RU2833597C2 - Fc VERSIONS WITH ALTERED BINDING TO FcRn - Google Patents

Abstract

FIELD: biotechnology.

SUBSTANCE: invention relates to biotechnology, specifically to anti-TNF antibodies, and can be used for their recombinant production. Disclosed is an anti-TNF antibody containing two identical heavy chain polypeptides, each having the amino acid sequence SEQ ID NO: 29; and two identical light chain polypeptides, each having the amino acid sequence SEQ ID NO: 23, as well as means for its recombinant production.

EFFECT: invention provides obtaining anti-TNF antibodies with improved pharmacokinetic properties.

5 cl, 23 dwg, 7 ex

Description

This application claims priority under 35 U.S.C. §119(e) to USSN 61/016793, filed December 26, 2007; USSN 61/031353, filed February 25, 2008; USSN 61/046353, filed April 18, 2008; USSN 61/050172, filed May 2, 2008; USSN 61/079779, filed July 10, 2008; and USSN 61/099178, filed September 22, 2008; and is a continuation-in-part of USSN 11/932,151, filed October 31, 2007, which is a continuation-in-part of USSN 11/436,266, filed May 17, 2006, which claimed priority under 35 U.S.C. §119(e) over USSN 60/951536, filed July 24, 2007, and is a continuation-in-part of USSN 11/274,065, filed November 14, 2005, which claimed priority under 35 U.S.C. §119(e) over USSN 60/627763, filed November 12, 2004; USSN 60/642,886, filed January 11, 2005; USSN 60/649508, filed February 2, 2005; USSN 60/662468, filed March 15, 2005; USSN 60/669311, filed April 6, 2005; USSN 60/681607, filed May 16, 2005; USSN 60/690200, filed June 13, 2005; USSN 60/696609, filed July 5, 2005; USSN 60/703018, filed July 27, 2005; and USSN 60/726453, filed October 12, 2005, the entire contents of which are incorporated herein by reference.

Field of technology

The present application relates to optimized variants of immunoglobulin IgG, methods for their production and their use, in particular for therapeutic purposes.

State of the art

Antibodies are immunological proteins, each of which binds to a specific antigen. In most mammals, including humans and mice, antibodies are formed from paired heavy and light polypeptide chains. Each chain is composed of individual immunoglobulin (Ig) domains, and so the collective term immunoglobulin is used for such proteins. Each chain is made up of two distinct regions called the "variable" and "constant" regions. The variable regions of the light and heavy chains exhibit significant sequence differences in antibodies and are responsible for binding to the target antigen. The constant regions exhibit less sequence differences and are responsible for binding to a variety of naturally occurring proteins to stimulate important biochemical reactions. In humans, there are five different classes of antibodies, including IgA (which includes the IgA1 and IgA2 subclasses), IgD, IgE, IgG (which includes the IgG1, IgG2, IgG3, and IgG4 subclasses), and IgM. The distinguishing feature of these classes of antibodies is their constant regions, although subtle differences may exist in the V region. IgG antibodies are tetrameric proteins consisting of two heavy chains and two light chains. The IgG heavy chain consists of four immunoglobulin domains linked from N- to C-terminal in the order VH-CH1-CH2-CH3 relative to the heavy chain variable domain, heavy chain constant domain 1, heavy chain constant domain 2, and heavy chain constant domain 3, respectively (also designated VH-Cγ1-Cγ2-Cγ3 relative to the heavy chain variable domain, gamma 1 constant domain, gamma 2 constant domain, and gamma 3 constant domain, respectively). The IgG light chain consists of two immunoglobulin domains linked from N- to C-terminal in the order VL-CL relative to the light chain variable domain and the light chain constant domain, respectively.

