, and it is advantageous in that there is no problem from the viewpoint of antigenicity when the antibody is used as a medicine. It is.

  Due to the above properties, the production method according to the present invention is a widely effective method for producing antitumor antibodies and antiviral antibodies having complement-dependent biological activity. In addition, a pharmaceutical composition containing a monoclonal antibody produced by the method of the present invention is effective in a small amount, and may increase efficacy, reduce side effects, or reduce medical costs.

The basic structure of an N-glycoside-linked sugar chain ((A)) and the typical structures ((B) to (E)) of N-glycoside-linked complex-type sugar chains found in IgG antibodies are shown. A table summarizing the results of sugar chain analysis of anti-CMV antibodies is shown. The (A) graph and (B) table showing the evaluation result of the neutralization activity of an anti-CMV antibody are shown. The evaluation result of the complement dependent cytotoxic activity (CDC) of Rituxan is shown.

  Hereinafter, a method for producing an antibody having high complement-dependent biological activity according to the present invention and an antibody produced by the method will be described in detail.

1. Explanation of terms (antibody)
In the present specification, “antibody” means a normal antibody (antibody), which is a glycoprotein molecule produced by B cells in lymphocytes, and recognizes and binds to a molecule (antigen) such as a specific protein. Means a molecule having The name “antibody” is a name that emphasizes the function of binding to an antigen, and the substance is called an immunoglobulin and is abbreviated as “Ig”. In particular, IgG used as an antibody drug is composed of four polypeptide chains, ie, two heavy (H) chains and two light (L) chains interconnected by disulfide bonds. Is an immunoglobulin molecule. The monoclonal antibody in the present invention is also composed of immunoglobulin molecules each containing two light chains (L chains) and heavy chains (H chains). Each heavy chain consists of a heavy chain variable region (sometimes referred to as “HCVR” or “VH”) and a heavy chain constant region (the heavy chain constant region is composed of three domains, “CH1”, “CH2”, It may be referred to as “CH3” (generic name: CH)). The H chain has a hinge region between CH1 and CH2. On the other hand, each L chain has an L chain variable region (sometimes referred to as “LCVR” or “VL”) and an L chain constant region (the L chain constant region is composed of one domain and may be referred to as “CL”). Is). In addition, when IgG is treated with papain, it is cleaved into two Fab fragments and one Fc fragment. When the Fc region is referred to, it mainly refers to the portion where the above “CH2” and “CH3” are combined. In the case of a human IgG antibody, one sugar chain is bound to the 297th Asn residue of each H chain by an N-glycoside bond.

  Antibodies are divided into several classes (isotypes) depending on the structure of the constant region. Mammals are classified into five classes of immunoglobulins, IgG, IgA, IgM, IgD, and IgE, depending on the structure of the constant region. In humans, IgG has four subclasses IgG1 to IgG4, and IgA has two subclasses IgA1 and IgA2, each having a slightly different structure. There are no subclasses in IgM, IgD, and IgE. In the present specification, the “antibody” means a generic name of the above-mentioned classes and subclasses, and further includes a mouse antibody, a chimeric antibody, a humanized antibody, a human antibody (fully human type), and a double antibody having an Fc region. Specific antibodies, multispecific antibodies and the like are included.

  The “effector function” of an antibody is a function carried by the Fc region of the antibody and largely depends on the antibody class. The function of activating complement is limited to IgM and IgG class antibodies, and the function of lysing cells to which antibody variable regions are bound is particularly called CDC (complement dependent cytotoxicity). In addition, Fc regions of IgG, IgE, and IgA class antibodies bind to specific Fc receptors, respectively, and activate cells having Fc receptors or act on intercellular transport of antibodies. In particular, it is known that IgG class antibodies activate these effector cells via Fc receptors on T cells, NK cells, neutrophils, and macrophages and kill target cells to which antibody variable regions are bound. Dependent cell damage). The function of activating NK cells may be used in distinction from ADCC and the function of activating macrophages from ADMC. (<URL: http://www.yodosha.co.jp/jikkenigaku/keyword/139.html> (Last browse date: January 27, 2012))

(protein)
In the present specification, “protein” means a molecule composed of a plurality of amino acid residues linked to each other by amide bonds (peptide bonds). An antibody is also a protein, and “recombinant antibody” means an antibody produced using genetic engineering techniques.

(Glycoprotein)
In the present specification, the “glycoprotein” means a normal glycoprotein, and a sugar chain is bound to a part of amino acids constituting the protein. In animals, it is said that most proteins secreted to the cell surface or extracellularly are glycoproteins. Two types of protein modifications are frequently observed: those bound to asparagine (N-linked type) and those linked to serine and threonine (O-linked type and mucin type).

  There are not so many kinds of sugars constituting the sugar chain bound to the glycoprotein, and common ones are glucose, galactose, mannose, fucose, N-acetylglucosamine, N-acetylgalactosamine, N-acetylneuraminic acid, There are about 7-8 types such as xylose. The structural style is also limited to some extent, and slight structural differences among them are identified, and various life phenomena are controlled. Antibodies are also representative glycoproteins, and research on their sugar chain structures and functions has been greatly advanced. In particular, regarding the structure and function of the N-glycoside-linked sugar chain present in the Fc region of IgG, there are many reports such as the effect on ADCC activity.

(Structure of N-glycoside-linked sugar chain)
In an N-glycoside-linked sugar chain (also referred to as N-linked sugar chain or N-linked sugar chain), N is attached to the nitrogen atom (N) of the Asn residue of the protein via N-acetylglucosamine (GlcNAc). -Sugar chains are linked by glycosidic bonds. Any Asn may be used, and the sequence is limited to Asn- (arbitrary amino acid) -Ser / Thr. In the case of a human IgG antibody, one sugar chain is bonded to the 297th Asn residue of each H chain by an N-glycoside bond, and there are many types of sugar chain structures.

  N-glycoside-linked sugar chains are roughly divided into “high mannose type”, “complex type” and “hybrid type”, but most of N-glycoside-linked sugar chains bound to the Fc region of an antibody are “complexed”. It has a sugar chain structure called “type”. FIG. 1 shows the basic structure of an N-glycoside-linked sugar chain ((A)) and representative structures ((B) to (E)) of N-glycoside-linked complex sugar chains found in IgG antibodies. . In the basic structure (A), two molecules of N-acetylglucosamine (GlcNAc) are bound to the reducing end side, one mannose (Man) molecule is ahead of it, and two mannose (Man) molecules are further ahead. It has a structure that is branched and connected. In the sugar chain of (B), fucose is bound to N-acetylglucosamine (GlcNAc) at the reducing end of the basic structure of the N-glycoside-linked sugar chain, and N-acetylated to two molecules of mannose (Man) at the non-reducing end. This is a structure in which one molecule of acetylglucosamine (GlcNAc) is bonded to each other, and galactose (Gal) is not present at the non-reducing end. The sugar chains of (C) to (E) have a structure in which one molecule or two molecules of Gal are bonded to GlcNAc at the non-reducing end of the sugar chain of (B). To add these Gals, B4Gal-T plays an important role. The IgG antibody produced in CHO cells has a high proportion of sugar chains in which fucose is bound to GlcNAc at the reducing end, but some of the complex sugar chains found in IgG antibodies do not have fucose bound . In addition to these, the complex type sugar chains found in IgG antibodies include one or two molecules of sialic acid (N-acetylneuraminic acid) in addition to Gal at the non-reducing end of the sugar chain structure (E). A sugar chain having a structure called “bisecting” in which GlcNAC is further bonded to a molecularly bonded sugar chain or a mannose where the sugar chain is branched is also included. In the present specification, in particular, a sugar chain structure in which one molecule of galactose is bonded to GlcNAc on the non-reducing end side is referred to as “G1 complex type sugar chain (or G1)” (typical example: (C) in FIG. 1 and (D)), and the sugar chain structure in which two molecules of galactose are bonded is the “G2 complex type sugar chain (or G2)” (typical example: (E) of FIG. 1) and no other galactose is bonded. The case is referred to as “G0 complex type sugar chain (or G0)”.

