JP2011182645A - Method for displaying animal cell and method for maturing whole human monoclonal antibody in vitro using the same - Google Patents

Abstract

A display method and screening method for a fully human monoclonal antibody using only an animal cell expression system is constructed and established as a method for antibody production and affinity maturation.
A vector designed to intervene a gene encoding an internal ribosome entry site between two structural genes each encoding a different protein and to express these genes bicistronically is represented by Flp. -An animal cell display method in which an animal cell having an In system is transfected and a protein expressed in a clone by introduction of one gene per cell is displayed on the cell membrane surface of the animal cell, and a fully human monoclonal antibody using the same In vitro maturation method.
[Selection] Figure 1

Description

  The present invention relates to an animal cell display method and an in vitro maturation method of a fully human monoclonal antibody using the same.

  An antibody (also called an immunoglobulin (Ig)) is a protein that specifically recognizes an antigen, and is called a “magic bullet” because of its nature. Antibodies are produced and secreted by B cells and bind to multiple antigens with a high degree of specificity for molecules and a wide variety. Ig genes are subject to different molecular processes by recombination, mutation and rearrangement in order to broaden the diversity of antibody production responses. Early in B cell development, V (D) J recombination occurs to bring together the V, D and J parts in different combinations. This increases the antibody repertoire. Later in B cell development, following antigenic stimulation, nucleotide changes can be introduced into the variable region by a process known as somatic hypermutation that confers increased antibody affinity and greater diversity to the antigen.

There are five known immunoglobulins: IgM, IgG, IgE, IgA, and IgD. From a biotechnological perspective, IgG is the most important type of antibody. IgG structurally has two identical heavy chains (H chains) and two identical light chains (L chains), which consists of one variable region (V _H ) and three constants. The region (C _H1 , C _H2 , C _H3 ) and the light chain are composed of one variable region (V _L ) and one constant region (C _L ). Functionally, antibodies are divided into two antigen-binding fragments (Fab) and constant region (Fc) linked via a mobile hinge site. The amino terminal portion of the variable region of the heavy and light chains recognizes the antigen. Each variable region is mainly involved in antigen recognition, and has three highly variable sequences called complementarity determining regions (CDRs). The amino acid remaining in the variable region acts as a scaffold for supporting the loop, and is called a framework residue (FR). The Fc part of an antibody mediates effector function, antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC).

  The characteristics of IgG are suitable for therapeutic methods using monoclonal antibodies, and monoclonal antibodies based on human IgG are mainly used for antibody treatment of cancer. Progress in monoclonal antibody production was made in 1975 by hybridoma technology. A hybridoma is a hybrid cell made by the fusion of B lymphocytes taken from the spleen of an animal immunized with a target antigen and immortalized myeloma cells. After limiting dilution cloning, the hybridoma is used as a pure and permanent source of monoclonal antibodies with specificity for the desired antigen. Once established in culture, hybridomas with confirmed antigen specificity can be propagated and produce monoclonal antibodies permanently. Such antibodies are expected to be applied in a wide range of fields such as test agents and therapeutic agents for various diseases including cancer. Among these, the use of human monoclonal antibodies is expanding as a molecular target drug.

  For human administration, human antibodies that do not exhibit antigenicity and have a long blood half-life are desired. As the first monoclonal antibody (mAb) approved by the FDA for human administration, mouse anti-CD3 mAb muromonab (OKT3) used for the treatment of transplanted organ rejection is known. However, such mouse antibodies have a short half-life of 15-30 hours and require frequent administration. In addition, it causes immune reactions with human anti-mouse antibodies (HAMA) and is not only rapidly eliminated from the blood but also has side effects ranging from hypersensitivity reactions to life-threatening anaphylactic reactions in repeated administration. cause. For this reason, the usefulness of mouse antibodies has been limited.

  For this reason, chimeric antibodies having rodent variable regions and human constant regions have been developed to reduce the immunogenicity of mouse antibodies when administered to humans. It was expected that such chimeric antibodies could improve biological activity in humans, have a longer half-life in blood, and be less immunogenic. However, the human anti-chimeric antibody (HACA) reaction was still observed, despite being actually less immunogenic than the murine antibody.

  Thus, humanized antibodies have been developed by protein engineering to further minimize the mouse components of antibodies. In such humanized antibodies, only the CDR loops of mouse antibodies involved in antigen binding are transplanted into the human variable region framework. The resulting humanized antibody was human up to 90%, was able to significantly reduce antigenicity over the chimeric antibody and tolerated repeated administration. However, CDR grafting is not an easy technique and often causes a decrease in affinity compared to the original mouse antibody.

  For humanized antibodies, the optimal framework site of the variable region must be selected in order to maintain the structure of the transplanted murine CDRs. Therefore, several experimental approaches, such as selection of human templates based on structural similarity between humans and mice, and computer-derived antibody variable region modeling, are necessary to humanize individual antibody clones. Met. Thus, the production of human monoclonal antibodies by hybridoma technology is not as easy as the production of mouse monoclonal antibodies, and many studies have been conducted to produce fully human antibodies for safer disease treatment.

Also, chimeric fused cells prepared by fusing mouse myeloma cells and human immunoglobulin producing cells tend to be unstable due to the gradual loss of human chromosomes. The human-human hybridoma technology has a problem of low fusion efficiency of about 0.1-1.0 clones per 10 ⁵ human-derived parent cells. Moreover, EBV (Epstein-Barr Virus) transformation of lymphocytes has long been used to obtain human monoclonal antibodies by immortalizing human B lymphocytes in vitro, the method of which is efficient. And the resulting transformed cells are often unstable and do not proliferate or secrete IgG. Even though these methods based on cell immortalization are mainly used for the production of human monoclonal antibodies, there are many problems associated with the efficiency of cell immortalization and the selection of antigen-specific antibody-producing cells.

  In order to solve these problems and obtain fully human antibodies, phage display libraries expressing human-derived artificial antibody fragments have been developed. In the mammalian immune system, high-affinity antibodies are mainly produced by recombination combining the germline antibody gene part and somatic hypermutation in the CDR part of the antibody gene, and by affinity maturation, higher affinity. B lymphocytes expressing antibodies with sex are selected and proliferated. Phage antibody technology can mimic the immune system by cloning a gene encoding an antibody specific for a target antigen from a large library of human antibody genes. In the phage antibody technique, only a variable region of an antibody is expressed in the form of a single chain antibody (Single Chain Fv, scFv) as a part of the phage coat protein. The antibody gene is linked to the phage coat protein gene. Upon expression, the encoded fusion protein is incorporated into the phage particle during normal phage production. The resulting phage particle expresses the antibody on its surface and contains the gene encoding the antibody. Since the protein and the gene are linked, the gene encoding the antibody can be selected as a result of isolation of a specific phage-displayed antibody through an affinity selection method for an antigen called panning. Once a library is created, it can be amplified and repeatedly used to isolate antibodies specific for multiple antigens.

  Other methods for generating human monoclonal antibodies have developed bacterial and yeast cell surface display methods. An important advantage of cell surface display over phage display is the ability to use flow cytometry sorting that can identify high affinity interactions and standardize and quantitatively analyze antibody protein expression. In recent years, full-length IgG antibodies can be expressed in E. coli to monitor their original function. Cell-free systems such as ribosome display are useful tools in that they can be used for the production of large libraries and antibody affinity maturation. However, due to problems with protein folding, post-translational modifications, and codon usage, these techniques actually have a limited number of antibodies with improved affinity.

  In recent years, human monoclonal antibody production methods often utilize mammalian expression systems or isolated B cells. In the scFv display method in human cells for affinity maturation of human monoclonal antibodies, the variable region is folded more efficiently, and the antibody gene is expressed in a human cell environment that is most similar to the in vivo site. (For example, Mitchell Ho et al., "Isolation of anti-CD22 Fv with high affinity by Fv display on human cells", Proc. Natl. Acad. Sci USA, Vol. 103, No. 25, (2006), pp. 9637-9642 (see Non-Patent Document 1). In this method, however, scFv is expressed on the cell membrane surface not by a monoclone but by a polyclone. In the Sindbis virus-mediated scFv display method on mammalian cells to isolate human monoclonal antibodies, animal cells have a low multiplicity of infection (in order to achieve expression of one antibody gene per infected cell ( The cells were infected with Sindbis virus containing scFv DNA by MOI (Multiplicity of Infection) (for example, Roger R. Beerli et al., “Isolation of human monoclonal antibodies by mammalian cell display”, PNAS, Vol. 105, No. 38, (2008), pp. 14336-14341 (Non-Patent Document 3)). However, with this method, it is difficult to ensure the expression of one antibody gene per cell. To isolate monoclonal antibodies with high affinity, antigen-specific memory B cells sensitized with the vaccine were utilized. In addition, in the method of selecting affinity matured antigen-specific antibody-producing cells naturally produced in the human body, the antigen-specific antibody gene is humanized by single-cell RT-PCR (Real Time-Polymerase Chain Reaction). Obtained from B cells isolated from. In these methods, an in vivo immune system is necessary to obtain a human monoclonal antibody having high affinity and high specificity. However, it is ethically difficult to immunize humans to obtain specific antibody-producing cells against any antigen, and many antigens are not suitable for immunization in human bodies.