In IgG, the Fc region between the Cγ2 and Cγ3 domains mediates interaction with the neonatal receptor FcRn. Binding to FcRn recycles endocytosed antibody from the endosome back into the circulation (Raghavan et al., 1996, Annu Rev Cell Dev Biol 12:181-220; Ghetie et al., 2000, Annu Rev Immunol 18:739-766, both incorporated herein by reference in their entireties). This process, coupled with impaired renal filtration due to the large volume of the full-length molecule, results in favorable serum half-lives of antibody, ranging from one to three weeks. Fc binding to FcRn also plays a key role in antibody trafficking. The binding site in the Fc region for FcRn is also the site at which bacterial proteins A and G bind. Tight binding by these proteins is typically used as a method for purifying antibodies by using protein A or protein G affinity chromatography during protein purification. Thus, the specificity of this Fc region is important for both the clinical properties of antibodies and their purification. The available structures of the rat Fc/FcRn complex (see Burmeister et al., 1994, Nature, 372:379-383; Martin et al., 2001, Mol Cell 7:867-877, both incorporated herein by reference in their entireties) and of Fc complexes with proteins A and G (Deisenhofer, 1981, Biochemistry 20:2361-2370; Sauer-Eriksson et al., 1995, Structure 3:265-278; Tashiro et al., 1995, Curr Opin Struct Biol 5:471-481, all incorporated herein by reference in their entireties) provide insight into the interaction of Fc with these proteins. The FcRn receptor is also responsible for the transfer of IgG into the neonatal intestine and into the lumen of the intestinal epithelium in adults (Ghetie and Ward, Annu. Rev. Immunol., 2000, 18:739-766; Yoshida et al., Immunity, 2004, 20(6):769-783, both incorporated herein by reference in their entireties).

Studies of the rat and human Fc domains have demonstrated the importance of certain Fc residues for FcRn binding. The rat and human sequences share approximately 64% sequence identity in the Fc regions (residues 237-443 according to EU index numbering). See Figs. 3, 4, and 5 for alignments of rat/human Fc, FcRn heavy chain, and FcRn light chain (beta-2-microglobulin). A model of the human Fc/FcRn complex was generated based on the available structure of the rat Fc/FcRn complex (Martin et al., 2001, Mol Cell 7:867-877, incorporated herein by reference in its entirety). The rat and human sequences contain some common residues that are critical for FcRn binding, such as H310 and H435 (Medesan et al., 1997 J. Immunol. 158(5):221-7; Shields et al., 2001 J. Biol. Chem. 276(9):6591-6604, both incorporated by reference in their entireties). However, the human and rat proteins contain different amino acids at many positions, giving these residues a different context and possibly different identity in the human sequence than in the rat sequence. This variability limits the ability to transfer features from one homolog to another.

In mouse Fc, random mutation and selection of phage display at sites T252, T254, and T256 results in a T252L/T254S/T256F triple mutant that has a 3.5-fold increase in affinity for FcRn and a 1.5-fold increase in serum half-life (Ghetie et al., 1997, Nat. Biotech. 15(7):637-640, incorporated herein by reference in its entirety). Disruption of the Fc/FcRn interaction by mutations at positions 253, 310, and 435 also results in decreased half-life in vivo (Medesan et al J. Immunol. 1997 158(5):2211-7, incorporated herein by reference in its entirety).

Mutation studies in human Fcγ have been performed at several residues important for binding to FcRn, and these studies have demonstrated increased serum half-life. In human Fcγ1, Hinton et al. individually mutated three residues into another 19 common amino acids. Hinton et al. found that some mutants, a double mutant, increased FcRn binding affinity (Hinton et al., 2004, J. Biol. Chem. 279(8): 6213-6216. Hinton et al. Journal of Immunology 2006, 176:346-356, both incorporated by reference in their entireties). Two mutations had increased half-life in monkeys. Shields et al. mutated residues almost exclusively to Ala and examined their binding to FcRn and FcγR's (Shields et al., 2001, J. Biol. Chem., 276(9):6591–6604, incorporated herein by reference in its entirety).