  In addition, in FIG. 1 in this specification, the structure of a typical N-glycoside-bonded complex type sugar chain found in an antibody derived from a mammal is shown, but N-acetylglucosamine ( In the case of a sugar chain in which fucose is not bound to (GlcNAc), it has already been shown that CDC activity is equivalent (Non-patent Document 6). In the case of “chain” or “G2 complex type sugar chain”, a sugar chain without fucose is also included. Although not shown in FIG. 1, the classification of G1 and G2 is the same as above even when the bisecting structure is provided. Further, as shown in FIG. 2, a sugar chain structure in which sialic acid is added to galactose at the non-reducing end of the sugar chain structure (E) is also included in “G2”.

  The structure of the Fc part of an antibody including these N-glycoside-linked sugar chains is considered to be an important part not only for affecting the antigenicity of the antibody but also for expressing an effector function.

As an antibody produced by the method for producing an antibody having complement-dependent biological activity of the present invention, for example,
(i) The proportion of G1 and G2 complex type sugar chains in the N-glycoside-linked complex type sugar chains bound to the Fc region of the antibody (at least of the two N-acetylglucosamines at the non-reducing ends of the sugar chains) On the other hand, the proportion of those having one molecule of galactose) is at least 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91% An antibody that is 92%, 93%, 94%, 95%, 96%, 97%, 98%, or more; and
(ii) The proportion of the G2 complex type sugar chain in the N-glycoside-bonded complex sugar chain bound to the Fc region of the antibody (galactose in both of the two N-acetylglucosamines at the non-reducing end of the sugar chain) At least 40%, 45%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85%, 90%, 91%, 92 %, 93%, 94%, 95%, 96%, 97%, 98%, or more.

(Biosynthetic pathway of N-glycoside-linked sugar chain)
The biosynthetic pathway of the N-glycoside-linked sugar chain bound to the antibody is considered to consist of three processes. That is, 1) a sugar chain precursor ((Glc) 3- (Man) 9- (GlcNAc) 2-PP-Dol) in which 14 sugars called “lipid intermediates or core oligosaccharides” are linked to dolichol phosphate 2) Biosynthetic process of 2) Transfer of sugar chain part ((Glc) 3- (Man) 9- (GlcNAc) 2) from lipid intermediate to protein asparagine residue by oligosaccharide transferase (oligosaccharide transfer) And 3) a process of modifying sugar chains on proteins with various glycosidases (sugar hydrolases) and glycosyltransferases, which proceed sequentially and continuously. Among these, the processes 1) and 2) are performed in an intracellular organelle called an endoplasmic reticulum, and the process 3) is performed in the endoplasmic reticulum and the Golgi apparatus (also one of the organelles).

  In particular, the modification in the process of 3) is species-specific and has a great difference among lower eukaryotes, plants, and animals because glycosyltransferases specific to each species act. The N-glycoside-linked sugar chain of animals is a high mannose type with 5-9 mannose residues added, GlcNAc, galactose (Gal), N-acetylgalactosamine (GalNAc), fucose (Fuc), sialic acid (Sia) It can be divided into composite types modified by the above, and mixed moldings having a structure like the intermediate. Among animals, complex sugar chains such as biantenna and bird antenna are not found except in vertebrates.

  In the process of 3) above, α-glucosidase I and II, α-mannosidase I and II, etc. act on the endoplasmic reticulum part first, then move to the Golgi part and α-mannosidase I acts, and the Man residue is 5 Glycoproteins reduced to individual are produced. Furthermore, N-acetylglucosaminyltransferase I (GnTI) that adds GlcNAc, α-mannosidase II that removes two Mans, and N-acetylglucosaminyltransferase II (GnTII) act on the Golgi body part to form a complex The basic structure of the type sugar chain ((GlcNAc) 2- (Man) 3- (GlcNAc) 2) is formed. Furthermore, in the Golgi portion, α-1,6-fucosyltransferase that adds fucose to N-acetylglucosamine at the reducing end, galactose transferase that adds galactose (Gal-T), and N-acetylneuron that adds sialic acid. It is known that sialyltransferases such as laminic acid exist, and N-glycoside-linked sugar chains are biosynthesized by the action of such enzymes (Biol. Pharm. Bull., 2009 ( vol.32) p767).

  In general, glycoprotein glycosylation can be modified by (i) application of inhibitors of glycosynthetic enzymes, (ii) selection of mutants, and (iii) genes associated with glycosynthetic biosynthesis. Introduction or deletion can be considered. Incidentally, examples of (i) include studies on inhibitors of various glycosidases involved in sugar chain processing (J. Appl. Glycosci., 2006 (vol. 53) p149). Examples of (ii) include a case of N-glycoside-linked sugar chain modification by a lectin resistant strain (Patent No. 4741011).

  In the present invention, in order to modify the structure of the sugar chain added to the antibody to be produced (increasing the ratio of G2 or G1 + G2 in the sugar chain composition of the antibody), the N-glycoside-linked sugar chain biosynthetic pathway It aims at stable expression of galactose transferase (Gal-T) related to sugar chain modification in “Process 3”. In the present specification, a method of introducing a gene encoding a protein having a human-derived β-1,4-galactosyltransferase (B4Gal-T) activity into a host cell derived from a mammal by a genetic engineering method is used. However, as long as it is a technique that can be expected to have the same effect as this purpose, a gene mutation technique, an enzyme reaction modification by an inhibitor / activator, and the like are also included in the present invention.

(Β-1,4-galactosyltransferase (B4Gal-T))
In the present specification, “β-1,4-galactosyltransferase (B4Gal-T)” means an enzyme that transfers Gal from UDP-Gal to a GlcNAc residue at the end of a sugar chain. This enzyme has a function of forming a structure of Galβ1 → 4GlcNAc. In the present specification, this enzyme activity is referred to as β-1,4-galactosyltransferase (B4Gal-T) activity. For this enzyme, many family genes have been cloned using amino acid sequence homology and sequence information of gene fragments registered in gene banks. In particular, for human-derived enzymes having B4Gal-T activity, there is a homologue of the amino acid sequence of human B4Gal-T1 (SEQ ID NO: 8) (Swiss-Prot Accession No .: P15291), which is identical to B4Gal-T1. B4Gal-T2 (Swiss-Prot Accession No .: O60909), T3 (GenBank Accession No .: CAH72145), T4 (GenBank Accession No .: AAQ89367), T5 (NCBI Accession No .: NP_004767), T6 (NCBI Accession) No .: NP_004766), and recently B4Gal-T7 (NCBI Accession No .: NM_007255) has also been reported. Their amino acid sequences are reported to be 55% to 25% identical compared to the amino acid sequence of human B4Gal-T1 (SEQ ID NO: 8) (Glycobiology, 1998 (vol. 8) p517; Curr. Drug Targets, 2008 (vol.9) p292).

  In addition, B4Gal-T1 homologues derived from various animals have also been reported. For example, B4Gal-T1 homologues (NCBI Accession No .: NP_445739 (rat), Swiss-, etc.) of various mammals such as rats, mice, cows, pigs, and pandas have been reported. Prot Accession No .: P15535 (mouse), Swiss-Prot Accession No .: P08037 (bovine), NCBI Accession No .: XP_003130728 (pig), NCBI Accession No .: XP_002926218 (panda)) is an amino acid of human B4Gal-T1 It has 80% or more identity with the sequence (SEQ ID NO: 8).