Mitchell Ho et al., "Isolation of anti-CD22 Fv with high affinity by Fv display on human cells", Proc. Natl. Acad. Sci USA, Vol. 103, No. 25, (2006), pp.9637-9642 Yoshiko Akamatsu et al., "Whole IgG surface display on mammalian cells: Application to isolation of neutralizing chicken monoclonal anti-IL-12 antibodies", Journal of Immunological Methods, Vol. 327, (2007), pp.40-52 Roger R. Beerli et al., "Isolation of human monoclonal antibodies by mammalian cell display", PNAS, Vol. 105, No. 38, (2008), pp.14336-14341 Paivi Pihkala et al., "An antigen-mediated selection system for mammalian cells that produce glycosylated single-chain Fv", Biochem. Biophys. Res. Commun, Vol. 324, (2004), pp.1165-1172 Ronald H. Hoess et al., "Interaction of the bacteriophage P1 recombinase Cre with the recombining site loxP", Proc. Natl. Acad, Sci USA, Vol. 81, (1984), pp. 1026-1029 Stephen O 'Gorman et al., "Recombinase-Mediated Gene Activation and Site-Specific Integration in Mammalian Cells", Science, Vol. 251, (1991), pp. 1351-1354

  The phage display method using a human antibody gene derived from peripheral blood B cells requires a great deal of time and effort to prepare a huge library. As a method for solving these problems, “in vitro immunization” using peripheral blood monocytes (PBMC) of healthy persons was originally developed by the present inventor. Induction of immune responses using soluble proteins and peptides as antigens. The inventor of the present invention allowed PBMCs to sensitize human juvenile basophils KU812F cells that highly express human high-affinity IgE receptor α chain (FcεRIα) in vitro to induce antigen-specific immune responses. Were subjected to a phage display method, and antigen-specific phage antibody genes were obtained by panning using an antigen protein. Among the obtained genes, a highly reactive clone was recombined into a human IgG type and expressed in a highly productive CHO cell to obtain a monoclonal antibody. The obtained antibody specifically recognized FcεRI and inhibited the binding between FcεRI and IgE.

  However, when an antigen-specific single chain antibody gene (scFv (single chain antibody) type) obtained by screening using the phage display method is recombined into a fully human antibody gene (human IgG format) and expressed in animal cells. Many mutations occur and the loss of specificity and affinity occurs, so a great deal of labor is required to obtain the target antibody. This was thought to be due to the fact that codon usage was different from animal cells in microbial expression systems, and that normal folding and post-translational modifications were not performed. This means that even if in vitro immunization and phage display are performed efficiently to obtain antigen-specific human monoclonal antibodies, the efficiency of the recombination process after screening using the phage display method Is very low. Furthermore, although the obtained antibody has specificity, there is also a problem that the affinity is low.

  The present invention has been made to solve the above-mentioned problems, and the object of the present invention is to construct a display method and a screening method for a fully human monoclonal antibody using only an animal cell expression system, and to produce an antibody. And establish it as a method for affinity maturation.

  The animal cell display method of the present invention is designed such that a gene encoding an internal ribosome entry site is interposed between two structural genes encoding different proteins, and these genes can be expressed bicistronically. The resulting vector is transfected into an animal cell having a Flp-In system, and a protein expressed in a clone by introducing one gene per cell is displayed on the cell membrane surface of the animal cell.

  In the animal cell display method of the present invention, the structural gene upstream of the internal ribosome entry site is a structural gene encoding an antibody heavy chain, and the structural gene downstream of the internal ribosome entry site is the antibody. A structural gene encoding a light chain, or the structural gene upstream of the internal ribosome entry site is a structural gene encoding a light chain of an antibody, and the structural gene downstream of the internal ribosome entry site Is a structural gene encoding the heavy chain of the antibody, and the antibody having the heavy chain and light chain expressed bicistronically is preferably displayed on the cell membrane surface of animal cells.

  In the animal cell display method of the present invention, the structural gene encoding the heavy chain sequentially encodes the variable region and the constant region from the upstream side, and the structural gene encoding the light chain includes the variable region and the constant region from the upstream side. It is preferable to code sequentially.

  The vector in the animal cell display method of the present invention is designed so that a gene encoding a transmembrane domain is interposed between a structural gene encoding a heavy chain and a structural gene encoding a light chain. Preferably there is. In this case, the vector is designed so that the loxP sequence on the 5 ′ end side of the gene encoding the transmembrane domain is hidden in the intron, and no extra amino acid sequence derived from the loxP sequence is added to the heavy chain constant region. It is preferable that

  In the present invention, the vector is between a structural gene encoding a heavy chain and a gene encoding a transmembrane domain, and between a gene encoding a transmembrane domain and a structural gene encoding a light chain. After the loxP site is mediated and the vector is transfected into the animal cell, the animal cell is transfected with a vector containing a gene encoding Cre recombinase, or the animal cell is treated with Cre recombinase as a lipid carrier. It is preferable that the gene is designed so that a gene encoding a transmembrane domain between two loxP sites is removed from the animal cell by introducing by the above. In this case, it is more preferable that the antibody is converted from a membrane type to a secreted type by removing a gene encoding a transmembrane domain between two loxP sites in the animal cell.

The antibody in the present invention is preferably human IgG.
The antibody in the present invention is preferably a human high-affinity IgE receptor α chain antibody, and obtained by sensitizing human peripheral blood mononuclear cells with human juvenile basophil KU812F cells in vitro. It is preferable that it is a thing.

  The animal cell in the present invention is preferably a CHO cell having an Flp-In system.

  The present invention also includes a step of randomly amplifying a gene encoding a variable region of an antibody heavy chain, and a structural gene encoding the obtained amplification product in order from the upstream to the variable region of the heavy chain and the constant region of the heavy chain. And a gene encoding the internal ribosome entry site between the structural genes encoding the light chain variable region and the light chain constant region in order from the upstream side, so that these genes can be expressed bicistronically A step of constructing a vector designed as a structural gene encoding a variable region of a heavy chain randomly mutated into the vector designed in the above, and transfection of the vector into an animal cell having a Flp-In system And constructing a library of animal cells in which the antibody expressed in the clone is displayed on the cell membrane surface by introducing one gene per cell Selecting an animal cell that expresses a high affinity antibody on the cell membrane from a library of animal cells by fluorescence activated cell sorting, and secreting a fully human monoclonal antibody to the selected animal cell. In vitro maturation methods for fully human monoclonal antibodies are also provided.

  By the animal cell display method of the present invention, a display method and screening method for a fully human monoclonal antibody using only an animal cell expression system has been established, and this method enables in vitro maturation of a fully human monoclonal antibody. .

FIG. 1 (a) is a diagram conceptually showing one specific example of a vector that can be constructed in the animal cell display method of the present invention, and FIG. 1 (b) uses the vector shown in FIG. 1 (a). It is the conceptual diagram which shows the animal cell display method which had been in steps. FIG. 2 (a) schematically shows between the downstream side of the structural gene encoding the heavy chain constant region and the gene encoding the transmembrane domain in the vector constructed by the present inventors in Experimental Example 1 described later. FIG. 2B is a diagram showing a specific base sequence and amino acid sequence. It is a schematic diagram which shows the procedure of construction | assembly of pcDNA5FRT / CL in Experimental example 1 in steps. It is a schematic diagram which shows the procedure of construction | assembly of pcDNA5FRT / CH in Experimental example 1 in steps. It is a schematic diagram which shows the procedure of insertion of the variable region gene in Experimental example 1 in steps. FIG. 6 (a) is a graph showing IgG molecule expression on the CHO cell surface in Experimental Example 1. FIG. 6 (a) shows a case where Flp-In-CHO cells were treated with a PE-conjugated mouse anti-human IgH antibody. It is a result when lp-In-CHO cells are treated with an anti-human Igκ antibody, and the shaded region indicates cells transfected with IgG plasmid, and the open region indicates cells not transfected. . FIG. 3 is a graph showing anti-high affinity IgE receptor α chain antigen binding activity of CHO cells expressing IgG on the surface in Experimental Example 1, in which shaded areas are cells transfected with IgG plasmid, and open areas are trans Each non-fected cell is shown. FIG. 8 (a) is a predicted view of recombination of the gene encoding the transmembrane domain in Experimental Example 1, where the dotted line indicates an intron and the arrow indicates a loxP site. FIG. 8 (b) is an electrophoretogram showing the results of the actual recombination of the gene encoding the transmembrane domain. FIG. 9A is a diagram schematically showing a template DNA state scheme and an amplification region by PCR in Experimental Example 1, and FIG. 9B shows an intron splicing reaction without a transmembrane domain. It is the electrophoresis photograph which shows this. It is a graph which shows the result of having detected IgG by ELISA from the culture supernatant after transfecting a cell with CEP4 / Cre, and a vertical axis | shaft is the light absorbency in 405 nm. It is a graph which shows the result of having detected the intracellular IgG level of the CHO cell in Experimental example 1 by ELISA, and a vertical axis | shaft is the light absorbency in 405 nm. It is a graph which shows the result of the cell sorting of the CHO cell in Experimental example 2, a vertical axis | shaft is SSC and a horizontal axis is the fluorescence intensity of FITC. FIG. 13 is a graph showing comparison of reactivity of anti-FcεRIα antibody from sorted cells in Experimental Example 2, FIG. 13 (a) shows the P1 region in FIG. 11, and FIG. 13 (b) shows the P2 region in FIG. It is a result, and as for all, a vertical axis | shaft is the light absorbency in 405 nm.

