JP2007525171A - Modification of antibody FcRn binding affinity or serum half-life by mutagenesis - Google Patents

Abstract

  The present invention provides that at least one amino acid from a heavy chain constant region selected from the group consisting of amino acid residues 250, 314 and 428 is substituted with another amino acid different from that present in the unmodified antibody. Thus, IgG class modified antibodies are provided that have altered binding affinity for FcRn and / or serum half-life compared to unmodified antibodies.

Description

  The present invention relates to the fields of immunology and protein engineering. In particular, the invention relates to modified antibodies of the IgG class that have a binding affinity for FcRn that has been altered as a result of one or more amino acid modifications within its Fc region, or an altered serum half-life.

An antibody is a protein that exhibits binding specificity for a particular antigen. Native (ie, naturally occurring or wild-type) antibodies are typically heterotetrameric glycoproteins of approximately 150,000 daltons, consisting of two identical light (L) chains and two identical heavy (H) chains. It is. As shown in FIG. 1, each light chain is linked to the heavy chain by one covalent disulfide bond, while the number of disulfide linkages varies between the heavy chains of different immunoglobulin isotypes. Each heavy chain has at one end a variable domain (V _H ) followed by a fixed number of constant domains. Each light chain has a variable domain (V _L ) at one end and a constant domain at the other end. The constant domain of the light chain is aligned with the first constant domain of the heavy chain.

  Certain portions of the variable domain vary widely in sequence between antibodies, and are responsible for the binding specificity of each particular antibody for that particular antigen. The constant domain is not directly involved in the binding of the antibody to the antigen and exhibits various effector functions. Depending on the amino acid sequence of the heavy chain CRG, antibodies or immunoglobulins may be assigned to different classes. In humans, there are five major classes (isotypes) of immunoglobulins: IgA, IgD, IgE, IgG and IgM, some of which are subclasses such as IgG1, IgG2, IgG3, and IgG4 and IgA1 and IgA2. It is further divided into (subtypes).

A schematic representation of the native IgG structure is shown in FIG. 1, where various portions of the native antibody molecule are shown. The heavy chain constant region includes C _H 1, the hinge region, C _H 2 and C _H 3. Papain digestion of the antibody produces two fragments, Fab and Fc. The Fc fragment consists of C _H 2, C _H 3 and part of the hinge region. The crystal structure of the human IgG1Fc fragment has been determined so far (Deisenhofer, Biochemistry 20: 2361-2370 (1981)). In human IgG molecules, Fc fragments are generated by papain cleavage of the hinge region N-terminal to Cys226. Thus, the human IgG heavy chain Fc region is usually defined as the extension from the amino acid residue at position 226 to the C-terminus (Kabat et al., “Immunologically Favored Protein Sequence”, 5th edition, Numbered according to the EU index of the US Health Organization, Bethesda. MD, (1991); the EU numbering scheme is used below).

  The Fc region is essential for the effector function of the antibody. Effector functions include initiation of complement dependent cytotoxicity (CDC), initiation of phagocytosis and antibody dependent cellular cytotoxicity (ADCC), and transport of antibodies across the cell barrier by transcytosis. Furthermore, the Fc region is extremely important for maintaining the serum half-life of IgG class antibodies (Ward and Ghetie, Ther. Immunol. 2: 77-94 (1995)).

Studies have shown that FC binding to the neonatal Fc receptor (FcRn) mediates the serum half-life of IgG antibodies. FcRn is a heterodimer consisting of a transmembrane α chain and a soluble β chain (β2-microglobulin). FcRn shares 22-29% sequence identity with class I MHC molecules and has a non-functional version of the MHC peptide binding groove. The α1 and α2 domains of FcRn interact with the CH ₂ and CH ₃ domains of the Fc region (Raghavan et al., Immunity 1; 303-315 (1994)).

  Models have been proposed for how FcRn can regulate the serum half-life of antibodies. As shown in FIG. 2, IgG is taken up by endothelial cells through nonspecific phagocytosis and then enters acidic endosomes. FcRn binds IgG in endosomes at acidic pH (<6.5) and releases IgG in the bloodstream at basic pH (> 7.4). Thus, FcRn salvages IgG from the lysosomal degradation pathway. If serum IgG levels are reduced, more molecules are available for IgG binding so that increased amounts of IgG are salvaged. Conversely, as serum IgG levels rise, FcRn becomes saturated and the proportion of phagocytosed IgG thus degraded increases (Ghetie and Ward, Annu, Rev. Immunol. 18: 739-766 (2000) ).

Consistent with the above model, the results of numerous studies support a correlation between the antibody's affinity for FcRn binding and serum half-life (Ghetie and Ward, ibid.). Notably, such a correlation has been extended to engineered antibodies with higher binding affinity for FcRn than their wild-type parent molecules.
Ghetie et al., Randomly mutagenized positions 252, 254, and 256 within the mouse IgG1 Fc-hinge fragment. One mutant showed a 3.5-fold higher affinity for mouse FcRn and a half-life of about 23% or 65% longer in the two mouse strains, respectively, compared to the wild type (Ghetie et al., Nat. Biotechnol. 15: 637-640 (1997)).

Shields et al., Used alanine scanning mutagenesis to modify residues within the Fc region of human IgG1 antibody. They found multiple mutants with higher binding affinity for human FcRn compared to wild type, but did not identify a mutation at position 250, 314 or 428 (Shields et al., J. Biol Chem. 276: 6591-6604 (2001)).
Martin et al., Proposed mutagenesis at a fixed number of positions within human IgG Fc to increase binding to FcRn, including positions 250, 314 and 428, among many others. However, none of the mutants proposed by Martin et al. Has been constructed or tested for binding to FcRn (Martin et al., Mol. Cell 7: 867-877 (2001)).

  Dall 'Acqua e al. Described random mutagenesis and screening of a human IgG1 hinge-Fc fragment phage display library against mouse FcRn. They disclosed random mutagenesis at positions 428-436, but did not identify mutagenesis at position 428 as having any effect on mouse FcRn binding affinity, and the wild-type methionine amino acid at this position was efficient. He stated that it was advantageous for good binding (Dall'Acqua et al., J. Immunol. 169-5171-5180 (2002)).

Kim et al. Mutagenized human IgG1 by amino acid substitution at position 253, position 310 or position 435 of the Fc region. They found that mutant Fc hinge fragments reduce serum half-life in mice compared to wild-type IgG1 Fc hinge fragments, and Ile253, His310 and His435 are central in regulating IgG serum half-life. (Kim et al., Eur. J. Immunol. 29: 2819-2825 (1999)).
Homick et al, showed that a single amino acid substitution at position 253 in the Fc region of a chimeric human IgG1 antibody accelerates clearance in the mouse body and improves immunoscintigraphy of solid tumors (Hormick et al , J. Nucl. Med. 41: 355-362 (2000)).

US Pat. No. 6,165,745 discloses a method for producing an antibody with reduced biological half-life by introducing mutations into the DNA segment encoding the antibody. The mutation includes an amino acid substitution at position 253, 310, 311, 433 or 434 of the Fc-hinge domain. The entire disclosure of US Pat. No. 6,165,745, as well as the entire disclosure of all other US patent references cited herein, is hereby incorporated by reference.
US Pat. Nos. 5,530,101; 5,585,089; 5.593,761; 5,693,762; and 6,180,370 disclose humanization of immunoglobulins.

US Pat. No. 6,277,375 B1 has an increased serum half-life in relation to wild IgG, including amino acid substitutions from threonine to leucine at position 252, serine to serine at position 254, or threonine to phenylalanine at position 256. Disclosed are compositions comprising mutant IgG molecules. Mutant IgG with an amino acid substitution at position 433, 435 or 436 is also disclosed.
US Patent Application No. 20020098193A1 and PCT Publication WO 97/34621 disclose mutant IgG molecules having at least one amino acid substitution in the Fc hinge region and having increased serum half-life in relation to IgG. Yes. However, no experimental support is provided for mutations at positions 250, 314 or 428.

US Pat.No. 6,528,624 included a human IgG Fc region comprising an amino acid substitution at one or more of amino acid positions 270, 322, 326, 327, 329, 331, 333 and 334 of the human IgG Fc region. Antibody variants are disclosed.
PCT publication WO 98/05787 discloses deletion or substitution of amino acids at positions 310 to 331 of the BR96 antibody to reduce its induced toxicity, but results in binding to modified FcRn Amino acid modifications are not disclosed.

  PCT publication WO00 / 42072 states that the numbering of residues in the Fc region is that of the EU index, and amino acid positions 238, 252, 253, 254, 255, 256, 265, 272, 286 in the Fc region, 288, 303, 305, 307, 309, 311, 312, 317, 340, 356, 360, 362, 376, 378, 380, 386, 388, 400, 413, 415, 424, 433, 434, 435, 436, Disclosed are polypeptides comprising variant Fc regions with altered FcRn binding affinity, including amino acid modifications at any one or more of positions 439 and 447 (Kabat et al., Supra). In the book).

  PCT Publication WO02 / 060919A2 has an increased half-life compared to the half-life of an IgG with a wild-type IgG constant domain, and one or more amino acid modifications are 251, 253, 255, 285-290, 308- Disclosed is a modified IgG comprising an IgG constant domain comprising one or more amino acid modifications in relation to a wild-type IgG constant domain at one or more of 314, 385-389 and 428-435 . However, no examples of mutations at positions 314 or 428 with altered binding to FcRn are disclosed.

  Martin, WL (Doctoral Dissertation entitled “Protein-Protein Recognition: Neonatal Fc Receptor and Immunoglobulin G”, California Institute of Technology (2001)) included positions 250 and 428, which can increase FcRn binding affinity. A theoretical mutation at multiple Fc positions in the rat gamma-2a constant region is proposed. Martin suggests, among other things, the possibility of substituting valine with isoleucine at position 250 or leucine with phenylalanine at position 428. Martin does not suggest any substitution for position 314. Martin has not demonstrated an increased binding affinity for FcRn for any of these proposed mutations.

  The publications described above have not shown that the serum half-life or FcRn binding affinity of IgG class antibodies can be altered by amino acid modifications at position 250, position 314 or position 428 of the Fc region. The present invention used molecular modeling to select Fc residues near the FcRn contact site that may have an effect on binding but may not be necessary for pH dependent binding. Amino acid modifications were made at position 250, 314 or 428 of the constant region of the IgG class immunoglobulin heavy chain. The serum half-life or FcRn binding affinity of the antibody containing the modification was altered and thus differed from that of the unmodified antibody.

  The present invention is based on the inventors' identification of multiple mutations in the constant domain of human IgG molecules that alter (ie increase or decrease) the affinity of IgG molecules for FcRn. The present invention provides modified antibodies having altered FcRn binding affinity and / or serum half-life in relation to the corresponding unmodified antibody. The in vivo half-life of antibodies and other biologically active molecules (ie, persistence in the subject's serum or other tissues) is an important factor in determining the amount and frequency of administration of the antibody (or any other pharmaceutical molecule). It is a clinical parameter. Accordingly, such molecules, including antibodies with increased (or decreased) half-life, are of great pharmaceutical importance.

The invention relates to a modified IgG constant domain (preferably derived from human IgG), wherein the IgG constant domain or fragment thereof is modified (preferably by amino acid substitution) to increase (or decrease) affinity for FcRn. It relates to a modified molecule (preferably an antibody) having an in vivo half-life that is increased (or decreased) by the presence of its FcRn binding moiety (preferably Fc or hinge-Fc domain).
In certain embodiments, the invention provides for the alteration of an amino acid residue at a position identified by structural studies as being directly or indirectly involved in the hinge-Fc domain interaction with the FcRn receptor. The in vivo half-life is extended (or shortened). It relates to a modified IgG class antibody.

In a preferred embodiment, the modified IgG class antibody is selected from the group consisting of daclizumab, fontolizumab, visilizumab and M200.
In a preferred embodiment, the constant domain (or fragment thereof) has a higher FcRc affinity at pH 6.0 than at pH 7.4. That is, the pH dependence of FcRn binding affinity mimics the wild type pH dependence. In an alternative embodiment, a modified antibody of the invention may exhibit a pH-dependent profile that has been altered in relation to an unmodified antibody. Such modified pH-dependent profiles can be useful in therapeutic or diagnostic applications.

In some embodiments, antibody modifications of the invention will alter FcRn binding and / or serum half-life without altering other antibody effector functions such as ADCC or CDC. In particularly preferred embodiments, the modified antibodies of the invention do not show any change in binding to Fc-gamma receptors or C1q. In an alternative embodiment, the antibody modification of the present invention may result in increased (or decreased) effector function as well as increased serum half-life. In particularly preferred embodiments, the modified antibodies of the invention may have increased (or decreased) ADCC activity as well as increased serum half-life.
It should be pointed out that the modification of the invention can also alter (ie increase or decrease) the bioavailability (eg mucosal surface or other target tissue) of the modified antibody (or other molecule). Should.

  In a preferred embodiment, the invention provides an amino acid that differs from that in which at least one amino acid from a heavy chain constant region selected from the group consisting of amino acid residues 250, 314 and 428 is present in an unmodified antibody. IgG class modified antibodies are provided that are substituted with residues. Preferably this substitution alters the FcRn binding affinity and / or serum half-life of the modified antibody in the context of an unmodified wild type antibody. The present invention further compares the amino acid residue 250 from the heavy chain constant region with glutamic acid or glutamine, or the amino acid residue 428 from the heavy chain constant region with phenylalanine or leucine, compared to an unmodified antibody. Modified antibodies with increased FcRn binding affinity and increased serum half-life are provided.

  The invention further includes (a) amino acid residue 250 from the heavy chain constant region substituted with glutamic acid and amino acid residue 428 from the heavy chain constant region substituted with phenylalanine; (b) from the heavy chain constant region Amino acid residue 250 is substituted with glutamine and amino acid residue 428 from the heavy chain constant region is substituted with phenylalanine; or (c) amino acid residue 250 from the heavy chain constant region is substituted with glutamine and the heavy chain constant Provided are modified antibodies having increased FcRn binding affinity and / or increased serum half-life compared to unmodified antibodies, wherein amino acid residue 428 from the normal region is replaced with leucine.

The present invention further provides a reduced FcRn binding affinity compared to an unmodified antibody, wherein amino acid residue 314 from the heavy chain constant region is replaced with another amino acid different from that present in the unmodified antibody. Provided are modified antibodies having a reduced serum half-life.
The present invention further provides that amino acid residue 250 from the heavy chain constant region is substituted with arginine, asparagine, aspartic acid, lysine, fernialanine, proline, tryptophan or tyrosine, or an amino acid residue from the heavy chain constant region. Reduced FcRn compared to unmodified antibody in which group 428 is replaced with alanine, arginine, asparagine, aspartic acid, cysteine, glutamic acid, glutamine, glycine, histidine, lysine, proline, serine, threonine, tyrosine, or valine Modified antibodies with binding affinity and / or reduced serum half-life are provided.

  The present invention also differs from that present in naturally occurring IgG class antibodies wherein at least one amino acid residue selected from the group consisting of residues 250, 314 and 428 is present, and thus naturally Also provided is an antibody having a constant region substantially identical to a naturally occurring IgG class antibody that alters the FcRn binding affinity and / or serum half-life of said antibody in relation to the generated antibody. In a preferred embodiment, a naturally occurring IgG class antibody comprises the heavy chain constant region of a human IgG1, IgG2, IgG2M3, IgG3 or IgG4 molecule. Similarly, in a preferred embodiment, amino acid residue 250 from the heavy chain constant region of an antibody having substantially the same constant region as a naturally occurring IgG class antibody is glutamic acid or glutamine, or the heavy chain constant region The derived amino acid residue 428 is phenylalanine or leucine. In other preferred embodiments, the antibody having a constant region that is substantially identical to a naturally occurring IgG class antibody has a glutamic acid residue at position 250 and a phenylalanine residue at position 428; or an amino acid residue 250 is glutamine and amino acid residue 428 is phenylalanine; or amino acid residue 250 is glutamine and amino acid residue 428 is leucine.

  In some embodiments, an antibody having a constant region substantially identical to a naturally occurring IgG class antibody constant region has an amino acid residue at position 314 that is different from that present in a naturally occurring antibody. Including and thus reducing FcRn binding affinity and / or reducing serum half-life in relation to naturally occurring antibodies. In embodiments, amino acid residue 314 is alanine, arginine, aspartic acid, asparagine, cysteine, glutamic acid, glutamine, glycine, histidine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, or valine. Antibodies are included. In one preferred embodiment, amino acid residue 314 is arginine.

  In other embodiments, an antibody having a constant region substantially identical to a naturally occurring IgG class antibody constant region is within the group consisting of arginine, asparagine, aspartic acid, lysine, phenylalanine, proline, tryptophan or tyrosine. Which contains an amino acid residue at position 250 selected from thus reducing FcRn binding affinity and / or reducing serum half-life in relation to naturally occurring antibodies. Similarly, the amino acid residue at position 428 is selected from the group consisting of alanine, arginine, asparagine, aspartic acid, cysteine, glutamic acid, glutamine, glycine, histidine, lysine, proline, serine, threonine, tyrosine, or valine. In the context of naturally occurring antibodies, thus reducing FcRn binding affinity and / or reducing serum half-life.

  The invention further substitutes at least one amino acid from a heavy chain constant region selected from the group consisting of amino acid residues 250, 314 and 428 with an amino acid different from that present in the unmodified antibody, Thus, there is provided a method of modifying an IgG class antibody comprising the step of altering the FcRn binding affinity and / or serum half-life of said unmodified antibody.

The present invention further includes
(A) at least one amino acid derived from a heavy chain constant region selected from the group consisting of amino acid residues 250, 314 and 428 is substituted with an amino acid different from that present in the unmodified antibody. Thus, an expression vector (preferably replicable) comprising an appropriate promoter operably linked to DNA encoding at least the constant region of an immunoglobulin heavy chain, thus causing alteration of FcRn binding affinity and / or serum half-life A suitable expression vector);
(B) transforming host cells with the vector; and (c) culturing the transformed host cells to produce the modified antibody;
There is also provided a method for producing an IgG class modified antibody having a modified binding affinity for FcRn and / or an altered serum half-life when compared to an unmodified antibody.

  Optionally, such methods further comprise preparing a second expression vector (preferably a replicable expression vector) comprising a promoter operably linked to the DNA encoding the complementary immunoglobulin light chain. And further comprising transforming the cell line with the second expression vector.

  The present invention also relates to pharmaceutical compositions using the modified immunoglobulins of the present invention (including immunoglobulins conjugated to toxins and radionuclides), proteins and other biologically active molecules having altered half-lives, and Also includes prevention and treatment methods. Also included are diagnostic methods using the modified immunoglobulins, proteins and other biologically active molecules of the present invention with altered half-lives. In preferred embodiments, the amino acid modifications of the present invention can be used to increase the serum half-life of therapeutic or diagnostic antibodies. For example, the present invention provides modified therapeutic or diagnostic antibodies of the IgG class that have an in vivo elimination half-life that is at least about 1.3 times longer than that of the corresponding unmodified antibody. A modified therapeutic or diagnostic antibody wherein at least one amino acid residue selected from the group consisting of residues 250, 314 and 428 is different from that present in the unmodified antibody. In preferred embodiments, the modified therapeutic or diagnostic antibody has an in vivo elimination half-life that is at least 1.3 times, 1.5 times, 1.8 times, 1.9 times or 2.0 times longer than that of the corresponding unmodified antibody. .

The present invention also provides modified therapeutic or diagnostic antibodies of the IgG class that have in vivo clearance that is at least about 1.3 times lower than that of the corresponding unmodified antibody. A modified therapeutic or diagnostic antibody wherein at least one amino acid residue selected from the group consisting of residues 250, 314 and 428 is different from that present in the unmodified antibody. In preferred embodiments, the modified therapeutic or diagnostic antibody is at least 1.3 times, 1.5 times, 1.8 times, 2.0 times, 2.3 times, 2.5 times, 2.8 times or 3.0 times that of the corresponding unmodified antibody. Has an in vivo clearance much more than double.
In a preferred embodiment, the therapeutic antibody is selected from the group consisting of daclizumab, fontlizumab, bicilizumab and M200.

The invention further provides modified therapeutic or diagnostic antibodies of the IgG class that have an in vivo area under a concentration-time curve that is at least about 1.3 times higher than that of the corresponding unmodified antibody. A modified therapeutic or diagnostic antibody wherein at least one amino acid residue selected from the group consisting of residues 250, 314 and 428 is different from that present in the unmodified antibody. In preferred embodiments, the modified therapeutic or diagnostic antibody is at least 1.3 times, 1.5 times, 1.8 times, 2.0 times, 2.3 times, 2.6 times, 2.8 times or 3.0 times that of the corresponding unmodified antibody. Has a very high in vivo elimination half-life.
In a preferred variant embodiment, the amino acid modifications of the present invention can also be used to reduce the serum half-life of therapeutic or diagnostic antibodies. Such therapeutic or diagnostic antibodies are well known in the art and are listed in the following description of the invention.

  The invention also provides a modified therapeutic antibody comprising a light chain amino acid sequence of SEQ ID NO: 118 and a heavy chain amino acid sequence selected from SEQ ID NOs: 119-128. The invention also provides a vector comprising a polynucleotide encoding one or more of those light or heavy chain amino acid sequences. Furthermore, the present invention provides a host cell comprising this vector.

  Further, the present invention provides a modified therapeutic antibody comprising a light chain amino acid sequence of SEQ ID NO: 129 and a heavy chain amino acid sequence selected from SEQ ID NOs: 130-134; a light chain amino acid sequence of SEQ ID NO: 135 and SEQ ID NO: 136 A modified therapeutic antibody comprising a heavy chain amino acid sequence selected from -140; and a light chain amino acid sequence selected from SEQ ID NO: 141 and a heavy chain amino acid sequence selected from SEQ ID NOs: 142-146 Supply therapeutic antibody. The invention also provides a vector comprising one or more heavy and / or light chain amino acid sequences of the modified therapeutic antibodies listed above; and the invention provides a vector of the vectors listed above. A host cell comprising either is provided.

I. Modified Antibodies with Modified FcRn Binding Affinity and / or Serum Half-Life In order to provide a more complete understanding of the present invention, some definitions are provided.
As used herein, the terms “immunoglobulin” and “antibody” mean a protein consisting of one or more polypeptides substantially encoded by immunoglobulin genes. The recognized immunoglobulin genes include kappa, lambda, alpha, gamma (γ1, γ2, γ3, γ4), delta, epsilon and mu constant region genes, as well as the myriad immunoglobulin variable region genes. A full-length immunoglobulin “light chain” (approximately 25 kd or 214 amino acids) is encoded by a kappa or lambda variable region at the NH2-terminus (approximately 110 amino acids) and a kappa or lambda constant region gene at the COOH terminus. The full-length immunoglobulin “heavy chain” (approximately 50 Kd or 446 amino acids) is one of the heavy chain variable region genes (approximately 116 amino acids) and other aforementioned constant region genes such as gamma (encoding approximately 330 amino acids). Coded by one.

One form of immunoglobulin constitutes the basic structural unit of an antibody. This form is a tetramer and consists of two identical immunoglobulin chain pairs, each with one light chain and one heavy chain. Within each pair, both the light and heavy chain variable regions are responsible for binding to one antigen and the constant region carries antibody effector functions. In addition to tetrameric antibodies, immunoglobulins include, for example, Fv, Fab and (Fab ′) ₂ and bifunctional hybrid antibodies (eg Lanzavecchia and Scheidegger, Eur. J. Immunol. 17: 105-111 (1987)). Various other forms and single strands, including Huston et al., Proc. Natl. Acad. Sci. USA, 85; 5879-5883 (1988), and Bird, incorporated herein by reference. et al., Science, 242: 423-426 (1998)). (See, generally, Hood et al., “Immunology”, 2nd edition, Benjamin, York (1984), and Hunkapiller and Hood, Nature, 323: 15-16 (1986)).

  The term “genetically modified antibody” means an antibody in which the sequence of amino acid residues has been changed from that of a native or wild-type antibody. Due to the degree of involvement of recombinant DNA technology, the present invention is not limited to amino acid sequence modifications found within natural antibodies. As described below, previously engineered antibodies can be redesigned according to the present invention to obtain desirable characteristics of altered FcRn binding affinity and / or serum half-life. There are many possible variants of modified antibodies useful in the present invention, from those that change as few as one or a few amino acids to complete redesigns of variable or constant regions, for example. Changes in the constant region will generally be made to improve or modify properties such as complement fixation, interactions with various Fc-gamma receptors and other receptor functional groups. Changes in the variable region will be made to improve antigen binding properties.

An antibody having a constant region that is substantially identical to a naturally occurring class IgG antibody constant region is that any constant region present is substantially identical to the amino acid sequence of the naturally occurring IgG class antibody constant region That is, an antibody that is at least about 85-90% and preferably at least 95% identical.
In many preferred uses of the invention, including in vitro detection assays and in vivo use of modified antibodies in humans, chimeras, primatized, modified according to the invention (ie, mutated), It may be preferred to use humanized or human antibodies.

  The term “chimeric antibody” means that the constant region is derived from an antibody of one species (typically human) and the variable region is derived from an antibody of another species (typically rodent) Refers to antibody. Methods for producing chimeric antibodies are well known in the art. For example, Morrison, Science 229: 1202-1207 (1985); Oi et al., Bio Techniques 4: 214-221 (1986); Gillies et al., See J. Immunol. Methods 125: 191-202 (1989); US Pat. Nos. 5,807,715; 4,816,567; and 4,816,397.

  The term “primatized antibody” means an antibody comprising a monkey variable region and a human constant region. Methods for producing primatized antibodies are known in the art. See, for example, US Pat. Nos. 5,658,570; 5,681,722; and 5,693,780, which are hereby incorporated by reference in their entirety.

  The term “humanized antibody” or “humanized immunoglobulin” includes at least one and preferably all complementarity-determining regions from a human framework, a non-human antibody, and any constant regions present. By an immunoglobulin that is substantially identical to a human immunoglobulin constant region, ie, at least about 85-90% and preferably at least 95% identical. Thus, with the exception of the CDRs, all portions of the humanized immunoglobulin are substantially identical to corresponding portions of one or more native human immunoglobulin sequences. Often, framework residues within the human framework regions will be substituted with the corresponding residue from the CDR donor antibody to alter, preferably improve, antigen binding. These framework substitutions are well known in the art, such as modeling CDR and framework residue interactions to identify framework residues important for antigen binding and unusual frames at specific positions. Identified by sequence comparison to identify work residues.

  See, for example, Queen et al., US Pat. Nos. 5,530,101; 5,585,089; 5,693,761; 5,693,762; 6,180,370, each of which is incorporated by reference in its entirety. Antibodies are all herein incorporated by reference in their entirety, eg, CDR-grafting (European Patent No. 239,400; PCT Publication No. WO91 / 09967; US Pat. Nos. 5,225,539; 5,530,101 and 5,585,089) veneering or Resurfacing (European Patent No. 592,106; European Patent No. 519,596; Padlan, Mol. Immunol., 28: 489-498 (1991); Studnicka et al., Prot. Eng. 7: 805-814 (1994); Roguska et al. Natl. Acad. Sci. 91: 969-973 (1994), and various techniques known in the art, including chain reshuffling (US Pat. No. 5,565,332) to humanize antibodies can do.

  For therapeutic treatment of human patients, fully human antibodies may be desirable. Human antibodies can be made by a variety of methods known in the art, including the above-described method of displaying the antibodies using antibody libraries derived from human immunoglobulin sequences. US Pat. Nos. 4,444,887; 4,716,111 and PCT publications WO98 / 46645; WO98 / 50433; WO98 / 24893; WO98 / 16654; WO96 / 34096; WO96 / 33735, each of which is incorporated herein by reference in its entirety. And see WO91 / 10741.