  In the present specification, the term “protein having β-1,4-galactosyltransferase (B4Gal-T) activity” is not limited to human-derived B4Gal-T1, but B4Gal-T2, 3, 4, 5, 6, etc. All enzymes having B4Gal-T1 activity derived from various animals are included. Such enzymes include at least 20%, 30%, 40%, 50%, 55%, 60%, 65%, 70%, 75%, 80%, 85% relative to the amino acid sequence of SEQ ID NO: 8. , 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99% of the amino acid sequence, which is substantially the same as the original protein Specifically, proteins having the same quality of activity (ie, B4Gal-T activity).

  In one embodiment of the invention, the protein having β-1,4-galactosyltransferase (B4Gal-T) activity is a protein having the amino acid sequence of SEQ ID NO: 8; or at least 80 relative to the amino acid sequence of SEQ ID NO: 8 % (Eg, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, or 99%) , A protein having B4Gal-T activity.

  “Substantially the same activity” indicates that these activities are qualitatively equivalent. Therefore, it is preferable that the enzyme activities are equivalent (for example, about 0.01 to 100 times, preferably about 0.5 to 20 times, more preferably about 0.5 to 2 times). And quantitative factors such as the molecular weight of the protein may be different. The measurement of B4Gal-T activity was performed according to Proc. Natl. Acad. Sci. USA 96: 4692-4697, 1999 and Glycobiology 12: 589-597, 2002, etc., or a method analogous thereto.

  The identity of the amino acid sequence and nucleotide sequence is determined by the algorithm BLAST (Proc. Natl. Acad. Sci. USA 87: 2264-2268, 1990; Proc. Natl. Acad. Sci USA 90: 5873, 1993 by Carlin and Arthur. ). Programs called BLASTN and BLASTX based on the BLAST algorithm have been developed (Altschul SF, et al: J Mol Biol 215: 403, 1990). When analyzing a nucleotide sequence using BLASTN, parameters are set to, for example, score = 100 and wordlength = 12. When analyzing an amino acid sequence using BLASTX, parameters are set to score = 50 and wordlength = 3, for example. When using BLAST and Gapped BLAST programs, the default parameters of each program are used.

(Galactose)
As used herein, “galactose” means normal galactose, and the chemical formula and molecular weight are C ₆ H ₁₂ O ₆ and 180.08, respectively, which are the same as glucose. The configuration is the 2nd position (second from the top in the Fischer projection formula) and the -OH at the 5th position is the same direction, the 3rd and 4th positions are the opposite direction, and D-galactose is the same as the 5th position D-glyceraldehyde Has an orientation. It is the 4-epimer of glucose. In nature, D-galactose is almost all.

(N-acetylglucosamine)
As used herein, “N-acetylglucosamine” means normal N-acetylglucosamine (N-acetyl-D-glucosamine, GlcNAc, NAG), and the chemical formula and molecular weight thereof are C ₈ H ₁₅ NO ₆ and 221.21. A monosaccharide derived from glucose and an important substance for several biochemical mechanisms. Chemically this material is an amide between glucosamine and acetic acid. N-acetylglucosamine is a component of glycosaminoglycan (mucopolysaccharide) such as glycoprotein and hyaluronic acid in mammals. N-acetylglucosamine forms the skeleton of an N-linked glycoprotein to which oligosaccharide chains centering on asparagine mannose bind (chitobiose structure) and is the main constituent sugar of sugar chains having a more complex structure.

(Fucose)
In this specification, “fucose” means normal fucose, which is 6-deoxy-galactose which is a kind of deoxy sugar, and has a chemical formula of C ₆ H ₁₂ O ₅ and a molecular weight of 164.16. It is classified into sugar and monosaccharide. Naturally, the L form is in the form of L-fucoside and exists widely in animals and plants. In mammals and plants, it is found on cell surface N-linked sugar chains.

(Polynucleotide)
As used herein, “polynucleotide” refers to a nucleoside phosphate ester (ATP, GTP, CTP, UTP; or dATP, dGTP, dCTP, dTTP) in which purine or pyrimidine is β-N-glycoside-linked to a sugar. It means a molecule that is bound more than one. A polynucleotide and another polynucleotide are “functionally linked” means that the function of each polynucleotide is not impaired and a state in which a desired function can be exerted by the linkage is ensured. Means. Specifically, it refers to a state in which the 3 ′ terminal nucleotide of one polynucleotide and the 5 ′ terminal nucleotide of the other polynucleotide are linked directly or via another linker sequence.

(Genes encoding proteins)
In the present specification, the “gene encoding a protein” is a polynucleotide containing a protein-coding region (open reading frame: ORF), for example, a protein gene cDNA. Therefore, the “gene encoding the H chain of an antibody” or the “gene encoding the L chain of an antibody” is a polynucleotide containing a region (open reading frame: ORF) encoding the H chain or L chain of an antibody. For example, cDNA of genes encoding these.

  The gene encoding the B4Gal-T protein used in the present invention includes a gene having the nucleotide sequence of SEQ ID NO: 7 or a gene having a nucleotide sequence that hybridizes to the nucleotide of SEQ ID NO: 7 under stringent hybridization conditions. Is included.

(Hybridization conditions)
Hybridization can be performed according to a known method or a method analogous thereto, for example, the method described in Molecular Cloning Third Edition, J. Sambrook et al., Cold Spring Harbor Lab. Press. 2001). Moreover, when using a commercially available library, it can carry out according to the method as described in an attached instruction manual. Here, the “stringent conditions” may be any of low stringent conditions, medium stringent conditions, and high stringent conditions. “Low stringent conditions” are, for example, conditions of 5 × SSC, 5 × Denhardt's solution, 0.5% SDS, 50% formamide, and 32 ° C. The “medium stringent conditions” are, for example, conditions of 5 × SSC, 5 × Denhardt's solution, 0.5% SDS, 50% formamide, and 42 ° C. “High stringent conditions” are, for example, conditions of 5 × SSC, 5 × Denhardt's solution, 0.5% SDS, 50% formamide, 50 ° C. Under these conditions, it can be expected that DNA having higher homology can be efficiently obtained as the temperature is increased. However, factors affecting the stringency of hybridization include multiple factors such as temperature, probe concentration, probe length, ionic strength, time, and salt concentration. Those skilled in the art will select these factors as appropriate. It is possible to achieve similar stringency.

  As a polynucleotide capable of hybridizing to the gene encoding the B4Gal-T1 protein used in the present invention, SEQ ID NO: 7 of SEQ ID NO: 7 was calculated using homology search software such as FASTA and BLAST using default parameters. Base sequence and at least 50%, 60%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, Alternatively, a polynucleotide having 99% identity can be mentioned.

(RT-PCR)
RT-PCR is a reverse transcription polymerase chain reaction (Reverse Transcription Polymerase Chain Reaction), in which reverse transcription is performed using mRNA as a template, and PCR is performed on the generated cDNA. In the PCR method, a primer is attached to DNA as a template, and DNA contained in the target primer sequence is specifically detected by DNA polymerase. PCR can be used to detect DNA, but it cannot detect RNA. Therefore, RNA is converted into cDNA by reverse transcription, and PCR is performed on the cDNA. This technique is used for the purpose of, for example, detection of RNA viruses such as retroviruses, detection of gene expression of specific genes (transcription to mRNA), and the like.

(Complement and complement dependent biological activity)
Complement is a group of blood proteins that mediate the immune response contained in animal blood. Complement components are represented by C1 to C9, and C1 further includes C1q, C1r, and C1s. There are two subtypes such as C5a and C5b. In addition to C1-C9 complement proteins, 16 proteins including factor B and factor D, 5 humoral regulators, and 4 regulators on the cell membrane also function complement. Are collectively referred to as the complement system. The complement system is a biochemical cascade that assists the organism in eliminating pathogens. There are three biochemical processes for activation of the complement system: the classical pathway, the alternative pathway, and the mannose-binding lectin pathway. In particular, the classical pathway is a pathway that begins with C1 activation, and C1 is activated by binding of complement C1q to the antibody-antigen complex. After that, it is basically activated in numerical order, but C4 comes exceptionally second. It is activated from “C1 → C4 → C2 → C3b → C5b”, and then C6 to C9 are coupled to C5b one after another, and finally C5b6789 is reached. C5b6789 is said to be a “cell membrane disorder complex” and attaches to the surface of pathogenic microorganisms and destroys the cell membrane. The complement-dependent cytotoxicity of anti-tumor antibodies such as rituximab also activates the similar complement system, resulting in pores in the cell membrane due to the action of the complement's “cytoplasmic complex”. It is thought to destroy cells.