[Animal cell display method]
FIG. 1 (a) is a diagram conceptually showing one specific example of a vector that can be constructed in the animal cell display method of the present invention, and FIG. 1 (b) uses the vector shown in FIG. 1 (a). It is the conceptual diagram which shows the animal cell display method which had been in steps. According to the present invention, a vector designed to intervene a gene encoding an internal ribosome entry site between two structural genes encoding different proteins from each other and to express these genes bicistronically is represented by Flp. The present invention provides an animal cell display method characterized by transfecting an animal cell having an In system and displaying a protein expressed in a clone by introduction of one gene per cell on the cell membrane surface of the animal cell. By such an animal cell display method of the present invention, a display method of a fully human monoclonal antibody using only an animal cell expression system and a screening method thereof are established, and by using this, in vitro maturation of a fully human monoclonal antibody can be achieved. It becomes possible.

  The vector used in the animal cell display method of the present invention includes a construct that expresses at least two genes in a single transcription cassette (ie, bicistronically) under the control of a single upstream promoter. Internal Ribosome Entry Sites (IRES) that are known to be used to design are used. By using such an IRES-containing vector, the proteins encoded by the structural gene on the upstream side and the structural gene on the downstream side of the IRES are composed of two types of independent constituent elements, respectively. One protein can be expressed.

  The protein encoded by the two structural genes in the vector used in the animal cell display method of the present invention is not particularly limited. However, human monoclonal antibody production, particularly in vitro maturation (affinity of human monoclonal antibody described later) It is preferable that an antibody is displayed on the cell membrane surface of an animal cell having a transfected Flp-In system.

  When the antibody is presented on the cell membrane surface of an animal cell, it is preferably either (A) or (B) below.

(A) The structural gene upstream of the IRES is a structural gene encoding an antibody heavy chain, and the structural gene downstream of the IRES is a structural gene encoding a light chain of the antibody.
(B) The structural gene upstream of the IRES is a structural gene encoding a light chain of an antibody, and the structural gene downstream of the IRES is a structural gene encoding a heavy chain of the antibody.

  With the configuration of (A) or (B), the antibody having the heavy and light chains expressed bicistronically can be presented on the cell membrane surface of animal cells.

  In the animal cell display method of the present invention, the structural gene encoding the heavy chain sequentially encodes the variable region and the constant region from the upstream side in any of the cases (A) and (B) described above. It is preferable that the structural gene encoding the light chain encodes the variable region and the constant region in this order from the upstream side.

  Although there is no restriction | limiting in particular as a vector used for this invention, A plasmid vector can be used conveniently.

  In the animal cell display method of the present invention, animal cells having the Flp-In system are used. Here, the Flp-In system is a system developed to generate a cell line that is stably expressed in order to overcome the limitation of screening in animal cells. Introduction of the Flp-In system into animal cells (Flp-In recombination) is known to be catalyzed by Flp recombinase derived from Saccharomyces cerevisiae. Flp recombinase is a member of the λ integrase family and recognizes a 34 base pair FRT (Flp Recombination Target) recognition target sequence. Flp recombinase also mediates insertion of plasmid DNA containing FRT sequences into the genome of a single FRT site or cell line into which cells are integrated through site-specific DNA recombination (eg, Stephen O 'Gorman et al. , "Recombinase-Mediated Gene Activation and Site-Specific Integration in Mammalian Cells", Science, Vol. 251, (1991), pp. 1351-1354 (Non-Patent Document 6)). In the present invention, by transfecting an animal cell having such a Flp-In system with a vector as described above, the introduction of a plurality of vectors into the animal cell was avoided, and the animal cell was expressed bicistronically. The protein is presented in clones on the cell membrane surface of the animal cell.

  The animal cell used in the present invention is not particularly limited as long as it is an animal cell into which the Flp-In system can be introduced. For example, Chinese Hamster Ovary (CHO) cell, 293 cell, BHK cell, Jurkat cell, CV-1 cells and the like can be mentioned, but besides these, appropriate animal cells into which the Flp-In system can be introduced can be used. Of these, CHO cells are particularly preferred because they are most widely used as host cells for antibody production in the industrial field. In an experimental example to be described later, a CHO cell (Flp-In-CHO cell) having a Flp-In system that includes a single integrated FRT site and stably expresses the lacZ-zeocin fusion gene encoded downstream of the FRT site. Are preferably used.

  The vector in the animal cell display method of the present invention is designed so that a gene encoding a transmembrane domain is interposed between a structural gene encoding a heavy chain and a structural gene encoding a light chain. Preferably there is. By including a gene encoding a transmembrane domain, a bicistronically expressed protein can be suitably presented on the cell membrane surface in animal cells transfected with the above-described vector. In particular, when the bicistronically expressed protein is an antibody, in order to retain the antibody expressed on the cell membrane surface of animal cells, a structural gene encoding a heavy chain (especially encoding a heavy chain constant region). It is preferable that a gene encoding a transmembrane domain is arranged downstream of the structural gene). As the transmembrane domain, for example, a transmembrane domain derived from human γ1 can be preferably used.

  Here, FIG. 2 (a) is a schematic diagram between the downstream side of the structural gene encoding the heavy chain constant region and the gene encoding the transmembrane domain in the vector constructed by the present inventors in Experimental Example 1 described later. FIG. 2 (b) shows a specific base sequence and amino acid sequence (SEQ ID NO: 1 for the base sequence). The gene encoding the transmembrane domain can be removed by Cre / loxP recombination after transfection into animal cells as described below, so that the loxP sequence on the 5 ′ end side is removed. It is preferable to have. This Cre / loxP recombination system has been widely used in multiple organisms, including yeast, plants, and animals, and can be used for gene inactivation or controlled ectopic gene expression to control the basic genes of a particular organ. Enable analysis. In the Cre / loxP recombination system, the P1 bacteriophage Cre recombinase is utilized to catalyze recombination between tandem loxP DNA sequences. Cre protein is a single polypeptide with a molecular weight of 35000 and binds to a loxP site to mediate a recombination reaction between two loxP sites. The loxP sequence also consists of two 13 base pair inverted repeats of the Cre binding region separated by an 8 base pair spacer site. This suggests that this 34 base pair is necessary for effective recombination.

  The animal cell display method of the present invention avoids frequent use of PCR and utilizes such Cre / loxP recombination, so that when the bicistronically expressed protein is an antibody, a membrane is used. Antibody expressed as a type molecule can be converted into a secreted type molecule, and it becomes possible to screen for antigen-specific fully human monoclonal antibodies and obtain production cells using animal cell expression platform, which is a natural environment . In the present invention, the vector has a loxP sequence on the 5 ′ end side of a gene encoding a transmembrane domain hidden in an intron so that an amino acid sequence derived from the loxP sequence is not added to the heavy chain constant region. It is preferable that it is designed. Thus, the 34 base loxP sequence avoids adding an extra amino acid to the C-terminus of the constant region of the antibody heavy chain expressed bicistronically in animal cells transfected with the vector. Can do.

  In the present invention, the vector is between a structural gene encoding a heavy chain and a gene encoding a transmembrane domain, and between a gene encoding a transmembrane domain and a structural gene encoding a light chain. It is preferable that a loxP site is interposed. Specifically, as shown in FIG. 2 (a), a 5 ′ splicing site followed by a BamHI site for restriction digestion is added to the 3 ′ end of the structural gene encoding the heavy chain constant region, As shown in FIG. 2 (b), on the downstream side from the BamHI site, the two loxP sequences are the downstream side of the branch point sequence (BPS) and the downstream side of the gene encoding the transmembrane domain. Respectively.

  The vector of the present invention is designed so that the gene encoding a transmembrane domain between two loxP sites is removed in the animal cell after the vector is transfected into the animal cell. ,preferable. The removal of the gene encoding the transmembrane domain between the two loxP sites is preferably either (a) or (b) below.

(A) transfecting an animal cell obtained by transfecting the animal cell with the vector with a vector containing a gene encoding Cre recombinase;
(B) Cre recombinase is introduced into the animal cell obtained by transfecting the animal cell with the vector using a lipid carrier.

  Thus, using Cre / loxP recombination, a gene encoding a transmembrane domain between two loxP sites is removed in an animal cell after the animal cell has been transfected with the vector. As a result, antibodies expressed bicistronically in animal cells are preferably converted from a membrane type to a secreted type.

  The construction of vectors used in the animal cell display method of the present invention, transfection into animal cells, and the like are based on the description of experimental examples and the like described later by appropriately combining known genetic engineering techniques in this field. Those skilled in the art can easily implement this.

  The antibody in the present invention is preferably human IgG. Thus, in the animal cell display method of the present invention, it is possible to suitably produce fully human monoclonal antibodies.