  Similarly, human antibodies can be produced using transgenic mice that are not capable of expressing functional endogenous immunoglobulins but are capable of expressing human immunoglobulin genes. To review this technique for producing human antibodies, see Lonberg and Huszar, Int. Rev. Immunol. 13: 65-93 (1995). For a detailed discussion of this technique for producing human antibodies and human monoclonal antibodies and protocols for producing such antibodies, see PCT publication WO 98/24893, herein incorporated by reference in its entirety. WO96 / 34096; WO96 / 33735; European Patent No. 0598877; U.S. Patent Nos. 5,413,923; 5,625,126; 5,633,425; 5,569,825; 5,661,016; 5,545,806; 5,814,318; In addition, it employs companies such as Abgenix, Inc. (Fremont, CA) and Medarex (Princeton, NJ) to provide human antibodies directed against selected antigens using techniques similar to those described above. Can be made.

Fully human antibodies that recognize selected epitopes can be generated using a technique called “inductive selection”. In this approach, selected non-human monoclonal antibodies, such as mouse antibodies, are used to induce selection of fully human antibodies that recognize the same epitope (Jespers et al., Biotechnology 12: 899-903 (1988).
As used herein, the term “modify” means “increase” or “decrease”.
The present invention replaces at least one amino acid from a heavy chain constant region selected from the group consisting of amino acid residues 250, 314 and 428 with a different amino acid residue from that present in the unmodified antibody. IgG-modified antibodies are provided.

Class IgG antibodies include IgG1, IgG2, IgG3 and IgG4 antibodies. The constant region of the heavy chain of an IgG molecule is shown in FIG. The numbering of the residues in the heavy chain is that of the EU index (Kabat et al., Supra). Replacement can be done at positions 250, 314 or 428 alone, but at any combination of positions 250 and 428 or positions 250 and 314, or positions 314 and 428, or positions 250, 314 and 428. Positions 250 and 428 are a preferred combination.
For each position, the substituting amino acid can be any amino acid residue that is different from that present at that position of the unmodified antibody.

  For position 250, the substituting amino acids include alanine, cysteine, aspartic acid, glutamic acid, phenylalanine, glycine, histidine, isoleucine, lysine, leucine, methionine, asparagine, proline, glutamine, arginine, serine, valine, tryptophan or tyrosine It can be (but is not limited to) any amino acid residue other than threonine.

  For position 314, the substituting amino acids include alanine, cysteine, aspartic acid, glutamic acid, phenylalanine, glycine, histidine, isoleucine, lysine, methionine, asparagine, proline, glutamine, arginine, serine, threonine, valine, tryptophan or tyrosine. It can be (but is not limited to) any amino acid residue other than leucine.

  For position 428, the substituting amino acids include alanine, cysteine, aspartic acid, glutamic acid, phenylalanine, glycine, histidine, isoleucine, lysine, leucine, asparagine, proline, glutamine, arginine, serine, threonine, valine, tryptophan or tyrosine. It can be (but is not limited to) any amino acid residue other than methionine.

  The present invention provides IgG class antibodies comprising at least one of the amino acid substitutions described above. For example, the present invention provides a mutated IgG2M3 constant region comprising two of the above substitutions at positions 250, 314 and / or 428. The amino acid sequences of some specific substitutions (ie mutations) of the constant region provided by the present invention are disclosed in Table 1 (SEQ ID NOs: 10-66) and FIG. 3B.

  In preferred embodiments, the present invention provides modified antibodies with altered serum half-life or FcRn binding affinity in relation to unmodified antibodies. The invention further replaces at least one amino acid from a heavy chain constant region selected from the group consisting of amino acid residues 250, 314, and 428 with a different amino acid residue from that present in the unmodified antibody. Thus, IgG class modified antibodies are provided that alter the FcRn binding affinity and / or serum half-life of the modified antibody relative to the binding affinity and / or serum half-life of the unmodified antibody. .

  The unmodified antibodies of the present invention include all species of natural antibodies. The term “natural antibody” means all antibodies produced by a host animal. Non-limiting examples of natural antibodies of the invention include transgenic rodents genetically engineered to produce human antibodies, including human, chicken, goat and rodents (eg, rats , Mice, hamsters and rabbits) (eg, Lonberg et al., WO 93/12227, incorporated herein by reference in its entirety); US Pat. No. 5,545,806; and Kucherlapati, et al., WO 91/10741; see US Pat. No. 6,150,584).

  Similarly, the unmodified antibody of the present invention includes a recombinant antibody having the same amino acid sequence as that of the natural antibody, or a genetically modified antibody having an amino acid sequence that is altered compared to the natural antibody. These can be made in any expression system, including both prokaryotic and eukaryotic expression systems, or using phage display methods (eg Dower et al., Which is incorporated herein by reference in its entirety). al., WO 91/17271 and McCafferty et al., WO 92/01047: See US Pat. No. 5,969,108).

  Unmodified antibodies of the present invention also include chimeric, primatized, humanized and human antibodies (see discussion above). As a result, the modified antibodies of the present invention are produced by substituting an amino acid residue at position 250, 314 or 428 in a humanized, primatized or chimeric antibody that was previously derived from a natural antibody. Is possible.

  Preferably, the chimeric antibody is derived from a variable region derived from a rodent and from a human such that it has a longer half-life and is relatively less immunogenic when administered to a human subject. Including a constant region. A primatized antibody comprises a variable region derived from a primate and a constant region derived from a human. Humanized antibodies typically comprise a donor antibody (eg, a mouse or chicken antibody) and at least one CDR from a heavy and / or light chain human framework. Occasionally, some amino acid residues within the human framework will be replaced by residues at equivalent positions in the donor antibody to ensure proper binding of the humanized antibody to its antigen. . Detailed guidelines for antibody humanization are disclosed in US Pat. Nos. 5,530,101; 5,585,089; 5,693,761,5,693,762 and 6,180,370, each incorporated herein by reference.

  Unmodified antibodies of the present invention can include genetically modified antibodies that are functionally equivalent to the corresponding natural antibodies. Unmodified antibodies that have been genetically altered to provide improved stability and / or therapeutic effects are preferred. Examples of modified antibodies include those with conservative substitutions of amino acid residues and one or more deletions or additions of amino acids that do not significantly alter antigen binding utility. Substitutions can range from changing or modifying one or more amino acid residues to complete redesign of a region as long as binding or functional utility is maintained. The antibody of the present invention can be modified after translation (for example, acetylation and phosphorylation), or can be modified synthetically (for example, attachment of a labeling group).

  The unmodified antibodies of the present invention can include antibodies having enhanced binding affinity for the antigen through genetic modification of the variable region (the United States of which is hereby incorporated by reference in its entirety). See patent 6,350,861). In one variant embodiment, the modified IgGs of the invention having a longer half-life compared to wild-type (or unmodified) IgG are also referred to as antigen binding sites, Fc receptor binding sites or complement binding sites. Such biologically active sites can also include IgGs that have been modified by genetic engineering to increase or decrease such activity relative to the wild type.

  The unmodified antibody and modified antibody of the present invention may be any of the recognized isotypes, but four IgG isotypes are preferred, and IgG1 and IgG2 are particularly preferred. Antibodies with constant regions mutated to have reduced effector function, eg, IgG2M3 and other IgG2 mutations described in US Pat. No. 5,834,597 (incorporated in its entirety) The body is included. In a preferred embodiment, the unmodified antibody R and the modified antibody in the present invention comprise the heavy chain constant region of human IgG.

  The present invention can be applied to any antibody comprising an IgG class heavy chain constant region, preferably IgG1, IgG2, IgG2M3, IgG3 and IgG4. The heavy chain variable region of such an antibody can be derived from any selected antibody. Examples of antibodies disclosed herein include the heavy and light chain variable regions of human anti-hepatitis B virus antibody OST577 (Ehrlich et al., Hum. Antibody hybridoma 3: 2-7 (1992)), OST577-IgG1 and OST577-IgG2M3 containing the light chain constant region of human lambda-2 and the heavy chain constant regions of human IgG1 and IgG2M3, respectively, are included. Also disclosed herein is the heavy and light chain variable regions of the humanized anti-HLA-DR β chain allele antibody HulD10 (Kostelny et al., Int J. Cancer 93: 556-565 (2001 )), HulD10-IgG1 and HulD10-IgG2M3 comprising the light chain constant region of human kappa and the above heavy chain constant regions of human IgG1 and IgG2M3, respectively.

  Further exemplifications of the invention disclosed herein include human IgG1, IgG2M3 (a genetically modified variant of IgG2), IgG3 and IgG4 antibodies that exemplify alteration of serum half-life of IgG class antibodies Mutants. The constant region of the heavy chain of IgG2M3 is derived from that of IgG2 by replacing residues 234 and 237 of the IgG2 heavy chain constant region with alanine. The generation of the heavy chain constant region of IgG2M3 is disclosed in US Pat. No. 5,834,597, incorporated herein by reference.

In general, the modified antibodies of the invention bind antigen (preferably as determined immunospecifically, ie by immunoassay methods well known in the art for assaying specific antigen-antibody binding. Including any immunoglobulin molecule containing FcRn-binding fragments. Such antibodies include polyclonal, monoclonal, bispecific, multispecific, human, humanized, chimeric antibody, single chain antibody, Fab fragment, F (ab ′) ₂ fragment, disulfide bonded Fvs, and Either a V _L or V _H domain or even a complementary determining region (CDR) that specifically binds an antigen engineered to contain or fused to this domain, optionally FcFn binding domain Containing fragments.

The modified IgG molecules of the invention can include any given animal IgG subclass. For example, in humans, the IgG class includes IgG1, IgG2, IgG3 and IgG4, the mouse IgG class includes IgG1, IgG2a, IgG2b and IgG3; the rat IgG class includes IgG1, IgG2a, IgG2b, IgG2c And IgG3. Certain IgG subclasses, for example rat IgG2b and IgG2c, have been found to have a higher clearance rate than, for example, IgG1 (Medesan et al., Eur. J. Immunol, 28: 2092-2100 (1998)). . Thus, when using an IgG subclass other than IgG1, one or more of the residues in the C _H 2 and C _H 3 domains that differ from the IgG1 sequence, in particular, are replaced with those of IgG1, thus other types of It may be advantageous to increase the in vivo half-life of IgG.

  The immunoglobulin of the present invention may be derived from any animal including birds and mammals. Preferably, the antibody is a human, rodent, donkey, sheep, rabbit, goat, guinea pig, camel, horse or chicken. As used herein, a “human” antibody includes an antibody having the amino acid sequence of a human immunoglobulin, from a human immunoglobulin library or as described below and by Kucherlapati et al. Antibodies isolated from animals that are transgenic for one or more human immunoglobulins as described in US Pat. No. 5,939,598 are included.

  Furthermore, the modified antibodies of the invention may be monospecific, bispecific, trispecific or of higher multispecificity. Multispecific antibodies can be specific for different epitopes of the polypeptide, or can be specific for heterologous epitopes, such as heterologous polypeptides or solid support materials. For example, PCT publications WO93 / 17715; WO92 / 08802; WO91 / 00360; WO92 / 05793; Tutt et al., J. Immunol. 147: 60-69 (1991); US Pat. Nos. 4,474,893; 4,714,681; No .; Kostelny et al., J. Immunol. 148: 1547-1553 (1992).

  The modified antibodies of the present invention may be obtained by covalent attachment of any type of molecule to the antibody in any other way, i.e., without preventing the antibody from binding the antigen and / or generating an anti-idiotypic response by covalent attachment. Includes derivatives to be modified. For example, although not limiting, antibody derivatives may be directed against, for example, glycosylation, acetylation, pegylation, phosphorylation, amidation, derivatization with known protecting / blocking groups, proteolytic cleavage, cellular ligands or other proteins Includes antibodies modified by linkage or the like. Any of a number of chemical modifications can be performed by known techniques including (but not limited to) specific chemical resolution, acetylation, formylation, metabolic tunicamycin synthesis, and the like. . In addition, the derivative may contain one or more conventional amino acids.

  Monoclonal antibodies useful in the present invention can be prepared using a variety of techniques known in the art, including the use of hybridoma, recombinant and phage display techniques or combinations thereof. For example, monoclonal antibodies are known in the art, eg, Harlow and Lane “Antibodies: Laboratory Manual”, Cold Spring Harbor Laboratory Press, New York (1988); Hammerling et al., “Monoclonal antibodies and T-cell hybridomas. Can be produced using hybridoma technology, including those taught within Elsevier New York (1981), p563-681, both of which are hereby incorporated by reference in their entirety. .

  The term “monoclonal antibody” as used herein is not limited to antibodies produced through hybridoma technology. The term “monoclonal antibody” means an antibody derived from a single clone, including any eukaryotic, prokaryotic, or phage clone, and not the method of production thereof.

  Production and screening methods for specific antibodies using hybridoma technology are routine and well known in the art. In a non-limiting example, mice can be immunized with the antigen of interest or cells expressing such antigen. Once an immune response is detected, eg, an antibody specific for that antigen is detected in mouse serum, the mouse spleen is harvested and splenocytes are isolated. The splenocytes are then fused to any suitable myeloma cells by well known techniques. Hybridomas are selected and cloned by limiting dilution. The hybridoma clones are then assayed by methods known in the art for cells that secrete antibodies capable of binding antigen. Ascites fluid, which generally contains high levels of antibodies, can be generated by inoculating mice with a positive hybridoma clone intraperitoneally.

Antibody fragments that recognize specific epitopes can also be useful in the present invention and can be generated by well-known techniques. For example, Fab and F (ab ′) by proteolytic cleavage of immunoglobulin molecules using enzymes such as papain (to produce Fab fragments) or pepsin (to produce F (ab ′) ₂ fragments) _Two fragments can be produced. The F (ab ′) ₂ fragment contains the complete light and heavy chain variable regions, the CH1 region and the hinge region.

  For example, antibodies can be generated using various display methods known in the art, including phage display. In the phage display method, the functional antibody domain is displayed on the surface of the phage particle carrying the polynucleotide sequence encoding it. In certain embodiments, such phage can be utilized to display antigen binding domains such as Fab and Fv expressed from repertoire or combinatorial antibody libraries (eg, human or mouse) or Fv stabilized with disulfide bonds. Phage expressing an antigen binding domain that binds the antigen of interest can be selected or identified with the antigen, eg, using labeled antigen or antigen bound or captured to a solid surface or bead. The phages used in these methods are typically linear phages, including fd and MB. The antigen binding domain is expressed as a recombinantly fused protein to either the phage gene III or gene VIII protein.

  Alternatively, the modified FcRn binding portion of the immunoglobulins of the invention can be expressed in a phage display system as well. Examples of phage display methods that can be used to make the immunoglobulins or fragments of the invention include Brinkman et al., J. Immunol. Methods 182; 41, each of which is incorporated herein by reference in its entirety. -50 (1995); Ames et al., J. Immunol, Methods 184: 177-186 (1995); Kettleborough et al., Eur. J. Immunol. 24: 952-958 (1994); Persic et al., Gene 187: 9-18 (1997); Burton et al., Advances in Immunology 57: 191-280 (1994); PCT Application No. PCT / GB91 / 01134; PCT Publications WO90 / 02809, WO91 / 10737; WO92 / 01047 WO92 / 18619; WO93 / 11236; WO95 / 15982; WO95 / 20401; U.S. Patent Nos. 5,698,426; 5,223,409; 5,403,484; 5,580,717; 5,427,908; 5,750,753; 5,821,047; 5,571,698; 5,427,908; Including those disclosed in.

As described in the above references, after phage selection, the antibody coding region from the phage is isolated and used to generate whole antibodies, including human antibodies, or any other desired fragment, For example, it is expressed in any desired host, including mammalian cells, insect cells, plant cells, yeast and bacteria, as detailed below. For example, techniques for recombinantly producing Fab, Fab ′, and F (ab ′) ₂ fragments, each of which is incorporated by reference in its entirety, PCT publication WO 92/22324; Mullinax et al., Bio Techniques 12: 864-869 (1992); Sawai et al., Amer. J. Reprod. Immunol. 34: 26-34 (1995); and Better et al., Science 240: 1041-1043 (1988). It can also be utilized using methods known in the art such as those described. Examples of techniques that can be used to produce single chain Fvs and antibodies include US Pat. Nos. 4,964,778 and 5,258,498; Huston et al., Methods in Enzymology, 203: 46-88 (1991); Shu et al. , Proc. Natl. Acad. Sci. 90: 7995-7999 (1993); and Skerra et al., Science 240: 1038-1040 (1988).

  In certain embodiments, the modified antibodies have therapeutic and / or prophylactic uses in vivo. Examples of such modifiable therapeutic and prophylactic antibodies include SYNAGIS ™ (Medimmune, MD), a humanized anti-respiratory syncytial virus (RSV) monoclonal antibody; treatment of patients with metastatic breast cancer HERCEPTIN ™ (Trastuzumab) (Genentech, CA), a humanized anti-HER2 monoclonal antibody for use; REMICADE ™ (infliximab), a chimeric anti-TNF-α-monoclonal antibody for the treatment of patients with Crohn's disease ) (Centocor, PA); REOPRO ™ (absiximab) (Centcor), an anti-glycoprotein IIb / IIIa receptor on platelets to prevent clot formation; immunosuppressive human for prevention of acute allogeneic transplant rejection This includes, but is not limited to, the typed anti-CD25 monoclonal antibody ZENA-PAX ™ (Daclizumab) (Roche Pharmaceuticals, Switzerland).

Other examples include humanized anti-CD18F (ab ′) 2 (Genetech); humanized anti-CD18F (ab ′) 2 CDP860 (Celltech, UK); anti-HIV gp120 antibody fused to CD4 PRO542 (Progenics / Genzyme Transgenis); OSTAVIR ^™ (Protein Design Labs / Novartis), a human anti-hepatitis B virus antibody; PROTOVIR ^™ (Protein Design Labs / Novaritis), a humanized anti-CMV IgG1 antibody; anti-CD14 Antibody IC14 (ICOS); Humanized anti-VEGF IgG1 antibody AVASTIN ^™ (Genetech); Chimeric anti-EGFR IgG antibody ERBITUX ^™ (Im Clone Systems); Humanized anti-αVβ3 integrin antibody VITAXIN ^™ (Applied Molecular Evolutions / Medimmune); Humanized anti-CD52 IgG1 antibody, Campath-1H / LDP-O3 (Leukosite); Humanized anti-CD33 IgG antibody, ZAMYL ^™ (Protein Design Labs / Kanebo); Chimeric anti-CD20 IgG1 antibody RITUXAN ^TM (IDEC Pharmaceuticals / Genentech, Roche / Zenyaku); humanized anti-CD22IgG anti LYMPHOCIDE ^TM (Immuno medics) which is the body;

Humanized anti-HLA-DR antibody REMITOGEN ^™ (Protein Design Labs); human anti-IL8 antibody ABX-IL8 (Abgenix); humanized IgG1 antibody RAPTIVA ^™ (Genetech / Xoma); human type Anti-ICAM3 antibody ICM3 (ICOS); Primate anti-CD80 antibody IDEC-114 (IDEC Pharmaceuticals / Mitsubishi); Humanized anti-CD40L antibody IDEC-131 (IDEC / Eisai); Primate anti-antibody -CD4 antibody IDEC151 (IDEC); primatized anti-CD23 antibody IDEC-152 (IDEC / Seikagaku); humanized anti-CD3 IgG NUV10N ^TM (Protein Design Labs); humanized anti-complement Factor 5 (C5) antibody 5G1.1 (Alexion Pharmaceuticals): human anti-TNF-α antibody HUMIRA ^™ (CAT / BASF); humanized anti-TNF-α Fab fragment CDP870 (Celltech); primate Anti-CD4 IgGl antibody, IDEC-151 (IDEC Pharmaceuticals / Smith-Kline Beecham); MDX-CD4, human anti-CD4 IgG antibody (Medarex / Eisai / Genmab); anti-TNF-αIgG4 anti-humanized CDP571 (Celltech) that is the body;

Humanized anti-α4β7 antibody LDP-02 (Leuko Site / Genetech); Human anti-CD4 IgG antibody Ortho Clone OKT4A (Ortho Biotech); Humanized anti-CD40L IgG antibody ANTOVA ^™ (Biogen); ANTEGREN ^™ (Elan), a humanized anti-VLA-4 IgG antibody; MDX-33 (Medarex / Centeon), a human anti-CD64 (FcγR) antibody; SCH55700, a humanized anti-IL-5IgG4 antibody Celltech / Schering); humanized anti-IL-5 and IL-4 antibody-240563 and cell-240683 (Smith Kline Beecham); humanized anti-IgEIgG1 antibody rhuMab-E25 (Genentech / Novartis / Tanox Biosystems); primate anti-CD23 antibody IDEC152 (IDEC Pharmaceuticals); chimeric anti-CD25 IgG1 antibody SIMULECT ^TM (Novartis Phamaceuticals); humanized anti-β2-integrin IgG antibody LDP-01 (Leukosite) ); CAT-152 (Cambridge Antibody Technology), a human anti-TGF-β2-antibody; and CorsevinM (C, a chimeric anti-factor VII antibody) entocor).

  The present invention allows the modification of these and other therapeutic antibodies to increase in vivo half-life and allows for the administration of less effective dosages and / or less frequent dosing of therapeutic antibodies. . Such modifications to increase in vivo half-life may also be useful to improve diagnostic immunoglobulins. For example, the increased serum half-life of a diagnostic antibody may allow for the administration of smaller doses to achieve sufficient diagnostic sensitivity. Alternatively, a reduced serum half-life may be advantageous in applications where rapid clearance of diagnostic antibodies is desired.

Disclosed herein is its heavy chain variable region (SEQ ID NO: 1) (OST577-VH) and constant region (SEQ ID NO: 2) (IgG2M3), with positions 250, 314 and 428 highlighted. -CH) and the amino acid sequence of OST577-IgG2M3, including the amino acid sequence of its light chain variable region (SEQ ID NO: 4) (OST577-VL) and constant region (SEQ ID NO: 5) (LAMDA2-CL) ( Figure 3A).
Disclosed herein is the amino acid sequence of the heavy chain constant region of OST577-IgG1 (SEQ ID NO: 3), with positions 250, 314, and 428 highlighted (FIG. 3C).

  The present invention provides increased FcRn binding compared to an unmodified antibody, wherein the amino acid residue 250 or 428 from the heavy chain constant region is replaced with another amino acid residue different from that present in the unmodified antibody. Modified antibodies with affinity and / or increased serum half-life are provided. Preferably, amino acid residue 250 from the heavy chain constant region is substituted with glutamic acid or glutamine. Alternatively, amino acid residue 428 from the heavy chain constant region is substituted with phenylalanine or leucine.

  In one example, the unmodified antibody comprises a heavy chain constant region of an IgG1 or IgG2 or IgG2M3 molecule, including but not limited to OST577-IgG2M3 or OST577-IgG1. IgG1, IgG2 and IgG2M3 have a threonine residue at position 250 and a methionine residue at position 428. Preferably, the threonine residue at position 250 is substituted with glutamic acid (T250E) or glutamine (T250Q), and the methionine residue at position 428 is substituted with phenylalanine (M428F) or leucine (M428L). FIG. 3B discloses the amino acid sequence of a modified IgG2M3 heavy chain constant region with amino acid substitutions of T250E (SEQ ID NO: 13), T250Q (SEQ ID NO: 23), M428F (SEQ ID NO: 52) or M428L (SEQ ID NO: 57). is doing. FIG. 3C shows the weight of a modified IgG1 with an amino acid substitution of T250D (SEQ ID NO: 67), T250E (SEQ ID NO: 68), T250Q (SEQ ID NO: 69), M428F (SEQ ID NO: 70), or M428L (SEQ ID NO: 71). The amino acid sequence of the constant region of the chain is disclosed.

The present invention provides modified antibodies with increased FcRn binding affinity and / or increased serum half-life compared to unmodified antibodies. The amino acid modification can be any one of the following substitutions;
1) Amino acid residue 250 from the heavy chain constant region is substituted with glutamic acid, and amino acid residue 428 from the heavy chain constant region is substituted with phenylalanine.
2) Amino acid residue 250 from the heavy chain constant region is substituted with glutamine, and amino acid residue 428 from the heavy chain constant region is substituted with phenylalanine.
3) Amino acid residue 250 from the heavy chain constant region is replaced with glutamine, and amino acid residue 428 from the heavy chain constant region is replaced with leucine.

The amino acid sequence of the modified heavy chain constant region of IgG2M3 with a double amino acid substitution of T250E / M428F (SEQ ID NO: 72), T250Q / M428F (SEQ ID NO: 73) or T250Q / M428L (SEQ ID NO: 74) is shown in FIG. 3B. It is disclosed.
The amino acid sequence of the modified heavy chain constant region of IgG1 with a double amino acid substitution of T250E / M428F (SEQ ID NO: 75) or T250Q / M428L (SEQ ID NO: 76) is disclosed in FIG. 3C.
Modified antibodies with double amino acid substitutions described at positions 250 and 428 display very high binding affinity for FcRn compared to that of unmodified antibodies.
In preferred embodiments of the invention, the FcRn binding affinity and / or serum half-life of the modified antibody is at least about 30%, 50%, 80%, 2 times, 3 times, 4 times, 5 times, 10 times. 15 times, 20 times, 25 times, 30 times, 40 times, 50 times, 60 times, 70 times, 80 times, 90 times or 100 times.

  The invention provides a reduced FcRn binding affinity compared to an unmodified antibody and / or wherein amino acid residue 314 from the heavy chain constant region is replaced with another amino acid that is different from that present in the unmodified antibody. Modified antibodies having a reduced serum half-life are provided. Modified antibodies with an amino acid substitution at position 314 have been shown to display reduced binding affinity, indicating that position 314 should be modified if a reduced serum half-life of one antibody is desired. Suggests. Preferably, the amino acid residue from the heavy chain constant region is alanine, arginine, aspartic acid, asparagine, cysteine, glutamic acid, glutamine, glycine, histidine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, or Substituted with valine. More preferably, the amino acid substitution is from leucine to alanine or arginine at position 314. As shown in the examples, the FcRn binding affinity of a modified OST577-IgG2M3 containing a leucine to arginine substitution is reduced to 11% of that of unmodified OST577-IgG2M3.

  As shown in FIG. 3B, L314A represents the amino acid sequence of the modified IgG2M3 heavy chain constant region with a leucine to alanine amino acid substitution at position 314 (SEQ ID NO: 29). L314R represents the amino acid sequence of a modified IgG2M3 heavy chain constant region with a leucine to arginine amino acid substitution at position 314 (SEQ ID NO: 42).

  In the present invention, (1) amino acid residue 250 derived from heavy chain constant region is substituted with arginine, asparagine, aspartic acid, lysine, phenylalanine, proline, tryptophan or tyrosine, or (2) derived from heavy chain constant region As compared to an unmodified antibody in which amino acid residue 428 is substituted with alanine, arginine, asparagine, aspartic acid, cysteine, glutamic acid, glutamine, glycine, histidine, lysine, proline, serine, threonine, tyrosine, or valine Modified antibodies are provided that have reduced FcRn binding affinity and / or reduced serum half-life. Preferably, amino acid residue 250 from the heavy chain constant region is substituted with aspartic acid, and amino acid residue 428 from the heavy chain constant region is substituted with glycine. Such amino acid substitutions can dramatically reduce the serum half-life of the antibody. As shown in the examples, the binding affinity for FcRn of modified OST577-IgG2M3 with such amino acid substitutions is reduced to about 5-7% of that of unmodified OST577-IgG2M3.