  In the present specification, “complement-dependent biological activity” refers to complement-dependent cytotoxicity, complement-dependent virus neutralization activity, etc., in which the antibody-antigen complex activates the complement system as described above. And the resulting complement system component refers to the biological activity observed by damaging cancer cells, pathogenic microorganisms, or cells infected by them.

(CDC activity)
In this specification, “complement-dependent cytotoxicity (CDC activity)” is one of complement-dependent biological activities and is a cytotoxic activity caused by the complement system activated by an antibody-antigen complex. Means. The “cytotoxic activity” may be to slow the growth of individual cells or to induce cell death. Those skilled in the art can evaluate the activity by a usual method for measuring the number of living cells.

(Virus neutralization activity)
In the present specification, the “virus neutralizing activity” of an antibody is an activity that inhibits the ability of the antibody to infect and proliferate the host cell of the virus. This is called body-dependent virus neutralizing activity. This activity can be measured by an ordinary evaluation method for virus neutralization activity. In the evaluation of the neutralizing activity of the virus, the strength of the activity is indicated by the antibody concentration (IC50) that inhibits the infectivity of the virus by 50%. The antiviral antibody preferable in the antibody pharmaceutical field usually has an IC50 of 10 μg / ml or less, preferably 5 μg / ml or less, more preferably 1 μg / ml or less.

(Antibodies with complement-dependent biological activity)
In the present specification, the “antibody having complement-dependent biological activity” is an antibody having complement-dependent cytotoxicity, complement-dependent virus neutralizing activity, etc., which are evaluated in vitro. It refers to a group of antibodies having neutralizing activity / cytotoxic activity in a complement-dependent manner. In various literatures, not only complement-dependent cytotoxicity but also the expression complement-mediated cytotoxicity may be described. These antibodies having complement-dependent biological activity include antibodies having many antitumor activities, antibodies having antiallergic / antiinflammatory activity, antibodies having antiinfection / antiviral activity, and the like.

  For example, examples of antibodies having antitumor activity include anti-CD20 antibody (Non-patent Document 10), anti-GD2 antibody (Cancer Res., 2005 (vol.65) p10562), anti-GD3 antibody (Cancer Immunity, 2002 (vol. .2) p13), anti-HER2 antibody (British J. Cancer, 2004 (vol.91) p1195), anti-CD52 antibody (Clin. Cancer Res., 2008 (vol.14) p569), anti-EGFR (Cancer Res., 1986 (vol. 46) p5592, anti-HLA-DR antibody (Mol. Cancer, 2011 (vol. 10); 42), etc. In addition, as an antibody having anti-allergic / anti-inflammatory activity, anti-TNF- α antibody (Arthritis & Rheumatism, 2008 (vol.58) p1248).

  Furthermore, anti-VZV antibodies (J. Gen. Virol., 1991 (vol. 72) p2065), anti-CMV antibodies (WO 2007/084423, Virology, 1993 (vol. 197) p143), anti-gp120 antibody (J. Virol., 1996 (vol. 70) p1100), anti-GM2 antibody (J. Immunol., 1999 (vol. 162) p533) and the like.

  The “complement-dependent biological activity” of an antibody can be measured, for example, by a method as shown in the examples described later.

  Examples of the antibody produced by the method for producing an antibody having complement-dependent biological activity of the present invention in which an N-glycoside-linked complex type sugar chain is modified by B4Gal-T activity include, for example, N-glycoside-linked complex type Compared to an antibody whose sugar chain is not modified by B4Gal-T activity, at least 1.25 times, 1.5 times, 1.75 times, 2 times, 3 times, 4 times, 5 times, 6 times, 7 times Antibodies that exhibit complement-dependent biological activity that are fold, 8 fold, 9 fold, 10 fold, 20 fold, or more are included.

(Cells that stably express)
In the present specification, “expression” of a gene or protein means that genetic information of a specific protein carried by the gene is transcribed into mRNA and translated as a protein. Further, “stable expression” of a certain gene or protein is used in the sense of contrasting to expressing a certain gene or protein only transiently. Therefore, a “cell stably expressing a gene or protein” means a cell that stably (continuously) express a gene or protein rather than transiently. Such cells include cells that have been artificially introduced with a gene that stably expresses the gene, and that have stably expressed the original (or endogenous) gene. Cells are included.

(Host cell)
Examples of the host cell of the present invention include yeast, plant cells, which are introduced using a genetic engineering technique, an enzyme or a group of enzymes involved in modification of an N-glycoside-linked sugar chain that binds to the Fc region of an antibody molecule, Cells such as insect cells and vertebrate cells can also be used as host cells. However, in lower eukaryotes such as yeasts, molds, plants, and insects, sugar chains that modify glycoproteins are so far away from mammals that cells derived from vertebrates (in this specification, simply “Animal cells” are sometimes preferred). Furthermore, in view of the similarity in the sugar chain structure of immunoglobulins derived from various animals, cells derived from mammals are even more preferable. In particular, among mammalian cells, human cells such as Namalwa cells, COS cells that are monkey cells, CHO cells that are cells derived from Chinese hamster ovary tissue, rat myeloma cells (eg, YB2 / 3HL cells) Mouse myeloma cells (for example, NSO cells, SP2 / 0 cells), Syrian hamster kidney-derived cells (for example, BHK cells), embryonic stem cells, fertilized egg cells, and the like are more preferable.

  In the following, with regard to the operation methods of the main parts of the present invention (acquisition of genes, production of stable expression strains, cell culture, protein purification, etc.), those limited examples will be outlined, but the methods described below are not necessarily limited. However, any method may be used as long as it can be used by those skilled in the art at the current scientific and technical level in order to realize the intention of the present invention.

2. Obtaining antibody genes Antibody genes are obtained using cells that produce antibodies. Examples of cells that can be used include cells that produce antibodies such as human B lymphocytes and mouse spleen lymphocytes, and hybridomas of such cells and myeloma cells can also be used as materials.

  For example, B lymphocytes are separated from human blood, and proliferation of the B lymphocytes is induced. Methods for inducing proliferation are known per se, such as the use of cytokines and stimulants necessary for induction, and “Epstein-Barr virus” (hereinafter referred to as “Epstein-Barr virus”), which is an inducing factor for cancer It can be carried out by a transformation method (referred to as EBV) (D. Kozbor et al.). That is, the above-mentioned B lymphocytes are infected with EBV to induce proliferation, and the proliferated cells are used as an antibody-producing cell library.

  The method of recovering the monoclonal antibody from the proliferation-induced cells can be performed by a well-known method commonly used in the production of monoclonal antibodies. A lymphocyte producing an antibody that binds to a specific antigen is selected from the antibody-producing cell library. Selection methods include cloning by limiting dilution, collection of single cells by cell microarray, and staining of cells producing the desired antibody by FACS sorting and collection by sorting. It is done.

  For detection of a fraction that binds to a specific antigen, ELISA or immunostaining using a specific antigen, immunostaining using cells that express a specific antigen, or an antibody expressed by binding to a specific antigen A method for measuring the activity (neutralizing activity) of the selenium can be employed.

  Total-RNA is extracted from the cell population producing the target antibody obtained above, and cDNA is synthesized by reverse transcription reaction using Oligo-dT primer. The antibody gene is amplified by PCR using this cDNA as a template. Primers used for PCR are designed based on a database of genetic information encoding antibody H and L chains. The full length of the antibody gene may be amplified, or only the variable region may be amplified and functionally linked to a known constant region gene.