  The antibody in the present invention is preferably a human high affinity IgE receptor α chain antibody (FcεRIα antibody), and human peripheral blood mononuclear cells (PBMC) are sensitized in vitro with human juvenile basophils KU812F cells. It is preferable that it is obtained. The human high-affinity IgE receptor is a receptor that recognizes the Fc region of IgE and is expressed in mast cells and the like, and allergy is induced when the receptor and IgE complex are cross-linked by an antigen. Therefore, obtaining an FcεRIα antibody capable of inhibiting the binding between this human high affinity IgE receptor and IgE is useful for application to allergy treatment.

  Antibodies against membrane proteins also contribute to controlling many important reactions (eg, rejection by immune cells and complement, reactions as ligands, inhibition of molecular interactions), but they are highly pure and naturally occurring. It is not so easy to prepare a large amount of membrane protein having a sequence. As described above, the present inventor has independently developed an “in vitro immunization” for such a problem, and an antigen-specific immune response is induced by using this method. PBMCs are obtained. Specifically, PBMCs isolated from peripheral blood were treated with L-Leucyl-L-Leucine methyl ester to remove immunosuppressive cells, and then appropriate cytokine conditions and adjuvant (immunopotentiator) The cells are cultured together and subjected to antigen sensitization, and after culturing for 1 week, antigen-specific antibody-producing cells are induced. This makes it possible to perform immunization against a membrane protein having a natural sequence without purifying the antigen.

[In vitro maturation of fully human monoclonal antibodies]
The present invention also provides a method for in vitro maturation of fully human monoclonal antibodies using the above-described animal cell display method of the present invention. That is, the in vitro maturation method of the fully human monoclonal antibody of the present invention comprises (1) a random amplification of a structural gene encoding the variable region of the heavy chain of the antibody, and (2) the obtained amplification product from the upstream side. An internal ribosome entry site is encoded between the structural gene that sequentially encodes the variable region of the heavy chain and the constant region of the heavy chain, and the structural gene that sequentially encodes the variable region of the light chain and the constant region of the light chain from the upstream side. (3) the step of constructing each of the incorporated vectors as a structural gene encoding the variable region of the heavy chain in a vector designed to intervene genes and express these genes bicistronically; Is transfected into an animal cell having the Flp-In system, and the antibody expressed in the clone is expressed by introduction of one gene per cell. A step of constructing a library of animal cells displayed on the surface of the membrane, and a step of (4) selecting an animal cell that expresses a high affinity antibody on the cell membrane from the library of animal cells by fluorescence activated cell sorting; 5) A method for in vitro maturation of fully human monoclonal antibodies, comprising the step of secreting fully human monoclonal antibodies into selected animal cells is also provided.

  In the in vitro maturation method of a fully human monoclonal antibody of the present invention, first, a structural gene encoding the variable region of the heavy chain of the fully human monoclonal antibody is randomly amplified. In this way, the variable region of the heavy chain of the antibody is artificially mutated to be diversified, and this is incorporated as a structural gene encoding the variable region of the heavy chain of the above-mentioned vector, and the resulting vector Is used to perform the above-described animal cell display method of the present invention. This makes it possible to construct an animal cell library containing various antibody genes in which fully human monoclonal antibodies are each cloned and presented on the cell membrane surface. Then, by isolating the clone with the highest affinity from this library using a cell sorter, it is possible to obtain animal cells that produce an antibody with higher affinity than the original antibody. Thus, the animal cell display method of the present invention described above is useful for maturation of affinity of fully human monoclonal antibodies.

  Hereinafter, although an example of an experiment is shown and the present invention is explained still more concretely, the present invention is not limited by these examples.

<Experimental example 1>
In Experimental Example 1, a plasmid vector for a new animal cell display system was prepared, and the efficiency and accuracy of a system for producing a human monoclonal antibody was analyzed.

[Materials and methods]
[1] Cell line As a cell line, a Chinese hamster ovary cell (Flp-In-CHO cell) (purchased from Invitrogen) containing an FRT site was used. The Flp-In-CHO cell line contains a single integrated FRT site and stably expresses the lacZ-zeocin fusion gene.

[2] Cell culture All solutions and equipment in contact with cells were sterilized before use. Appropriate aseptic techniques were always used in the laminar flow hood. Flp-In-CHO cells are treated with 100 U / ml penicillin (Meiji Seika), 100 g / ml streptomycin (Meiji Seika), 100 g / ml Zeocin (Invitrogen), and 10 in a humid environment of 37 ° C., 5% CO _2. % It was cultured in Ham'sF12 medium (Nissui Pharmaceutical) containing fetal bovine serum.

[3] β-Galactosidase staining Gene insertion into FRT was confirmed by β-galactosidase staining. Cells were washed with PBS and fixed with 2% formaldehyde / 0.2% glutaraldehyde in PBS for 5 minutes. After fixation, the cells were washed twice with PBS, and β-galactosidase staining solution (150 mM NaCl, 2 mM MgCl) containing 1 mg / ml 5-bromo-4-chloro-3-indolyl-D-β-galactoside (X-gal). _2. Incubated with 5 mM potassium ferricyanide, 5 mM potassium ferricyanide / sodium phosphate buffer, pH 7.4) at 37 ° C. for 2-4 hours. Cells with DNA inserted into the FRT site were not stained, and cells with no DNA inserted into the FRT site were identified as blue.

[4] Construction of a plasmid vector that expresses complete IgG on the cell surface (1) PCR
Using a specific primer pair consisting of a primer containing the restriction digestion site NotI-Ld (SEQ ID NO: 2) and a primer containing the restriction digestion site CL-ApaI (SEQ ID NO: 3), the pSecTag2A / CL to Igκ chain using KOD plus DNA polymerase The constant region of kappa chain DNA after the leader sequence was amplified. PCR reaction solution is 0.5 μl of pSecTag2A / CL, 5 μl of 10 × KOD plus PCR buffer, 3 μl of the above specific primer pair, 2 μl of 25 mM MgSO ₄ , 1 μl of KOD plus DNA polymerase in 50 μl of distilled water. Was added and prepared. Amplification was performed in 30 PCR cycles (94 ° C. for 15 seconds, 60 ° C. for 30 seconds, 68 ° C. for 1 minute). Amplified products were stained with 5 μl ethidium bromide and subjected to 1.5% agarose gel electrophoresis. The gel was stained with ethidium bromide and the band containing DNA was detected under ultraviolet light. The DNA fragment after electrophoresis was purified with Wizard SV Gel and PCR cleanup system (Promega). In this way, an amplification product leader-Cκ was obtained.

Using a specific primer pair consisting of a primer containing the restriction digestion site NheI-Ld (SEQ ID NO: 4) and a primer containing the restriction digestion site IgGc-BamHI (SEQ ID NO: 5), the pSecTag2A / IgGc to gκ chain using KOD plus DNA polymerase. The constant region of the heavy chain DNA after the leader sequence was amplified. Amplification was performed by adding 0.5 μl of pSecTag2A / IgGc, 5 μl of 10 × KOD plus PCR buffer, 3 μl of the above specific primer pair, 2 μl of 25 mM MgSO ₄ , 1 μl of KOD plus DNA polymerase to 50 μl, Prepared. Amplification was performed in 30 PCR cycles (94 ° C. for 15 seconds, 60 ° C. for 30 seconds, 68 ° C. for 1 minute). Amplified products were stained with 5 μl ethidium bromide and subjected to 1.5% agarose gel electrophoresis. The gel was stained with ethidium bromide and the band containing DNA was detected under ultraviolet light. The DNA fragment after electrophoresis was purified with Wizard SV Gel and PCR cleanup system (Promega). In this way, an amplification product leader-IgGc was obtained.

Using a specific primer pair consisting of a primer containing the restriction digestion site BstXI-IRES (SEQ ID NO: 6) and a primer containing the restriction digestion site IRES-NotI (SEQ ID NO: 7), pRESSbleo (Clontech) from IRES was used with KOD plus DNA polymerase. DNA was amplified. Amplification was prepared by adding 0.5 μl of pIRESbleo, 5 μl of 10 × KOD plus PCR buffer, 3 μl of the above specific primer pair, 2 μl of 25 mM MgSO ₄ , 1 μl of KOD plus DNA polymerase to 50 μl. . Amplification was performed in 30 PCR cycles (94 ° C. for 15 seconds, 60 ° C. for 30 seconds, 68 ° C. for 1 minute). Amplified products were stained with 5 μl ethidium bromide and subjected to 1.5% agarose gel electrophoresis. The gel was stained with ethidium bromide and the band containing DNA was detected under ultraviolet light. The DNA fragment after electrophoresis was purified with Wizard SV Gel and PCR cleanup system (Promega). In this way, an amplification product IRES was obtained.

(2) A-tailing 15 μl of each PCR product leader-Cκ, leader-IgGc and IRES with 5 μl 10 × PCR buffer (Promega), 5 μl 25 mM MgCl ₂ , 1 μl 100 mM dATP, 0.5 μl Taq polymerase Mix and add distilled water to 50 μl. The mixture was reacted at 70 ° C. for 2 hours. The reaction product was mixed with the staining solution and subjected to agarose gel electrophoresis. The gel was stained with ethidium bromide and the band containing DNA was detected under ultraviolet light. DNA fragments were purified with Wizard SV Gel and PCR cleanup system. Thus, in order to ligate pGEM-T easy (Promega), d (A) was added to the blunt end of the three amplification products, leader-Cκ, leader-IgGc and IRES.