As shown in FIG. 3B, T250D represents the amino acid sequence of the modified IgG2M3 heavy chain constant region with a threonine to aspartic acid amino acid substitution at position 250 (SEQ ID NO: 12). M428G represents the amino acid sequence of a modified IgG2M3 heavy chain constant region with a methionine to glycine amino acid substitution at position 428 (SEQ ID NO: 53).
In preferred embodiments of the invention, the FcRn binding affinity and / or serum half-life of the modified antibody is at least about 30%, 40%, 50%, 60%, 70%, 80%, 85%, 90 Decrease by%, 95%, 97%, 98% or 99%.

The present invention relates to a modified IgG antibody described herein, preferably a heavy chain constant region, Fc region of a modified IgG1, IgG2 or IgG2M3 antibody having an amino acid substitution described herein. Or it includes the C _H 2-C _H 3 region.
The present invention also includes a polypeptide comprising the amino acid sequence of any one of SEQ ID NOs: 10 to 76. In a preferred embodiment, these polypeptides are mutated IgG1, IgG2, or IgG2M3 constant regions.

  The heavy chain constant region of a modified antibody of the invention can be linked to the heavy chain variable region of any selected antibody to produce the desired hybrid heavy chain. Examples of selected antibodies include, but are not limited to antibodies to IL-2, IL-4, IL-10, IL-12, HSV, CD3, CD33, CMV and IFN-γ. Absent. Furthermore, the variable region can be of any kind of natural antibody, such as human, primate or rodent. Alternatively, these include humanized antibodies, antibodies that have increased binding affinity for that antigen through genetic modification, or fully human antibodies, including but not limited to antibodies Can be. Such hybrid heavy chains can be linked to various light chains to produce the desired antibody. The light chain can be either a lambda or kappa light chain. Since the serum half-life of an antibody is primarily determined by its heavy chain constant region, the desired serum half-life of the antibody produced is achieved through amino acid substitutions within the heavy chain constant region described herein. Is possible.

II. Production of modified antibodies with altered FcRn binding affinity and / or serum half-life The present invention provides methods for producing proteins, particularly antibodies, with altered FcRn binding affinity and / or serum half-life . Preferably, the present invention provides a method for modifying an antibody of a given IgG class at one or more of the positions disclosed herein. This can be accomplished chemically, or alternatively by recombinant production and random or site-directed mutagenesis using any known production method.

The present invention replaces at least one amino acid from the heavy chain constant region selected from the group consisting of amino acid residues 250, 314 and 428 with an amino acid different from that present in the unmodified antibody, thus Provided is a method of modifying an IgG class antibody comprising causing an alteration of the FcRn binding affinity and / or serum half-life of the unmodified antibody.
The substitution can be done alone at position 250, 314 or 428, or any combination thereof, such as positions 250 and 428.

  In order to increase the FcRn binding affinity of an antibody and / or increase its serum half-life, amino acid residue 250 from the heavy chain constant region is substituted with glutamic acid or glutamine, or an amino acid from the heavy chain constant region Residue 428 is replaced with phenylalanine or leucine. Alternatively, amino acid residue 250 from the heavy chain constant region is replaced with glutamic acid and amino acid residue 428 from the heavy chain constant region is replaced with phenylalanine; or amino acid residue 250 from the heavy chain constant region is replaced with glutamine. Substituted and amino acid residue 428 from the heavy chain constant region is replaced with phenylalanine; or amino acid residue 250 from the heavy chain constant region is replaced with glutamine and amino acid residue 428 from the heavy chain constant region is replaced with leucine. Is replaced. Modifications at both 250 and 428 are preferred because antibodies with such double mutations display exceptionally high FcRn binding affinity.

  To produce a modified antibody with reduced FcRn binding affinity and / or reduced serum half-life compared to an unmodified antibody, amino acid residue 314 from the heavy chain constant region is present in the unmodified antibody. Substituted with another amino acid that is different from the existing one. Preferably, the amino acid residue 314 from the heavy chain constant region is alanine, arginine, aspartic acid, asparagine, cysteine, glutamic acid, glutamine, glycine, histidine, lysine, methionine, phenylalanine, proline, serine, threonine, tryptophan, tyrosine, Or substituted with valine. More preferably, amino acid residue 314 from the heavy chain constant region is substituted with alanine or arginine.

  To produce a modified antibody with reduced FcRn binding affinity and reduced serum half-life compared to an unmodified antibody, amino acid residue 250 from the heavy chain constant region contains arginine, asparagine, aspartic acid, Substituted with lysine, phenylalanine, proline, tryptophan or tyrosine, otherwise the amino acid residue 428 from the heavy chain constant region is alanine, arginine, asparagine, aspartic acid, cysteine, glutamic acid, glutamine, glycine, histidine, lysine , Proline, serine, threonine, tyrosine or valine. More preferably, amino acid residue 250 is substituted with aspartic acid or amino acid 428 is substituted with glycine.

  The amino acid substitutions described herein are accomplished by standard recombinant DNA techniques. In one embodiment, site-directed mutagenesis can be used to introduce amino acid substitutions into DNA encoding an unmodified antibody. The resulting DNA of the modified antibody is then delivered into the host cell, thus producing the modified antibody. The desired alteration of the FcRn binding affinity of the modified antibody can be selected using phage display techniques or any other suitable method known in the art and can be confirmed by measuring the binding affinity.

Preferably, a method of producing an IgG class modified antibody having a modified binding affinity for FcRn and / or an altered serum half-life when compared to an unmodified antibody comprises:
(A) selected from the group consisting of amino acid residues 250, 314 and 428, comprising an appropriate promoter operably linked to DNA encoding at least the constant region of an immunoglobulin heavy chain Preparing a replicable expression vector in which at least one amino acid from the heavy chain constant region is replaced with a different amino acid than that present in the unmodified antibody, thus causing a change in serum half-life;
(B) transforming host cells with the vector; and (c) culturing the transformed host cells to produce the modified antibody;
Comprising.

  Such a method optionally comprises, after step (a), preparing a second replicable expression vector comprising a promoter operably linked to DNA encoding a complementary immunoglobulin light chain. Further comprising, wherein the cell line is further transformed with the vector.

  In order to generate DNA in step (a), amino acid substitution is performed by site-directed mutagenesis (Kunkel, Proc. Natl. Acad. Sci. USA 82: 488-492 (1985)), PCR mutagenesis (Higuchi, “PCR Protocol; Methods and Applications Guide, ”Academic Press, San Diego (1990) pp. 177-183), and cassette mutagenesis (Wells et al., Gene 34: 315-323 (1985)). It can be introduced by mutagenesis. Preferably, site-directed mutagenesis is performed by the overlap extension PCR method disclosed in the examples (Higuchi, “PCR technology: its principles and application to DNA amplification, Stockton Press, New York ( 1989), p61-70).

  The technique of overlap extension PCR (Higuchi, ibid) can be used to introduce any desired mutation (s) into the target sequence (starting DNA). For example, as shown in FIG. 4, the first round of PCR in the overlap extension method involves an external primer (primer 1) and an internal mutagenesis primer (primer 3), and a separate second external primer. (Primer 4) and internal primer (primer 2) are involved in amplifying the target sequence and generating two PCR segments (segments A and B). The internal mutagenesis primer (Primer 3) is designed to include mispairing to the target sequence that defines the desired mutation (s). In the second round of PCR, the products from the first round of PCR (segments A and B) are amplified by PCR using two external primers (primers 1 and 4). The resulting full length PCR segment (segment C) is digested with restriction enzymes and the resulting restriction fragments are cloned into an appropriate vector.

  As a first step in mutagenesis, the starting DNA is operably cloned into the mutagenesis vector. Primers are designed to reflect the desired amino acid substitution (see further details in the examples). In one example, a vector used for in vitro mutagenesis can be used to direct protein expression. Thus, the DNA obtained as a result of the overlap extension PCR can be cloned back into the mutagenesis vector in such a way that an expression vector containing the DNA with the desired mutation is created. Examples of mutagenesis vectors containing the starting DNA include, but are not limited to, pVAg2M3-OST577.

  For example, the mutation at position 250 amplifies the region surrounding the PinAI-BamHI fragment of pVAg2M3-OST577 (see restriction map in Figure 5A) in a two-step process using the overlap extension method described above. The resulting PCR segment was then digested with PinAI and BamHI and the resulting restriction fragment was cloned into pVAg2M3-OST577. In a similar manner, the mutation at position 314 or 428 amplifies the region surrounding the Pm1I-BamHI fragment by overlap extension PCR and then digests the resulting PCR segment with Pm1I and BamHI, resulting in This was done by cloning the resulting restriction fragment into pVAg2M3-OST577.

As long as the starting DNA contains the ACR to be modified, the entire unmodified antibody, the entire immunoglobulin heavy chain of the unmodified antibody, the heavy chain constant region or a portion of the heavy chain constant region of the unmodified antibody Can be DNA encoding.
If the DNA encoding the entire unmodified antibody is used as the starting DNA for mutagenesis, the entire modified antibody is converted into steps (a), (b) and (c) of the method described herein Can be produced. The step between steps (a) and (b) of the method for producing a complementary light chain is not necessary.

  If the starting DNA for mutagenesis is DNA encoding the entire heavy chain of an unmodified antibody, mutagenesis will generate a vector containing DNA encoding the entire modified heavy chain. Steps (a) and (b) of the methods disclosed herein are performed to produce the entire modified antibody. That is, another replicable expression vector containing an appropriate promoter operably linked to DNA encoding a complementary immunoglobulin light chain is cotransfected into the same host cell. As a result, both the complementary light chain and the modified heavy chain are expressed in the same host cell and properly assembled to generate the entire modified antibody. An example of the expression vector containing DNA encoding an immunoglobulin light chain includes, but is not limited to, pVAλ2-OST577.

If the starting DNA for mutagenesis is DNA encoding a portion of a heavy chain constant region, such as a C _H 2-C _H 3 segment or an Fc domain, the result is that it encodes such a modified partial heavy chain. The resulting DNA is first ligated in the same frame with the remaining unmodified heavy chain, thus generating DNA encoding the entire heavy chain with the modifications described herein in step (a). Will be. At this time, the entire modified antibody is produced by co-transfecting host cells with a vector containing DNA encoding a complementary light chain and a vector containing DNA encoding such a modified heavy chain. Ligation of the DNA encoding the decorated partial heavy chain and the remaining undecorated heavy chain can be accomplished by using standard molecular cloning techniques known in the molecular biology arts such as restriction digestion and ligation ( Sambrook and Russell, “Molecular Cloning; Laboratory Manual, 3rd Edition Cold Spring Harbor Laboratory Press, New York (2001)).

  The light and heavy chains can be cloned in the same or different expression vectors. A DNA segment encoding an immunoglobulin chain is operably linked to control sequences in expression vector (s) that ensure expression of the immunoglobulin polypeptide. Such control sequences include signal sequences, promoter enhancers, and transcription termination sequences (Queen et al., Proc. Natl, which is hereby incorporated by reference in its entirety for all purposes). Acad. Sci. USA 86: 10029-10033 (1989); WO90 / 07861; Co et al., J. Immunol. 148: 1149-1154 (1992); “Antibody Engineering: A Practice Guide”, Borrebaeck, Ed., (See Freeman, New York (1997)).

Host cells are transformed using techniques known in the art such as liposomes, calcium phosphate, electroporation and the like (Sambrook and Russell, supra). Preferably, the host cell is transiently transfected using the liposome method.
Host cells used to produce the modified antibodies of the invention can be cultured in a variety of media known in the art.
The modified antibodies described herein can be produced intracellularly, in the periplasmic space, or otherwise secreted directly into the medium. Preferably, the modified antibody in the present invention is secreted into the medium. The host cell culture medium producing the modified antibody is collected and the cell debris is spun by centrifugation. The supernatant is collected and subjected to a protein expression assay (see further details in the examples).

The expression of the modified antibody is confirmed by gel electrophoresis using SDS-PAGE reducing or non-reducing protein gel analysis or any other technique known in the art. An ELISA can also be used to detect both the expression of the modified antibody and the quantity of that antibody.
The modified antibody should retain appropriate antigen binding compared to the unmodified antibody. Thus, proper antibody-antigen binding is tested by procedures known in the immunology art such as ELISA. Additional experiments can be performed to confirm that the modified antibody has structural properties similar to those of the unmodified antibody. These experiments include, but are not limited to, SDS-PAGE, SEC, ELISA, and protein A binding assays. Protein A binding assays are preferred because protein A binds to the same region of the C _H 2-C _H 3 junction, such as FcRn, although different residues are involved in binding.

  Modified antibodies prepared from host cells include (but are not limited to) gel filtration and column chromatography (eg, protein A affinity chromatography, cation exchange chromatography, anion exchange chromatography and gel filtration). ), And can be purified using techniques known in the art. The minimum acceptable purity of an antibody for use in a pharmaceutical formulation is 90%, preferably 95%, more preferably 98%, most preferably 99% or more.

  The binding affinity of the produced antibody for FcRn can be detected by performing a competitive binding assay at pH 6.0, which is the optimal condition for binding to FcRn. Binding affinity can be tested by immobilizing FcRn on a solid substrate such as Sepharose ™ beads. Alternatively, binding affinity can be assessed using an ELISA. Preferably, the present invention tests binding affinity by performing a competitive binding assay in a cell-based system. A dilution series of modified and unmodified antibodies produced is compared for binding to FcRn expressed on one cell line, preferably the NS0 cell line. Experimental procedures for performing competitive binding assays are detailed in the examples.

Experiments in the present invention have shown that similar binding affinity results can be achieved with culture supernatants of cells producing antibodies or purified antibodies. Thus, the supernatant can be used directly to test the FcRn binding affinity of the produced antibody in order to confirm that the desired modification of binding affinity has been achieved. After such confirmation, the produced antibodies are subjected to more complex purification procedures.
A direct binding assay should also be performed to confirm that the modified antibody binds to FcRn in a pH-dependent manner. In particular, the binding affinity of the modified antibody for FcRn is tested at both pH 6.0 and pH 8.0 (see further details in the examples). In general, the binding affinity at pH 6.0 should exceed that at pH 8.0.

For example, by measuring serum radioactivity levels as a function of time using a radiolabeled protein; or using ELISA as a function of time to determine the level of intact antibody (of known specificity) present in the serum. Although biological stability (or serum half-life) can be measured by various in vitro or in vivo means such as assaying, a particularly preferred measure of increased biological stability is increased serum Evidenced by half-life and reduced clearance rate.
The present invention relates to a polynucleotide molecule encoding a modified antibody with the mutations described herein, or a modified antibody such as a constant region, Fc region or C _H 2 -C _H 3 region A polynucleotide molecule encoding the decorated partial or complete heavy chain.

The present invention relates to polynucleotide molecules encoding modified antibodies with the mutations (substitutions) described herein, or modifications such as constant regions, Fc regions or C _H 2 -C _H 3 regions. A vector comprising a polynucleotide molecule encoding a decorated partial or complete heavy chain of a modified antibody is provided.
The invention includes a host cell containing the vector comprising the nucleic acid molecule as described herein. Suitable host cells for expression of the modified antibodies described herein are derived from prokaryotes such as Escherichia coli or eukaryotic multicellular organisms including yeast, plants, insects and mammals. .

  E. coli is one prokaryotic host that is particularly useful for cloning and / or expressing the DNA sequences of the present invention. Other microbial hosts suitable for use include bacilli such as Bacillus subtilis and other enteric bacterial species such as Salmonella, Serratia and various Pseudomonas species. Among these prokaryotic hosts, expression vectors that normally contain expression control sequences (eg, origins of replication) that are compatible with the host cell can also be made. In addition, there can be any number of various well-known promoters such as lactose promoter system, tryptophan (trp) promoter system, beta-lactamase promoter system, or promoter system from phage lambda. Promoters typically have expression, optionally together with operator sequences, including ribosome binding site sequences to initiate and complete transcription and translation.

  Other bacteria such as yeast can also be used for expression. Saccharomyces is the preferred host, and suitable vectors have expression control sequences such as promoters including 3-phosphoglycerate kinase or other glycolytic enzymes and origins of replication, termination sequences, etc. as desired.

  Plants and plant cell cultures can be used for expression of the DNA sequences of the present invention. (Larrick and Fry, Hum. Autibodies Hybridomas 2; 172-189 (1991); Benvenuto et al., Plant Mol. Biol. 17: 865-874 (1991); During et al., Plant Mol. Biol. 15: 281 -293 (1990); Hiatt et al., Nature 342: 76-78 (1989)). Preferred plant hosts include, for example, Arabidopsis, Nicotiana tabaceum, Nicotiana rustica and Solanum tuberosum. A preferred expression cassette for expressing a polynucleotide sequence encoding a modified antibody of the present invention is such that the inserted polynucleotide sequence encoding the modified antibody is operable to a CaMV35S promoter with an overlap enhancer. The linked plasmid pMOG18, which is used according to the method of Sijmons et al., Bio / Technology 8: 217-221 (1990). Alternatively, a preferred embodiment for expression of a modified antibody in plants is a polynucleotide encoding the modified antibody of the invention in place of the immunoglobulin sequence used by Hiatt et al., Supra. The sequence is used to follow the method of Hiatt et al., Described above. To express the DNA sequences of the present invention, Agrobacterium tumifaciens T-DNA based vectors can be used; preferably, such vectors include a marker gene encoding spectinomycin resistance or another Contains selectable markers.

  Insect cell cultures can also be used to produce the modified antibodies of the invention using standard baculovirus-based expression systems. Modified antibodies can be produced by expressing a polynucleotide sequence that encodes a modified antibody according to the method of Putlitz et al., Bio / Technology 8: 651-654 (1990).

  In addition to microorganisms and plants, mammalian cell cultures can also be used to express and produce the polypeptides of the invention (see Winnacker “Gene to Clone”, VCH Publishers, New York (1987)). ). Since a certain number of suitable heavy chain lines with the ability to secrete intact immunoglobulins have been developed in the art, mammalian cells are currently preferred, including the CHO cell line, Various COS cell lines, HeLa cells, preferably myeloma cell lines, etc. or transformed B-cells or hybridomas are included. Expression vectors for these cells include expression control sequences such as origins of replication, promoters, enhancers (Queen et al., Immunol. Rev. 89; 49-68 (1986)) and necessary processing information sites such as ribosome binding. It may include sites, RNA splice sites, polyadenylation sites and transcription terminator sequences. Preferred expression control sequences are promoters derived from immunoglobulin genes, SV40, adenovirus, bovine papilloma virus, cytomegalovirus and the like. In general, a selectable marker such as a neo expression cassette is included in the expression vector.

  The present invention has been modified by conjugating or otherwise binding the agent to a moiety identified as having an increased or decreased serum half-life through interaction with FcRn. Methods are provided for making agents with FcRn binding affinity and / or serum half-life. Such moieties include, but are not limited to, decorated IgGs or decorated partial or complete heavy chains, including amino acid substitutions described herein. Such agents include antibodies, antibody fragments, hormones, receptor ligands, immunotoxin conjugates, all types of therapeutic agents, T-cell receptor binding antigens and other that can be bound to the increased serum half-life portion of the present invention. Any active agent will be included, but is not limited to this.

  To create a fusion protein with modified in vivo stability, upstream at one location to confer to the vector the ability to express a fusion protein comprising such a protein operably linked to a constant region. Alternatively, a DNA segment encoding such a protein can be operably incorporated into the recombinant vector within the same frame as the modified antibody constant region, either downstream. For example, by genetic engineering using restriction endonucleases; techniques for manipulating DNA segments in this manner will be apparent to those skilled in the art in light of both this disclosure and references such as Sambrook and Russell, discussed above. is there. The methods described above have been proposed for use in the production of a series of therapeutic compounds with improved biological stability.

  Such compounds include, for example, interleukin-2, insulin, interleukin-4, and interferon gamma or even T cell receptors. The recombinant Fc domains of the present invention are also considered to be used in the stabilization of a wide range of drugs that are likely to alleviate the need for repeated administration. However, the method is not limited to the production of proteins for human administration, and has been augmented, for example, as can be used in immunization protocols, veterinary animal treatment or rodent in vivo therapy models. It can also be used to produce a large amount of any protein with stability.

III. Use of Modified Antibodies with Modified FcRn Binding Affinity and / or Serum Half Life The present invention provides a composition comprising a modified antibody described herein and a pharmaceutically acceptable carrier. is doing. Compositions for parenteral administration generally comprise a solution of the antibody or cocktail thereof dissolved in an acceptable carrier, preferably an aqueous carrier. A variety of aqueous carriers can be used, such as water, buffered water, 0.4% saline, 0.3% glycine. These solutions are sterile and generally free of particulate matter. The composition is pharmacologically as required to approximate physiological conditions such as pH adjusters and buffers, toxicity modifiers, etc. such as sodium acetate, sodium chloride, potassium chloride, calcium chloride, sodium lactate. Acceptable auxiliary substances can be included. The concentration of antibody in these formulations can vary widely, i.e., less than about 0.01%, usually from at least less than about 0.1% to 5% by weight, with the particular mode of administration selected primarily According to viscosity and fluid volume.

A standard composition for intravenous infusion may be configured to contain 250 ml of sterile Ringer's solution and 10 mg to 100 mg of antibody (“Remington's Pharmacy” 15th edition, Mack Publishing Company, Easton, PA (1980 )checking).
The modified antibodies in the present invention can be used for various non-therapeutic purposes. These can be used as affinity purification agents. They can also be useful in diagnostic assays, such as detecting the expression of antigens of interest in specific cells, tissues or sera. In diagnostic applications, the antibody will typically be labeled with a detectable moiety, including radioisotopes, fluorescent labels, and various enzyme substrate labels. The antibodies can also be used in any known assay method, such as competitive binding assays, direct and indirect sandwich assays and immunoprecipitation assays. The antibodies can also be used for in vivo diagnostic assays. In general, antibodies are labeled with a radionuclide in such a way that immunoscintigraphy can be used to locate the antigen or cells that express it.

  Kits for use with modified antibodies may also be provided for protection against or detection of cellular activity or for the presence of selected cell surface receptors or diagnosis of disease. Thus, the compositions that are the subject of this invention can be provided in lyophilized form, usually in containers, either alone or in combination with additional antibodies specific for the desired cell type. Modified antibodies that may or may not be conjugated to a label or toxin include buffers such as Tris, phosphate, carbonate, etc., stabilizers, biocides, inactive proteins such as serum. It is included in the kit along with a buffer solution such as albumin and a set of instructions for use. In general, these materials are present in less than about 5% by weight based on the amount of active antibody, and usually again in a total amount of at least about 0.001% by weight based on antibody concentration. Often it will be desirable to include an inert bulking agent or excipient to dilute the active ingredient, in which case the excipient may be present at about 1-99% by weight of the total composition. . If a second antibody with the ability to bind to the modified antibody is utilized in the assay, it will usually be in a separate vial. The second antibody is typically conjugated and formulated to the label in a manner similar to the antibody formulation described above.

Modified antibodies have a variety of therapeutic applications. Modified antibodies can be used to treat patients suffering from or susceptible to certain diseases or disorders that may benefit from administration of such antibodies. Physical conditions that can be treated with the antibody include cancer; inflammatory conditions such as asthma; autoimmune diseases; and viral infections.
Cancers that can be treated with the antibodies described herein include breast cancer, squamous cell carcinoma, small cell lung cancer, non-small cell lung cancer, gastrointestinal cancer, pancreatic cancer, glioblastoma, cervical cancer, ovarian cancer Bladder cancer, liver cancer, colon cancer, colorectal cancer, endometrial cancer, salivary gland cancer, kidney cancer, liver cancer, prostate cancer, vulvar cancer, thyroid cancer, liver cancer and various types of head and neck cancer Including but not limited to this.

  Autoimmune diseases include Addison's disease, ear autoimmune disease, ocular autoimmune diseases such as uveitis, autoimmune hepatitis, Crohn's disease, diabetes (type I), epididymis, glomerulonephritis, Graves' disease , Guillain-Barre syndrome, Hashimoto's disease, hemolytic anemia, systemic lupus erythematosus, multiple sclerosis, myasthenia gravis, pemphigus vulgaris, psoriasis, rheumatoid arthritis, sarcoidosis, scleroderma, Sjogren's syndrome, spine Includes or is not limited to arthropathy, thyroiditis, ulcerative colitis, and vasculitis.

  The modified antibodies with reduced serum half-life in the present invention can be used in the treatment of diseases or disorders where it is desirable to destroy or remove tissue or exogenous microorganisms. For example, antibodies can be used to treat cancer, inflammatory disorders; infections and other body conditions where removal of tissue is desired. Antibodies are generally useful in that a faster biological clearance time will result in a decrease in the immunogenicity of any antibody administered. Other areas of application would include antibody-based imaging methods, antibody-based drug removal or the creation of shorter-lived immunotoxins.

Modified antibodies with increased serum half-life can be anti-tissue factor (TF) antibodies, anti-IgE antibodies and anti-integrin antibodies. The desired mechanism of action may be to block the ligand-receptor binding pair. Modified antibodies with increased serum half-life can be agonist antibodies as well. Antibodies can also be used as therapeutic agents such as vaccines. The dosage and immunization frequency of such vaccines will be reduced due to the increased serum half-life of the antibody.
The composition comprising the antibody is administered by any suitable means, including parenteral subcutaneous, intraperitoneal, pulmonary, and intranasal administration and, if desired for local immunosuppressive therapy, intralesional administration. Is done. Parenteral infusion includes intramuscular, intravenous, intraarterial, intraperitoneal or subcutaneous administration. In addition, the antibody is suitably administered by pulse infusion, particularly at decreasing antibody doses.

  The composition containing the antibody or a cocktail thereof can be administered for prophylactic and / or therapeutic treatments. In therapeutic applications, a composition is administered to a patient already suffering from a particular disease in an amount sufficient to cure or at least partially prevent the physical condition and its complications. . An appropriate amount to accomplish this is defined as a “therapeutically effective amount”. Effective amounts for this use will depend on the severity of the physical condition and the overall condition of the patient's own immune system, but generally from about 0.01 to about 100 mg of modified antibody per dose A dosage of 1-10 mg per patient is more commonly used.

  In the prophylactic application field, a composition containing a modified antibody or a cocktail thereof is administered to a patient who is not already in a disease state to enhance the patient's resistance. Such an amount is defined as a “prophylactically effective amount”. In this application, the exact amount will again depend on the patient's health and overall immunity level, but is generally in the range of 0.1-100 mg per dose, in particular a dosage of 1-10 mg per patient It is.

Single or multiple administrations of the compositions can be carried out, but the dose level and pattern are selected by the treating physician. In any case, the pharmaceutical formulation must provide a sufficient amount of the mutant antibody of the invention to effectively treat the patient.
The following examples are provided for purposes of illustration, not limitation. The disclosures of all citations are expressly incorporated herein by reference.

Example 1 .
This example describes an antibody expression vector used in the present invention.
The components of the heavy chain expression plasmid pVAg2M3-OST577, which is one derivative of the M3 variant of pVg2.D.Tt (Core et al., J. Immunol. 159: 3613-3621 (1997)), are as follows. As shown in FIG. 5A, proceeding clockwise from the EcoRI site, the heavy chain unit is the human cytomegalovirus (hCMV) major immediate early (IE) promoter and enhancer (Boshart et al.) As an EcoRI-XbaI fragment. al., Cell 41; 521-530 (1985)). The hCMV region is followed by the OST577V _H region as an XbaI fragment containing the signal sequence, J segment and splice donor sequence.