Dall'Acqua et al. used phage display to select Fc mutations that bound FcRn with increased affinity (Dall' Acqua et al. 2002, J. Immunol. 169:5171-5180, incorporated herein by reference in its entirety). The DNA sequences selected were predominantly double and triple mutants. Dall'Acqua et al. expressed proteins encoded by many of the selected sequences and found several that bound FcRn more tightly than wild-type Fc.

Administration of antibodies and Fc fusion proteins as drugs requires injections at a given frequency relative to the clearance and half-life parameters of the protein. A longer in vivo half-life allows for less frequent injections or lower doses, which is obviously beneficial. Although mutations in the Fc domain have resulted in some proteins with increased FcRn binding affinity and increased in vivo half-life, these mutations have not determined the optimal mutations and increased in vivo half-life.

One of the hallmarks of the Fc region is the conserved N-linked glycosylation occurring at N297. This carbohydrate, or oligosaccharide as it is sometimes called, plays a key structural and functional role for the antibody and is one of the main reasons why antibodies should be produced using mammalian expression systems (Umaña et al., 1999, Nat Biotechnol 17: 176–180; Davies et al., 2001, Biotechnol Bioeng 74: 288–294; Mimura et al., 2001, J Biol Chem 276: 45539–45547.; Radaev et al., 2001, J Biol Chem 276: 16478–16483; Shields et al., 2001, J Biol Chem 276: 6591–6604; Shields et al., 2002, J Biol Chem 277: 26733-26740; Simmons et al., 2002, J Immunol Methods 263: 133-147; Radaev et al., 2001, J Biol Chem 276: 16469-16477; and Krapp et al., 2003, J Mol Biol 325: 979-989, all of which are incorporated herein by reference in their entireties).

It is known to create antibodies for therapeutic use. Such therapies are described, for example, in Chamow et al., 1996, Trends Biotechnol 14: 52-60; Ashkenazi et al., 1997, Curr Opin Immunol 9: 195-200, Cragg et al., 1999, Curr Opin Immunol 11: 541-547; Glennie et al., 2000, Immunol Today 21: 403-410, McLaughlin et al., 1998, J Clin Oncol 16: 2825-2833 and Cobleigh et al., 1999, J Clin Oncol 17: 2639-2648, each of which is incorporated herein by reference in its entirety. Currently, in cancer therapy, any small reduction in mortality is considered a success. Certain IgG variants described herein enhance the ability of antibodies to limit further growth or destroy, at least in part, target cancer cells.

The antitumor activity of antibodies is exerted by enhancing their ability to mediate cytotoxic effector functions such as ADCC, ADCP, and CDC. See, e.g., Clynes et al., 1998, Proc Natl Acad Sci U S A 95: 652-656; Clynes et al., 2000, Nat Med 6: 443-446; and Cartron et al., 2002, Blood 99: 754-758, both of which are herein incorporated by reference in their entireties.

Human IgG1 is the most commonly used antibody for therapeutic purposes, and most of the exploratory studies have been conducted in this regard. However, different isotypes of the IgG class, including IgG1, IgG2, IgG3 and IgG4, have unique physical, biological and clinical properties. There is a need in the art for improved variants of IgG1, IgG2, IgG3 and IgG4. There is also a need for these variants to improve binding to FcRn and/or increase half-life in vivo, compared to native IgG polypeptides. In addition, there is a need to combine variants having improved pharmacokinetic properties with variants containing modifications to improve efficacy through altered FcgammaR binding. The present application addresses these and other needs.

Brief description of the invention

The present application relates to Fc variants of a parent polypeptide comprising at least one modification in the Fc region of said polypeptide. In various embodiments, the variant polypeptides exhibit altered binding to FcRn compared to the parent polypeptide. In some embodiments, the modification may be selected from the group consisting of: 428L, 434M, and 434S, wherein the numbering is given according to the EU Index of Kabat et al.

In another embodiment, the Fc variant comprises at least two modifications selected from the group consisting of: 252Y/428L, 428L/434H, 428L/434F, 428L/434Y, 428L/434A, 428L/434M and 428L/434S.