  The above is an example of cloning a new human antibody. Regarding a monoclonal antibody whose antibody gene sequence is known, those skilled in the art can, for example, describe the gene information registered in GenBank and published patent information. It is possible to build based on the above.

3. Determination of antibody amino acid sequence based on nucleotide sequence
The antibody H chain and L chain cDNAs amplified by the PCR method are inserted into a plasmid vector, and each base sequence is confirmed by an ABI sequencer. The amino acid sequences of the antibody H chain and L chain are determined from the obtained base sequence.

4). Acquisition of β1,4-galactosyltransferase (B4Gal-T) gene
The B4Gal-T gene is amplified by PCR from cDNA of animal cells expressing B4Gal-T. A primer suitable for amplifying a gene encoding the enzyme is prepared by the enzyme to be selected, and amplified by PCR. Insert the B4Gal-T gene into an appropriate plasmid vector, and confirm the nucleotide sequence using a sequencer.

5. Method for preparing antibody stable expression strain The obtained H chain and L chain genes are inserted into a vector constructed for the preparation of a double gene vector. The double gene vector is introduced into an appropriate cell, and a cell clone having been successfully introduced into the vector is selected using a selection marker on the double gene vector.

  For example, after inserting the H chain and L chain genes into a vector constructed for the production of a double gene vector, the gene is introduced into a cell suitable for antibody production (eg, CHO-K1 or NS0), and a selection marker on the double gene vector. In (eg, GS selection), a cell clone that has successfully introduced the vector is selected.

  The selected cells are confirmed to produce the desired antibody. For confirmation of antibody production, a method for detecting a fraction that binds to the specific antigen shown above or an antibody quantitative ELISA can be used.

6). Method for preparing B4Gal-T stable expression strain The obtained B4Gal-T gene is inserted into an appropriate expression vector. The B4Gal-T expression vector is introduced into an appropriate cell, and a cell clone that has been successfully introduced into the vector is selected using a selection marker on the vector.
For example, the B4Gal-T gene is inserted into a vector suitable for gene expression (pcDNA3.1, Invitrogen) and introduced into a cell suitable for antibody production (for example, CHO-K1 or NS0) in the same manner as described above. Cell clones that have successfully introduced the vector are selected using the above selection marker (Geneticin).
Check whether the selected cells express the B4Gal-T gene.

7). Confirmation of β1,4-galactosyltransferase (B4Gal-T) gene expression
RT-PCR is used as a method for confirming the expression of the B4Gal-T gene. Specifically, a B4Gal-T expression vector is introduced, total-RNA is isolated from the selected cells, converted to cDNA by reverse transcriptase, and then amplified by PCR to detect the B4Gal-T gene.

  For example, Cells-to-cDNA II Kit (Ambion) is used for cDNA synthesis, and the B4Gal-T1 gene is amplified by PCR using the synthesized cDNA as a template. At the same time, the GAPDH gene is amplified by PCR. Since the expression level of the GAPDH gene is constant, it is used as a control for comparing the expression level in RT-PCR.

The primers used for amplification are shown below (Table 1). For amplification of B4Gal-T1 gene, use two types of Primers (SEQ ID NO: 1 / SEQ ID NO: 2) used for cloning to amplify the full length, or use Primer (SEQ ID NO: 3 / SEQ ID NO: 4) A part of the B4Gal-T1 gene is amplified. For amplification of GAPDH, Primer (SEQ ID NO: 5 / SEQ ID NO: 6) is used to amplify a part of the GAPDH gene. Primer (SEQ ID NO: 5 / SEQ ID NO: 6) is for human GAPDH amplification, but can be applied to amplification of CHO-derived GAPDH.

8). Method for preparing antibody / B4Gal-T co-stable expression strain Antibody / B4Gal-T co-stable expression strain is divided into two steps: the method for producing antibody stable expression strain and the method for producing B4Gal-T stable expression strain. That is, the B4Gal-T gene is introduced into the produced antibody stable expression strain and selected, or the antibody gene is introduced into the produced B4Gal-T stable expression strain and selected. At this time, use different selection markers for the two-stage selection. However, when all of the antibody H chain / L chain gene and the B4Gal-T gene are inserted into one expression vector, the antibody / B4Gal is selected with a one-step 1 selection marker in the same manner as in the above-described method for producing an antibody stable expression strain. -T Simultaneously stable expression strains can be selected.

9. Antibody production and purification method As a step of producing and purifying an antibody using the antibody stable expression strain obtained above, a uniform antibody can be obtained by using a general antibody production and purification method. it can. For example, the antibody stable expression strain or the antibody / B4Gal-T simultaneous stable expression strain is cultured in a serum-free medium, and the culture supernatant containing the antibody is collected and used as a purification material. The antibody is purified by affinity purification using a Protein A or Protein G column. The purification conditions are those recommended by the column manufacturer. For antibody purification on a small scale, use Protein A or Protein G magnetic beads. The purification conditions are those recommended by the magnetic bead manufacturer.

10. Sugar chain analysis method The analysis of sugar chains in the present invention includes identification and quantification of sugar chains. The sugar chain identification is to specify one or a plurality of sugar chain components contained in the sample to be analyzed. As a method for identifying the sugar chain of the glycoprotein, the sugar chain is separated from the glycoprotein, mass analysis of the obtained sugar chain is performed, and then the sugar chain analysis is performed based on the obtained detection result. A method for identification is given.

  In addition, quantification of sugar chains includes, for example, determining the amount ratio (molar ratio) of two or more sugar chains contained in a sample to be analyzed. Specifically, it refers to determining the ratio of two or more sugar chains that bind to IgG or its Fc.

  The detection of a sugar chain in the present invention means, for example, quantification of IgG or a sugar chain binding to the Fc thereof by calculating a ratio of peak sizes of glycopeptides derived from the Fc part detected by mass spectrometry. Can do. The size of the peak may be measured using, for example, the area of the mass range having the peak or the area of the entire peak, or may be measured using the height of the peak as an index.

  The method of mass spectrometry of glycans used in the present invention is well known in the art (Niwa, latest mass spectrometry, Chemical Dojin (1995)), electrospray (ESI) method, matrix-assisted laser desorption ionization ( Any ionization technique can be used, including the MALDI method. For mass spectrometry, any mass separation method such as time-of-flight (TOF), quadrupole, or magnetic field can be used, but a method using TOF is preferably used.

  In the present invention, MALDI-TOF-MS is an abbreviation for Matrix Assisted Laser Deposition Ionization Time-of-Flight Mass Spectrometer. MALDI is a mass spectrometry technique found by Tanaka et al. And developed by Hillenkamp et al. In this method, after mixing a sample and a matrix solution, the mixed solution is dried on a plate to obtain a crystalline state. Large energy is given to the matrix by pulse laser irradiation, and sample-derived ions such as (M + H) + and (M + Na) + are desorbed from the matrix-derived ions.

  MALDI-TOF-MS uses MALDI to measure mass based on time of flight, and may use Shimadzu's MALDI-TOFMS AXIMA series models or a mass spectrometer such as Bruker's AUTOFLEX. it can.

11. Method for evaluating activity of antibody The activity measurement method is determined by what kind of antigen the antibody binds to and what kind of biological activity it has. For example, as a method for evaluating the activity of an antibody against a virus, a virus neutralizing activity evaluation method by “immunostaining method” and “plaque method” is a typical method. In the case of an antibody that recognizes an antigen present on the cell membrane, such as an anti-tumor antibody, the activity is evaluated using cells that specifically express the antigen. These are described below, but only one case. When determining complement-dependent biological activity, the complement is added to the following system at the same time as the antibody or after the addition of the antibody and evaluated.