After a V _H region, the human complement gene C2 as BamHI-EcoRIEG (Ashfield et al, EMBO J. 10:. 4197-4207 (1991)) transcription terminator is followed, mRNA processing following the C _H 3 polyadenylation (poly a) signal, and part of the intron preceding the C _H 1 for, to entailment the C _H 1, hinge _(H), C H 2 and C _H 3 exons with the intervening introns, XbaI A decorated genomic DNA fragment containing the human gamma-2M3 heavy chain constant region (Cole et al., Supra) as a -BamHI fragment follows. The heavy chain unit is followed by a mutant form of dihydrofolate reductase (dhfr) along with regulatory elements (enhancer, promoter, splice signal and poly A signal) from simian virus 40 (SV40) required for transcription. The gene continues. Taken as a BamHI-EcoRI fragment from plasmid pVgI (Co et al., Supra), this region was decorated by converting the BamHI site into an EcoRI site.

  When moving counterclockwise from the original EcoRI site within this unit, first the bacterial origin of replication was replaced with the corresponding segment from pUC18 to increase the copy number of the vector in the bacterial host ( Plasmid pBR322 (Suteliffe, Cold Spring Harbor Symp) containing a bacterial replication origin for selection in E. coli and an ampicillin resistance gene, except that Yanisch-Perron et al., Gene 33: 103-119 (1985)) Quant. Biol. 43: 77-90 (1979)). Next, there is an SV40 segment (Reddy et al., Science 200: 494-502 (1978)) that contains an SV40 enhancer and early promoter to ensure strong transcription initiation. This segment is followed by the coding sequence of the E. coli dhfr gene (Simonsen and Levinson, Proc. Natl. Acad. Sci. USA 80: 2495-2499 (1983)). The dhfr gene is followed by an SV40 segment containing a small t-antigen intron that is thought to increase mRNA levels, when the plasmid contains another SV40 segment containing a polyA signal to terminate the mRNA transcript. Including.

The components of the heavy chain expression plasmid pVAgl.N-OST577, which is a derivative of pVgl (Co et al., supra), are as follows. As shown in Figure 5B, proceeding from the EcoRI site in the clockwise direction, the heavy chain unit, the hCMV IE promoter and enhancer that was used in the pVAg2M3-OST577 into vectors OST577V _H region as XbaI fragment followed Start with the same EcoRI-XbaI fragment containing. PolyA signal for mRNA processing following the C _H 3, and part of the intron preceding the C _H 1, entailment the C _H 1, hinge _(H), C H 2 and C _H 3 exons with the intervening introns To do. The _VH region is followed by a genomic DNA fragment containing the human gamma-1 heavy chain constant region (Ellison et al., Nucleic Acids Res. 10: 4071-4079 (1982)) as an XbaI-BamHI fragment. . To facilitate subsequent manipulation of the coding region, overlap extension PCR mutagenesis (Higuchi, supra) was used to create a NheI site in the intron between the hinge and the C _H 2 exon. The heavy chain unit is followed by the same BamHI-EcoRI restriction fragment encoding the dhfr with part of the plasmid pBR322 and the regulatory elements used in the pVAg2M3-OST577 vector.

The components of the light chain expression plasmid pVAλ2-OST577, which is a derivative of pVK (Co et al., supra), are as follows. In the heavy chain vector, proceeding clockwise from the EcoRI site, followed by the OST577V _L region as an Xbal fragment, including the signal sequence, J segment and splice donor sequence, as shown in FIG. Start with the same EcoRI-XbaI fragment containing the hCMV IE promoter and enhancer used in The _VL region includes a polyA signal for mRNA processing from the human lambda-1 light chain and a portion of the 3 'untranslated region from human lambda-2 light chain and a human lambda-2 light chain constant region exon (Cλ2), human lambda Human lambda as an XbaI-Sau3A fragment decorated by PCR to encode the human lambda-2 light chain constant region, including the -1 light chain intron (Hieter et al., Nature 294: 536-540 (1981)) Followed by a genomic DNA fragment containing one light chain constant region (Hieter et al., Ibid).

  The light chain gene is followed by a gene (gpt) encoding xanthine guanine phosphoribosyltransferase, together with regulatory elements from SV40 that are required for transcription. The function of this region taken as a BamHI-EcoRI fragment from plasmid pSV2-gpt (Mulligan and Berg. Supra) provides a selectable drug resistance marker after transfection of the plasmid into mammalian cells. There is. Moving counterclockwise within this unit from the EcoRI site first of all replaced the bacterial origin of replication with the corresponding segment from pUC18 to increase the copy number of the vector in the bacterial host (Yanisch- Except for Perron et al., Supra), there is a portion of plasmid pBR322 (Sutcliffe, supra) that contains a bacterial origin of replication for selection in E. coli and an ampicillin resistance gene.

  Next, there is an SV40 segment (Reddy et al., Supra) containing an SV40 enhancer and early promoter to ensure a strong transcription initiation. This segment is followed by the coding sequence of the E. coli gpt gene (Richardson et al., Nucleic Acids Res. 11: 8809-8816 (1983)). The gpt gene is followed by an SV40 segment containing a small t antigen intron that is thought to increase mRNA levels, when the plasmid contains another SV40 segment containing a polyA signal to terminate the mRNA transcript. Including.

The components of the heavy chain expression plasmid pVAg2M3-HuID10 (see FIG. 7A) is, OST577V _H region plasmid pHuID10.IgG1.rgpt.dE (Kostelny et al., ( 2001) op.cit) are replaced with HuID10V _H region from Is identical to that of pVAg2M3-OST577 from which it was derived. The components of the heavy chain expression plasmid pVAg1.N-HulD10 (see FIG. 7B), OST577V _H region plasmid pHuID10.IgG1.rgpt.dE (Kostelny et al., ( 2001) ibid.) Replaced by a HulD10V _H region from Is identical to that of pVAg1.N-OST577 from which it was derived.

The components of the heavy chain expression plasmid pHuHCg3.Tt.D-Hu1D10 (see FIG. 7C) are identical to those of pVg2.D.Tt (Cole et al., Supra) except that human γ-2M3 The XbaI-BamHI fragment containing the heavy chain constant region is replaced as the XbaI fragment by the Hu1D10V _H region from the plasmid pHu1D10.IgG1.rgpt.dE (Kostelng et al. (2001), supra) followed by C _H 1 4 hinges (H), C _H 2 and C _H 3, exons containing intervening introns, part of the intron before C _H 1, and polyadenylation for poly (adenylation) following C _H 3 (poly A) There is a genomic DNA fragment containing the human γ-3 heavy chain constant region (Huck et al., Nucleic Acids Res. 14: 1779-1789 (1986)) as an Xba-BamHI fragment containing the signal.

The components of the heavy chain expression plasmid pHuHCg4.Tt.D-Hu1D10 are identical to those of pHuHCg3.Tt.D-Hu1D10 (see FIG. 7D), except that the human γ-3 heavy chain constant region is C _H 1, hinge (H), C _H 2 and C _H 3, exons containing intervening introns, part of the intron before C _H 1, and polyadenylation (polyA for mRNA processing following C _H 3 ) Replaced by a genomic DNA fragment containing the human γ-4 heavy chain constant region (Ellison et al., Supra) as an Xba-BamHI fragment containing the signal.
The components of the light chain expression plasmid pVk-HulD10, which is a derivative of pVk (Co et al., supra), are as follows. As shown in FIG. 8, the HulD10V _L region as an Xbal fragment (Kostelny et al. (2001) proceeding clockwise from the EcoRI site and including the signal sequence, J segment and splice donor sequence. ) Followed by the same EcoRI-XbaI fragment containing the hCMV IE promoter and enhancer used in the heavy chain vector. Following the _VL region is a human kappa light chain constant region exon (Cκ), a portion of an intron preceding Cκ, and a polyA signal for mRNA processing following Cκ, and a human kappa light chain constant as an XbaI-BamHI fragment. It is followed by a genomic DNA fragment containing the normal region (Hieter et al., Cell 22: 197-207 (1980)). The light chain unit is followed by the same BamH-EcoRI fragment encoding gpt with the regulatory elements required for transcription and a portion of the plasmid pBR322 used in the pVK vector.

Example 2 .
This example describes the plasmid used in the present invention.
OST577 heavy and light chain cDNAs were cloned in their entirety by PCR from trioma cell lines expressing human monoclonal anti-HBV antibodies (Ehrich et al., Supra). The heavy and light chain variable regions contain signal sequences, V, (D), and J segments, splice donor sequences and corresponding intron portions, as outlined in Co et al., Supra. It was converted by PCR into a mini-exon flanked by XbaI at the end of. The expression vector pVAg2M3-OST577 (see Figure 5A), which is a derivative of the M3 variant of pVg2.D.Tt (Cole et al., supra) is an OST577-V _H miniexon and contains the OKT3-V _H miniexon. Constructed by replacing the XbaI fragment.

The PciI-FspI fragment containing the bacterial origin of replication is then replaced with the corresponding PciI-FspI fragment from pUC18 (Yanisch-Perron et al., Supra) to increase the copy number of the vector in the bacterial host. It was. The expression vector pVAgl.N-OST577 (see FIG. 5B), a derivative of pVgl (Co et al., supra), inserts an XbaI fragment containing the OST577-V _H miniexon into the unique XabI site of pVgl, Overlapping extension PCR decorates the hinge-CH ₂ intron to create a unique NheI site, creating a unique NheI site, and the HindIII-XhoI fragment containing the bacterial origin of replication with the corresponding HindIII-XhoI fragment from pVAg2M3-OST577 It was constructed by replacing and increasing the copy number of the vector in the bacterial host.

The expression vector pVλ-OST577, a derivative of pVK (Co et al., supra), first contains the XbaI-Bam HI fragment of pVk containing the genomic human kappa constant region and the XbaI- containing the genomic human lambda-1 constant region. Constructed by replacing with Bgl II PCR product. The coding region and a portion of the 3 'untranslated region were replaced by PCR with the corresponding fragment from the OST577 light chain cDNA, essentially generating a genomic human lambda-2 constant region exon. Finally, an OST577-V _L miniexon was inserted into the XabI site of the vector. The expression vector pVAλ2-OST577 (see Figure 6), a derivative of pVλ-OST577, replaces the SapI-FspI fragment containing the bacterial origin of replication with the corresponding SapI-FspI fragment from pVAg2M3-OST577. Built by increasing copy number.

The expression vector pVAg2M3-HulD10 (see FIG. 7A) contains the XhoI-XbaI fragment from the plasmid pVAg2M3-OST577 containing the hCMV promoter and enhancer (Boshart et al. Supra) and the OST577V _H region, the hCMV promoter and enhancer and HulD10V _H It was constructed by replacing with the corresponding XhoI-XbaI fragment from the plasmid pHulD10.IgG1.rgpt.dE containing the region (Kostelny et al. (2001) supra). The expression vector pVAg1.N-Hul D10 (see FIG. 7B) contains the hCMV promoter and enhancer (Boshart et al., Supra) and the XhoI-XbaI fragment from plasmid pVAgl.N-OST577 containing the OST577V _H region and the hCMV promoter. It was constructed by replacing with the corresponding XhoI-XbaI fragment from plasmid pHulD10.IgG1.rgpt.dE (Kostelny et al. (2001) supra) containing the enhancer and the HulD10V _H region.

The expression vector pHuHCg3.Tt.D-Hu1D10 (see FIG. 7C) contains the hCMV promoter and enhancer, and the human γ-2M3 heavy chain constant chain region of pVg2.D.Tt (Cole et al. Supra). The XhoI-BamHI fragment from the M3 variant is an XhoI-XbaI fragment from plasmid pHu1D10.IgG1.rgpt.dE (Kostelny et al. (2001), supra) containing the hCMV promoter and enhancer, and the Hu1D10 V _H region, And an XbaI-BamHI fragment containing the human γ-3 heavy chain constant region (Huck et al., Supra), respectively.

The expression vector pHuHCg4.Tt.D-Hu1D10 (see FIG. 7D) contains the hCMV promoter and enhancer, and the human γ-2M3 heavy chain constant chain region of pVg2.D.Tt (Cole et al. Supra). The XhoI-BamHI fragment from the M3 variant is an XhoI-XbaI fragment from plasmid pHu1D10.IgG1.rgpt.dE (Kostelny et al. (2001), supra) containing the hCMV promoter and enhancer, and the Hu1D10 V _H region, And an XbaI-BamHI fragment containing the human γ-4 heavy chain constant region (Huck et al., Supra), respectively.

The expression vector pVK HulD10 (see FIG. 8) replaces the XhoI-XbaI fragment (Co et al., Supra) from the plasmid pVk containing the hCMV promoter and enhancer (Boshart et al. Supra) with the hCMV promoter and enhancer and HulD10V _L It was constructed by replacing with the XhoI-XbaI fragment from the plasmid pHulD10.IgG2.rgpt.dE containing the region (Kostelny et al. (2001) supra).

  The basic EX VpDL172, a derivative of pVk.rg (Cole et al., supra), contains an XbaI-SphI fragment containing the genomic human kappa constant region, an XbaI-NheI fragment containing the N-terminal portion of the M195 heavy chain signal sequence ( Co et al., Supra), 0.7 kb NheI-AgeI fragment, mouse monoclonal antibody E10 followed by GPI linkage signal from human decay accelerator (Caras et al., Nature 325; 545-549 (1987)) (Evan et al., Mol. Cell. Biol. 5; 3610-3616 (1985)), a synthetic AgeI-EagI fragment encoding a human c-myc decapeptide flanked by a linker peptide, and human immunity It was constructed by replacement with an XbaI-SphI fragment consisting of an EagI-SphI fragment containing the polyA signal of the globulin gamma-1 gene (Ellison et al., Supra).

  The extracellular domains of human beta-2-microglobulin (β2m) and human neonatal Fc receptor (FcRn) alpha chain were cloned by PCR from cDNA libraries prepared from human peripheral blood mononuclear cells. The human FcRn alpha chain gene is decorated by PCR, using a flanking NheI site and M195 heavy chain signal sequence (Co et al., Supra) at the 5 'end and a flanking AgeI site at the 3' end. Thus, the NheI-AgeI fragment of pDL172 was replaced, resulting in the expression vector pDL172 + HuFcRn.

  The human β2m gene was modified by PCR to add flanking XbaI and SalI sites at the 5 ′ and 3 ′ ends, respectively, and to remove the internal EcoRI site. The resulting XbaI-SalI fragment is a polyadenylation signal of the mouse immunoglobulin gamma-2a gene followed by a Bam HI-EcoRI fragment containing the transcription terminator of human complement gene C2 (Ashfield et al., Supra). The SalI-Bam HI fragment containing Kostelny et al. (1992) supra and on its 3 'end and the EcoRI-XbaI fragment containing the hCMVIE promoter and enhancer (Boshart et al. Supra) Subcloning into an intermediate vector flanked at the ends. The resulting EcoRI-EcoRI fragment containing the functional human β2m transcription unit was cloned into the unique EcoRI site of pDL172 + HuFcRn, resulting in the expression vector pDL172 + HuFcRn + Huβ2m, hereinafter referred to as pDL208 (see FIG. 9A). Got.

  The extracellular domain of rhesus monkey β2m and rhesus monkey FnRn alpha chain was cloned by PCR from a cDNA library prepared from rhesus monkey peripheral blood mononuclear cells. The rhesus monkey β2m gene was decorated with PCR, adding flanking XbaI and SalI sites at the 5 ′ and 3 ′ ends, respectively, and removing the internal EcoRI site. The resulting XbaI-SalI fragment is a polyadenylation signal of the mouse immunoglobulin gamma-2a gene followed by a Bam HI-EcoRI fragment containing the transcription terminator of human complement gene C2 (Ashfield et al., Supra). The SalI-Bam HI fragment containing Kostelny et al. (1992) supra and on its 3 'end and the EcoRI-XbaI fragment containing the hCMVIE promoter and enhancer (Boshart et al. Supra) Subcloning into an intermediate vector flanked at the ends.

  The resulting EcoRI-EcoRI fragment containing the functional rhesus monkey β2m transcription unit was used to replace the EcoRI-EcoRI fragment (containing the human β2m transcription unit) of pDL172 + HuFcRn + Huβ2m, resulting in pDL172 + HuFcRn + Rhβ2m was obtained. The rhesus monkey FcRn alpha chain gene was modified by PCR to include a flanking NheI site at the 5 ′ end and the C-terminal portion of the M195 heavy chain signal sequence (Co et al., Supra) and a flanking AgeI site at the 3 ′ end. Is used to replace the NheI-AgeI fragment of pDL172 + HuFcRn + Rhβ2m (including the human FcRn alpha chain gene), resulting in the expression vector pDL172 + RhFcRn + Rhβ2m, hereinafter referred to as pDL410 (see FIG. 9B). Obtained.

Example 3 .
This example describes mutagenesis of the Fc region of the human γ2M3 heavy chain gene.
Molecular modeling :
Based on the low-resolution crystal structure of the rat Fc / FcRn complex, an initial model of the human Fc / FcRn complex was generated (Burmeister et al., Nature 372: 379-383 (1994); RCSB protein databank code IFRT ). First, the complex rat β2m was replaced by superposition of the same configuration to that of human β2m taken from the high-resolution crystal structure of the human histocompatibility antigen HLA-A2 (Saper et al. J. Mol. Biol. 219: 277-319 (1991)); RCSP code 3HLA). Subsequently, the alpha chain of rat FcRn was replaced by superposition of human alpha chains taken from the high resolution crystal structure of human FcRn in the same coordination as the rat alpha chain in the complex (West and Bjorkman, Biochemistry 29: 9698-9708 (2000); RCSB code (1EXU).

  The rat residues in complex Fc are then replaced with the corresponding residues from human IgG1Fc (Kabat et al. Supra) to generate a model of the human IgG1Fc / FcRn complex (SEGMOD and ENCAD programs ( Energy minimization calculations were performed using Levitt, J. Mol. Biol. 226: 507-533 (1992); Levitt, J. Mol. Biol. 168: 595-620 (1983)). Finally, the model human IgG1 Fc residue is replaced with the corresponding residue from human IgG2M3Fc (Cole et al., Supra), and the energy minimization calculation is repeated, so that human IgG2M3 Fc / A model of FcRn complex was generated.

A second human IgG2M3 Fc / FcRn complex model, hereinafter referred to as model 2, is a model of the rat Fc / FcRn complex (Weng et al., J. Mol. Biol. 282: 217-225 (1998); RCSB code 2FRT Was generated as described above.
A third model, referred to below as model 3, is a high-resolution crystal structure of the heterodimeric rat Fc / FcRn complex (Martin et al., Mol. Cell 7: 867-877 (2001); RCSB code 1 / 1A) Was generated as described above.

Mutagenesis ;
Overlap extension polymerase chain reaction (PCR) to generate random amino acid substitutions at positions 250, 314 and 428 of the OST577-IgG2M3 heavy chain (numbered according to the EU index of Kabat et al., Supra). The method (Higuchi, supra) was used. To generate a random mutant at position 250, mutagenesis primer JY24 (5′-GAC CTC AGG GGT CCG GGA GAT as M = A or C and N = A, C, G or T. CAT GAG MNN GTC CTT GG-3 ′) (SEQ ID NO: 77) and JY25 (5′-CTC ATG ATC TCC CGG ACC CCT GAG GTC-3 ′) (SEQ ID NO: 78) were used. In the first round of overlap extension PCR, the external primer msc g2-1 (5′-CCA GCT CTG TCC CAC ACC G-3 ′) (SEQ ID NO: 79) and JY24 are used for the left fragment, and the right fragment For this, the external primers kco8 ′ (5′-GCC AGG ATC CGA CCC ACT-3 ′) (SEQ ID NO: 80) and JY25 were used.
PCR reactions were cycled in GeneAmp ™ PCR System 9600 (Applied Biosystems ™, Foster City, Calif.) First at 94 ° C. for 5 minutes, then at 94 ° C. for 5 seconds, 55 ° C. for 5 seconds and 72 ° C. for 60 seconds. This was done using the Expand ^™ high fidelity PCR system (Roche Diagnostics Corporation, Indianapolis, IN) according to the manufacturer's recommendations by incubating in batches and then incubating for 7 minutes at 72 ° C. PCR products were run on a low melting agarose gel, excised from the gel and melted at 70 ° C.

The second round of PCR to combine the left and right fragments was performed as described above using 35 cycles, external primers mscg2-1 and kco8. The final PCR product was run on a low melting point agarose gel and the expected size DNA fragment was excised and purified using the QIAEX ^™ II gel extraction kit (QIAGEN ™, Valencia, Calif.). The purified fragment was digested with PinAI and BamHI, gel purified as described above, and cloned between the corresponding sites in pVAg2M3-OST577.

  To generate T250I and T250L mutants, the mutagenic primer KH4 (5′-GAC CTC AGG GGT CCG GGA GAT CAT GAG AAK GTC CTT GG-3 ′) (SEQ ID NO: 81) is used, where K = G or T. ) And KH3 (5′-CTC ATG ATC TCC CGG ACC CCT GAG GTC-3 ′) (SEQ ID NO: 82) were used. To generate T250C and T250G mutants, the mutagenic primer KH5 (5′-GAC CTC AGG GGT CCG GGA GAT CAT GAG GCM GTC CTT GG-3 ′), assuming M = A or C SEQ ID NO: 83) and KH3 were used. To generate mutants of T250N and T250Q, the mutagenic primer KH6 (5′-GAC CTC AGG GGT CCG GGA GAT CAT as K = G or T and N = A, C, G or T GAG NTK GTC CTT GG-3 ′) (SEQ ID NO: 84) and KH3 were used.

In the first round of PCR, the external primers msc g2-1 and KH4, KH5 or KH6 are used for the left fragment, and the external primers MGD-1 (5′-GCC AGG ATC CGA CCC ACT- 3 ') (SEQ ID NO: 85) and KH3 were used. The PCR reaction was first performed at 94 ° C for 5 minutes, followed by 25 cycles of 94 ° C for 5 seconds, 60 ° C for 5 seconds and 72 ° C for 60 seconds followed by 7 minutes at 72 ° C for Expand ^™ high fidelity Performed using a PCR system (Roche Diagnostics Corporation). PCR products were run on low melting point agarose, excised from the gel and melted at 70 ° C.

The second round of PCR to combine the left and right fragments, using external primers mscg2-1 and MGD-1, first at 94 ° C for 5 minutes, then 94 ° C for 5 seconds, 60 ° C and 105 seconds for 5 seconds. This was done as described above by incubating with 35 cycles of 72 ° C. per second followed by 7 minutes at 72 ° C. The final PCR product was run on a low melting point agarose gel, the expected size DNA fragment was excised and purified using the QIAquick ^™ gel extraction kit (QIAGEN ™). The purified fragment was digested with PinAI and BamHI, gel purified as described above, and cloned between the corresponding sites in pVAg2M3-OST577.

  To generate a random mutant at position 314, the mutagenic primer kco78 (5′-ACC GTT GTG CAC CAG GAC TGG, assuming that K = G or T and N = A, C, G or T. NNK AAC GGC AAG GAG-3 ′) (SEQ ID NO: 86) and kco79 (5′-CCA GTC CTG GTG CAC AAC GG-3 ′) (SEQ ID NO: 87) were used. In the first round of PCR, the external primer ks g2-5 (5′-CTC CCG GAC CCC TGA GGT C-3 ′) (SEQ ID NO: 88) and kco79 for the left fragment and the external primer for the left fragment kco8 and kco78 were used. All subsequent steps show that the second round of PCR used the external primers ksg2-5 and kco8 and the final PCR fragment was digested with Pm1I and BamHI and cloned into the corresponding sites in pVAg2M3-OST577. Except as described above for random mutagenesis at position 250.

  To generate the L314I mutant, mutagenic primers MGD-10 (5′-ACC GTT GTG CAC CAG GAC TGG ATC AAC GGC AAG GA-3 ′) (SEQ ID NO: 89) and kco79 were used. To generate the L314Y mutant, mutagenic primers MGD-11 (5′-ACC GTT GTG CAC CAG GAC TGG TAT AAC GGC AAG GA-3 ′) (SEQ ID NO: 90) and kco79 were used. To generate the L314H mutant, mutagenic primers MGD-12 (5′-ACC GTT GTG CAC CAG GAC TGG CAC AAC GGC AAG GA-3 ′) (SEQ ID NO: 91) and kco79 were used. To generate the L314M mutant, mutagenic primers MGD-13 (5′-ACC GTT GTG CAC CAG GAC TGG ATG AAC GGC AAG GA-3 ′) (SEQ ID NO: 92) and kco79 were used. To generate the L314N mutant, mutagenic primers MGD-14 (5′-ACC GTT GTG CAC CAG GAC TGG AAT AAC GGC AAG GA-3 ′) (SEQ ID NO: 93) and kco79 were used.

In the first round of PCR, the external primer jt240 (5′-GGA CAC CTT CTC TCC TCC C-3 ′) (SEQ ID NO: 94) and kco79 are used for the left fragment and the external primer for the right fragment. kco41 (5′-ATT CTA GTT GTG GTT TGT CC-3 ′) (SEQ ID NO: 95) and MGD-10, MGD-11, MGD-12, MGD-13 or MGD-14 were used. PCR reactions, first, at 5 minutes 94 ° C., then 5 seconds 94 ° C., by incubation for 5 seconds 60 ° C. and 60 seconds 72 ° C. for 25 cycles, and then incubated at 7 minutes 72 ° C., Expand ^TM High Performed using a fidelity PCR system (Roche Diagnostics Corporation).

PCR products were run on a low melting agarose gel, excised from the gel and melted at 70 ° C. The second round of PCR to combine the left and right fragments uses the external primers jt240 and kco41, first at 94 ° C for 5 minutes, then at 94 ° C for 5 seconds, 60 ° C for 5 seconds and 72 ° C for 90 seconds. This was done as described above by incubating in 35 cycles followed by 7 minutes at 72 ° C. The final PCR product was run on a low melting point agarose gel, the expected size DNA fragment was excised and purified using the QIAquick ^™ gel extraction kit (QIAGEN ™). The purified fragment was subcloned in pCR ™ 4Blunt-TOPO ™ (Invitrogen ^™ , Carlsbad, Calif.), Then digested with PmII and Bam HI, gel purified as described above, and contained in pVAg2M3-OST577. Cloned between corresponding sites.

  To generate a random mutant at position 428, the mutagenesis primer JY22 (5′-GAA CGT CTT CTC ATG is first assumed as K = G or T and N = A, C, G or T. CTC CGT GNN KCA TGA GGC TCT G-3 ′) (SEQ ID NO: 96) and JY23 (5′-CAC GGA GCA TGA GAA GAC GTT C-3 ′) (SEQ ID NO: 97) were used. In the first round of PCR, the outer primers ksg2-5 and JY23 were used for the left fragment and the outer primers kco8 and JY22 were used for the right fragment. All subsequent steps were performed as described above for position 314 random mutagenesis.

To generate additional random mutants at position 428, the mutagenic primer MGD-2 (5′-GTG TAG TGG TTG TGC AGA GCC TCA TGM NNC ACG GAG CAT GAG AAG-3 ′) (SEQ ID NO: 98) and KH1 (5′-CAT GAG GCT CTG CAC AAC CAC TAC AC-3 ′) (SEQ ID NO: 99) were used . In the first round of PCR, external primers msc g2-1 and MGD-2 were used for the left fragment and external primers MGD-1 and KH1 were used for the right fragment. The PCR reaction was first performed at 25 ° C for 5 minutes followed by 25 cycles of 94 ° C for 5 seconds, 60 ° C for 5 seconds and 72 ° C for 90 seconds followed by 7 minutes at 72 ° C for Expand ^™ high fidelity. This was done using a PCR system (Roche Diagnostics Corporation).