In another embodiment, the Fc variant comprises at least one modification selected from the group consisting of: M428L/N434S, V308F/M428L/N434S.

In another embodiment, the Fc variant comprises at least one modification selected from the group consisting of: 259I/434S, 308F/434S, 308F/428L/434S, 259I/308F/434S, 307Q/308F/434S, 250I/308F/434S, and 308F/319L/434S.

In another embodiment, the Fc variant comprises at least one modification selected from the group consisting of:

In another embodiment, the present invention includes a method of treating a patient in need of such treatment, comprising administering an effective amount of an Fc variant as described herein.

In another embodiment, the present invention includes a method for increasing the half-life of an antibody or immunoadhesin by modifying the Fc according to the modifications described herein.

In another embodiment, the present invention includes an Fc variant having enhanced binding of FcRn to additional Fc variants that modulate effector function.

Brief description of the figures

Fig. 1. Sequence alignments of the human IgG constant heavy chains. Gray indicates differences from IgG1, and squared residues indicate common allotypic variations in the human population.

Fig. 2. (SEQ ID NO: 1-6). Amino acid sequences of constant regions used according to the present invention.

Fig. 3. (SEQ ID NO: 7-12). Amino acid sequences of constant regions of typical variants.

Fig. 4. (SEQ ID NO: 13-22). Amino acid sequences of the variable regions VH and VL used according to the present invention.

Fig. 5. (SEQ ID NO: 23-29). Amino acid sequences of typical variant antibodies.

Fig. 6. Relative binding of VEGF by wild type and selected variant IgG1 anti-VEGF antibodies. The graph shows the Biacore reaction units (RU) at the end of the association phase for binding of the analyte antibody to the immobilized VEGF antigen. IgG1 anti-Her2 antibody was used as a negative control.

Fig. 7. Biacore sensorgrams of wild-type and variant IgG1 antibodies to immobilized human FcRn at low (6.0) and high (7.4) pH.

Fig. 8. FcRn binding affinities of wild-type IgG1 and selected variant IgG1 antibodies for human FcRn at pH 6.0 as determined by Biacore. The pseudoaffinity constant (Ka*) is plotted on a log scale.

Fig. 9. Relative binding of IgG1 anti-VEGF variant antibodies to human FcRn as determined by Biacore. The table shows the fold Ka* of each variant relative to wild-type (native) human IgG1. n indicates the number of times each variant was tested, “Mean” and “SD” indicate the mean and standard deviation, respectively, for each variant in n binding experiments. The fold of FcRn was calculated for all variants relative to wild-type IgG1 within each respective binding experiment. “NB” indicates that no binding was detected. “ND” indicates that binding was not determined for this particular variant. “NF” indicates that a match was not possible based on the binding data.

Fig. 10. Relative binding of IgG2 and IgG1/2 anti-VEGF variant antibodies to human FcRn as determined by Biacore. The table is as described in Fig. 9.

Fig. 11. Analysis of additive and synergistic substitution combinations. Fig. 11a shows a plot of the experimentally determined fold binding of each variant to human FcRn versus the predicted fold binding of FcRn as determined by the output of the individual variants. The base points of the variants are marked, with the line representing absolute additivity. Fig. 11b shows the difference between the experimental and predicted folds for each combination variant. Fig. 11c shows the synergy of each combination of variants. % synergy is calculated as 100×[(experimental fold/predicted fold) – 1)].

Fig. 12. Relative binding of anti-TNF, -CD25, -EGFR and –IgE variant antibodies to human FcRn as determined by Biacore. The table is as described in Fig. 9.

Fig. 13. In vivo pharmacokinetic characteristics of wild-type and variant antibodies in mFcRn−/− hFcRn+ mice. Graphs show serum antibody concentration versus time after single intravenous dose administration. Fig. 13a shows data from one of four studies conducted using IgG1 antibodies (Study 3), and Fig. 13b shows data from a study conducted using IgG2 antibodies (Study 5).