(1) Immunostaining method As one of the antiviral antibody neutralizing activity evaluation methods, an immunostaining method is known. The method is summarized as follows. Purified antibody and complement are added to the virus solution for 1 hour at an appropriate temperature, and then inoculated into host cells. Incubate for an appropriate time after inoculation and then wash the cells twice. Subsequently, after culturing for an appropriate time, immunostaining is performed with an antibody against a specific protein expressed in virus-infected cells, and the stained cells are counted as the number of infected cells. The infection inhibition rate is evaluated by reducing the number of infected cells.

(2) Plaque method The plaque method is known as one of the methods for evaluating the neutralizing activity of a virus. In a classical way, it is used that viruses grown on cells-coated plates form plaques. The method is summarized as follows. Purified antibody and complement are added to the virus solution for 1 hour at an appropriate temperature, and then inoculated into host cells. After incubation for an appropriate time, the cells are washed twice. Next, the cells are overlaid with agar or a medium containing methylcellulose and cultured.

  After culturing until clear plaque formation due to the death of virus-infected cells is confirmed, the cells are fixed and stained with an appropriate staining reagent. After staining, the number of plaques is counted, and the inhibition rate of infection is evaluated by the reduction of plaques.

(3) CDC activity evaluation method of antibody The CDC activity evaluation method of an antibody is performed on an antibody that recognizes an antigen present on a cell membrane, and the activity is evaluated using cells that specifically express the antigen. The target cells are washed twice with a medium, and then an appropriate number is seeded on a plate. After adding the antibody to the cells, place it in an incubator at an appropriate temperature for 30 minutes to 1 hour. Thereafter, complement is added and placed in the incubator for 2-3 hours. Thereafter, the number of living cells is evaluated by MTT assay or WST-1 assay, and the number of cells killed by the complement-dependent cytotoxic activity of the antibody is evaluated.

  Hereinafter, the present invention will be described in more detail by way of examples using two kinds of antibodies against completely different antigens, but the present invention is not limited to these examples. Unless otherwise stated, procedures used in this example can be referred to in Molecular Cloning: A Laboratory Manual (Third Edition) (Sambrook et al., Cold Spring Harbor Laboratory Press, 2001).

(Effect of modification of constant region N glycoside-linked sugar chain in anti-CMV antibody)
Materials and methods
1． Antibody gene cloning and production of stable antibody expression strain Total-RNA was extracted from anti-CMV antibody-producing LCL (lymphoblastoid Cell Line) (TRIzol reagent, Invitrogen), and cDNA was synthesized by reverse transcription (SuperScript II, Invitrogen). Antibody genes (H chain and L chain) amplified by PCR were TA cloned (pGEM-T easy, Promega), and the antibody gene sequence was determined (see the sequence of “G3D No. 9” in WO2007 / 094423). Thereafter, the antibody genes (H chain and L chain) were inserted into a vector constructed for preparing a double gene vector. The double gene vector was introduced into CHO-K1 cells, selected with Glutamine Synthetase (GS) selective medium (SAFC), and an antibody stable expression strain was cloned. For selection of CHO-K1 cells, GS inhibitor Methionine Sulfoximine (MSX, SIGMA) (final 2.5 μM) was added.

2． Sugar chain modification of CHO-K1 cell-produced anti-CMV antibody The β1,4-galactosyltransferase (B4Gal-T1) gene was amplified by PCR from cDNA of anti-CMV antibody-produced LCL. The primer sequences used are as follows.
B1,4GT-Fw: 5'-CACCCTTCTTAAAGCGGCGG-3 '(SEQ ID NO: 1)
B1,4GT-Rv: 5'-GTACCAAAACGCTAGCTCGG-3 '(SEQ ID NO: 2)

  The PCR product is cloned into pcDNA3.1D / V5-His-TOPO vector (Invitrogen), the sequence is analyzed (SEQ ID NOs: 7 and 8), and the gene encoding B4Gal-T1 protein (Swiss-Prot Accession No .: P15291) (NCBI Accession No .: NM_001497). The B4Gal-T1 expression vector is introduced into an anti-CMV antibody stable expression clone (CHO-K1 cell) and selected with Geneticin (Gibco) (final 800 μg / ml) in MSX-added GS selection medium, and anti-CMV antibody and B4Gal- A clone was obtained that expressed T1 simultaneously and stably. The expression of B4Gal-T1 was confirmed by RT-PCR. Among them, one clone with good expression of B4Gal-T1 was selected, and the produced antibody was purified. The antibody produced by the anti-CMV antibody stable expression clone before introduction of the B4Gal-T1 expression vector was purified in the same manner.

3． Antibody purification Antibody purification was carried out using an affinity column (HiTrap rProtein A FF, GE Healthcare) by the recommended method. The antibody after column elution was dialyzed against PBS (-) and the concentration was determined by the UV method. The purified antibody was subjected to SDS-PAGE to confirm the antibody heavy chain (about 50 kDa) and light chain (about 25 kDa).

Four. Sugar chain analysis The analysis of antibody sugar chains was outsourced to Glience Corporation. In the analysis, the antibody was pretreated (desalted and dried under reduced pressure), the protein was denatured, and glycosylation from the protein by glycopeptidase A and proteolysis by pepsin were simultaneously performed. Furthermore, after proteolysis with pronase, sugar chains and amino acids were separated by a P2 gel filtration column. The sugar chain fraction was confirmed by the orcinol-sulfuric acid method, recovered and dried under reduced pressure. Since the separated sugar chain was pyridylaminated (PA-ized), the sugar chain and 2-aminopyridine were combined, and after the reduction, the PA sugar chain and excess 2-aminopyridine were separated using a Sephadex-G15 gel filtration column.

  As the primary analysis, PA sugar chains were separated from neutral sugars, monosialylated sugars, and disialylated sugars by DEAE anion exchange chromatography. Furthermore, each fraction was dried under reduced pressure, and then sugar chains were separated by ODS chromatography, and each peak was subjected to MALDI-TOF-MS to confirm whether or not it was a sugar chain from the molecular weight.

  As a secondary analysis, each peak determined to be a sugar chain was separated by normal phase Amide chromatography. Glc. Determined by ODS column and Amide column. Candidate sugar chains were narrowed down by the sugar chain analysis software GALAXY from the Unit (GU) value and molecular weight. Finally, ODS chromatography was used to strike each peak with the standard product, and the sugar chain structure was identified.

Five. Preparation of CMV
HEL cells were cultured in 75 ml flasks using Eagle MEM supplemented with 10% FCS. After removing the medium, 1 ml of a virus solution (CMV AD169 strain) diluted 5 to 10 times was added, and the virus was adsorbed in an incubator for 1 hour. After removing the virus solution, 15 ml of 5% FCS-added Eagle MEM was added, and cultured while changing the medium every 2-3 days until a cytopathic effect (CPE) was observed. When CPE was observed in 90% or more of the cells, the medium was changed, and the cells were further cultured for 1 day. Infected cells were removed from the flask with sterile glass beads after removing half of the medium, and transferred to a plastic tube. Thereafter, the cells were disrupted by sonication for 30 seconds while cooling on ice, and the supernatant (virus solution) was recovered after centrifugation at 3000 rpm for 15 minutes. Sorbitol was added to the collected virus solution to a final concentration of 35%, and after dispensing, it was stored at -80 ° C.