(3) Ligation and transformation Each A-tailed amplification product was ligated into a pGEM T easy vector. The ligation product was transfected into competent bacteria (E. coli JM109). Transfected cells were incubated on LB plates containing 50 μg / ml ampicillin overnight at 37 ° C. After incubation, the formed colonies were picked and the insertion of the DNA fragment into the vector was confirmed by PCR. Single colonies carrying the inserted vector were picked and incubated with 2 ml of LB containing 50 μg / ml ampicillin for 18 hours at 37 ° C. After incubation, a plasmid vector pGemTeasy / NotI-Ld-Cκ-ApaI was prepared with GenElute ™ Plasmid Miniprep Kit (Sigma-aldrich).

(4) Determination of base sequence The base sequences of these inserted genes were confirmed using Big Dye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems). 200 ng plasmid vector, 4 μl 5 × sequencing buffer, 3.2 pmol T7-2 primer (5′-TAATACGACTCACTATAGGGG-3 ′ (SEQ ID NO: 8)), 1 μl Ready Reaction Premix with distilled water to a final volume of 20 μl The reaction mixture was prepared by addition.

  PCR reaction was carried out under conditions of 25 cycles, with one cycle consisting of denaturation at 96 ° C. for 10 seconds, annealing at 50 ° C. for 5 seconds, and extension at 60 ° C. for 4 minutes. The obtained amplification product was purified by ethanol precipitation. The amplification product was mixed with 2 μl of 125 mM EDTA, 2 μl of 3M sodium acetate, 50 μl of 100% ethanol and stored at room temperature for 15 minutes. Mixing was performed by centrifugation at 20000 xg for 20 minutes at 4 ° C. After removing the supernatant and washing with 70 μl 70% ethanol, the precipitate was centrifuged at 20000 × g for 10 minutes at 4 ° C. The pellet was dried under vacuum conditions for 10 minutes and suspended in 20 μl Hi-diformamide. The product was incubated at 94 ° C. for 2 minutes, then chilled on ice for 5 minutes and analyzed with a 3100-Avant Genetic Analyzer (Applied Biosystems).

(5) Construction of pcDNA5FRT / CL pcDNA5FRT / CL was constructed by the procedure schematically shown in FIG. First, 1 μg of DNA (pGemTeasy / NotI-Ld-Cκ-ApaI or pcDNA5FRT (Invitrogen)), 5 μl of 10 × NEB buffer 4, 0.5 μl of 100 × BSA (NEB), 1 μl of NotI, 1 μl of ApaI Prepared and added distilled water to 50 μl. The mixture was incubated at 37 ° C. for 2 hours, and pGemTeasy / NotI-Ld-Cκ-ApaI, pcDNA5FRT was digested with the restriction endonucleases NotI and ApaI. The product after restriction digestion was subjected to 1.5% agarose gel electrophoresis and purified with Wizard SV Gel and PCR cleanup system (Promega).

  Next, the products of restriction digestion of pGemTeasy / NotI-Ld-Cκ-ApaI, pcDNA5FRT were mixed in a 3: 1 molecular ratio. This mixture was mixed with 2 × ligation mix, mixed at 16 ° C. for 30 minutes and incubated. The ligation product (pcDNA5FRT / CL) was transfected into competent bacteria (E. coli JM109) on ice for 10 minutes. Transfected cells were incubated on LB plates containing 50 μg / ml ampicillin overnight at 37 ° C. After the incubation, the formed colonies were picked, and insertion of NotI-Ld-Cκ-ApaI DNA fragment into the vector was confirmed by PCR.

  Single colonies carrying the pcDNA5FRT / CL vector were picked and incubated with 2 ml LB containing 50 μg / ml ampicillin at 37 ° C. for 18 hours. Following incubation, plasmid vectors were prepared with GenElute ™ Plasmid Miniprep Kit (Sigma-aldrich).

(6) Construction of pcDNA5FRT / CH pcDNA5FRT / CH was constructed by the procedure schematically shown in FIG. 1 μl of DNA (pGemTeasy / IRES or pcDNA5FRT) was mixed with 5 μl of 10 × NE Buffer 3, 0.5 μl of 100 × BSA, 1 μl of BstXI, and distilled water was added to 49 μl. After incubation at 55 ° C. for 2 hours, 1 μl of NotI was added to the mixture, reacted at 37 ° C. for 2 hours, and plasmid DNAs pGemTeasy / IRES and pcDNA5FRT were digested with restriction endonucleases NotI and BstXI. The product after restriction digestion was subjected to 1.5% agarose gel electrophoresis and purified with Wizard SV Gel and PCR cleanup system (Promega).

  Next, the product of restriction digest of pGemTeasy / IRES, pcDNA5FRT was mixed in a 3: 1 molecular ratio. This mixture was mixed with 2 × ligation mix, mixed at 16 ° C. for 30 minutes and incubated. The ligation product (pcDNA5FRT / IRES) was transfected into competent bacteria (E. coli JM109) in the same manner as described above. Transfected cells were incubated on LB plates containing 50 μg / ml ampicillin overnight at 37 ° C. After incubation, a plasmid vector was prepared with GenElute ™ Plasmid Miniprep Kit (Sigma-aldrich) to obtain pcDNA5FRT / IRES.

  Next, 1 μl of DNA (pcDNA5FRT / IRES or pGemTeasy / NheI-Ld-IgGc-BamHI) was mixed with 5 μl 10 × NE Buffer 2, 0.5 μl 100 × BSA, 1 μl NheI, 0.5 μl BamHI Then, distilled water was added to make 50 μl. The mixture was incubated at 37 ° C. for 2 hours, and pcDNA5FRT / IRES and pGemTeasy / NheI-Ld-IgGc-BamHI were digested with the restriction endonucleases NheI and BamHI. The product after restriction digestion was subjected to 1.5% agarose gel electrophoresis and purified with Wizard SV Gel and PCR cleanup system (Promega).

  Next, the restriction digested Ld-IgGc fragment and pcDNA5FRT / IRES were mixed at a molecular ratio of 3: 1. This mixture was combined with 2 × ligation mix and incubated at 16 ° C. for 30 minutes. The ligation product was transfected into competent bacteria (E. coli JM109) as described above. Transfected cells were incubated on LB plates containing 50 μg / ml ampicillin overnight at 37 ° C. After incubation, the formed colonies were picked and the insertion of the DNA fragment into the vector was confirmed by PCR.

  A single colony carrying the inserted vector was picked and incubated with 2 ml LB containing 50 mg / ml ampicillin for 18 hours at 37 ° C. After incubation, a plasmid vector was prepared with GenElute ™ Plasmid Miniprep Kit (Sigma-aldrich) to obtain pcDNA5FRT / Ld-IgGc-IRES.

  1 μg of pUC / TMi or pcDNA5FRT / Ld-IgGc-IRES was mixed with 5 μl of 10 × NE Buffer 2, 0.5 μl of 100 × BSA, 1 μl of BstX1, and a maximum of 49 μl of distilled water was added. After 2 hours incubation at 55 ° C., 1 μl of BamHI was added to the mixture and allowed to react at 37 ° C. for 2 hours. The reaction product was subjected to 1.5% agarose gel electrophoresis. The DNA fragment was purified with Wizard SV Gel and PCR cleanup system (Promega).

  The digested TMi fragment and pcDNA5FRT / Ld-IgGc-IRES were mixed at a molecular ratio of 3: 1. This mixture was mixed with 2 × ligation mix and incubated at 16 ° C. for 30 minutes. The ligation product was transfected into competent bacteria (E. coli JM109) as described above. Transfected cells were incubated on LB plates containing 50 μg / ml ampicillin overnight at 37 ° C. After incubation, the formed colonies were picked and the insertion of the DNA fragment into the vector was confirmed by PCR. A single colony carrying the inserted vector was picked and incubated at 37 ° C. for 18 hours with 2 ml of LB containing 50 μg / ml ampicillin. After the incubation, a plasmid vector was prepared with GenElute (trademark) Plasmid Miniprep Kit (Sigma-aldrich) to obtain pcDNA5FRT / CH. Here, the plasmid vector pUC was designed to contain a DNA fragment in the transmembrane region between the loxP sequences hidden in the intron (TMi).

(7) Insertion of variable region gene The variable region gene was inserted by the procedure schematically shown in FIG. C9 anti-FcεRIα antibody variable region genes (VH and Vκ) were used for the construction of a model whole IgG expression vector. 1 μl of DNA (plasmid vector pSecTag2A / C9H or pcDNA5FRT / IgH) was mixed with 5 μl of 10 × NE Buffer 2, 0.5 μl of 100 × BSA100, 1 μl of SfiI, and distilled water was added to make 50 μl. After 2 hours incubation at 0 ° C., 1 μl of XhoI was added to the mixture and incubated at 37 ° C. for 2 hours, and the plasmid vectors pSecTag2A / C9H and pcDNA5FRT / IgH were digested with the restriction endonucleases SfiI and XhoI. The product after restriction digestion was mixed with 5 μl of staining solution and subjected to 1.5% agarose gel electrophoresis. The gel was stained with ethidium bromide and the band containing DNA was detected under ultraviolet light. The DNA fragment was purified with Wizard SV Gel and PCR cleanup system.