Fig. 14. Fitted PK parameters for all in vivo PK studies performed in mFcRn−/− hFcRn+ mice using variant and wild-type antibodies. n represents the number of mice per group with mean and standard deviation (SD) data predicted by the PK parameters. “Half-life” represents the beta phase characterizing the elimination of antibody from serum. Cmax represents the maximum observed serum concentration, AUC represents the area under the concentration-time curve, and “clearance” represents the clearance of antibody from serum. Half-life multiples are calculated as the half-life of the variant antibody relative to the half-life of the parental wild-type IgG1 or IgG2 within each study.

Fig. 15. Correlation between half-lives of IgG1 (Fig. 15a) and IgG2 (Fig. 15b) variant antibodies in mFcRn−/−hFcRn+ mice and fold FcRn binding relative to wild-type IgG1. Y-axis data are from Fig. 14 and x-axis data are from Figs. 9 and 10. Selected variants are marked and data from replicate variants are circled. Fig. 15c shows both IgG1 and IgG2 correlation data, with black and gray lines representing the correspondence of IgG1 and IgG2 data, respectively.

Fig. 16 (SEQ ID NO: 30-35). Amino acid sequences of variant and parental anti-TNF Fc immunoadhesins used according to the present invention.

Fig. 17. Binding of anti-TNF immunoadhesins to TNF antigen as determined by Biacore.

Fig. 18. Relative binding of variant Fc immunoadhesins to human FcRn as determined by Biacore. The table shows the fold Ka* of each variant relative to human wild-type (native) IgG1. n indicates the number of times each variant was tested, and “Mean” and “SD” indicate the mean and standard deviation, respectively, for each variant in n binding experiments. The fold of FcRn was calculated for all variants relative to the corresponding parental IgG1 within each respective binding experiment.

Fig. 19. In vivo pharmacokinetic characteristics of parental and variant Fc immunoadhesins in mFcRn-/- hFcRn+ mice. Graphs show the serum Fc hybrid concentration versus time after single intravenous administration.

Fig. 20. Fitted PK parameters of the Fc hybrid obtained from an in vivo PK study in mFcRn-/- hFcRn+ mice. The parameters are as described in Fig. 14. The % increase in half-life is calculated as the half-life of the variant Fc hybrid relative to the half-life of the parental wild-type IgG1 or IgG2 multiplied by 100.

Fig. 21. Relative binding of variant anti-VEGF IgG1 antibodies to cynomolgus monkey and human FcRn as determined by Biacore. Fig. 21a presents the data in tabular form. The description of this figure corresponds to the description of Fig. 9, and the binding data to human FcRn are taken from Fig. 9. Fig. 21b presents a graph of these data.

Fig. 22. In vivo pharmacokinetic characteristics of wild-type and variant antibodies in cynomolgus macaques. Graphs show serum antibody concentration versus time after single intravenous administration.

Fig. 23. Fitted PK parameters obtained from an in vivo PK study in cynomolgus macaques using variant and wild-type antibodies. The parameters are as described in Fig. 14.

Detailed description of the invention

The present invention describes the production of novel Fc domain variants, including variants contained in antibodies, Fc hybrids, and immunoadhesins, that have increased binding to the FcRn receptor. As described herein, binding to FcRn results in longer serum retention in vivo.