6． Neutralization activity evaluation
Plaque assay : CMV (AD169 strain) was diluted with 10% FCS-added Eagle MEM (10% FCS-MEM) to 1000 PFU / 100 μl. The purified anti-CMV antibody was diluted 4-fold with 10% FCS-MEM. In the case of measurement in the presence of complement, the antibody was diluted in the same manner with 10% FCS-MEM supplemented with 10% guinea pig complement. Virus and antibody were mixed in equal amounts and reacted at 37 ° C for 1 hour. HEL cells were cultured in a 35 mm dish using 10% FCS-MEM, the medium was removed before virus inoculation, and 0.3 ml of 10% FCS-MEM was added. Viral and antibody reaction solutions are inoculated into HEL cells in 50 μl aliquots, adsorbed at 37 ° C for 1 hour, washed with 10% FCS-MEM, and overlaid with 10% FCS-MEM containing 1% methylcellulose for 10 days. did. After removing methylcellulose, the measurement was stained with methylene blue, and the number of plaques was counted under a microscope (reference: Masuho, Y. et al., J. Gen. Virol., 68, p1457-1461 (1987)). The assay was performed in triplicate and the average number of plaques was calculated.

Immunostaining : CMV (AD169 strain) was diluted with a dilution buffer [PBS containing 5% FCS (−)] to 2000 PFU / 100 μl. The purified anti-CMV antibody was diluted 4-fold with dilution buffer. Guinea pig complement was diluted 5-fold with dilution buffer. An equal amount of antibody and complement was mixed with the virus solution and reacted at 4 ° C. for 1 hour. MRC-5 cells were cultured at 100 μl / well in 96-well half area plates using 10% FCS-added MEM-α. The virus / antibody reaction solution was inoculated into 25 μl each of MRC-5 cells, adsorbed at 37 ° C. for 1 hour, washed twice with 10% FCS-added MEM-α, and cultured for 20 hours. Infected cells were detected by CMV-IE1 staining. The medium was removed from the cells, fixed with 100% ethanol for 20 minutes, submerged with PBS (−) for 10 minutes, and blocked with blocking reagent [PBS (−), 1% BSA, 0.1% Tween20] for 10 minutes. IE1 antibody (MAB810R, Millipore) is diluted to 0.1 μg / ml with antibody dilution buffer [PBS (−), 0.1% Goat Serum, 0.1% Tween200], and 25 μl of 1 well is added as a primary antibody at 37 ° C. Reacted for hours. After washing 3 times with washing buffer [PBS (-), 0.1% Tween200], biotinylated labeled anti-mouse IgG (115-065-062, Jackson) as secondary antibody was diluted 1000 times with antibody dilution buffer. 25 μl of wells were added and allowed to react at 37 ° C. for 30 minutes. After washing 3 times with washing buffer, Streptavidin-peroxidase (SIGMA) was diluted 1000 times with antibody dilution buffer, and 25 μl of 1 well was added at 37 ° C. Reacted for 30 minutes. After washing 3 times with the washing buffer, True Blue peroxidase substrate (KPL) was added in a volume of 25 μl per well and allowed to develop for 15 minutes at room temperature. After washing with distilled water, it was washed once with 100% ethanol and dried in the dark. Infected cells (CMV-IE1-positive cells) were counted under a microscope (reference: Abai, et al., J. of Immunol. Methods 322 p82-93 (2007)). The assay was performed twice and the average number of infected cells was calculated.

Results N of antibody constant region of anti-CMV antibody (G3D-LCL: LCL-derived antibody, G3D-CHO: CHO-K1 cell-derived antibody, G3D-CHO + Gal: CHO-K1 cell-derived antibody with galactose added to the sugar chain) The structure of glycosidic linkage sugar chain was analyzed. The results are shown in FIG. The ratio of each sugar chain is shown as a percentage. Sugar chains of 1% or less were classified as other. The sugar chain structure was classified into G0, G1, and G2 according to the number of galactose terminals. The sugar chains estimated to be hybrid sugar chains were classified into one for each category. The sugar chain structure is expressed as fucose (*), N acetylglucosamine (●), mannose (■), galactose (▲), sialic acid (N acetylneuraminic acid) (♦). The description of the binding mode of the complex type sugar chain was omitted. The binding mode of acetylneuraminic acid was α2-6 bond.

  In the LCL-derived antibody (G3D-LCL), G2 accounted for most of the terminal structure of the N glycoside-linked sugar chain that binds to the antibody constant region. In addition, 25% monosialylated saccharide was present. On the other hand, most of the antibodies derived from CHO-K1 cells (G3D-CHO) were G0, but after sugar chain modification (G3D-CHO + Gal), more than 70% became G2, and the sugar chain modification was successful.

  Anti-CMV antibody (G3D-LCL: LCL-derived antibody, G3D-CHO: CHO-K1 cell-derived antibody, G3D-CHO + Gal: CHO-K1 cell-derived antibody with galactose added to the sugar chain) and control antibody (hIgG) Neutralizing activity was evaluated in the presence of complement (plaque assay and immunostaining). The number of plaques and the number of infected cells were calculated as percentages with the number in the control assay being 100.

  In FIG. 3, the result of neutralization activity evaluation is shown. The IC50 of G3D-LCL was 1.2 μg / ml, whereas the IC50 of G3D-CHO was 46.2 μg / ml, indicating a large difference in neutralizing activity. However, with G3D-CHO + Gal, the neutralizing activity recovered to 2.5 μg / ml with an IC50. Therefore, it was shown that neutralization activity can be enhanced in B4Gal-T1 introduced cells by adding galactose to CHO-K1 cell-derived antibodies.

(Effect of modification of constant region N glycoside-linked sugar chain in anti-CD20 chimeric antibody)
Materials and Methods Synthesis of anti-CD20 chimeric antibody (Rituxan) gene and production of antibody stable expression strain

In order to clone the gene encoding the anti-CD20 chimeric antibody (Rituxan), the nucleotide sequence (GenBank Accession No. AR000013) of the variable region of the antibody H chain was synthesized as 16 DNA fragments (Table 2). Each DNA fragment was composed of about 45 bases, the odd-numbered DNA fragment was in the sense direction, and the even-numbered DNA fragment was in the antisense direction, and each adjacent DNA fragment was designed to overlap 15 bases in a complementary manner. All of these DNA fragments were mixed, and the full length was amplified by PCR using RTX-H01 and RTX-H16 as amplification primers. Similarly, the nucleotide sequence of the variable region of the antibody L chain (GenBank Accession No. AR015962) and the partial constant region of the human κ chain was synthesized in the same manner (Table 3), and PCR was performed using RTX-K01 and RTX-K20 as amplification primers. Amplified by the method. The amplified fragment of H chain was subjected to restriction enzyme treatment (HindIII, ApaI), then recombined into a vector containing the constant region of human IgG1 H chain, and cloned as a chimeric antibody gene. On the other hand, the amplified fragment of the L chain was subjected to restriction enzyme treatment (HindIII, BbvCI), recombined into a vector containing a part of the constant region of human κ chain, and cloned as a chimeric antibody gene. Thereafter, the chimeric antibody gene (H chain and L chain) was inserted into a vector constructed for preparation of a double gene vector.

Next, the double gene vector was introduced into CHO-K1 cells and selected with Glutamine Synthetase (GS) selective medium (SAFC) to clone a Rituxan stable expression strain. For selection of CHO-K1 cells, GS inhibitor Methionine Sulfoximine (MSX, SIGMA) (final 2.5 μM) was added.

2． Glycosylation of Rituxan produced by CHO-K1 cells
B4Gal-T1 expression vector (Example 1, Materials and Methods 2.) was introduced into Rituxan stable expression clone (CHO-K1 cells) and selected with Geneticin (Gibco) (final 1200 μg / ml) in MSX-added GS selection medium And a clone expressing Rituxan and B4Gal-T1 simultaneously and stably was obtained. Confirmation of B4Gal-T1 expression was performed by RT-PCR. Among them, 2 clones (RTX-1, RTX-2) with good expression of B4Gal-T1 were selected, and the produced antibody was purified. In addition, the antibody produced by the Rituxan stable expression clone before introduction of the B4Gal-T1 expression vector was similarly purified.