PCR products were run on a low melting agarose gel, excised from the gel and melted at 70 ° C. The second round of PCR to combine the left and right fragments, using external primers g2-1 and MGD-1, first at 94 ° C for 5 minutes, then 94 ° C for 5 seconds, 60 ° C and 75 seconds for 5 seconds. This was done as described above by incubating with 35 cycles of 72 ° C. per second followed by 7 minutes at 72 ° C. The final PCR product was run on a low melting point agarose gel and the expected size DNA fragment was excised and purified using the QIA quick ^™ gel extraction kit (QIAGEN ™). The purified fragment was digested with PinAI and Bam HI, gel purified as described above, and cloned between the corresponding sites in pVAg2M3-OST577.

  To generate M428E mutants, MGD-8 (5′-GTG TAG TGG TTG TGC AGA GCC TCA TGT TCC ACG GAG CAT GAG AAG-3 ′) (SEQ ID NO: 100) and KH1 were used. In the first round of PCR, external primers mscg2-1 and MGD-8 were used for the left fragment and external primers MGD-1 and KH1 were used for the right primer. All subsequent steps were performed as described above for position 428 random mutagenesis.

  The T250E / M428F double mutant replaces the PmlI-Bam HI fragment in the pVAg2M3-OST577 plasmid containing the T250E mutation with the corresponding PmlI-Bam HI fragment from the pVAg2M3-OST577 plasmid containing the M428F mutation. Generated by. The T250Q / M428F and T250Q / M428L double mutants are the PmlI-Bam HI fragment in the pVAg2M3-OST577 plasmid containing the T250Q mutation and the corresponding PmlI from the pVAg2M3-OST577 plasmid containing the M428F and M428L mutations, respectively. -Generated by replacing with Bam HI fragment.

Similarly, multiple amino acid substitutions were made at positions 250 and 428 of the HulD10-IgG2M3 heavy chain. To generate the M428L mutant, the XhoI-XbaI fragment from the hCMV promoter and enhancer (Boshart et al. Supra) and the plasmid pVAg2M3-OST577 (M428L) containing the OST577V _H region was transformed into the hCMV promoter and enhancer and Hul. It was replaced with the corresponding XhoI-XbaI fragment from plasmid pHulD10IgG1.rgpt.dE (Kostelny et al. (2001) supra) containing the D10V _H region. To generate the T250Q / M428L mutant, the XCM-XbaI fragment from the plasmid pVAg2M3-OST577 (T250Q / 428L) containing the hCMV promoter and enhancer (Boshart et al. Supra) and the OST577V _H region was It was replaced with the corresponding XhoI-XbaI fragment from plasmid pHulD18.IgG1.rgpt.dE (Kostelny et al, (2001), supra) containing the enhancer and the HulD10V _H region.

Plasmid DNA was prepared using the QIAprep ^™ Spin Miniprep kit (QIAGEN ™) and nucleotide substitutions were identified by sequencing. Large-scale plasmid DNA preparations were made using the Endo Free ™ Plasmid Maxi kit (QIAGEN ™). The coding region of the OST577-IgG2M3 expression plasmid was confirmed by nucleotide sequence.

Result :
To isolate human IgG mutants with higher or lower affinity for the neonatal Fc receptor (FcRn) that would be expected to have an altered serum half-life (Kabat et al., Supra. Random amino acid substitutions were generated at positions 250, 314, and 428 of the human γ2M3 heavy chain (numbered according to the EU index). These three positions represent the X-ray crystal structure of the rat Fc / FcRn complex (computer modeling of the human IgG2M3Fc and human FcRn complex deduced from Burmeister et al., Supra, model 1, described at the beginning of the example, (See 2 and 3.) The wild-type amino acids at positions 250, 314 and 428 are near the Fc / FcRn interface, but these residues have a pH-dependent interaction between Fc and FcRn. Does not seem to contribute directly to the action.

  Thus, amino acid substitutions at these positions may increase (or decrease) the affinity of Fc for FcRn while maintaining pH-dependent binding. Single mutants generated by PCR-based mutagenesis include wild type T at position 250, A, C, D, E, F, G, H, I, K, L, M, N, P , Q, R, S, V, W, or 19 mutants converted to wild type L at position 314 A, C, D, E, F, G, H, I, K, M, N 19 mutants converted to P, Q, R, S, T, V, W, or Y; and the wild type M at position 428 is A, C, D, E, F, G, H, I, There were 19 mutants converted to K, L, N, P, Q, R, S, T, V, W, or Y (see Table 1). Some of the single mutants with increased binding to FCRn were combined to generate multiple double mutants including T250E / M428F, T250Q / M428F and T250Q / M428L.

Example 4 .
This example describes mutagenesis of the Fc region of the human γ1 heavy chain gene.
Mutagenesis :
Use the overlap extension PCR method (Higuchi, supra) to generate amino acid substitutions at positions 250 and 428 of the OST577-IgG1 heavy chain (numbered according to the EU index of Kabat et al, supra). did. To generate T250E mutants, mutagenic primers JX076 (5′-AAC CCA AGG ACG AAC TCA TGA TCT CCC G-3 ′) (SEQ ID NO: 101) and JX077 (5′-GGA GAT CAT GAG TTC GTC CTT GGG TTT TG-3 ′) (SEQ ID NO: 102) was used. In the first round of overlap extension PCR, the external primers JX080 (5′-CCT CAG CTC GGA CAC CTT CTC-3 ′) (SEQ ID NO: 103) and JX077 are used for the left fragment and for the right fragment. The external primers NT244 (5′-GCC TCC CTC ATG CCA CTC A-3 ′ (SEQ ID NO: 104)) and JX076 were used.

The PCR reaction was incubated in GeneAmp ™ PCR System 9600 (Applied Biosystems ™), first for 5 minutes at 94 ° C, then for 35 cycles of 94 ° C for 20 seconds, 55 ° C for 20 seconds and 72 ° C for 90 seconds, This was then performed using the Expand ^™ high fidelity PCR system (Roche Diagnostics Corporation) according to the manufacturer's recommendations by incubating for 7 minutes at 72 ° C. PCR products were run on a low melting agarose gel, excised from the gel and melted at 70 ° C. The second round of PCR to combine the left and right fragments was performed as described above using 35 cycles, external primers JXO80 and NT244. The final PCR product was run on a low melting point agarose gel and the expected size DNA fragment was excised and purified using the QIAquick ^™ II gel extraction kit (QIAGEN ™). The purified fragment was digested with NheI and EagI, gel purified as described above, and cloned between the corresponding sites in pVAg.N-OST577.

  To generate T250D mutants, mutagenic primers JX087 (5′-AAC CCA AGG ACG ACC TCA TGA TCT CCC G-3 ′) (SEQ ID NO: 105) and JX088 (5′-GGA GAT CAT GAG GTC GTC CTT GGG TTT TG-3 ′) (SEQ ID NO: 106) was used. In the first round of PCR, external primers JX080 and JX088 were used for the left fragment, and external primers NT244 and JX087 were used for the right fragment. All subsequent steps were performed as described above.

  To generate M428F mutants, mutagenic primers JX078 (5′-CTC ATG CTC CGT GTT CCA TGA GGC TCT GC-3 ′) (SEQ ID NO: 107) and JX079 (5′-AGA GCC TCA TGG AAC ACG GAG CAT GAG-3 ′) (SEQ ID NO: 108) was used. In the first round of PCR, external primers JX080 and JX079 were used for the left fragment and external primers NT244 and JX078 were used for the right fragment. All subsequent steps were performed as described above.

  To generate M428L mutants, mutagenic primers JXM428L1 (5′-CTC ATG CTC CGT GTT GCA TGA GGC TCT GC-3 ′) (SEQ ID NO: 109) and JXM428L2 (5′-AGA GCC TCA TGC AAC ACG GAG CAT GAG-3 ′) (SEQ ID NO: 110) was used. In the first round of PCR, external primers JX080 and JXM428L2 were used for the left fragment and external primers NT244 and JXM428L1 were used for the right fragment. All subsequent steps were performed as described above.

  To generate T250Q mutants, mutagenic primers JXT250Q1 (5′-AAC CCA AGG ACC AAC TCA TGA TCT CCC G-3 ′) (SEQ ID NO: 111) and JXT250Q2 (5′-GGA GAT CAT GAG TTG GTC CTT GGG TTT TG-3 ′) (SEQ ID NO: 112) was used. In the first round of PCR, external primers JX080 and JXT250Q2 were used for the left fragment, and external primers NT244 and JXT250Q1 were used for the right fragment. All subsequent steps were performed as described above.

  To generate a T250E / M428F double mutant, mutagenic primers JX076 and JX077 were used to create a T250E mutation in a template containing the M428 mutation. In the first round of PCR, external primers JX080 and JX077 were used for the left fragment, and external primers NT244 and JX076 were used for the right fragment. All subsequent steps were performed as described above.

  To generate the T250Q / M428L double mutant, mutagenic primers JXT250Q1 and JXT250Q2 were used to create the T250Q mutation, and mutagenic primers JXM428L1 and JXM428L2 were used to create the M428L mutation. In the first round of PCR, external primers JX080 and JXT250Q2 were used to create the left fragment, JXT250Q1 and JXM428L2 were used to create the middle fragment, and external primers NT244 and JXM428L1 were used to create the right fragment. All subsequent steps were performed as described above.

Similarly, multiple amino acid substitutions were made at positions 250 and 428 of the HulD10-IgG1 heavy chain. To generate the 428L mutant, the XhoI-XbaI fragment from the plasmid pVAg1.N-OST577 (M428L) containing the hCMV promoter and enhancer (Boshart et al., Supra) and the OST577V _H region was transformed into the hCMV promoter and enhancer. And the corresponding XhoI-XbaI fragment from plasmid pHulD10IgG1.rgpt.dE (Kostelny et al. (2001) supra) containing the Hul D10V _H region. To generate the T250Q / M428L mutant, an XhoI-XbaI fragment from the plasmid pVAg1.N-OST577 (T250Q / 428L) containing the hCMV promoter and enhancer (Boshart et al. Supra) and the OST577V _H region is It was replaced with the corresponding XhoI-XbaI fragment from plasmid pHulD18.IgG1.rgpt.dE (Kostelny et al, (2001), supra) containing the promoter and enhancer and the HulD10V _H region.

Plasmid DNA was prepared using the QIAprep ^™ Spin Miniprep kit (QIAGEN ™) and nucleotide substitutions were confirmed by sequencing. Large-scale plasmid DNA preparations were made using the Endo Free ™ Plasmid Maxi kit (QIAGEN ™). The coding region of the OST577-IgG1 expression plasmid was confirmed by nucleotide sequence.

Result :
To identify human IgG mutants with altered affinity for the neonatal Fc receptor (FcRn) that would be expected to have an altered serum half-life (EU index of Kabat et al., Supra). Multiple amino acid substitutions were generated at positions 250 and 428 of the human γ1 heavy chain (numbered according to These two positions were selected based on the identification of mutations at these positions within the human γ2M3 heavy chain that resulted in increased or decreased binding to FcRn. Although the wild type amino acids at positions 250 and 428 are near the Fc / FcRn interface, these residues do not appear to contribute directly to the pH-dependent interaction between Fc and FcRn. Thus, amino acid substitutions at these positions may increase (or decrease) the affinity of Fc for FcRn while maintaining pH-dependent binding. Both single and double mutants that showed increased binding in the context of the human γ2M3 heavy chain are the single mutants T250E, T250Q, M428F and M428L and the double mutants T250E / M428F and T250Q. It was evaluated in human γ1 heavy chain including / M428L. Single mutants that showed reduced binding in the context of the human γ2M3 heavy chain (T250D) were also evaluated in the human γ1 heavy chain.

Example 5 .
This example describes the characterization of mutant IgG2M3 and IgG1 antibodies.
Cell culture :
Human kidney cell line 293-H (Life Technologies ™, Rockville, MD) was obtained in 10% fetal bovine serum (FBS) (HyClone ™, hereinafter referred to as 293 medium at 37 ° C. in a 7.5% CO ₂ incubator. ), Logan, UT), DMEM (Bio Whittaker ^™ , Walkersville, MD) containing 0.1 mM MEM-essential amino acid (Invitrogen ^™ ) and 2 mM L-glutamine (Invitrogen ^™ ). For the expression and purification of monoclonal antibodies following transient transfection, 293-H cells are referred to as low-IgG293 medium, 10% low-IgG FBS (HyClone ™), 0.1 mM MEM-essential amino acid. And incubated in 2 mM L-glutamine. The mouse myeloma cell line Sp2 / 0 (US Standard Culture Collection, Manassus, VA) was maintained in DMEM containing 10% FBS and 2 mM L-glutamine. Sp2 / 0 cells were adapted for growth on hybridoma-SFM (HSFM) (Life Technologies ™) for purification of monoclonal antibodies after stable transfection.

Transient transfection :
Of 293-H cells, appropriate light chain plasmids and appropriate wild type heavy chain plasmids, or heavy chain plasmids with various mutations containing single or double amino acid substitutions at positions 250, 314 or 428. One of the cells was transiently co-transfected. For small-scale transient transfections, plate approximately 1 × 10 ⁶ cells per transfection in 3 ml of 293 medium in a 6-well plate and confluent overnight Grown up to. The next day, 2 μg light chain plasmid and 2 μg wild type or mutated heavy chain plasmid were combined with 0.25 ml HSFM.

In separate tubes, 10 μl Lipofectamine ^™ 2000 reagent (Invitroge ^™ ) and 0.25 ml HSFM were combined and incubated for 5 minutes at room temperature. 0.25 ml of Lipofectamine ^™ 2000-HSEM mixture was gently mixed with 0.25 ml of DNA HSFM mixture and incubated for 20 minutes at room temperature. Aspirate the medium covering 293-H cells, replace with low-IgG293 medium, then add Lipofectamine-DNA complex dropwise to the cells, mix gently by swirling, and add the cells to 7.5% CO ₂ The supernatant was harvested after incubation in an incubator at 37 ° C. for 5-7 days.

For large-scale transient transfections, plate approximately 7 × 10 ⁶ cells in 25 ml 293 medium in a T-75 flask per transfection and confluent overnight I grew it until. The next day, 12 μg light chain plasmid and 12 μg wild type or mutated heavy chain plasmid were combined with 1.5 ml HSFM. In separate tubes, 60 μl Lipofectamine ^™ 2000 reagent and 1.5 ml HSFM were combined and incubated for 5 minutes at room temperature. 1.5 ml of Lipofectamine ^™ 2000-HSEM mixture was gently mixed with 1.5 ml of DNAHSFM mixture and incubated for 20 minutes at room temperature. Aspirate the medium covering the 293-H cells and replace with low-IgG273 medium, then add the lipofectamine-DNA complex dropwise to the cells, mix gently by swirling, and then add the cells to 7.5% CO ₂ The supernatant was harvested after incubation in an incubator at 37 ° C. for 5-7 days.

Antibody concentration The supernatant from the small-scale transient transfection was harvested by centrifuging at 1200 rpm for 5 minutes and 0.22 μm Millex ™ -GV microfilter (Millipore ™ Corporation, Bedford, Mass.) Aseptically filtered. Specimens were concentrated approximately 6-fold to 0.5 ml using a 6 ml Vivaspin ™ concentrator (50,000 MWCO) (Vivascience ™ AG, Hannover, Germany) by centrifugation at 3000 rpm. The concentrated protein was resuspended in 5 ml PBS, pH 6.0 and concentrated to a volume of 0.5 ml as described above. The antibody concentration in each specimen was measured using the ELISA method described below.

Stable transfection :
Sp2 / 0 cells as one of the appropriate light chain plasmid and the appropriate wild type heavy chain plasmid or various mutated heavy chain plasmids containing single or double amino acid substitutions at positions 250 or 428. And transfected in a stable manner. Approximately 1 × 10 ⁷ cells were washed with 10 ml PBS and resuspended in 1 ml PBS. Approximately 25-30 μg light chain plasmid and 50-60 μg heavy chain plasmid were linearized with FspI and added to the cells. Cells and DNA were mixed gently and transferred to Gene Pulser Cuvelte (Bio-Rad ™ Laboratories, Hercules, CA) on ice. Cells were electroporated using Gene Pulser II (Bio Rad ™ Laboratories) set at 0.360 kV, 25 μF and returned to ice for 10-20 minutes. Cells were diluted in 40 ml DMEM, 10% FBS, 2 mM L-glutamine and identified in 4 96-well plates at 100 μl / well.

After 48 hours, 100 μl / well of 2 × mycophenolic acid (MPA) selection medium (DMEM, 10% FBS, 1 × HT Media Supplement Hybri-Max ™ (Sigma ™, St. Louis, MO), 300 μg / ml xanthine (Sigma ™), 2 μg / ml mycophenolic acid (Life Technologies ™) and 2 mM L-glutamine) were added. Supernatants from wells that apparently contained a single colony were screened by ELISA after 10-14 days. The best antibody producing clones were selected for expansion and adaptation to HSFM (Life Technologies ™). Matched clones are expanded into roller bottles in 450 ml HSFM, gassed with 5% CO ₂ in air, and 50 ml Protein Free Feed Medium-2 (PFFM-2) after 2 months (Sauer et al Biotechnol. Bioeng. 67: 585-597 (2000)) was supplemented and grown to exhaustion.

A portion of the best antibody-producing clone was similarly adapted to Protein Free Basal Medium-1 (PFBM-1) (Protein Design Labs ^™ , Inc.), expanded to a 10 L spinner flask, and 1/10 volume of PFFM-2 was supplemented after 2 days and grown until exhaustion.

ELISA :
An ELISA was performed to quantify the amount of OST577 or HulD10 antibody present in the culture supernatant. Immulon ^™ 4 plates (DYNEX ™ Technologies, Inc., Chantilly, VA) at 4 ° C. overnight at 1.0 μg / ml in 100 μl / well 0.2 M carbonate / bicarbonate buffer at pH 9.4. Coated with a goat F (ab ′) ₂ anti-human IgG gamma chain antibody (BioSource International, Camarillo, Calif.) Or AffiniPure ^™ goat anti-human IgG Fcy fragment specific antibody (Jackson Immuno Research Laboratories, Inc. West Grove, PA) . The next day, the plates were washed with ELISA Wash Buffer (EWB) (PBS, 0.1% Tween 20) and blocked with Super Block ™ Blocking Buffer in 300 μl / well TBS for 20-30 minutes at room temperature. Plates were washed with EWB and appropriately diluted test specimens were added to each well. Purified OST577 or HulD10 antibody was serially diluted 2-fold in 100 μl / well ELISA Buffer (EB) (PBS, 1% bovine serum albumin, 0.1% Tween20) starting from 0.2 μg / ml as appropriate.

The culture supernatant was first diluted 1:10 in 100 μl / well EB and then diluted 2-fold in 100 μl / well EB. Plates are incubated for 1-2 hours at room temperature, then washed with EWB and 100 μl / well goat anti-human lambda light chain HRP-conjugated antibody (BioSource International, or Southern Biotechnology Associates, Inc., Birmingham, AL) or goat anti-human kappa Light chain HRP-conjugated antibody (Southern Biotechnology Associates, Inc.) was added at 1.0 μg / ml in EB as appropriate. After incubating for 1 hour at room temperature, the plates were washed with EWS and then 100 μl / well ABTS Peroxidase Substrate / Peroxidase Solution B (Kirkegaard & Perry Laboratories, Gaithersburg, MD) was added. The reaction was stopped with 100 μl / well 2% oxalic acid and the absorbance at 415 nm was measured using a VER SAmax ^™ microtiter plate reader (Molecular Devices Corporation ™, Sunnyvale, Calif.).

Antibody purification :
Culture supernatants from transient transfections were harvested by centrifugation and sterile filtered. The pH of the filtered supernatant was adjusted by adding 1/50 volume of 1M sodium citrate pH 7.0. The supernatant was run on a 1 ml HiTrap ™ Protein A HP column (Amersham Biosciences ^™ Corporation, Piscataway, NJ) pre-equilibrated with 150 mM NaCl, 20 mM sodium citrate, pH 7.0. The column was washed with the same buffer and the bound antibody was eluted with 20 mM sodium citrate pH 3.5. After neutralization by adding 1/50 volume of 1.5M sodium citrate at pH 6.5, the pooled antibody fractions were pre-equilibrated with 5 ml of 120 mM NaCl, 20 mM sodium citrate at pH 6.0. On a HiTrap ™ Desalting column (Amersham Biosciences ^™ Corporation). The influent was collected and fractions with an OD ₂₈₀ > 0.1 were pooled and concentrated to a maximum of 0.5-1.0 mg / ml using a 2 ml Vivaspin ™ concentrator (50,000 Dalton MWCO) (Vivascience ™ AG) . The specimens were then filter sterilized using a 0.2 μm Millex ™ -GV microfilter (Millipore ™ Corporation). The concentration of the purified antibody was determined by UV spectroscopy by measuring the absorbance at 280 nm (1 mg / ml = 1.4A ₂₈₀ ).

For small scale purification of antibodies from stable transfection, the culture supernatant was harvested by centrifugation and sterile filtered. The supernatant was run on a 5 ml POROS ™ 50A Protein A column (Applied Biosystems ™) pre-equilibrated with PBS pH 7.4. The column was washed with the same buffer and the bound antibody was eluted with 0.1 M NaCl, 0.1 M glycine, pH 3.0. After neutralization by adding 1/20 volume of 1M Tris base, the pooled fractions were collected in PBS pH 7.4 using a PD-10 desalting column (Amersham Biosciences ^™ Corporatio) or by dialysis. The buffer was exchanged. The specimens were then filter sterilized using a 0.2 μm Millex ™ -GV microfilter (Millipore ™ Corporation). The concentration of the purified antibody was determined by UV spectroscopy by measuring the absorbance at 280 nm (1 mg / ml = 1.4A ₂₈₀ ).

For large-scale purification of antibodies from stable transfectants, cell culture harvests were clarified by dead-end filtration using Sartorius ™ filtration capsules (Sartorius ™ AG, Goettingen, Germany) . The clarified harvest was concentrated from about 10 L to 750 ml using a Pellicon ™ 2 cassette (30,000 Dalton MWCO) (Millipore ™ Corporation) and then rProteinA Sepharose FF using the citrate buffer system described above. Purified by protein A affinity chromatography on a column (Amersham Biosciences ^™ Corporation). The Protein A eluate is concentrated in an Amicon ™ stirred cell instrument using a YM30 membrane (Millipore ™ Corporation) and then pH 6.0 using a Superdex ^™ 200 column (Amersham Biosciences ^™ Corporation). Samples were buffer exchanged into 120 mM NaCl, 20 mM sodium citrate. The pH and osmotic pressure were measured and the concentration of the purified antibody was determined by UV spectroscopy by measuring the absorbance at 280 nm (1 mg / ml = 1.4A ₂₈₀ ).

SDS-PAGE
NuPAGE (TM) Novex4-12% Bis-Tris gels ( Invitrogen TM) under reducing or non-reducing conditions on run a purified antibody samples 5 [mu] g, using ^{Simply Blue TM SafeStain Kit (Invitrogen TM} ) according to the manufacturer's recommendations Stained.

Result :
IgG2M3Fc and IgG1Fc mutants are the light chain and heavy chain variable regions of OST577 (Ehrlich et al., Supra) and the light chain constant region of human lambda-2 (Hieter et al. (1981), supra, respectively. ) And human gamma-2 mutant 3 (IgG2M3) (Cole et al., Supra) and IgG (Ellison et al., Supra) and expressed as an anti-HBV antibody containing the heavy chain constant region. The IgG2M3 variant contains two amino acid substitutions within the C _H 2 region (V234A and G237A) and exhibits very low residual binding to the human Fcy receptor (Cole et al., Supra). The IgG2M3Fc and IgG1Fc mutants are similarly described in the HulD10 light and heavy chain variable regions (Kostelny et al. (2001) supra), the human kappa light chain constant region (Hieter et al. (1980) supra, respectively. ) And human IgG2M3 (Cole et al., Supra) and IgG1 (Ellison et al., Supra) and also expressed as an anti-HLA-DRβ chain allele antibody containing the heavy chain constant region.

  As described above, the appropriate wild type or mutant weight chain expression vector was transiently co-transfected with the appropriate light chain expression vector into 293-H cells for expression of the OST577 or HulD10 monoclonal antibody. ELISA analysis of culture supernatants harvested after transient transfection showed that antibody expression levels were typically 5-50 μg / ml in 25 ml supernatant. OST577 or HulD10 antibody was purified by protein A affinity chromatography to give a final yield of approximately 100-1000 μg. Stable expression of OST577 or HulD10 antibody in Sp2 / 0 cells typically resulted in expression levels of 5-50 μg / ml as determined by ELISA. A yield of approximately 50-80% of the antibody present in the culture supernatant was obtained by small scale protein A affinity chromatography.

  The purified antibody was characterized by SDS polyacrylamide gel electrophoresis (SDS-PAGE) under non-reducing and reducing conditions. SDS-PAGE analysis under non-reducing conditions shows that the purified antibody has a molecular weight of about 150-160 kD (data not shown) Analysis under reducing conditions shows that the purified antibody has a molecular weight of about 50 kD It was shown to be composed of a heavy chain and a light chain with a molecular weight of approximately 25 kD (see FIGS. 10A and 10B). SDS-PAGE analysis of antibodies purified from stable Sp2 / 0 transfectants provided results similar to those observed with purified antibodies from transient 293-H transfection.

Example 6 .
This example describes the competitive binding analysis of mutant IgG2M3 and IgG1 antibodies.
Cell culture :
The mouse myeloma cell line NS0 (European Collection of Animal Cell Cultures, Salisbury, Wiltshire, UK) was maintained in DMEM with 10% FBS. NS0 transfectants expressing recombinant GPI-linked human or rhesus monkey FcRn on the surface were added to mycophenolic acid (MPA) selective medium (DMEM, 10% FBS, 1xHT Media Supplement Hybri-Max ™ (Sigma (Trademark)), 250 μg / ml xanthine (Sigma ™), 1 μg / ml mycophenolic acid (Life Technologies ™ and 2 mM L-glutamine) or 2 × MPA selective medium.

Human FcRn cell line
NS0 cells were transfected stably with pDL208. Approximately 1 × 10 ⁷ cells were washed once, resuspended in 1 ml of simple DMEM, transferred to Gene Pulser ^™ Cuvette (Bio-Rad ™ Laboratories) and incubated on ice for 10 minutes. 40 μg of plasmid pDL208 is linearized with FspI, gently mixed with cells on ice and then pulsed twice using Gene Pulser ^™ II (Bio-Rad ™ Laboratories) set to 1.5 kV, 3 μF The cells were electroporated with and returned to ice for 10 minutes. Cells were diluted in 20 ml DMEM, 10% FBS and fixed in 2 96-well plates at 100 μl / well. The medium was replaced with MPA selection medium after 48 hours. Mycophenolic acid resistant NS0 transfectants from wells containing apparently single colonies were expanded in MPA selection medium and screened by FACS ^™ after approximately 3 weeks.