To enhance the retention of Fc proteins in vivo, increased binding affinity must occur at a pH of about 6, while lower affinity maintenance occurs at a pH of about 7.4. Although studies are ongoing, it is thought that Fc regions have a longer half-life in vivo because binding to FcRn at pH 6 in the endosome sequesters Fc (Ghetie and Ward, 1997 Immunol Today. 18(12): 592-598, incorporated by reference in its entirety). The endosomal compartment then returns Fc to the cell surface. Once the compartment opens to the extracellular space, the higher pH of ~7.4 induces the release of Fc back into the blood. In mice, Dall' Acqua et al. showed that Fc mutants that had increased FcRn binding at pH 6 and pH 7.4 actually had reduced serum concentrations and half-lives similar to wild-type Fc (Dall' Acqua et al. 2002, J. Immunol. 169: 5171-5180, incorporated herein by reference in its entirety). It is believed that the increased affinity of Fc for FcRn at pH 7.4 does not allow Fc to be released back into the blood. Therefore, Fc mutations that would increase Fc half-life in vivo would ideally increase FcRn binding at lower pH while allowing Fc to be released at higher pH. The amino acid histidine changes its charged state over the pH range of 6.0-7.4. It is therefore not surprising that His residues are located in positions of importance in the Fc/FcRn complex (Fig. 6).

An additional aspect of the present invention is an increase in FcRn binding compared to wild type, in particular at a lower pH, around pH 6.0, to facilitate Fc/FcRn binding in the endosome. Also described are Fc variants having altered FcRn binding and altered binding to another class of Fc receptors, the FcγR's (sometimes written FcgammaR's), as differential binding to FcγRs, in particular increased binding to FcγRIIIb and decreased binding to FcγRIIb, has been shown to result in increased potency.

Definitions

For a more complete understanding of this application, certain definitions are provided below. It is understood that the said definitions include grammatical equivalents.

As used herein, the term "ADCC" or "antibody-dependent cell-mediated cytotoxicity" refers to a cell-mediated reaction in which non-specific cytotoxic cells expressing FcγRs recognize bound antibody on a target cell and then cause lysis of the target cell.

As used herein, the term "ADCP" or "antibody-dependent cell-mediated phagocytosis" refers to a cell-mediated reaction in which non-specific cytotoxic cells expressing FcγRs recognize bound antibody on a target cell and then cause phagocytosis of the target cell.

As used herein, the term "modification" means substitution, insertion and/or deletion of amino acids in a polypeptide sequence or transformation into a fragment chemically linked to a protein. For example, a modification may be an altered carbohydrate structure or PEG attached to a protein. As used herein, the term "amino acid modification" means substitution, insertion and/or deletion of amino acids in a polypeptide sequence.

As used herein, the term "amino acid substitution" or "substitution" refers to the replacement of an amino acid at a particular position in a parent polypeptide with another amino acid. For example, the substitution E272Y refers to a variant polypeptide, in this case an Fc variant, in which glutamic acid at position 272 is replaced with tyrosine.

As used herein, the term "amino acid insertion" or "insertion" refers to the addition of an amino acid sequence at a specific position in a parent polypeptide sequence. For example, -233E or ^∧233E refers to the insertion of glutamic acid after position 233 and before position 234. In addition, -233ADE or ^∧233ADE refers to the insertion of AlaAspGlu after position 233 and before position 234.

As used herein, the term "amino acid deletion" or "deletion" refers to the deletion of an amino acid sequence at a specific position in a parent polypeptide sequence. For example, E233- or E233# refers to a deletion of glutamic acid at position 233. In addition, EDA233- or EDA233# refers to a deletion of the GluAspAla sequence beginning at position 233.