3． Purification of Rituxan and glycosylated Rituxan All antibodies were purified using ProteinA Magnetic Beads (PureProteome, Millipore) as recommended. Specifically, antibody-producing cells were cultured in a serum-free medium, beads were added to the collected culture supernatant, and reacted at room temperature for 1 hour. The beads were collected, washed 3 times with PBS (-), and the antibody was eluted with 0.1 M Citrate Buffer pH 3.0 and neutralized with 1 M Tris-HCl pH 9.0. The concentration of the purified antibody was determined by UV method and ELISA method.

Four. Rituxan's complement-dependent cytotoxic activity (CDC activity) evaluation
Rituxan (Original) from Rituxan stably expressing clones and Rituxan (RTX-1, RTX-2) with galactose added to sugar chains derived from Rituxan and B4Gal-T1 co-stable expression clones Evaluation was performed. Daudi cells, which are CD20 positive cells, were used as subjects (BioTechnol. Prog. 2005, 21 1644-1652).

Daudi cells were washed once with 10% FCS-added RPMI1640 (SIGMA), washed again with RPMI, suspended in RPMI, and seeded in a 96-well plate at 25 μl / well to 5 × 10 ⁴ cells / well. Next, a 2-fold dilution series of Rituxan (Original, RTX-1, RTX-2) was prepared with RPMI1640 (7 steps from the final concentration of 2 μg / ml), added to Daudi cells at 25 μl / well, and incubator (37 ° C, The reaction was carried out at 5% CO ₂ ) for 30 minutes. Next, human complement (SIGMA) diluted to 20% with RPMI1640 was added at 50 μl / well (final concentration 10%), and cultured in an incubator (37 ° C., 5% CO ₂ ) for 3 hours. In addition, antibody-free wells were prepared as controls, and cell-free wells were prepared as blanks. Thereafter, WST-1 (Cell counting kit, Dojindo Chemical Co., Ltd.) was added to each well at 10 μl / well and reacted in an incubator (37 ° C., 5% CO ₂ ) for 3-5 hours. After completion of the reaction, the absorbance (OD450nm-OD620nm) was measured. The assay was performed in triplicate and results are shown as average values. CDC activity (%) was calculated using the following formula.
CDC (%) = 100 × {1- (SB) / (CB)}
S: Absorbance of sample well, B: Absorbance of blank well, C: Absorbance of control

Results Fig. 4 shows the CDC evaluation results of Rituxan (Original) without sugar chain modification and Rituxan (RTX-1, RTX -2) with galactose added to the sugar chain. As a result of CDC activity evaluation, RTX-1 and RTX-2 had higher CDC activity than Original, and were about 4-5 times higher at an antibody concentration of 0.125 μg / ml.

  In the above-described examples, one example of each of the anti-CMV antibody (fully human antibody) and the anti-CD20 antibody (chimeric antibody) was taken up and explained. However, the antibody obtained as shown in Example 1 is human. The sugar chain composition is very similar to the sugar chain in the lymphoblastoid cell line (LCL) derived from the cells, and the complement-dependent biological activity is 4 as shown in Examples 1 and 2. An increase in activity of -5 to several tens of times was observed. That is, in the present invention, an excellent host cell system for producing an active antibody having a high complement-dependent biological activity and a sugar chain composition similar to a human-derived host cell by a genetic recombination technique, and It can be said that a production system that can be easily industrialized has been found.

  In addition, regarding the origin of the antibody, the acquisition history of the antibody gene varies depending on the mouse-derived antibody, hybridoma-derived antibody, chimeric antibody, and humanized antibody, but the present invention can be applied regardless of the origin of these antibodies. It will be apparent to those skilled in the art.

  According to the present invention, a host cell system that produces a recombinant antibody that exhibits high complement-dependent biological activity compared to conventional antibody drugs, a production method that facilitates large-scale production of antibodies using the cells, And antibodies with high complement dependent biological activity are provided.

  The antibody production method of the present invention, the antibody produced by the method, and the host cell system that stably expresses such an antibody can be used for research or antibody drugs such as preventive and / or therapeutic agents for various diseases, etc. It is useful in the field of

Sequence number 1: Primer sequence (B1,4GT-Fw)
SEQ ID NO: 2: Primer sequence (B1,4GT-Rv)
SEQ ID NOs: 3 and 4: Primer sequence (for partial amplification of B4Gal-T1 gene)
SEQ ID NOs: 5 and 6: Primer sequence (for partial amplification of GAPDH gene)
SEQ ID NO: 7: B4Gal-T1 gene SEQ ID NO: 8: Amino acid sequence of B4Gal-T1
SEQ ID NOs: 9-24: Nucleotide sequences for cloning the heavy chain variable region of Rituxan
SEQ ID NOs: 25-44: Nucleotide sequences for cloning the light chain variable region and partial constant region of Rituxan

Claims (12)

a step of culturing a host animal cell expressing a gene encoding a protein having β-1,4-galactosyltransferase (B4Gal-T) activity and a gene encoding an antibody having complement-dependent biological activity; Complement-dependent biological activity comprising the step of purifying the antibody in which the N-glycoside-linked complex sugar chain bound to the Fc region of the antibody is modified from the host animal cell by the expressed B4Gal-T activity. A method for producing an antibody comprising: The culturing step comprises:
(i) a gene encoding an antibody having complement-dependent biological activity in a host animal cell in which a gene encoding a protein having β-1,4-galactosyltransferase (B4Gal-T) activity is stably expressed. Culturing the cells introduced in an expressible manner, or
(ii) a host animal cell into which both a gene encoding a protein having β-1,4-galactosyltransferase (B4Gal-T) activity and a gene encoding an antibody having complement-dependent biological activity can be expressed. Culturing,
The method of claim 1 comprising: The gene encoding the protein is
(i) the polynucleotide of SEQ ID NO: 7;
(ii) a polynucleotide that hybridizes with the polynucleotide of SEQ ID NO: 7 under a stringent condition and encodes a protein having B4Gal-T activity;
(iii) a polynucleotide encoding the amino acid sequence of SEQ ID NO: 8; or
(iv) The method according to claim 1 or 2, comprising a polynucleotide comprising an amino acid sequence having 80% or more identity with the amino acid sequence of SEQ ID NO: 8, and encoding a protein having B4Gal-T activity.   The method according to any one of claims 1 to 3, wherein the host animal cell is a vertebrate cell, a mammalian cell, or a CHO cell. further,
(i) The proportion of G1 and G2 complex type sugar chains in the N-glycoside-linked complex type sugar chains bound to the Fc region of the antibody (of the two N-acetylglucosamines at the non-reducing end of the sugar chain) Isolating an antibody having a ratio of at least one molecule of galactose bound to one molecule) of 50% or more; or
(ii) The proportion of the G2 complex type sugar chain in the N-glycoside-linked complex type sugar chain bound to the Fc region of the antibody (both the two N-acetylglucosamines at the non-reducing end of the sugar chain) The method according to any one of claims 1 to 4, which comprises a step of isolating an antibody having a ratio of 50% or more of each galactose bonded to one molecule.   The complement-dependent biological activity includes complement-dependent cytotoxic activity, complement-dependent virus neutralizing activity, complement-dependent CMV neutralizing activity, and complement-dependent cytotoxic activity of anti-CD20 antibodies. 6. The method according to any one of claims 1 to 5, wherein the activity is selected from the group consisting of: further,
(i) isolating an antibody exhibiting complement-dependent biological activity that is at least 3 times higher than an antibody whose sugar chain is not modified; or
(ii) The method according to any one of claims 1 to 6, comprising a step of isolating an antibody exhibiting complement-dependent biological activity that is 10 times or more higher than an antibody whose sugar chain is not modified. .   The method according to any one of claims 1 to 7, wherein the antibody is a monoclonal antibody.   The method according to claim 1, wherein the antibody is IgG or IgG1.   An antibody produced by the method according to any one of claims 1-9.   An antibody produced by the method according to claim 5 or 7. A host cell system stably expressing the antibody of claim 11.