Approximately 1.5 × 10 ⁵ cells per test, 100 μl FACS containing 10 μg / ml biotinylated mouse anti-human B2-microglobulin antibody (Chromaprobe, Inc., Aptos, Calif.) For 1 hour on ice Incubation was performed in Staining Buffer (FSB) (PBS, 1% FBS, 0.1% NaN ₃ ). The cells were washed once with 4 ml FSB and then incubated in 25 μl FSB (Southern Biotechnology Assosciates, Inc.) containing 20 μg / ml streptavidin-FITC conjugate. Cells were washed once with 4 ml FSB and resuspended in 1% formaldehyde. Samples were analyzed for antibody binding to human β2m using FACS can flow cytometry (BD ™ Biosciences, San Josc, Calif.). Multiple clones with the best apparent staining were subcloned using a FACStar cell sorter (BD ™ Biosciences), expanded in DMEM, 10% FBS, 2 mM L-glutamine and as described above. Retested by FACS ^™ . One subclone, designated NS0 HuFcRn (memb), clone 7-3, was used in subsequent binding assays.

Rhesus monkey FcRn cell line :
NS0 cells were transfected stably with pDL410. Approximately 6 × 10 ⁵ cells were transfected by electroporation as described above. Cross-react with rhesus monkey FcFn alpha chain, stained with 100 ng of mouse anti-human FcRn alpha chain antibody (Protein Design Labs ^™ , Inc.) per test and goat anti-mouse kappa FITC conjugated antibody (Southern Biotechnology Associates, Inc. The transfectants were identified by FACS ^{™ as} described above. Subsequent binding assays used a cell line designated NS0RhFcRn, clone-3.

Single point competitive binding test :
Concentrated OST577-IgG2M3 supernatant was tested in a single point competitive binding assay for binding to human FcRn on cell line NS0 HuFcRn (memb), clone 7-3. Approximately 2 x 10 ⁵ cells per test are performed once in pH8.0 FACS Binding Buffer (FBB) (PBS containing 0.5% BSA, 0.1% NaN ₃ ), pH 6.0 FBB. In premixed biotinylated OST577-IgG2M3 antibody (8.3 μg / ml) and concentrated supernatant (containing 8.3 μg / ml competing antibody) in 120 μl, washed once in FBB at pH 6.0 Resuspended.

Cells were incubated on ice for 1 hour, washed twice in pH 6.0 FBB, diluted to 2.5 μg / ml in pH 6.0 FBB, 25 μl streptavidin-RPE conjugate (Bio Source International) Resuspended in. After incubating on ice for 30 minutes in the dark, the cells were washed twice in FBB pH 6.0 and resuspended in 1% formaldehyde. Samples were analyzed for antibody binding to FcRn by FACS ^™ using a FACS Calibur flow cytometer (BD ™ Biosciences). The mean channel fluorescence (MCF) of each mutant was compared to that of the wild type antibody and plotted using Excel (Microsoft ™ Corporation, Redmond, WA).

Competitive binding test :
A dilution series of each purified OST577-IgG2M3 antibody was obtained from the biotinylated HuEP5C7-IgG2M3 antibody (He et al., J. Immunolol) for binding to human FcRn on cell line NS0 HuFuRn (memb) clone 7-3. 160; 1029-1035 (1998)). For initial screening experiments, approximately 2 × 10 ⁵ cells per test were washed once in pH 6.0 FSB and pre-mixed biotinylated HuEP5C7-IgG2M3 antibody in pH 6.0 FSB ( 10 μg / ml) and OST577-IgG2M3 competing antibody (2 × serial dilution from 208 μg / ml to 0.102 μg / ml) resuspended in 100 μl.

Cells were incubated with the antibody mixture for 1 hour on ice, washed twice in pH 6.0 FSB, diluted to 2.5 μg / ml in pH 6.0 FSB, 25 μl streptavidin-RPE conjugate (Bio Source International). After 30 minutes incubation on ice in the dark, the cells were washed twice in pH 6.0 FSB and resuspended in 1% formaldehyde. Samples were analyzed for antibody binding to FcRn by FACS ^™ using a FACScan flow cytometer (BD ™ Biosciences). Mean channel fluorescence (MCF) was plotted against competitor concentration and IC50 values were calculated using Graph Pad Prism ™ (Graph Pad ^™ Software, Inc., San Diego, Calif.). For consistency, the IC50 values shown in the table are based on the final competitive concentration.

  Subsequent competition binding experiments consisted of washing the cells once in pH 8.0 FBB, once in pH 6.0 FBB, and then premixed biotinylated HuEP5C7-IgG2M3 antibody in pH 6.0 FBB. (10 μg / ml) and OST577-IgG2M3 competing antibody (2 × serial dilution from 208 μg / ml to 0.102 μg / ml) except that resuspended in 100 μl. All subsequent incubations and washes were performed with FBB at pH 6.0 as described above. 120 μl premixed biotinylated OST577-IgG2M3 antibody (8.3 μg / ml) and OST577-IgG2M3 competing antibody (2 × serial dilution from 208 μg / ml to 0.102 μg / ml) in FBB pH 6.0 as described above One group of experiments was conducted. Another group of experiments was performed as described above, premixed biotinylated OST577-IgG1 antibody (5.0 μg / ml) and OST577-IgG1 competitor antibody (2 × serial dilution from 125 μg / ml) in pH 6.0 FBB. Or three-fold serial dilution from 250 μg / ml) in 200 μl.

Yet another group of experiments was performed as described above, 3 × starting from premixed biotinylated HuEP577-IgG1 antibody (5.0 μg / ml) and OST577-IgG1 competitor antibody (750 μg / ml) in pH 6.0 FBB. Serial dilution) was performed in 200 μl. A dilution series of each purified OST577-IgG2M3 antibody was competed against biotinylated OST577-IgG2M3 antibody for binding to rhesus monkey FcRn on cell line NS0 RhFc, clone R-3. In a group of experiments, about 2 × 10 ⁵ cells per test are washed once in pH 8.0 FSB and once in pH 6.0 FBB, then in pH 6.0 FSB Resuspended in 120 μl of premixed biotinylated OST577-IgG2M3 antibody (8.3 μg / ml) and OST577-IgG2M3 competitor antibody (2 × serial dilution from 208 μg / ml to 0.102 μg / ml).

Cells were incubated with the antibody mixture for 1 hour on ice, washed twice in pH 6.0 FBB, diluted to 2.5 μg / ml in pH 6.0 FBB, 25 μl streptavidin-RPE conjugate (Bio Source International). After incubating on ice for 30 minutes in the dark, the cells were washed twice in FBB pH 6.0 and resuspended in 1% formaldehyde. Samples were analyzed for antibody binding to FcRn by FACS ^™ using a FACSCalibur flow cytometer (BD ™ Biosciences). Another group of experiments was performed as described above, premixed biotinylated OST577-IgG1 antibody (5.0 μg / ml) and OST577-IgG1 competitor antibody (starting from 500 μg / ml) in pH 6.0 FBB. Serial dilution) was performed in 200 μl.

Result :
Determine relative binding of wild-type OST577-IgG2M3 or OST577-IgG1 antibody and its various mutants to FcRn using a transfected NS0 cell line that stably expresses human FcRn on its surface did. As described above, the concentrated supernatant was tested for binding to human FcRn according to a single point competition assay, and the purified antibody was tested for FcRn binding in a competitive binding assay. Increasing concentrations of unlabeled competing antibodies were incubated with cells in the presence of saturating labeled IgG2M3 or IgG1 antibodies in pH 6.0 FSB or FBB.

  The results of a standard experiment with concentrated OST577-IgG2M3 supernatant are shown in FIGS. 11A, 11B and 11C. As shown in FIG. 11A, some of the mutants at position 250 (eg, T250E, T250Q) are stronger competitors than wild type, and these mutants are increased compared to wild type antibodies. Suggested that it has binding to human FcRn. Other mutants at this position (eg T250D, T250F, T250K, T250N, T250P, T250R, T250W, T250Y) are weaker competitors than the wild type, and these mutants are compared to the wild type antibody Suggesting that it has reduced binding to human FcRn. As shown in FIG. 11B, none of the mutants at position 314 are stronger competitors than the wild type, and none of these mutants has increased binding to human FcRn compared to the wild type antibody. It was suggested that it does not have.

  Most of the mutants at this position (e.g., L314A, L314C, L314D, L314E, L314F, L314G, L314H, L314K, L314M, L314N, L314P, L314Q, L314R, L314S, L314T, L314V, L314W, L314Y) It was a weaker competitor than the wild type, suggesting that these mutants have reduced binding to human FcRn compared to the wild type antibody. As shown in FIG. 11C, some of the mutants at position 428 (eg, M428F, 428L) are stronger competitors than wild type and these mutants are increased compared to wild type antibodies. Suggested that it has binding to human FcRn. Other mutants in this position (eg M428A, M428C, M428D, M428E, M428G, M428H, M428K, M428N, M428P, M428Q, M428R, M428S, M428T, M428V, M428Y) are less competitive than the wild type Material, suggesting that these mutants have reduced human FcRn compared to wild-type antibody.

  Table 2 shows IC50 values of purified wild type OST577-IgG2M3 antibody and its mutants (amount of competing antibody required to inhibit 50% of labeled antibody binding to FcRn) for competition against human FcRn. It is summarized. Relative binding values were calculated as the ratio between the IC50 value of wild type OST577-IgG2M3 antibody and the IC50 value of each mutant. At amino acid position 314, none of the purified antibodies showed increased binding to human FcRn in relation to the wild type antibody. In fact, all four of the purified mutants at position 314 showed reduced binding in relation to the wild type antibody. However, at amino acid position 250, one of the mutants (T250E) showed about 6-fold better binding to human FcRn than the wild-type antibody. Multiple mutants at position 250 show slightly reduced binding in relation to wild-type antibody, and one mutant (T250D) against human FcRn substantially reduced in relation to wild-type antibody. Showed binding. At amino acid position 428, one of the mutants (M428) also showed about three times better binding to human FcRn than the wild type antibody, while the other mutant (M428G) Showed substantially reduced binding to human FcRn.

  Since two amino acid substitutions at two different positions, i.e. T250E, T250Q, M428F and M428L, each showing increased binding to human FcRn, were identified, the double mutants T250E / M428F, T250Q / M428F and T250Q / M428L was constructed, transiently transfected into 293-H cells, purified, and tested for binding to human FcRn. As described above, increasing concentrations of unlabeled antibody were incubated with cells expressing human FcRn in the presence of an unsaturated concentration of labeled HuEP5C7-IgG2M3 or OST577-IgG2M3 antibody in pH 6.0. .

  As shown in Figure 12A, the double mutant (T250E / M428F) showed better binding to human FcRn than any of the single mutants (T250E / M428F), compared to the wild-type antibody About 15 times better binding to human FcRn. As shown in FIG. 12B, the double mutant (T250Q / M428L) is a better human than either the single mutant (T250Q or M428L) or the double mutant (T250Q / M428F). It showed binding to FcRn and about 28 times better binding to human FcRn than the wild-type antibody.

  As summarized in Table 3, in one experimental group, the IC50 for the wild-type OST577-IgG2M3 antibody is up to 9 μg / ml, while for each of the single mutants (T250E and M428F) IC50 of up to 3 μg / ml and IC50 for double mutant (T250E / M428F) is less than 1 μg / ml. As summarized in Table 4, in another experimental group, the IC50 for wild-type OST577-IgG2M3 antibody was up to 12 μg / ml, while each of the single mutants (T250Q and M428L) The IC50 for each of the double mutants (T250Q / M428F) is up to 2-4 μg / ml and the IC50 for the double mutant (T250Q / M428L) is less than 1 μg / 1 ml.

  Wild type OST577-IgG1, single mutants T250E, T250Q, M428F, M428L and double mutants T250E / M428F and T250Q / M428L were created as well. As summarized in Table 5, in one experimental group, the IC50 for the wild type OST577-IgG1 antibody was up to 14 μg / ml, while the IC50 for each of the single mutants (T250E and M428F). Is up to 3-5 μg / ml and the IC50 for the double mutant (T250E / 428F) is less than 1 μg / ml. As summarized in Table 6, in another experimental group, the IC50 for the wild-type OST577-IgG1 antibody was up to 10 μg / ml, while the IC50 for the single mutant (T250Q) was up to 3 μg. IC50 for single mutant (M428L) and double mutant (T250Q / 428L) is less than 1 μg / ml.

  Binding of a portion of OST577-IgG2M3 and its mutants to rhesus monkey FcRn was tested in competitive binding experiments. As summarized in Table 7, the IC50 for wild-type OST577 / IgG2M3 antibody is up to 15 μg / ml, whereas single mutant (T250Q and M428L) and double mutant (T250Q / 428F) The IC50 for each of these is up to 2-4 μg / ml and the IC50 for the double mutant (T250Q / 428L) is less than 1 μg / ml. Binding of a portion of OST577-IgG1 and its mutants to rhesus monkey FcRn was also tested in competitive binding experiments. As summarized in Table 8, the IC50 for the wild type OST577-IgG1 antibody is up to 9 μg / ml, while the IC50 for the single mutant (T250Q) is up to 3 μg / ml, The IC50 for the single mutant (M428L) and the double mutant (T250Q / M428L) have an IC50 of less than 1 μg / 1 ml.

Example 7 .
This example describes confirmation of the properties of IgG2M3 and IgG1 mutants in FcRn binding.
Direct binding test :
The purified OST577-IgG2M3 antibody was tested for binding to untransfected NS0 cells in pH 6.0 FBB or to human FcRn on cell line NS0HuFcRn (memb), clone 7-3. Wash approximately 2 x 10 ⁵ cells once per test in pH 8.0 FBB, once in pH 6.0 FBB, then 11 μg / ml in pH 6.0 FBB Resuspended in 100 μl of antibody at a concentration of

Cells were incubated with antibody for 1 hour on ice, washed twice in pH 6.0 FBB, 25 μl goat anti-human IgG RPE conjugated antibody (Southern, diluted to 5 μg / ml in pH 6.0 FBB) Biotechnology Associates, Inc). After 30 minutes incubation on ice in the dark, cells were washed twice in pH 6.0 FBB and resuspended in 1% formaldehyde. Samples were analyzed for antibody binding to FcRn by FACS ^™ using a FAC Scan flow cytometer (BD ™ Biosciences). The mean channel fluorescence (MCF) of each mutant was plotted using Excel (Microsofto ™ Coaporation).

Competitive binding test :
Each purified OST577-IgG2M3 antibody dilution series was competed against biotinylated OST577-IgG2M3 antibody for binding to human FcRn on cell line NS0 HuFcRn (memb), clone 7-3 at 37 ° C. Approximately 2 x 10 ⁵ cells per test are washed once in pH 8.0 FBB, once in pH 6.0 FBB and then pre-mixed in pH 6.0 FBB. Biotinylated OST577-IgG2M3 antibody (10 μg / ml) and OST577-IgG2M3 competitor antibody (2 × serial dilution from 208 μg / ml to 0.102 μg / ml) were resuspended in 100 μl.

Cells were incubated with antibody mixture for 1 hour at 37 ° C., washed twice in pH 6.0 FSB, diluted to 2.5 μg / ml in pH 6.0 FBB, 25 μl streptavidin-RPE conjugate ( Bio Source International). After 30 minutes incubation in the dark, the cells were washed twice in pH 6.0 FBB and resuspended in 1% formaldehyde. Samples were analyzed for antibody binding to FcRn by FACS ^™ using a FACScan flow cytometer (BD ™ Biosciences). Mean channel fluorescence (MCF) was plotted against competing concentrations and IC50 values were calculated using GraphPad Prism ™ (GraphPad ^™ Software).

pH Dependent Binding and Release Assay Purified OST577-IgG2M3 and OST577-IgG1 mutant antibodies are compared to their respective wild type antibodies for binding to human FcRn, followed by NS0 HuFcRn (memb), clone 7-3 Release at various pH values in a single point binding and release assay. Approximately 2 × 10 ⁵ cells per test were washed once in pH 8.0 FBB, once in pH 6.0 FBB, then purified antibody (10 μg in pH 6.0 FBB) / ml) resuspended in 100 μl. Cells are incubated on ice for 1 hour, washed twice in FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0, and goat F (ab ′) diluted to 1.25 μg / ml in FBB at appropriate pH ₂ Resuspended in 25 μl of anti-human IgG FITC conjugated antibody (Southern Biotechnology Associates, Inc). After 30 minutes incubation on ice in the dark, the cells were washed twice in FBB at the appropriate pH and resuspended in 1% formaldehyde. Samples were analyzed for antibody binding to FcRn by FACS ^™ using a FACS Calibur flow cytometer (BD ™ Bioscience). The mean channel fluorescence (MCF) of each mutant was plotted using Excel (Microsoft ™ Corporation).

  Purified OST577-IgG2M3 and OST577-IgG1 mutant antibodies are compared to their respective wild-type antibodies for binding to rhesus monkey FcRn and then in a one-point binding and release assay using NS0 RhFcRn clone R-3 as described above. Release at various pH values.

Result :
The binding properties of wild-type OST577-IgG2M3 or OST577-IgG1 antibody to human FcRn were confirmed using a transfected NS0 cell line that stably expresses human FcRn on its surface. Confirm that the binding of the mutant antibody was specific for human FcRn on the transfected NS0 cell line rather than through some other receptor or by an unknown mechanism Therefore, the antibodies were tested for binding to untransfected NS0 cells as opposed to transfected NS0 cell lines that stably express human FcRn. As described above, cells were incubated with an unsaturated concentration of antibody in FBB at pH 6.0 and binding was analyzed by FACS ^™ . As shown in FIG. 13, the results show that there is no apparent binding to the parent NS0 cell line and the antibody specifically binds to untransfected cells via human FcRn. It meant that it was suggested.

  In order to confirm that each of the mutants generated in the present invention behaved in a physiologically relevant manner, the effects of temperature and pH on binding to human FcRn were examined more closely. Since the initial competitive binding assay was performed at 4 ° C., the experiment was repeated at a physiologically more appropriate 37 ° C., indicating that the mutant was still active at this temperature. As described above, increasing concentrations of unlabeled competing antibodies are incubated with cells expressing human FcRn in the presence of subsaturating labeled OST577-IgG2M3 antibodies in FBB at pH 6.0 at 37 ° C. I let you. As shown in FIG. 14, the results indicated that the antibody maintained its relative pattern of binding to human FcRn at 37 ° C.

It has been found that IgG binding to FcRn is pH dependent. That is, IgG binds strongly to FcRn at pH 6.0 or only weakly at pH 8.0. To engineer mutant antibodies with longer serum half-life, it is desirable to increase binding to FcRn at pH 6.0 while retaining pH-dependent release from FcRn at pH 8.0 . To confirm that the binding was pH dependent, the antibody was tested for binding to a transfected NS0 cell line that stably expresses human FcRn, and then from pH 6.0 to pH 8.0. This was released at pH values in the range of. As described above, cells were incubated with an unsaturated concentration of antibody in FBB at pH 6.0, washed in FBB at pH 6.0, 6.5, 7.0, 7.5, or 8.0, and binding was analyzed by FACS ^™ . As shown in FIG. Was shown to exhibit strong binding to human FcRn at pH 6.0, decreasing binding as pH increased to 8.0.

  As shown in FIG. However, it showed that it showed strong binding to human FcRn at pH 6.0. The OST577-IgG1 antibody with the T250D mutation showed weaker binding (relative to the wild type) to human FcRn at pH 6.0, decreasing binding as the pH value increased to pH 8.0. As shown in FIG. , Indicating strong binding to human FcRn at pH 6.0. These results indicated that the antibody binding to human FcRn was indeed pH dependent.

  In a similar manner, antibodies are tested for binding to a transfected NS0 cell line that stably expresses rhesus monkey FcRn, which is then released at pH values in the range of pH 6.0 to pH 8.0. It was. As shown in FIG. It showed strong binding to rhesus monkey FcRn at pH 6.0 while decreasing binding as it increased to pH 8.0. As shown in Figure 15E, the results show that T250Q, M428L or T250Q / M428L, a modified OST577-IgG1 antibody with a mutation, all decreases binding as the pH value increases to pH 8.0. In other words, it showed strong binding to rhesus monkey FcRn at pH 6.0. These results indicated that antibody binding to rhesus FcRn was also pH dependent.

Example 8 .
This example describes the confirmation of additional properties of IgG2M3 and IgG1 mutants.
Cell culture :
Human Burkitt lymphoma cell line Raji (US Standard Culture Collection) was purchased from RPMI 1640 (Bio) with L-glutamine containing 10% FBS (HyClone ™) and 1% penicillin-streptomycin (Life Technologies ™). Whittaker ^™ ).

Antigen binding assay :
The antigen binding activity of OST577 wild type and mutant antibodies was confirmed in a competitive binding ELISA. Immulon ^™ 2 plates (DYNEX ™ Technologies) were coated overnight at 4 ° C. with recombinant hepatitis B surface antigen (HBsAg) 1.0 μg / ml (Advanced Immuno Chemical, Inc., Long Beach, Calif.). The next day, the plates were washed with EWB and blocked with 300 μl / well of SuperBlock ™ Blocking Buffer (Pierce Chemical Company) in TBS for 30 minutes at room temperature. Plates were washed with EWB and premixed biotinylated 577-IgG2M3 antibody (0.25 μg / ml) and competitor OST577-IgG2M3 antibody (33 μg / ml to 0.033 μg / ml) in 100 μl EB for each well. Two-fold serial dilution), or premixed biotinylated OST577-IgG1 antibody (0.25 μg / ml) and competitor OST577-IgG1 antibody (67 μg / ml to 0.067 μg / ml serial dilution) were added to each well .

Plates were incubated at room temperature for 1 hour, then washed with EWB, and 100 μl / well streptavidin-HRP conjugate (Pierce Chemical Company) was added at 1 μg / ml in EB. After incubating at room temperature for 30 minutes, the plates were washed with EWB and then 100 μl / well ABTS peroxidase substrate / peroxidase solution B (Kirkegaard & Perry Laboratories) was added. The reaction was stopped with 100 μl / well oxalic acid and the absorbance at 415 nm was measured using a VER SAmax ^™ microtiter plate reader (Molecular Devices Corporation ™).

Antigen binding activity of HulD10-IgG2M3 wild type and mutant antibodies was confirmed in a FACS ^TM binding assay using Raji cells expressing alleles of the HLA-DRβ chain recognized by HulD10 (Kostelny et al , (2001) supra). 2.5 × 10 ⁵ cells per test were washed once in pH 7.4 FBB and HulD10-IgG2M3 antibody in pH 7.4 FBB (3-fold serial dilution from 60 μg / ml to 0.027 μg / ml) ) Resuspended in 140 μl. Cells are incubated with antibody for 1 hour on ice, washed twice in FBB at pH 7.4, and 25 μl goat F (ab ′) ₂ anti-human kappa RPE conjugate diluted to 10 μg / ml in pH 7.4 FBB Resuspended in type antibodies (Southern Biotechnology Associates, Inc.). After 30 minutes incubation on ice in the dark, the cells were washed twice in pH 7.4 FBB and resuspended in 1% formaldehyde. Samples were analyzed for antibody binding to the HLA-DR β chain allele by FACS ^™ using a FACS Calibur flow cytometer (BD ™ Biosciences).

In a similar manner, the antigen binding activity of HulD10-IgG1 wild type and mutant antibodies was confirmed in a FACS ^™ binding assay using Raji cells. Approximately 2.0 × 10 ⁵ cells per test were washed once in pH 7.4 FBB, and HulD10-IgG1-antibody in pH 7.4 FBB (2 × 25 μg / ml to 12.5 μg / ml) Serial dilution, 3 × serial dilution from 12.5 μg / ml to 0.0020 μg / ml) was resuspended in 100 μl. A dilution series of HuFd79-IgG1 antibody (Co et al., Proc. Natl. Acad. Sci. 88: 2869-2873 (1991)) was prepared as described above and used as a negative control. Cells are incubated with antibody for 1 hour on ice, washed twice in FBB at pH 7.4, and 25 μl goat F (ab ′) ₂ anti-human IgG FITC diluted to 20 μg / ml in pH 7.4 FBB Resuspended in conjugated antibody (Southern, Biotechnology Associates, Inc). After 30 minutes incubation on ice in the dark, the cells were washed twice in pH 7.4 FBB and resuspended in 1% formaldehyde. Samples were analyzed for antibody binding to the HLA-DRβ chain allele by FACS ^™ using a FAC S. Calibur flow cytometer (BD ™ Biosciences).

ADCC assays published methods (Shields et al. Op. Cit.) By measuring the lactate dehydrogenase (LDH) release using Raji cells as human peripheral blood mononuclear cells (PBMC) and target as effectors according to, Hu1D10 The antibody-dependent cell-mediated cytotoxicity (ADCC) activity of wild type and mutant antibodies was confirmed. PBMCs were prepared from fresh whole blood using a Ficoll-Paque ™ Plus gradient (Amersham Biosciences ^™ Corporation) and resuspended at a cell density of 8 × 10 ⁶ cells / ml in assay medium (RPMI1640, 1% BSA). Raji cells were washed 3 times in assay medium and resuspended at a cell density of 0.4 × 10 ⁶ cells / ml in assay medium. HulD10 wild type and mutant antibodies were diluted to 4 μg / ml, 0.25 μg / ml, and 0.016 μg / ml in assay medium.

Raji cells (50 μl / well) and HulD10 antibody (50 μl / well ie 200 ng, 12.5 ng or 0.8 ng per test) are combined in wells of a Falcon 96 well U-bottom assay plate (BD ™ Biosciences) at room temperature Incubate for 30 minutes. PBMC (100 μl / well, ie 40: 1 effector / target ratio) was added to the opsonized cells and allowed to incubate at 37 ° C. for 4 hours in a CO ₂ incubator. Antibody-dependent cell-mediated cytotoxicity (AICC) was measured by incubating effector and target cells in the absence of antibody. Spontaneous release was measured by incubating target cells (SRtarget) or effector cells (SR effector) in the absence of antibody. Maximum release (MR) was measured by adding 2% Triton X-100 to the target cells.

The plate was gently centrifuged and the supernatant (100 μl / well) was transferred to a Falcon 96-well-flat bottom plate. LDH activity was measured by incubating the supernatant with 100 μl / well LDH reaction mixture from the Cytotoxicity Detection Kit (Roche Diagnostics Corporation) for 30 minutes at room temperature. The reaction was stopped with 50 μl / well of 1N HCl and the absorbance at 490 nm was measured using a VERSAmax ^™ microtiter plate reader (Molecular Devices Corporation ™). The percentage of cytotoxicity was calculated as (LDH release _sample- SR _effector- SR _target ) / MR _target- SR _target ) × 100.

Result :
The antigen binding properties of the wild type and mutant OST577 and HulD10 antibodies to their respective antigens were confirmed using appropriate binding assays. As described above, binding of OST577 antibody to HBsAg was determined in a competitive ELISA. As shown in FIG. 16A, the binding of wild type and mutant OST577-IgG2M3 antibodies to HBsAg was essentially identical. Similarly, as shown in FIG. 16B, the binding of wild type and mutant OST577-IgG1 antibodies to HBsAg was essentially identical.

  As described above, the binding of the HulD10 antibody to the HLA-DRβ chain allele was determined in a FACS binding assay. As shown in FIG. 17A, the binding of wild type and mutant HulD10-IgG2M3 antibodies to the HLA-DRβ chain allele was essentially identical. Similarly, as shown in FIG. 17B, the binding of wild type and mutant HulD10-IgG1 antibodies to HLA-DRβ was essentially identical. These results indicate that the mutations described at positions 250 and 428 do not affect antibody binding as expected.