  C9VH DNA containing the SfiI and XhoI digested sites was ligated into the pcDNA5FRT / IgH vector digested with restriction enzymes. The ligation product was transfected into competent bacteria (E. coli JM109) as described above. After incubation on LB plates containing 50 μg / ml ampicillin overnight at 37 ° C., single colonies carrying the pcDNA5FRT / C9IgH vector were picked and incubated with 2 ml of 50 μg / ml ampicillin LB for 18 hours at 37 ° C. Following incubation, plasmid vectors were prepared with GenElute ™ Plasmid Miniprep Kit (Sigma-aldrich).

  1 μl of DNA (pSecTag2A / C9L or pcDNA5FRT / Cκ) was mixed with 5 μl 10 × NE Buffer 2, 0.5 μl 100 × BSA, 1 μl AscI, 1 μl KasI, and distilled water was added to make 50 μl. The mixture was incubated at 37 ° C. for 2 hours and the plasmid vectors pSecTag2A / C9L and pcDNA5FRT / Cκ were digested with the restriction endonucleases AscI and KasI. The product after restriction digestion was subjected to agarose gel electrophoresis. The gel was stained with ethidium bromide and the band containing DNA was detected under ultraviolet light. The DNA fragment was purified with Wizard SV Gel and PCR cleanup system.

  C9 VL DNA containing restriction sites digested with AscI and KasI was ligated into pcDNA5FRT / Cκ vector digested with enzymes, and the ligation product was transfected into competent bacteria (E. coli JM109) as described above. After incubation on LB plates containing 50 μg / ml ampicillin overnight at 37 ° C., single colonies carrying pcDNA5FRT / C9L vector were picked and incubated with 2 ml of 50 μg / ml ampicillin LB for 18 hours at 37 ° C. Following incubation, plasmid vectors were prepared with GenElute ™ Plasmid Miniprep Kit (Sigma-aldrich).

  1 μl of DNA (pcDNA5FRT / C9IgH or pcDNA5FRT / C9L) was mixed with 5 μl 10 × NE Buffer 2, 0.5 μl 100 × BSA100, 1 μl NheI, 1 μl NotI to make 50 μl. The mixture was incubated at 37 ° C. for 2 hours and digested with the restriction endonucleases NheI and NotI. The product after restriction digestion was subjected to agarose gel electrophoresis, and then the gel was stained with ethidium bromide, and a band containing DNA was detected under ultraviolet light. DNA fragments were purified with Wizard SV Gel and PCR cleanup system.

  The resulting restriction digested DNA fragment NheI-leader-VH-Gc-TM-IRES-Not I was ligated to the restriction digested pcDNA5FRT / C9L plasmid. The ligation product was transfected into competent bacteria (E. coli JM109) as described above. After incubation with 50 μg / ml ampicillin-containing LB plates overnight at 37 ° C., single colonies carrying the pcDNA5FRT / C9 IgG vector were picked and incubated with 2 ml of 50 μg / ml ampicillin-containing LB for 18 hours at 37 ° C. Following incubation, plasmid vectors were prepared with GenElute ™ Plasmid Miniprep Kit (Sigma-aldrich).

[5] Expression of all IgG molecules (1) Transfection using FLP-In system Transfection was performed using an artificial lipid carrier (provided by Fukuoka Industrial Technology Center). Flp-In-CHO cells were seeded in 60 mm culture dishes the day before transfection so that the confluence was 70-80% on the day of transfection. The Flp-In-CHO cell line is a cell that contains a single integrated FRT site and stably expresses the lacZ-zeocin fusion gene encoded downstream of the FRT site. After Flp-In vector DNA transfection with pOG44, followed by hygromycin treatment, cells in which Flp-In has been inserted into the FRT site acquire hygromycin resistance and lack lacZ-zeocin.

  500 μl Ham F12 medium and 10 μg DNA (1 μg pcDNA5FRT / IgG and 9 μg pOG44 for Flp-In recombination) were gently mixed to prepare a transfection reagent mixture. 50 μl of artificial lipid carrier was added to the medium-DNA mixture and incubated for 15 minutes at room temperature. While the complex was formed, the culture medium was replaced with fresh 5 ml Ham F12 containing 10% FBS. The transfection reagent mixture was added to the culture dish.

  Flp-In recombination was confirmed using pcDNA5FRT / GFP. pcDNA5FRT / GFP was transfected into Flp-In-CHO cells with or without pOG44 and cultured in Ham's F12 medium containing 10% FBS and hygromycin. Cells co-transfected with Flp-In vector and pOG44 acquired hygromycin resistance and fluoresced indicating GFP expression. To confirm region-specific recombination mediated by the Flp-In system, β-galactosidase staining was performed. Cells stably expressing GFP after Flp-In recombination lacked lacZ activity. These results indicated that the pcDNA5 FRT vector was specifically inserted into the FRT site by the Flp-In system.

(2) Flow cytometry analysis The expression of IgG in Flp-In-CHO cells transfected with pcDNA5FRT / IgG plasmid was measured by flow cytometry. Cells (2 × 10 ⁵ ) were collected in 1.5 ml Eppendorf tubes and washed with PBE. Cells were incubated with 2 μl PE-conjugated mouse anti-human IgG (H) antibody (Beckman Coulter) or 2 μl PE-conjugated mouse anti-human Igκ antibody (Beckman Coulter) in 50 μl PBE. After 15 minutes incubation at 4 ° C., the cells were washed with PBE and resuspended with 500 μl PBE. The fluorescence emitted by the cells was then measured with a FACSAria ™ Cell Sorter (BD Biosciences). As a result of flow cytometry analysis of Flp-In-CHO cells, the fluorescence intensity of PE increased, indicating that H and L chains were expressed on the cell membrane surface (FIG. 6). Next, cells were treated with FITC-labeled FcεRIα to measure membrane IgG antigen recognition. IgG expressing cells specifically bound to the antigen (FIG. 7).

(3) Cre / loxP recombination As described above, plasmid DNA pCEP4 / Cre encoding Cre recombinase was transfected into cells expressing membrane-type IgG on the surface. Recombination mediated by Cre at the loxP site was expected to change the form of IgG expression. The presence or absence of transgene recombination in cells by Cre / loxP recombination was analyzed by extracting total RNA, synthesizing cDNA, and subjecting it to PCR by the following procedure.

(4) Extraction of total RNA from Flp-In-CHO cells Total RNA from Flp-In-CHO cells was prepared with Fast PureU ™ RNA kit (TAKARA). Cultured Flp-In-CHO cells were collected and suspended in 350 μl of lysis buffer containing 2-mercaptoethanol. After pipetting, 50 μl lysis buffer was added and vortexed for 15 seconds. After adding 170 μl of ethanol (> 99%), the mixture was vortexed for 1 minute and transferred to a cartridge. After centrifugation at 8000 × g for 1 minute, 750 μl of washing buffer was added and centrifuged. The washing process was repeated 3 times. The cartridge was transferred to a collection tube and 50 μl of elution buffer was added. After 4 minutes incubation at room temperature, the tubes were centrifuged at 8000 × g for 1 minute and the eluted fractions were collected as RNA solution.

(5) cDNA synthesis 1 μg of total RNA obtained by synthesizing cDNA from total RNA using the Reverse Tra Ace qPCR RT kit (TOYOBO) was denatured at 65 ° C. for 5 minutes and transferred to ice. After cooling for 5 minutes, the RNA solution was mixed with 2 μl of 5 × RT buffer, 0.5 μl of primer mix, 0.5 μl of RT enzyme, and distilled water was added to make 10 μl. The mixture was reacted at 37 ° C. for 15 minutes, 98 ° C. for 5 minutes and 4 ° C.

(6) PCR
PCR was performed with KOD FX DNA polymerase (TOYOBO). Flp-In-CHO cells (1 × 10 ⁵ cells) were collected in an Eppendorf tube and washed with PBS. The pellet was suspended in 100 μl PBS and used as a template for genomic PCR. 0.5 μl cDNA or 1 μl cell solution as template, 5 μl 2 × KOD-FX buffer, 2 μl 2 mM dNTPs, 0.3 μl each of sense primer IgGc-md (SEQ ID NO: 9) and antisense primer IRES-md (SEQ ID NO: 10), 0.2 μl of KOD FX DNA polymerase was added to distilled water to make 10 μl, and a PCR reaction solution was prepared. The primers were designed to detect DNA before and after the gene encoding the deleted transmembrane domain. 30 cycles of PCR reaction were performed with 94 ° C. for 15 seconds, 60 ° C. for 30 seconds, and 68 ° C. for 1 minute as one cycle. The amplification product was mixed with 1 μl of a staining solution, subjected to 1.5% agarose gel electrophoresis, the gel was stained with ethidium bromide, and a band containing DNA was detected under ultraviolet light.

  A transgene successfully recombined with the original transgene encoding the transmembrane domain was expected to give amplified fragments of 696 bp and 464 bp, respectively (FIG. 8 (a)). That is, it was considered that three bands of 696 bp derived from the original transgene, 556 bp derived from the mature mRNA, and 464 bp derived from the recombined Cre / loxP genomic region were detected. In fact, a shorter 464 bp long recombinant DNA fragment was detected in pCEP4 / Cre transfected cells (FIG. 8 (b)). Furthermore, splicing reactions of these transgenes were also analyzed by PCR. The template DNA state scheme and the PCR amplification region are schematically shown in FIG. 9 (a). After recombination, an intron splicing reaction without a transmembrane domain was confirmed at around 300 bp (FIG. 9 (b)).

(7) ELISA
Next, the amount of IgG in the culture supernatant is measured by ELISA (sandwich ELISA) to confirm whether the cells with the recombined transgene expressed and secreted total IgG in the culture medium as expected. did. A 96-well microtiter plate (Nunc) was coated with 100 μl of a 1000-fold diluted IgG antibody (TAGO) in 0.1 M sodium carbonate buffer (pH 9.6) and incubated at 37 ° C. for 2 hours. After that, it was blocked with 1% fish gelatin (FG) in PBS for 2 hours at 37 ° C. After washing the plate 3 times with PBST, 50 μl of serially diluted culture supernatant was added and incubated at 37 ° C. for 2 hours. After washing the wells with PBST, the 2000-fold diluted HRP-conjugated goat anti-human IgG was added to the captured IgG, followed by incubation at 37 ° C. for 2 hours. After washing 3 times with TPBS, 100 μl substrate solution, 0.1 M citrate buffer containing 0.003% H ₂ O ₂ and 0.3 mg / ml 2,2′-azido-bis (3-ethylbenzothiazolidine-6-sulfone Acid) Diammonium salt (ABTS: Wako Pure Chemical Industries) was added and incubated for 20 minutes. Absorbance at 405 nm was measured using an ELISA reader.

  As a result, IgG was clearly detected in the culture supernatant after transfection of cells with pCEP4 / Cre (FIG. 10). On the other hand, IgG in the Flp-In-CHO cell lysate was also measured by ELISA. IgG secretion was not detected in Flp-In-CHO cells transfected with pcDNA5FRT / IgG alone (FIG. 10). However, intracellular IgG molecules were detected (FIG. 11). These results confirmed that Cre-mediated recombination can convert antibody-presenting cells into antibody-secreting cells.

[Discussion]
Microorganisms currently used in antibody engineering have problems with protein folding, post-translational modifications and codon usage. In order to exert the biological activity of the antibody, it is necessary to convert the scFv fragment into a whole IgG molecule and express it in mammalian cells. These serious problems often arise when antibody genes selected using phage display methods that express proteins on microbial platforms are expressed in mammalian expression systems.

  The novel animal cell display method of the present invention systematically presents human total IgG on the surface of animal cells, and enables production and screening of antigen-specific human monoclonal IgG with appropriate folding and post-transcriptional modification. Proved. A plasmid vector for IgG expression was constructed with the variable region gene of the C9 anti-FcεRIα antibody and transfected into Flp-In-CHO cells. After selection with antibiotics, cell surface heavy and light chain expression was detected by flow cytometry, and antibodies expressed on the cell surface captured FcεRIα antigen. These results indicated that the bicistronic vector can express heavy and light chains from mRNA transcribed using the same promoter to produce membrane-type whole IgG on the cell surface. .

  In order to convert the whole membrane IgG displayed on the cell surface into a secreted IgG form, the downstream of the heavy chain constant region of the vector is removed after screening by site-specific recombination with Cre recombinase. It was designed to break into the transmembrane region gene between two loxP sequences. A pcDNA5FRT / IgG vector with the loxP sequence hidden in the intron was constructed and used for transfection. This is because the conversion system expected that secreted IgG might not be able to form the native structure by adding unnecessary amino acids from the transcribed loxP element. As mentioned above, all human antibodies were displayed on the cell membrane by transfection of pcDNA5FRT / IgG vector. The displayed IgG was successfully converted to secreted form by Cre / loxP recombination. If artificially constructed intron splicing, including loxP sequences, does not work well, IgG will not be presented on the cell membrane surface of animal cells due to incorrect positioning of the transmembrane region downstream of the heavy chain constant region.

  It was also confirmed that the splicing reaction of the synthetic intron occurred at the expected appropriate position. Since a sequence similar to the original 3'SS present in the transmembrane region may be recognized as 3'SS, irregular splicing may occur. However, according to ELISA results, IgG secretion due to irregular splicing without Cre / loxP recombination was not observed. The rate of splicing can alternatively be controlled by inducing mutations near the 3'SS. That indicates that the efficiency of IgG secretion may be improved further by further studies. Promoter-activated production (PAP) systems have already been developed to enhance recombinant protein productivity in cultured cells by activating cytomegalovirus promoters via transcriptional activators such as oncogenes. Mammalian cells such as CHO cells are the primary system used for the production of the majority of therapeutic antibodies in the pharmaceutical industry.

  The PAP system has been shown to be very effective in CHO cells, and it would be possible to construct highly productive Flp-In-CHO cells. The IgG expression and secretion levels of this study may be increased by the combination of the PAP system and the animal cell display method of the present invention.

  Furthermore, the construction of the plasmid vector can be optimized. First, in the construction of the expression vector pcDNA5FRT / IgG, it was necessary to perform VL recombination by two-step recombination due to the presence of a KasI restriction digestion site in the heavy chain constant region and the IRES element. That is, variable region genes were inserted into separate vectors for the H and L chains, and these two vectors were ligated together to construct a complete expression plasmid vector. This problem can be solved by optimizing the plasmid vector. Second, the bicistronic expression of the IgG antibody gene affects the balance of heavy and light chain expression levels. The antibody heavy chain is not secreted in the free antibody heavy chain state unless it binds to the light chain into an Ig molecule. On the other hand, most light chains can be secreted as free monomers or dimeric molecules. In a bicistronic expression vector having an IRES, the translation efficiency of the second cistron gene located behind the IRES element is reduced to about 1/10 that of the first cistron gene. Therefore, it is considered that the efficiency of Ig molecule secretion can be increased by placing the light chain gene upstream of the IRES element and the heavy chain gene downstream.

  pcDNA5FRT / IgG was constructed for bicistronic expression using the heavy chain as the first cistron and the light chain as the second cistron. As noted above, reconstructing IgG expression vectors could improve the efficiency of recombination and total IgG production. Usually, a lot of work is required to produce a monoclonal antibody by the phage display method. Cells that express human total IgG on the cell membrane surface and show a high binding activity to the antigen are easily screened with a cell sorter. After screening, analysis of antibody production and antibody characteristics can be easily performed by converting membrane-type IgG into secretory IgG molecules using Cre / loxP recombination. Since these operations do not require sophisticated genetic recombination operations, robotization is possible, and the animal cell display method of the present invention is considered to be extremely useful industrially.

<Experimental example 2>
The human monoclonal C9 anti-FcεRIα antibody isolated from the in vitro immunized phage library clearly recognized FcεRIα and detected the receptor expressed on the surface of KU812 cells. However, the affinity of the antibody was not high. In Experimental Example 2, for the purpose of demonstrating the usefulness of the animal cell display method constructed in the above-mentioned Experimental Example 1 in in vitro affinity maturation of C9 antibody, CDRH3 of C9 antibody gene is randomly amplified and affinity optimized. The antibody was isolated by animal cell display method.

[Materials and methods]
[1] Mutagenesis of CDR3 region of C9 heavy chain variable region gene Mutation of CDR3 VH was carried out by PCR using a specific primer containing an NNS denatured codon in substantially the same manner as described above. The C9 anti-FcεRIα heavy chain constructed in pSecTag2A (pSecTag2A / hIgH C9) was used as a template for PCR. 50 ng of template, 3 μl of 5 mM C9 VHCDR3 primer (SEQ ID NO: 11), 3 μl of 5 mM T7-2 primer (SEQ ID NO: 12), 5 μl of 10 × KOD-plus buffer, 5 μl of 2 mM dNTPs, 2 μl of 25 mM MgSO ₄ , 1 μl Was mixed with KOD plus DNA polymerase, and distilled water was added to make 50 μl. Amplification was performed for 30 PCR cycles (94 ° C. for 15 seconds, 60 ° C. for 30 seconds, 68 ° C. for 1 minute). The PCR product was mixed with 5 μl of staining solution and subjected to 1.5% agarose gel electrophoresis. The gel was stained with ethidium bromide and the band containing DNA was detected under ultraviolet light. The DNA fragment was purified with Wizard SV Gel and PCR cleanup system.

[2] Library construction The amplified C9 VH CDR3 randomized (C9VHCDR3r) DNA and pcDNA5FRT / C9IgG vector were restriction digested with restriction endonucleases SfiI and XhoI. 1 μl of DNA was mixed with 5 μl of 10 × NE Buffer 2, 0.5 μl of 100 × BSA, 1 μl of SfiI, and distilled water was added to make 50 μl. After 2 hours incubation at 50 ° C., 1 μl of XhoI was added to the mixture and incubated at 37 ° C. for 2 hours. The product was mixed with 5 μl of staining solution and subjected to 1.5% agarose gel electrophoresis. The gel was stained with ethidium bromide and the band containing DNA was detected under ultraviolet light. The DNA fragment was purified with Wizard SV Gel and PCR cleanup system. C9VHCDR3r DNA containing restriction digests SfiI and XhoI was ligated into the restriction digest pcDNA5FRT / C9IgG vector.