The ADCC activity of HulD10 wild type and mutant antibodies was confirmed in an LDH release assay using human PBMC as an effector and Raji cells as a target. As shown in FIG. 18A, using a donor carrying the homozygous 158V / VF _Cγ RIII allele, the ADCC activity of the double mutant (T250Q / M428L) HulD10-IgG1 antibody is that of the wild-type antibody. While very similar to the one, ADCC activity of the single mutant (M428L) HulD10-IgG1 antibody was slightly reduced compared to the wild-type antibody. As expected, wild type and mutant HulD10-IgG2M3 antibodies lacked ADCC activity.

  In a similar manner, using a donor carrying the homozygous 158F / F FcγRIII allele, as shown in FIG. 18B, the ADCC activity of the double mutant (T250Q / M428L) HulD10-IgG1 antibody is While very similar to the wild type antibody, the ADCC activity of the single mutant (M428L) HulD10-IgG1 antibody was somewhat reduced compared to the wild type antibody. Wild type and mutant HulD10-IgG2M3 antibodies lacked ADCC activity. These results indicate that the T250Q / M428L mutation described in the present invention does not affect the ADCC activity of the IgG1 form of the antibody, whereas M428L MUT slightly reduces the ADCC activity of the IgG1 form. Represents. The described mutants at positions 250 and 428 do not affect the ADCC activity of the IgG2M3 form of the antibody.

Example 9 .
This example describes in vitro and in vivo serum half-life assays.
It is expected that human IgG antibodies with higher (or lower) affinity for FcRn in vivo will each have a longer (or shorter) serum half-life in vivo. The affinity of human IgG mutants for FcRn can be determined by various methods such as surface plasmon resonance (SPR) using soluble FcRn conjugated to an appropriate biosensor chip, or on the surface of transfected cells. Can be measured in vivo by conducting competitive binding experiments using FcRn expressed in <RTIgt; The FcRn used in the in vitro affinity experiment can be derived from mouse, rhesus monkey or human. The serum half-life of a human IgG mutant with the desired properties is determined by injecting an antibody dose in the range of 0.1-10 mg antibody / kg body weight into an appropriate laboratory animal (eg mouse or monkey) or human and then It can be measured in vivo by removing serum samples at various time intervals spanning the expected serum half-life and testing the samples for the presence of intact IgG by an appropriate technique such as ELISA.

Rhesus monkey pharmacokinetic study :
A non-GLP pharmacokinetic study entitled "Comparison of pharmacokinetics of three variants of OST577" was conducted at the California National Primate Research Center (CNPRC) at the University of California, Davis. Twelve male macaque rhesus monkeys were randomly selected by weight and assigned to one of three study groups. The four animals comprising each study group each received a single intravenous dose of one of the wild type or two variants of OST577 at a rate of 1 mg / kg over 15 minutes. The OST antibody was wild type OST577-IgG2M3, one variant of OST577-IgG2M3 containing a single mutation M428L and one variant of OST577-IgG2M3 containing a double mutation T250Q / M428L. All three antibodies were expressed by transfection of Sp2 / 0 cells and purified as described in Example 5.

  Blood samples were taken before dosing on day 0, 1 and 4 hours and 1 day after dosing, on day 7, day 14, day 21, day 28, day 42 and day 56 . At each time point, 4 ml blood was drawn from the saphenous vein, serum was prepared, and two aliquots were frozen and kept at -20 ° C until use. For serum chemistry and hematology determination, blood samples were collected 16 days before the study, before dosing on day 0 and at the end of the study on day 56.

ELISA :
The concentration of OST577-IgG2M3 in rhesus monkey serum specimens was determined by ELISA using a conditional assay. Pooled normal rhesus serum (PNRS) was obtained from CNPRC. The same PNRS lot was used to prepare calibrators, positive serum controls and for predilution of rhesus monkey serum specimens. Prepare calibrator by standard dilution of OST577-IgG2M3 in PNRS at 3000, 1500, 750, 375, 187.5, 93.75, 46.88, 23.44 and 0 ng / ml, equilibrate at room temperature for 2 hours, and form aliquots at -20 ° C And frozen. Positive serum control by spiking PNRS with OST577-IgG2M3 at 0.2 μg / ml for low positive serum control, 0.4 μg / ml for medium positive serum control, 0.8 μg / ml for high positive serum control Was prepared, equilibrated at room temperature for 2 hours, and frozen in aliquots at -20 ° C. Pre-dose serum specimens from each animal were used as negative serum controls.

Immulon ^™ 2 plate (DYNEX ™ Technologies, Inc.) was incubated overnight at 2-8 ° C. at 100 μl / well of mouse anti-OST577-IgG1 idiotype monoclonal antibody (OST577-γ ¹ anti-OST577-γ ¹ anti-PBS) in PBS. -id, Protein Design Labs ^™ , Inc). The next day, the plate is washed 3 times with 300 μl / well PBS / Tween (phosphate buffered saline, 0.1% Tween20), patted on paper towels to dry, and 300 μl / well in PBS for 60 ± 5 minutes at room temperature. Blocked with Super Block ™ Blocking Buffer. Calibrators, positive and negative serum controls and serum specimens were thawed and brought to room temperature before use. Calibrators and positive and negative serum controls were diluted 1:10 in Super Block ™ Blocking Buffer in PBS.

  Serum specimens were pre-diluted appropriately in PNRS (1: 10-1: 80) and then diluted 1:10 in Super Block Buffer in PBS. The plate was washed 3 times with 300 μl / well PBS / Tween and patted dry on a paper towel. Diluted calibrators, positive and negative serum controls and serum specimens were then added at 100 μl / well into duplicate wells and incubated at room temperature for 60 ± 5 minutes. The plate was washed 3 times with 300 μl / well PBS / Tween and dried by tapping on a paper towel. Prepare goat anti-human lambda light chain HRP-conjugated antibody (Southern Biotechnology Associates, Inc.) by 1: 1000 dilution in PBS / BSA / Tween (phosphate buffered saline, 0.5% bovine serum albumin, 0.1% Tween20) And added at 100 μl / well and allowed to incubate at room temperature for 60 ± 5 minutes.

The plate was washed 3 times with PBS / Tween 300 μl / well and dried by tapping on a paper towel. ABTS peroxidase substrate / peroxidase solution B (Kirkegaard & Perry Laboratories) was added at a rate of 100 μl / well and incubated for 7 ± 1 minutes. Progress was stopped by adding Substrate Stop Solution (2% oxalic acid) at a rate of 100 μl / well. Absorbance values at 415 nm were measured within 30 minutes of adding Substrate Stop Solution using a VERSA max ^™ microtiter plate reader (Molecular Devices Corporation ™).

  A calibration curve was generated by fitting data to a 41 parameter logistic regression curve with SOFTmax ™ PRO, version 4.0 (Molecular Devices Corporation ™), using average absorbance values obtained from calibrators. The average absorbance value for the negative serum control (ie the pre-dose sample average for each animal) was subtracted from each absorbance value obtained for the calibrator. After subtracting the average absorbance value obtained for the negative serum control from each absorbance value obtained for the positive serum control, the positive serum control concentration was determined. The concentration corresponding to the resulting average absorbance value was derived by interpolation from the calibration curve. The concentration of the serum sample is the concentration corresponding to the average absorbance by subtracting the average absorbance value of the negative serum control from the absorbance value of each sample, averaging the resulting absorbance values, and interpolating from the calibration curve. And, where applicable, determined by multiplying the resulting concentration by the predilution factor to arrive at the final concentration for each sample.

  The estimated quantification range of the assay was 0.10-0.90 μg / ml. The test showed that (1) the mean back-calculated concentration of all three calibrators within the quantitative range was in the range of 20% of its nominal value, and (2) the average of four of the six positive serum controls It was considered appropriate when two conditions were met: the result was within 30% of its nominal value and at least one average result from each concentration level was within 30% of its nominal value. Data from plates that did not meet these criteria were rejected. Data from individual serum specimens were: (1) Absorbance values in duplicate wells differed from each other by more than 40%, (2) Mean calculated concentration was lower limit of quantification of assay (LLQQ) (0.10 μg (3) The average calculated concentration was higher than the upper limit of quantification (ULOQ) (0.90 μg / ml) of the test, and one of the three conditions Rejected if satisfied.

Result :
Serum antibody concentration data was fitted with a two-compartment model using WinNonlin ™ Enterprise Edition, version 3.2 (Pharsight ™ Corporation, Mountain View, CA). This model assumes a first order distribution and a first removal rate and fits the data well. Modeled data (simulated based on the geometric mean of each group of primary pharmacokinetic parameters) and the observed mean serum antibody concentration (μg / ml) and standard deviation for each group of 4 animals GraphPad Prism ™, version 3.02 (GraphPad ^™ Software, Inc.) and plotted as a function of time (days after infusion). As shown in FIG. 19, the data represent that the mean serum antibody concentrations of OST577-IgG2M3 M428L and T250Q / M428L mutants were maintained at higher levels than wild-type OST577-IgG2M3 at all time points. Yes.

Various pharmacokinetic parameters were calculated from data using WinNonlin ™ Enterprise Edition, version 3.2 (Pharsight ™ Corporation). (GraphPad Prism ™ Version 3.02 (GraphPad ^™ Software, Inc.) was used to calculate the statistical analysis of pharmacokinetic parameters. As shown in Table 9, the mean maximum serum antibody concentration (Cmax) is Were very similar among the three test groups, indicating that the administered antibody was distributed into the circulation in a similar manner, thus higher than the mutant IgG2M3 antibody after the distribution phase Antibody concentration is due to its increased persistence in serum, and analysis of mean clearance (CL) showed that this was the case.

  The mean CL, the amount of serum antibody cleared per unit time, is approximately 1.8 times lower (0.0811 ± 0.0384 ml) for the M428L variant compared to wild-type OST577-IgG2M3 (0.144 ± 0.047 ml / hr / kg) / hr / kg; p = 0.057), about 2.8 times lower for the T250Q / M428L mutant (0.0514 ± 0.0075 ml / hr / kg; p = 0.029), compared to the wild type, OST577 from rhesus monkey circulation -Clearly decreased clearance of IgG2M3 M428L and T250Q / M428L mutants (Table 9).

  The PK profile of the OST577-IgG2M3 variant was further analyzed by calculating with other parameters (Table 9). Since the AUC (area under the curve) is inversely proportional to CL, the mean AUC, the area under the concentration-time curve extrapolated from time zero to infinity, is the wild type OST577-IgG2M3 (7,710 ± 3,110 hours * μg / Compared to ml), the M428L mutant is about twice as high (15,200 ± 8,700 hours * μg / ml: p = 0.057), and the T250Q / M428L mutant is about 2.6 times higher (19,800 ± 2,900 hours * μg / ml). : P = 0.029), indicating that the complete exposure of OST577-IgG2M3 M428L and T250Q / M428L mutants is significantly increased compared to the wild type (Table 9).

  Finally, the mean elimination (β-phase) half-life is about 1.8 times longer for the M428L variant (642 ± 205 hours) than the wild-type OST577-IgG2M3 (351 ± 121 hours), and the T250Q / M428L variant Was about 1.9 times longer (652 ± 28 hours; p = 0.029) (Table 9). The elimination half-life for wild-type OST577-IgG2M3 in this study is similar to that of OST577-IgG1 (324 ± 85 hours) in previous PK studies in rhesus monkeys (Ehrlich et al., Supra).

Example 10 .
This example describes the application of the in vitro serum half-life assay described in Example 9 to IgG1 antibody variants.
Rhesus monkey pharmacokinetic study:
A non-GLP pharmacokinetic study called “Comparison of Pharmacokinetics of Two Variants of OST577” was performed at the California National Primate Research Center (CNPRC) at the University of California, Davis. Eight male rhesus monkeys were randomized by weight and assigned to one of two studies. Four animals, including individual study groups, each received a single intravenous dose of wild type or mutant OST577 at 1 mg / kg administered over 15 minutes. The OST577 antibody was a variant of OST577-IgG1 containing wild type 0ST577-IgG1 and the double mutation T250Q / M428L. Both antibodies were expressed by Sp2 / 0 transfection and purified as described in Example 5.

  Blood samples were taken on day 0, before dosing, at 1 and 4 hours after dosing, and at 1, 7, 14, 21, 28, 42 and 56 days. At individual time points, 4 ml of blood was collected from the saphenous vein, serum was prepared, and two aliquots were frozen and maintained at -20 ° C until use. Blood samples were taken on day 0, before dosing, and at the end of the 56 day study for serum chemistry and hematology determinations.

ELISA:
The concentration of OST577-IgG1 antibody in rhesus monkey serum samples was determined by ELISA using a validated assay. Pooled normal rhesus monkey serum (PNRS) was obtained from CNPRC. PNRS sample lots were used to prepare calibrators, positive and negative controls, and for pre-dilution of rhesus monkey serum samples. Calibrators were prepared by standard dilutions of OST577-IgG1 into PNRS at 3200, 1600, 800, 400, 200, 100, 50, 25 and 0 ng / ml, equilibrated at room temperature for 2 hours, and at −80 ° C. Frozen in aliquots. The positive serum control was PNRS with OST577-IgG1 at 0.2 μg / ml for the low positive serum control, 0.4 μg / ml for the medium positive serum control, and 0.8 μg / ml for the high positive serum control. Was prepared by spiking, equilibrated at room temperature for 2 hours, and frozen in aliquots at -80 ° C. PNRS was used as a negative serum control.

Immulon ^™ 2 plates (DYNEX ™ Technologies, Inc.) were washed with 100 μl / well of mouse anti-OST577-IgG1 indiotype monoclonal antibody (OST577-γ1 anti-id, Protein Design Labs, 1.0 μg / ml in PBS). ^TM , Inc.) at 2-8 ° C. overnight. The next day, the plate is washed 3 times with 300 μl / well PBS / Tween (phosphate buffer solution, 0.1% Tween 20), dabbed on a paper towel and dried, and at room temperature for 60 ± 5 minutes, PBS Inhibited with 300 μl / well SuperBlock ™ Blocking Buffer (Pierce Chemical Company). Calibrators, positive and negative serum controls, and serum samples were thawed and returned to room temperature before use.

  The calibrator and positive and negative serum controls were diluted 1:10 with SupeBlock ™ Blocking Buffer in PBS. Serum samples were roughly prediluted with PNRS (1: 5 to 1:80) and then diluted 1:10 with SuperBlock ™ Blocking Buffer in PBS. Plates were washed 3 times with 300 μl / well PBS / Tween and patted dry on a paper towel. Next, diluted calibrators, positive and negative serum controls, and serum samples were added to two wells at 100 μl / well and incubated at room temperature for 60 ± 5 minutes. Plates were washed 3 times with 300 μl / well PBS / Tween and patted dry on a paper towel. Dilute goat anti-human λ light chain HRP-conjugated antibody (Southern Biotechnology Associates, Inc.) 1: 1000 with PBS / BSA / Tween (phosphate buffer solution, 0.5% bovine serum albumin, 0.1% Tween20) And added at 100 μl / well and incubated at room temperature for 60 ± 5 minutes.

Plates were washed 3 times with 300 μl / well PBS / Tween and patted dry on a paper towel. ABTS peroxidase substrate / peroxidase solution B (Kirkegaard & Perry Laboratories) was added at 100 μl / well and incubated for 7 ± 1 minutes. Progression was stopped by the addition of substrate stop solution (2% oxalic acid) at 100 μl / well. Absorbance values at 415 nm were measured using a VERSAmax ^™ microtiter plate reader (Molecular Devices Corporation ™) within 30 minutes after addition of the substrate stop solution.

  Calibration curve using the mean absorbance value obtained from the calibrator and by fitting the data to a four parameter logistic regression curve using SOFTmax ™ PRO, Version 4.0 (Molecular Devices Corporation ™) Prepared. The average absorbance value for the 0.0 ng / ml calibrator was subtracted from the individual absorbance values obtained for the remaining calibrators. The positive serum control concentration was determined after subtracting the average absorbance value obtained for the negative serum control from the individual absorbance value obtained for the positive serum control. The concentration corresponding to the average absorbance value obtained is obtained by interpolation from the calibration curve. Concentrate the serum sample concentration by subtracting the average absorbance value of the appropriate pre-dose sample from the absorbance value of the individual study sample, averaging the resulting absorbance values, and interpolating from the calibration curve to the concentration corresponding to the average absorbance value Was determined by multiplying the resulting concentration by the predilution factor to reach the final concentration for each individual sample.

  The estimated quantification range of the assay was 0.10-0.90 μg / ml. The assay appeared to be appropriate if the following two conditions were met: (1) The average back-calculated concentration of all four calibrators in the quantification range was within 20% of their nominal value And (2) the calculated average results of four of the six positive serum controls are within 30% of their nominal values, and at least one average result from each concentration level is designated Within 30% of the value.

  Data from plates that did not meet the above criteria were rejected. Data from individual serum samples were rejected if any of the following conditions were met: (1) The calculated mean concentration is below the lower limit of quantitation (LLOQ) of the assay (0.10 μg / ml) Or (2) The calculated average concentration is above the upper limit of quantification (ULOQ) of the assay (0.90 μg / ml). If the calculated results of the duplicate samples differed from each other by more than 40%, the samples were retested in a second independent assay. If the calculated average result of the second assay was within 15% of the calculated average result of the first assay for the sample in question, the calculated average result of the first assay was used. On the other hand, the sample was retested by a third independent assay and if one value was not removed by the outlier test, the results of all three assays were averaged. In this case, the average of the two remaining values was reported.

result:
Serum antibody concentration data was fitted to a two-compartment model using WinNonlin Enterprise Edition, version 3.2 (Pharsight ™ Corporation, Mountain View, CA). The model assumes primary distribution and primary removal rate and fits the data well. Modeled data (simulated based on the median value of the primary pharmacokinetic parameters of each group), as well as the observed mean serum antibody concentration (μg / ml) and for each group of 4 animals Standard deviations were plotted as a function of time (days after infusion) using GraphPad Prism ™ version 3.02 (GraphPad Software, Inc.). As shown in FIG. 20, the data indicate that the mean serum antibody concentration of the T250Q / M428L variant of OST577-IgG1 was maintained at a higher level than wild type OST577-IgG1 at all time points.

Various pharmacokinetic parameters were calculated from data using WinNonlin ™ Enterprise Edition, version 3.2 (Pharsighte ™ Corporation). Statistical analysis of pharmacokinetic parameters was calculated using GraphPad Prism ™, version 3.02 (GraphPad ^™ Software, Inc.). As shown in Table 10, C _max is very similar between both test groups, indicating that the administered antibody was distributed to the circulation in a similar manner. Thus, the higher antibody concentration of the variant IgG1 antibody following the partitioning phase is due to its increased persistence in serum. Analysis of average CL showed that this was the case. Mean CL was about 2.3 times lower for the T250Q / M428L mutant (0.0811 ± 0.0191 ml / hr / kg; p = 0.029) compared to wild type OST577-IgG1 (0.190 ± 0.022 ml / hr / kg) (Table 10) This shows a marked decrease in clearance of the OST577-IgG1 T250Q / M428L mutant from rhesus monkey circulation compared to wild type.

The PK profile of the OST577-IgG1 variant was further analyzed by calculating other parameters (Table 10). The average AUC is about 2.4 times higher for the T250Q / M428L mutant (12,900 ± 3,000 hours ^* μg / ml; p = 0.029) compared to wild-type OST577-IgG1 (5,320 ± 590 hours ^* μ / ml) (Table 10) This indicates a marked increase in complete exposure of the OST577-TgG1 T250Q / M428L mutant compared to the wild type.

  Finally, the mean elimination (β-phase) half-life was approximately 2.5 times for the T250Q / M428L variant (838 ± 187 hours; p = 0.029) compared to wild type OST577-IgG1 (336 ± 34 hours) It was long (Table 10). The elimination half-life for wild-type OST577-IgG1 in this study is similar to the half-life for OST577-IgG1 (324 ± 85 hours) in a previous PK study in rhesus monkeys (Ehrlich et al., Supra).

Example 11 .
This example describes the application of the various binding assays described in Examples 6 and 7 to IgG3 and IgG4 antibody variants.
Mutagenesis:
The overlap-extension PCR method (Higuchi, supra) was used to generate specific site amino acid substitutions at positions 428 of Hu1D10-IgG3 heavy chain or at positions 250 and 428 of Hu1D10-IgG4 heavy chain (Kabat et al. , Numbered according to the EU indicators mentioned above). M428L mutant was generated in Hu1D10-IgG3 heavy chain. Both M428L and T250QM428L variants were generated in the Hu1D10-IgG4 heavy chain.

Transfection:
The wild type or mutant Hu1D10-IgG3 or Hu1D10-IgG4 heavy chain expression vector, described in detail in Examples 1 and 2, together with the pVk-Hu1D10 light chain expression vector, was added to the human kidney cell line 293-H (Life Technologies ™ )) Temporarily co-transfected. Hu1D10-IgG3 or Hu1D10-IgG4 expression vectors were also stably co-transfected into Sp2 / 0 cells with the pVk-Hu1D10 expression vector as described in Example 5. For stable transfection in Sp2 / 0, the IgG3 expression vector is linearized with FspI; however, the IgG4 expression vector has two FspI sites in pHuHC.g4.It.D-Hu1D10, so Bstz171 Was linearized.

Antibody purification:
Culture supernatants containing human IgG4 antibody were quantified by ELISA, harvested by centrifugation, sterile filtered and purified by protein A affinity chromatography as described in Example 5.

  Culture supernatants containing human IgG3 antibodies were quantified by ELISA, harvested by centrifugation and sterile filtered as described in Example 5. The pH of the filtered supernatant was adjusted by adding 1/75 volume of 1M Tris-HCl (pH 8.0). The supernatant was tested on a 1 ml HiTrapW ™ Protein G HP column (Amersham Biosciences Corporation) pre-equilibrated with 20 mM sodium phosphate, pH 7.0. The column was washed with the same buffer and the bound antibody was eluted with 100 mM glycine-HCl (pH 2.7). After neutralization by adding about 1/50 volume of 1M Tris-HCl (pH 8.0), the pooled protein fractions were washed overnight in 20 mM sodium citrate, 120 mM NaCl (pH 6.0). Tested on a 5 ml HiTrap ™ Desalting column (Amersham Biosciences Corporation) dialyzed or pre-equilibrated with 20 mM sodium citrate 120 mM NaCl, pH 6.0.

The flow through the desalting column is collected and the fractions with OD280> 1 are pooled and about 0.5-1.0 mg using a 2 ml Vivaspin ™ concentrator (50,000 Dalton MWCO) (Vivascience ™ AG) Concentrated to / ml. The dialyzed material was concentrated in the same manner. The sample was then filter sterilized using 0.2 μm Millex ™ -GV microfilters (Milliporee Corporation). The concentration of the purified antibody was determined by UV spectroscopy by measuring the absorbance at 280 nm (1 mgl / ml = 1.4A ₂₈₀ ).

SDS-PAGE:
A 5 μg sample of purified antibody was tested under reducing or non-reducing conditions as described in Example 5.

Competitive binding assays:
NSO transformants expressing recombinant GPI-bound human or rhesus FcRn on the surface as described in Example 6 were supplemented with mycophenolic acid (MPA) selection medium (DMEM, 10% FBS, lxHT medium). Hybri-Maxo ™ (Sigma ™), 250 μg / ml xanthine (Sigma ™), 1 μg / ml mycophenolic acid (Life Technologies ™), and 2 mM L-glutamine, or Maintained in 2 × MPA selective medium.

A series of diluted solutions of individual purified HU1D10-1 gG3 antibodies compete against human IgG (Sigma-Aldrich, St. Louis, MO) labeled with biotin (Pierce Biotechnology, Inc., Rockford, IL) I was damned. The antibodies were tested for binding to cell line NS0 HuFcRn (memb), ie human FcRn on clone 7-3, and cell line NS0 RhFcRn, ie rhesus monkey FcRn on clone R-3. Approximately 2 × 10 ⁵ cells per test are washed once with FBB (pH 8.0) and once with FBB (pH 6.0) and then 120 μl of pre-filled in FBB (pH 6.0) Resuspended in mixed, biotinylated human IgG antibody (8.3 μg / ml) and Hu1D10-IgG3 competitive antibody (2-fold serial dilution from 625 μg / ml to 0.305 μg / ml).

Cells are incubated with the antibody mixture for 1 hour on ice, washed twice with FBB (pH 6.0), and streptavidin-RPE conjugate (BioSource) diluted to 2.5 μg / ml in FBB (pH 6.0). International) and resuspended in 25 μl. After 30 minutes incubation on ice in the dark, the cells were washed twice with FBB (pH 6.0) and resuspended in 1% formaldehyde. Samples were analyzed for antibodies binding to FcRn by FACS ^™ using a FACSCalibur flow cytometry (BD ™ Biosciences).

  A series of diluted solutions of individual purified Hu1D10-IgG4 antibodies were competed against human IgG (Sigma-Aldrich) labeled with biotin (Pierce Biotechnology, Inc.). IgG4 antibodies were tested for binding to cell line NS0 HuFcRn (memb), ie FcRn on clone 7-3, and cell line NS0 RhFcRn, ie rhesus monkey FcRn on clone R-3, as described above for IgG3 antibody. did.

Ph-dependent binding and release assays:
Purified Hu1D10-IgG3 and Hu1D10-IgG4 variant antibodies are compared to their respective wild type antibodies for binding to human or rhesus monkey FcRn, and then cell line NS0 HuFcRn as described in Example 7. (memb), ie clone 7-3, and NS0 RhFcRn, ie clone R-3, were released at various pH values in a single point binding and release assay. To ensure that both subtypes of IgG3 and IgG4 were equally labeled, 25 μl of goat F (ab ′) ₂ anti-human κFTTC− diluted to 1.25 μg / ml in FBB at the appropriate pH Conjugated antibody (Southern Biotechnology Associates, Inc.) was used as a detection reagent.

result:
Amino acid substitutions occurred at position 428 of the human γ3 heavy chain and at positions 250 and 428 of the human γ4 heavy chain (numbered according to the EU index, Kabat, et al., Supra). These two positions were selected based on the identification of mutations at those positions in the human γ2M3 heavy chain that resulted in increased binding to FcRn. M428L mutant was evaluated in human γ3 heavy chain. Both M428L mutant and T250Q / M428L mutant were evaluated in human γ4 heavy chain.

  IgG3 and IgG4Fc variants are the light and heavy chain variable regions of Hu1D10 (Kostelny et al. (2001), supra), the light chain constant region of human kappa (Hieter et al. (1980), supra), and It was expressed as an anti-HLA-DR β chain allele antibody comprising the heavy chain constant regions of human IgG3 (Huck et al., Supra) and IgG4 (Ellison et al., Supra). As described above, the appropriate wild type or mutant weight chain expression vector was transiently co-transfected into 293-H cells with the appropriate light chain expression vector for expression of the Hu1D10 monoclonal antibody.

  ELISA analysis of culture supernatants temporarily harvested 5-7 days after transfection showed that antibody expression levels were typically 5-25 μg / ml in 25 ml supernatant. Indicated. Hu1D10-IgG3 and Hu1D10-IgG4 antibodies were purified by affinity chromatography using protein G and A, respectively, for a final yield of about 100-500 μg. Stable expression of Hu1D10 antibody in Sp2 / 0 cells typically resulted in an expression level of 30-100 μg / ml as determined by ELISA. A yield of about 50-80% antibody present in the culture supernatant was obtained by small scale protein G or A affinity chromatography.

  The purified antibody was characterized by SDS polyacrylamide gel electrophoresis (SDS-PAGE) under non-reducing and reducing conditions. SDS-PAGE analysis under non-reducing conditions showed that the purified antibody had a molecular weight of about 150-170 kDa (data not shown); analysis under reducing conditions showed that the purified antibody It was shown to be composed of a heavy chain with a molecular weight of about 50-60 kD and a light chain with a molecular weight of about 25 kD (data not shown). SDS-PAGE analysis of antibodies purified from stable Sp2 / 0 transfectants gave results similar to those observed with antibodies purified from transient 293-H transfections.

  Relative binding of wild-type Hu1D10-IgG3 and Hu1D10-IgG4 antibodies and their variants to FcRn is determined using a transfected NSO cell line that stably expresses human FcRn on its surface did. Purified antibodies were tested for FcRn binding in a competitive binding assay as described above. Increasing concentrations of unlabeled competitive antibody were incubated with cells in FBB (pH 6.0) in the presence of subsaturating biotinylated human IgG antibody (Sigma-Aldrich).

  The binding of Hu1D10-IgG3 wild type and M428L mutant to human FcRn was tested in competitive binding experiments. As summarized in Table 11, the IC50 for the wild-type Hu1D10-IgG3 antibody is about 15 μg / ml and the IC50 for the M428L single mutant is about 2 μg / ml. The binding of Hu1D10-IgG4 and its variants to human FcRn was also tested in competitive binding experiments. As summarized in Table 12, the IC50 for the wild-type Hu1D10-IgG antibody is about 76 μg / ml, and the IC50 for the M428M single mutant is about 5 μg / ml, and the T250Q / M428L double mutation The IC50 for the body is about 1 μg / ml.

  The binding of Hu1D10-IgG3 wild type and M428L mutant to rhesus monkey FcRn was tested in competitive binding experiments. As summarized in Table 13, the IC50 for the wild type Hu1D10-IgG3 antibody is about 14 μg / ml and the IC50 for the M428L single mutant is about 3 μg / ml. Binding of Hu1D10-IgG4 and its variants to rhesus monkey FcRn was also tested in competitive binding experiments. As summarized in Table 14, the IC50 for the wild-type Hu1D10-IgG antibody is about 98 μg / ml, and the IC50 for the M428M single mutant is about 7 μg / ml, and the T250Q / M428L double mutation The IC50 for the body is about 1 μg / ml.

IgG binding to FcRn is known to be pH-dependent; IgG binds strongly to FcRn at pH 6.0, but weakly at pH 8.0. To construct a variant antibody with a longer serum half-life, it is desirable to enhance binding to FcRn at pH 6.0 while retaining pH-dependent release from FcRn at pH 8.0. To ensure that the binding is pH-dependent, the antibodies were tested for binding to a transfected NSO cell line that stably expresses human FcRn and a pH in the range of 6.0-8.0. Released by value. As above, cells are incubated with subsaturating antibodies in FBB (pH 6.0), washed with FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0, and binding is analyzed by FACS ^™. did.

  As shown in FIG. 21A, the results show that the modified Hu1D10-IgG3 antibody with the M428L mutation shows strong binding to human FcRn at pH 6.0 and the pH value is increased to pH 8.0. As shown, the bond decreased. As shown in FIG. 21B, the results show that the modified Hu1D10-IgG4 antibody with the M428L or T250Q / M428L mutation shows strong binding to human FcRn at pH 6.0 and the pH value is pH 8.0. It was shown that the binding decreased as the value increased. The results showed that the binding of IgG3 and IgG4 antibodies to human FcRn was pH-dependent.

  Similarly, antibodies were tested for binding to transfected NSO cell lines that stably express rhesus monkey FcRn and then released at pH values in the range of 6.0-8.0. As shown in FIG. 21C, the results show that the modified Hu1D10-IgG3 antibody with M428L shows strong binding to rhesus monkey FcRn at pH 6.0, and binding increased as the pH value increased to pH 8.0. Showed a decline. As shown in FIG.21D, the results show that the modified Hu1D10-IgG4 antibody with H428L or T250Q / M428L mutation has strong binding to rhesus monkey FcRn at pH 6.0, and the pH value is pH 8. As it rose to 0, it showed a decrease in binding. The results indicated that antibody binding to FcRn in rhesus monkeys was also pH-dependent.

Example 12 .
This example describes the application of the in vitro and in vivo serum half-life described in Examples 9 and 10 to IgG3 and IgG4 antibody variants.
The protocol of “Rhesus monkey pharmacokinetic study” as described in Examples 9 and 10 was performed on the IgG3 and IgG4 variants, and the effect of the mutation on in vivo serum half-life and various pharmacokinetic parameters. It was confirmed.

Example 13 .
This example designs and generates FcRn binding variants of well-known therapeutic antibodies to allow improved patient treatment rules, thereby extending individual serum half-lives ( (Or reduce).

Daclizumab:
Daclizumab is a human-adapted, clinically developed for many autoimmune and inflammatory disease symptoms such as asthma, diabetes, uveitis, multiple sclerosis, rheumatoid arthritis and ulcerative colitis Anti-CD25 monoclonal antibody. Daclizumab is currently marketed under the trademark Zenapax ™ for transplantation indications only.

The amino acid sequences of the light and heavy chain variable regions for daclizumab are disclosed in US Pat. No. 5,530,101, incorporated herein by reference.
Daclizumab in its current clinical embodiment is an IgG1 isotype antibody, however, an IgG2M3 isotype version of daclizumab that exhibits similar therapeutic characteristics can be generated.

  In order to increase the serum half-life of daclizumab, its constant region amino acid sequence was modified with any of the FcRn binding mutations described above. For example, the T250Q / M428L mutation can be generated in daclizumab using the vector design and mutagenesis methods described in Examples 1-4 above. The sequence of T250Q / M428L daclizumab is shown in Figure 22 (SEQ ID NO: 122).

The enhanced serum half-life of T250Q / M428L daclizumab can be determined using the in vitro binding assay method described in Examples 5-7 or using the in vivo assay method described in Examples 9-10.
It is anticipated that T250Q / M428L daclizumab will provide the benefits of increased frequency and dose of administration to the therapeutic effects of unchanged daclizumab.

As shown in FIG. 22 (SEQ ID NOs: 119-123), other FcRn binding mutations in the daclizumab (IgG1 isotype) constant region sequence can also be generated according to the methods of Examples 1-3 above. It is expected that these variants will also affect the serum half-life of daclizumab as desired based on the effects of the OST577 and Hu1D10 antibody examples above.
Similarly, FcRn binding mutations in the IgG2M3 version of the daclizumab constant region sequence, as shown in FIG. 22 (SEQ ID NOs 124-128), can also be generated according to the methods of Examples 1-4 above.

Font Rizumab:
Fontlizumab is a humanized anti-interferon-γ (IFN-γ) monoclonal antibody of the IgG1 isotype, also known as HuZAF ^™ . Fontlizumab is currently being clinically developed as a therapeutic treatment for Crohn's disease. The variable region of fontlizumab is disclosed in US Pat. No. 6,329,511, which is incorporated herein by reference. As described above for daclizumab, the serum half-life of fontlizumab can also be determined using the method described in Examples 1-4 above, its steady state as shown in FIG. 23 (SEQ ID NOs: 130-134). It can be altered by mutagenizing the region sequence.

Bicilizumab:
Bicilizumab is a human adapted anti-CD3 monoclonal antibody of the IgG2 isotype, also known as Nuvion ™. Bicilizumab targets the CD3 molecule that forms an antigen-receptor complex on all mature T cells. Bicilizumab is currently being clinically developed as a treatment for steroid-refractory ulcerative colitis. As described above for daclizumab and fontlizumab, the serum half-life of bicilizumab is also as shown in FIG. 24 (SEQ ID NOs: 136-140) using the methods described in Examples 1-4 above. The constant region sequence can be altered by mutagenesis.

M200:
M200 is a chimeric IgG4 antibody directed against α5β1 integrin that is currently being developed as an angiogenesis inhibitor therapeutic for a variety of proliferative disorders. As described above for daclizumab, fontlizumab and bicilizumab, the serum half-life of M200 is also shown in FIG. 25 (SEQ ID NOs: 142-146) using the method described in Examples 1-3 above. Such constant region sequences can be altered by mutagenesis.

Although the invention has been described with reference to the presently preferred embodiment, it should be understood that various modifications can be made within the scope of the invention.
All publications, patents, patent applications and websites are at the same level as if each individual publication, patent, patent application and website was specifically and individually incorporated by reference in its entirety. , The entirety of which is incorporated herein by reference.

FIG. 1 illustrates the structure of an IgG molecule. FIG. 2 shows the salvage pathway of IgG molecules. FIG. 3A shows the amino acid sequences of OST577-IgG2M3 and OST577-IgG1, with heavy chain positions 250, 314, and 428 highlighted. “OST577-VH” (SEQ ID NO: 1) represents the amino acid sequence of the heavy chain variable region of OST577-IgG2M3 or OST577-IgG1. “IgG2M3-CH” (SEQ ID NO: 2) represents the amino acid sequence of the heavy chain constant region of OST577-IgG2M3. “IgG1-CH” (SEQ ID NO: 3) represents the amino acid sequence of the heavy chain constant region of OST577-IgG1. “OST577-VL” (SEQ ID NO: 4) represents the amino acid sequence of the light chain variable region of OST577-IgG2M3 or OST577-IgG1. “LAMBDA2-CL” (SEQ ID NO: 5) represents the amino acid sequence of the light chain constant region of OST577-LgG2M3 or OST577-IgG1. FIG. 3B-1 shows the amino acid sequence of the constant region of modified OST577-IgG2M3 compared to unmodified OST577-IgG2M3 (see Table 1 for the sequence number of each disclosed amino acid sequence). FIG. 3B-2 shows the amino acid sequence of the constant region of modified OST577-IgG2M3 compared to unmodified OST577-IgG2M3 (see Table 1 for the sequence number of each disclosed amino acid sequence). FIG. 3B-3 shows the amino acid sequence of the constant region of modified OST577-IgG2M3 compared to unmodified OST577-IgG2M3 (see Table 1 for the sequence number of each disclosed amino acid sequence). FIG. 3B-4 shows the amino acid sequence of the constant region of modified OST577-IgG2M3 compared to unmodified OST577-IgG2M3 (see Table 1 for the sequence number of each disclosed amino acid sequence). FIG. 3B-5 shows the amino acid sequence of the constant region of modified OST577-IgG2M3 compared to unmodified OST577-IgG2M3 (see Table 1 for the sequence number of each disclosed amino acid sequence). FIG. 3B-6 shows the amino acid sequence of the constant region of modified OST577-IgG2M3 compared to unmodified OST577-IgG2M3 (see Table 1 for the SEQ ID NO of each disclosed amino acid sequence). FIG. 3C shows the amino acid sequence of the constant region of modified OST577-IgG1 compared to unmodified OST577-IgG1. FIG. 3D shows the amino acid sequences of HulD10-IgG2M3 and HulD10-IgG1, with heavy chain positions 250, 314 and 428 highlighted. “Hul10-VH” (SEQ ID NO: 6) represents the amino acid sequence of the heavy chain variable region of HulD10-IgG2M3 or HulD10-IgG1. “IgG2M3-CH” (SEQ ID NO: 2) represents the amino acid sequence of the heavy chain constant region of HulD10IgG2M3. “IgGl-CH” (SEQ ID NO: 7) represents the amino acid sequence of the heavy chain constant region of HulD10-IgG1. “HulD10-VL” (SEQ ID NO: 8) represents the amino acid sequence of the light chain variable region of HulD10-IgG2M3 or HulD10-IgG1. “KAPPA-CL” (SEQ ID NO: 9) represents the amino acid sequence of the light chain constant region of HulD10-IgG2M3 or HulD10-IgG1. FIG. 3E shows the amino acid sequence of the constant region of modified HulD10-IgG2M3 compared to unmodified HulD10-IgG2M3. FIG. 3F shows the amino acid sequence of the constant region of modified HulD10-IgG1 compared to unmodified HulD10-IgG1. FIG. 3G shows the amino acid sequences of Hu1D10-IgG3 and Hu1D10-IgG4 with heavy chain positions 250 and 428 highlighted. “Hu1D10-VH” (SEQ ID NO: 6) represents the amino acid sequence of the heavy chain variable region of HuID10-IgG3 or Hu1D10-IgG4. “IgG-CH” (SEQ ID NO: 113) represents the amino acid sequence of the heavy chain constant region of Hu1D10-IgG3. “IgG4-CH” (SEQ ID NO: 114) represents the amino acid sequence of the heavy chain constant region of Hu1D10-IgG4. “Hu1D10-VL” (SEQ ID NO: 8) represents the amino acid sequence of the light chain variable region of Hu1D10-IgG3 or Hu1D10-IgG4. “KAPPA-CL” (SEQ ID NO: 9) represents the amino acid sequence of the light chain constant region of Hu1D10-IgGM3 or Hu1D10-IgG4. FIG. 3H shows the amino acid sequence of the constant region of the M428LHulD10-IgG3 variant (SEQ ID NO: 115) compared to unmodified HulD10-IgG3. FIG. 3I shows the amino acid sequences of the constant regions of M428LHulD10-IgG4 variant (SEQ ID NO: 116) and T250Q / M428LHulD10-IgG4 (SEQ ID NO: 117) compared to unmodified HulD10-IgG4. FIG. 4 illustrates the overlap extension PCR method. FIG. 5A shows a restriction map of heavy chain vector pVAg 2M3-OST577. FIG. 5B shows a restriction map of heavy chain vector pVAgl.N-OST577. FIG. 6 shows a restriction map of the light chain vector pVAλ2-OST577. FIG. 7A shows a restriction map of heavy chain vector pVAg2M3-HulD10. FIG. 7B shows a restriction map of heavy chain vector pVAgl.N-HulD10. FIG. 7C shows a restriction map of the heavy chain vector pHuHCg3.Tt.D-Hu1D10. FIG. 7D shows a restriction map of the heavy chain vector pHuHCg4.Tt.D-Hu1D10. FIG. 8 shows a restriction map of the light chain vector pVk-HulD10. FIG. 9A shows a restriction map of the human FcRn vector pDL208. FIG. 9B shows a restriction map of rhesus FcRn vector pDL410. FIG. 10A shows SDS-PAGE analysis of OST577-IgG2M3 wild type and mutant antibodies. Purified OST577-IgG2M3 wild type and mutant antibodies were analyzed by SDS-PAGE under reducing conditions as described in Example 5. FIG. 10B shows SDS-PAGE analysis of OST577-IgG1 wild type and mutant antibodies. Purified OST577-IgG1 wild type and mutant antibodies were analyzed by SDS-PAGE under reducing conditions as described in Example 5. FIG. 11A shows a single point comparative binding assay of various mutants at position 250 of OST577-IgG2M3 against human FcRn. Binding of biotinylated OST577-IgG2M3 antibody to human FcRn on NS0 cells transfected in the presence of wild type or position 250 mutant OST577-IgG2M3 competitive antibody in FBB at pH 6.0 As described in Example 6, detected with streptavidin-conjugated RPE and analyzed by flow cytometry. FIG. 11B shows a single point comparative binding assay of various mutants at position 314 of OST577-IgG2M3 against human FcRn. Binding of biotinylated OST577-IgG2M3 antibody to human FcRn on NS0 cells transfected in the presence of wild type or position 314 mutant OST577-IgG2M3 competitive antibody in FBB at pH 6.0 As described in Example 6, detected with streptavidin-conjugated RPE and analyzed by flow cytometry. FIG. 11C shows a single point comparative binding assay of various mutants at position 428 of OST577-IgG2M3 against human FcRn. Binding of biotinylated OST577-IgG2M3 antibody to human FcRn on NS0 cells transfected in the presence of wild type or position 428 mutant OST577-IgG2M3 competitive antibody in FBB at pH 6.0 As described in Example 6, detected with streptavidin-conjugated RPE and analyzed by flow cytometry. FIG. 12A shows a competitive binding assay of OST577-IgG2M3 wild type and mutant antibodies against human FcRn. of biotinylated HuEP5C7-IgG2M3 antibody against human FcRn on NS0 cells transfected in the presence of increasing concentrations of wild-type or mutant OST577-IgG2M3 competitive antibody in FBB at pH 6.0 Binding was detected with streptavidin-conjugated RPE and analyzed by flow cytometry as described in Example 6. FIG. 12B shows a competitive binding assay of OST577-IgG2M3 wild type and mutant antibodies against human FcRn. of biotinylated OST577-IgG2M3 antibody against human FcRn on NS0 cells transfected in the presence of increasing concentrations of wild-type or mutant OST577-IgG2M3 competitive antibody in FBB at pH 6.0 Binding was detected with streptavidin-conjugated RPE and analyzed by flow cytometry as described in Example 6. FIG. 13 shows antibody binding to cells transfected with human FcRn versus untransfected cells. Binding of wild-type or mutant OST577-IgG2M3 to human FcRn on NS0 cells transfected with pH 6.0 or to non-transfected NS0 cells flowed as described in Example 7. Analyzed by cytometry. FIG. 14 shows a competitive binding assay of OST577-IgG2M3 wild type and mutant antibodies against human FcRn at 37 ° C. of biotinylated HuEP5C7-IgG2M3 antibody against human FcRn on NS0 cells transfected in the presence of increasing concentrations of wild-type or mutant OST577-IgG2M3 competitive antibody in FBB at pH 6.0 Binding was detected with streptavidin-conjugated RPE and analyzed by flow cytometry as described in Example 7. All incubations were performed at 37 ° C. FIG. 15A shows pH-dependent binding and release of OST577-IgG2M3 wild type and mutant antibodies to human FcRn. Binding and release of wild-type or mutant OST577-IgG2M3 antibody to human FcRn on NS0 cells transfected in FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0 as described in Example 7 Were analyzed by flow cytometry. FIG. 15B shows the pH-dependent binding and release of OST577-IgG1 wild type and mutant antibodies to human FcRn. Binding and release of wild-type or mutant OST577-IgG1 antibody to human FcRn on NS0 cells transfected in FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0 as described in Example 7 Were analyzed by flow cytometry. FIG. 15C shows the pH dependent binding and release of OST577-IgG1 wild type and mutant antibodies to human FcRn. Binding and release of wild-type or mutant OST577-IgG1 antibody to human FcRn on NS0 cells transfected in FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0 as described in Example 7 Were analyzed by flow cytometry. FIG. 15D shows the pH-dependent binding and release of OST577-IgG2M3 wild type and mutant antibodies to rhesus monkey FcRn. Binding and release of wild type or mutant OST577-IgG2M3 antibody to rhesus monkey FcRn on NS0 cells transfected in FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0 as described in Example 7. Were analyzed by flow cytometry. FIG. 15E shows pH-dependent binding and release of OST577-IgG1 wild type and mutant antibodies to rhesus monkey FcRn. Binding and release of wild type or mutant OST577-IgG1 antibody to rhesus monkey FcRn on NS0 cells transfected in FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0 as described in Example 7. Were analyzed by flow cytometry. FIG. 16A shows a competitive binding assay of OST577-IgG2M3 wild type and mutant antibodies against HBsAg. Binding of wild-type or mutant OST577-IgG2M3 antibody to HBsAg was analyzed in an ELISA competition as described in Example 8. FIG. 16B shows a competitive binding assay of OST577-IgG1 wild type and mutant antibodies against HBsAg. Binding of wild type or mutant OST577-IgG1 antibody to HBsAg was analyzed in an ELISA competition as described in Example 8. FIG. 17A shows the binding assay of HulD10-IgG2M3 wild type and mutant antibodies against the HLA-DR β chain allele. Binding of wild-type or mutant HulD10-IgG2M3 antibody to large cells was analyzed in a FACS binding assay as described in Example 8. FIG. 17B shows the binding assay of HulD10-IgG1 wild type and mutant antibodies against the HLA-DR β chain allele. Binding of wild type or mutant HulD10-IgG1 antibody to large cells was analyzed in a FACS binding assay as described in Example 8. FIG. 18A shows ADCC assay of HulD10-IgG1 and HulD10-IgG2M3 wild type and mutant antibodies using PBMC from a 158V / V donor. The ADCC activity of wild-type or mutant HulD10-IgG1 and HulD10-IgG2M3 on large cells was isolated from a donor carrying a homozygous 158V / V FcγRIII allele as described in Example 8. Determined using PBMC. FIG. 18B shows ADCC assay of HulD10-IgG1 and HulD10-IgG2M3 wild type and mutant antibodies using PBMC from 158F / F donor. The ADCC activity of wild type or mutant HulD10-IgG1 and HulD10-IgG2M3 on large cells was isolated from a donor carrying a homozygous 158F / F FcγRIII allele, as described in Example 8. Determined using PBMC. FIG. 19 shows the pharmacokinetics of OST577-IgG2M3 wild type and mutant antibodies in rhesus macaques. Observed and modeled mean serum concentrations (μg / ml) and standard deviations of OST577-IgG2M3 wild type and mutant antibodies administered by infusion at a dose of 1 mg / kg to a group of 4 rhesus macaques are examples Plotted as a function of time (days after infusion) as described in 9. FIG. 20 shows the pharmacokinetics of wild type and mutant antibodies of OST577-IgG1 in rhesus macaques. The observed modeled mean serum concentration (μg / ml) and standard deviation of OST577-IgG1 wild-type and mutant antibodies administered by ring fluid at a dose of 1 mg / kg to a group of 4 rhesus macaques is , Plotted as a function of time (number of days after fluid), as described in Example 10. FIG. 21A shows the pH-dependent binding and release of Hu1D10-IgG3 wild type and mutant antibodies to human FcRn. pH-dependent binding and release of Hu1D10-IgG3 wild type and mutant antibodies to human FcRn on NSO cells transfected in FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0 as described in Example 11. As described above, analysis was performed by a flow cytometry method. FIG. 21B shows the pH-dependent binding and release of Hu1D10-IgG4 wild type and mutant antibodies to human FcRn. pH-dependent binding and release of Hu1D10-IgG4 wild type and mutant antibodies to human FcRn on NSO cells transfected in FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0 are described in Example 11. As described above, analysis was performed by a flow cytometry method. FIG. 21C shows the pH-dependent binding and release of Hu1D10-IgG3 wild type and mutant antibodies against rhesus monkey FcRn. The pH-dependent binding and release of Hu1D10-IgG3 wild type and mutant antibodies to rhesus monkey FcRn on NSO cells transfected in FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0 is described in Example 11. As described above, analysis was performed by a flow cytometry method. FIG. 21D shows the pH-dependent binding and release of Hu1D10-IgG4 wild type and mutant antibodies against rhesus monkey FcRn. The pH-dependent binding and release of Hu1D10-IgG4 wild type and mutant antibodies to rhesus monkey FcRn on NSO cells transfected in FBB at pH 6.0, 6.5, 7.0, 7.5 or 8.0 is described in Example 11. As described above, analysis was performed by a flow cytometry method. FIG. 22-1 shows the amino acid sequence of daclizumab with various FcRn binding mutations. FIG. 22-2 shows the amino acid sequence of daclizumab with various FcRn binding mutations. FIG. 23 shows the amino acid sequence of fontlizumab with various FcRn binding mutations. FIG. 24 shows the amino acid sequence of bicilizumab with various FcRn binding mutations. FIG. 25 shows the amino acid sequence of M200 with various FcRn binding mutations.

Claims (27)

  At least one amino acid residue from a heavy chain constant region selected from the group consisting of amino acid residues 250, 314 and 428 is different from that present in the unmodified IgG class antibody, Modified antibodies of the IgG class, having modified FcRn binding affinity and / or serum half-life in relation to one another.   2. The modified antibody of claim 1, wherein the unmodified IgG class antibody is selected from the group consisting of daclizumab, fontolizumab, visilizumab, and M200.   A naturally occurring IgG wherein at least one amino acid residue from a heavy chain constant region selected from the group consisting of amino acid residues 250, 314 and 428 is different from that present in a naturally occurring IgG class antibody An antibody having a constant region substantially the same as that of a class antibody, wherein the FcRn binding affinity and / or serum half-life of said antibody is altered in relation to said naturally occurring antibody.   4. The modified antibody of claim 3, wherein the naturally occurring IgG class antibody is selected from the group consisting of daclizumab, fontlizumab, bicilizumab and M200. (A) the amino acid residue 250 from the heavy chain constant region is glutamic acid or glutamine; or (b) the amino acid residue 428 from the heavy chain constant region is phenylalanine or leucine;
The antibody according to claim 3.   4. The antibody of claim 3, wherein amino acid residue 250 from the heavy chain constant region is glutamine.   4. The antibody of claim 3, wherein amino acid residue 428 from the heavy chain constant region is leucine. (A) the amino acid residue 250 from the heavy chain constant region is glutamic acid and the amino acid residue 428 from the heavy chain constant region is phenylalanine;
(B) the amino acid residue 250 from the heavy chain constant region is glutamine and the amino acid residue 428 from the heavy chain constant region is phenylalanine; or (c) the amino acid residue 250 from the heavy chain constant region is Glutamine, amino acid residue 428 from the heavy chain constant region is leucine,
The antibody according to claim 3.   An IgG class modified therapeutic antibody having an in vivo elimination half-life that is at least about 1.3 times longer than that of the corresponding unmodified IgG class antibody.   The at least one amino acid residue from a heavy chain constant region selected from the group consisting of residues 250, 314 and 428 is different from that present in the unmodified antibody. IgG class modified therapeutic antibody.   10. The IgG class modified therapeutic antibody of claim 9, wherein the antibody is selected from the group consisting of daclizumab, fontlizumab, bicilizumab and M200. (A) the amino acid residue 250 from the heavy chain constant region is glutamic acid and the amino acid residue 428 from the heavy chain constant region is phenylalanine;
(B) the amino acid residue 250 from the heavy chain constant region is glutamine and the amino acid residue 428 from the heavy chain constant region is phenylalanine; or (c) the amino acid residue 250 from the heavy chain constant region is Glutamine, amino acid residue 428 from the heavy chain constant region is leucine,
10. A modified therapeutic antibody of the IgG class according to claim 9.   An IgG class modified therapeutic antibody having an in vivo clearance phase that is at least about 1.3 times lower than that of the corresponding unmodified IgG class antibody.   At least one amino acid residue from a heavy chain constant region selected from the group consisting of residues 250, 314 and 428 is different from that present in an unmodified IgG class antibody. A modified therapeutic antibody of the described IgG class.   14. The IgG class modified therapeutic antibody of claim 13, wherein the antibody is selected from the group consisting of daclizumab, fontlizumab, bicilizumab and M200.   A modified therapeutic antibody comprising a light chain amino acid sequence of SEQ ID NO: 118 and a heavy chain amino acid sequence selected from SEQ ID NOs: 119-128.   A vector comprising a polynucleotide encoding the light or heavy chain amino acid sequence of claim 16.   A host cell comprising the vector of claim 16.   A modified therapeutic antibody comprising a light chain amino acid sequence of SEQ ID NO: 129, and a heavy chain amino acid sequence selected from SEQ ID NOs: 130-134.   20. A vector comprising a polynucleotide encoding the light or heavy chain amino acid sequence of claim 19.   20. A host cell comprising the vector of claim 19.   A modified therapeutic antibody comprising a light chain amino acid sequence of SEQ ID NO: 135 and a heavy chain amino acid sequence selected from SEQ ID NOs: 136-140.   23. A vector comprising a polynucleotide encoding the light or heavy chain amino acid sequence of claim 22.   23. A host cell comprising the vector of claim 22.   A modified therapeutic antibody comprising a light chain amino acid sequence of SEQ ID NO: 141 and a heavy chain amino acid sequence selected from SEQ ID NOs: 142-146.   26. A vector comprising a polynucleotide encoding the light or heavy chain amino acid sequence of claim 25.   26. A host cell comprising the vector of claim 25.

Images