  The resulting ligation product was transfected into competent bacteria (E. coli JM109) as described above. After overnight incubation at 37 ° C. on LB plates containing 50 μg / ml ampicillin, single colonies carrying the pcDNA5FRT / C9IgG-VHCDR3r vector were picked and incubated at 37 ° C. for 18 hours on 2 ml of LB containing 50 μg / ml ampicillin. After incubation, a plasmid vector was prepared using the GenElute ™ Plasmid Miniprep Kit (Sigma-aldrich). Plasmid vectors were transfected into Flp-In-CHO cells by lipofection.

  Flp-In-CHO cells were seeded in 60 mm dishes the day before transfection so that the confluency was 70-80% on the day of transfection. 500 μl Ham F12 medium and 10 μg DNA (1 μg pcDNA5FRT / IgG and 9 μg pOG44 for Flp-In recombination) were gently mixed to prepare a transfection reagent mixture. 50 μl of artificial lipid carrier was added to the medium-DNA mixture and incubated for 15 minutes at room temperature. During this time, the medium was replaced with fresh Ham F12 containing 5 ml of 10% FBS, and the transfection reagent mixture was added to the culture dish. Hygromycin was added to the media 48 hours after transfection to select antibiotic resistant cells.

In this way, a CHO cell surface display library was constructed, where endogenous antibody hot spots were randomized. There are 6 CDRs (3 in the heavy chain and 3 in the L) containing antigen binding in the antibody. CDR3 of the C9 VH gene was randomized by PCR with primers to mutate the CDR3 sequence (CACAACTGGTATCACTTTTGACTCC) to NNSNNSNSNSNSNSNSNSNSNSNS and replaced with the parent C9 VH gene of pcDNA5FRT / C9IgG. The CDRH3 randomized IgG (CDRH3r) library was transduced into E. coli cells and the library complexity was composed of 1.0 × 10 ⁶ independent clones. The library was then transfected into Flp-In-CHO cells with pOG44 and mutant IgG with artificially randomized hot spots was expressed on the cell surface. Isolation of CHO cell surface display library of mutant IgG with high reactivity to FcεRIα Transfection of one well in a 12-well plate resulted in 1 of transfected Flp-In-CHO cells after antibiotic selection. 3 × 10 ⁴ clones could be produced.

[3] Flow cytometry Transfected Flp-In-CHO cells (2 × 10 ⁵ cells) were collected in a 1.5 ml Eppendorf tube and washed with PBE. Cells were incubated in 50 μl PBE with 2 μl PE-conjugated mouse anti-human IgG (H) antibody (Beckman Coulter) or 2 μl PE-conjugated mouse anti-human Igκ antibody (Beckman Coulter). After 15 minutes incubation at 4 ° C., the cells were washed with PBE and resuspended with 500 μl PBE. The fluorescence emitted by the cells was measured with a FACSAria ™ Cell Sorter (BD Biosciences).

[4] Cell sorting related to fluorescence intensity Cells (1 × 10 ⁶ cells) were collected in a 1.5 ml Eppendorf tube and washed with 1 ml PBE. Cells were incubated in 250 μl PBE with 5 μmol FITC conjugated rFcεRIα. After 15 minutes incubation at 4 ° C., the cells were washed with PBE and resuspended in 2.5 ml PBE. Among the fluorescent cells, as shown in FIG. 12, each population of P1 and P2 was sorted into a 96-well plate by FACSAria ™ Cell Sorter (BD Biosciences) and cultured. Of all 960 wells sorted, 360 cells grew from a single cell as a clone.

[5] ELISA
The amount of IgG expressed in the cytoplasm was measured by ELISA (sandwich ELISA) in the same manner as described above.

[6] Direct ELISA
As described above, the cells producing IgG (having high reactivity with FcεRIα-presenting CHO cells) were selected by sorting the fluorescence-activated cells. After culturing each clonal cell to about 1.0 × 10 ⁶ cells, the cell was lysed with a lysis buffer, and the amount of immunoglobulin was measured by sandwich ELISA. A 96-well microtiter plate was coated with 100 μl of 10 μg / ml rFcεRIα in 0.1 M sodium carbonate buffer (pH 9.6) and incubated at 4 ° C. overnight. Subsequently, it was blocked with 1% FG in PBS for 2 hours at 37 ° C. After washing the plate 3 times with PBST, whole cell lysate containing 100 ng immunoglobulin was added and incubated at 37 ° C. for 2 hours. Wells were washed 3 times with TPBS. Bound IgG was detected by adding 2000-fold diluted HRP-conjugated goat anti-human IgG and then incubated at 37 ° C. for 2 hours. After washing 3 times with TPBS, 100 μl of substrate solution, 0.1 M citrate buffer containing 0.003% H ₂ O ₂ and 0.3 mg / ml 2,2 ′ azidobis (3-ethylbenzothiazoline-6-sulfonic acid ) Diammonium salt (ABTS; WAKO) was added and incubated for 20 minutes. Using an ELISA reader, the absorbance at 405 nanometers was measured.

  As a result, IgG production was confirmed in 38 clones among the 360 clones that grew. The immunoglobulins contained in the cell lysate were measured for reactivity to FcεRIα by direct ELISA. As a result, many clones producing IgG with higher affinity than the parent antibody C9 were sorted in both P1 and P2. (Figure 13). Further, the FR3, CDR3 and FR4 amino acid sequences (SEQ ID NO: 13) of the parent antibody C9 and the P101 and P217 obtained from P1, and P216 and P217 obtained from P1, respectively. 14, 15, 16).

  It should be considered that the embodiments and experimental examples disclosed herein are illustrative in all respects and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.

Claims (12)

  The Flp-In system is a vector designed to intervene genes encoding internal ribosome entry sites between two structural genes encoding different proteins, and to express these genes bicistronically. A method for animal cell display, which comprises transfecting an animal cell having a protein and expressing a protein expressed in a clone by introduction of one gene per cell on the cell membrane surface of the animal cell.   The structural gene upstream of the internal ribosome entry site is a structural gene encoding an antibody heavy chain, and the structural gene downstream of the internal ribosome entry site is a structural gene encoding the antibody light chain. Or the structural gene upstream of the internal ribosome entry site is a structural gene encoding a light chain of an antibody, and the structural gene downstream of the internal ribosome entry site is a structure encoding a heavy chain of the antibody The animal cell display method according to claim 1, wherein the antibody having a heavy chain and a light chain that are genes and are expressed bicistronically is displayed on a cell membrane surface of the animal cell.   The structural gene encoding the heavy chain encodes the variable region and the constant region in order from the upstream side, and the structural gene encoding the light chain encodes the variable region and the constant region in order from the upstream side. The animal cell display method according to 2.   4. The vector according to claim 2, wherein the vector is designed so that a gene encoding a transmembrane domain is interposed between a structural gene encoding a heavy chain and a structural gene encoding a light chain. Animal cell display method.   The vector is designed so that the loxP sequence on the 5 ′ end side of the gene encoding the transmembrane domain is hidden in the intron, and an extra amino acid sequence derived from the loxP sequence is not added to the constant region of the heavy chain. The animal cell display method according to claim 4.   The vector has a loxP site interposed between a structural gene encoding a heavy chain and a gene encoding a transmembrane domain, and between a gene encoding a transmembrane domain and a structural gene encoding a light chain, By transfecting the animal cell with the vector and then transfecting the animal cell with a vector containing a gene encoding Cre recombinase, or by introducing Cre recombinase into the animal cell with a lipid carrier, The animal cell display method according to claim 4 or 5, wherein the animal cell is designed such that a gene encoding a transmembrane domain between two loxP sites is removed from the animal cell.   The animal cell display method according to claim 6, wherein the antibody is converted from a membrane type to a secreted type by removing a gene encoding a transmembrane domain between two loxP sites in the animal cell.   The animal cell display method according to claim 1, wherein the antibody is human IgG.   The animal cell display method according to any one of claims 1 to 8, wherein the antibody is a human high affinity IgE receptor α chain antibody.   The animal cell display method according to claim 9, which is obtained by in vitro antigen sensitization using human juvenile basophil KU812F cells on human peripheral blood mononuclear cells.   The animal cell display method according to claim 1, wherein the animal cell is a CHO cell having a Flp-In system. Random amplification of the gene encoding the variable region of the heavy chain of the antibody;
The obtained amplification product is composed of a structural gene that sequentially encodes the heavy chain variable region and the heavy chain constant region from the upstream side, and a structural gene that sequentially encodes the light chain variable region and the light chain constant region from the upstream side. A structure that encodes the variable region of a heavy chain that is randomly mutated in a vector designed to intervene genes encoding internal ribosome entry sites and express these genes bicistronically. A step of constructing a vector incorporated as a gene,
Transfecting an animal cell having the Flp-In system with the vector and constructing a library of animal cells in which antibodies expressed in clones are introduced on the cell membrane surface by introducing one gene per cell;
Selecting animal cells expressing a high affinity antibody on the cell membrane from a library of animal cells by fluorescence activated cell sorting;
In vitro maturation of a fully human monoclonal antibody comprising secreting the fully human monoclonal antibody into selected animal cells.