As used herein, the term "variant protein" or "protein variant" or "variant" means a protein that differs from a parent protein by having at least one amino acid modification. The term "protein variant" may refer to the protein itself, a composition comprising said protein, or an amino acid sequence encoding it. Preferably, a protein variant comprises at least one amino acid modification compared to the parent protein, such as from about one to about seventy amino acid modifications, and preferably from about one to about five amino acid modifications compared to the parent protein. As used herein, the sequence of a protein variant will preferably have at least about 80% homology to the sequence of the parent protein, and more preferably at least about 90% homology, most preferably at least about 95% homology. The term "variant protein" may refer to the variant protein itself, compositions comprising said protein variant, or a DNA sequence encoding it. Accordingly, as used herein, the term "antibody variant" or "variant antibody" means an antibody that differs from a parent antibody by having at least one amino acid modification, the term "IgG variant" or "variant IgG" means an antibody that differs from a parent IgG by having at least one amino acid modification, and the term "immunoglobulin variant" or "variant immunoglobulin" means an immunoglobulin sequence that differs from a parent immunoglobulin sequence by having at least one amino acid modification. As used herein, the term "Fc variant" or "variant Fc" means a protein that comprises a modification in the Fc domain. Fc variants of the present invention are defined according to the amino acid modifications they comprise. Thus, for example, N434S or 434S is an Fc variant that comprises a substitution, serine, at position 434 relative to the parent Fc polypeptide, wherein the numbering is given according to the EU index. Similarly, M428L/N434S defines an Fc variant comprising the substitutions M428L and N434S relative to a parent Fc polypeptide. The wild-type amino acid identity may not be determined, in which case the above variant is designated 428L/434S. It is noted that the order in which the substitutions are provided is arbitrary, i.e., for example, 428L/434S is the same Fc variant as M428L/N434S, etc. For all positions described according to the present invention, numbering is according to the EU index. The term "EU index" or "EU index according to Kabat" or "EU numbering scheme" refers to the EU numbering of an antibody (Edelman et al., 1969, Proc Natl Acad Sci USA 63:78-85, hereby incorporated by reference in its entirety). The modification may be an addition, deletion, or substitution. Substitutions may include naturally occurring amino acids and non-naturally occurring amino acids. Variants may comprise unnatural amino acids. Examples include US 6,586,207; WO 98/48032; WO 03/073238; US2004-0214988A1; WO 05/35727A2; WO 05/74524A2; J. W. Chin et al., (2002), Journal of the American Chemical Society 124: 9026-9027; J. W. Chin, & P. G. Schultz, (2002), ChemBioChem 11: 1135-1137; J. W. Chin, et al., (2002), PICAS United States of America 99: 11020-11024; and, L. Wang, & P. G. Schultz, (2002), Chem. 1-10, all of which are incorporated herein by reference in their entireties.

As used herein, the term "protein" means at least two covalently attached amino acids, and includes proteins, polypeptides, oligopeptides, and peptides. The peptidyl group may comprise naturally occurring amino acids and peptide bonds or synthetic peptidomimetic structures, i.e., "analogs," such as peptoids (see Simon et al., PNAS USA 89(20):9367 (1992), which is incorporated herein by reference in its entirety). The amino acids may be either naturally occurring or non-naturally occurring, as will be apparent to those skilled in the art. For example, homophenylalanine, citrulline, and norleucine are considered amino acids for the purposes of the present invention, and amino acids of both the D- and L- (R or S) configurations may be used. Variants of the present invention may comprise modifications involving the use of unnatural amino acids using, for example, the techniques developed by Schultz and colleagues, including, but not limited to, the methods described by Cropp & Shultz, 2004, Trends Genet. 20(12): 625-30, Anderson et al., 2004, Proc Natl Acad Sci USA 101(2): 7566-71, Zhang et al., 2003, 303(5656): 371-3 and Chin et al., 2003, Science 301(5635): 964-7, all of which are incorporated herein by reference in their entireties. In addition, polypeptides may include synthetic derivatization of one or more side chains or terminals, glycosylation, PEGylation, cyclic rearrangement, cyclization, linkers for attachment to other molecules, hybridization into proteins or protein domains, and the addition of peptide tags or labels.

As used herein, the term "residue" refers to a position in a protein and the associated amino acid identity. For example, Asparagine 297 (also referred to as Asn297 or N297) is the residue at position 297 in the human IgG1 antibody.

As used herein, the term "Fab" or "Fab region" means a polypeptide comprising the VH, CH1, VL, and CL domains of an immunoglobulin. The term "Fab" may refer to this region alone or in the context of a full-length antibody, an antibody fragment, or a Fab fusion protein. As used herein, the term "Fv" or "Fv fragment" or "Fv region" means a polypeptide comprising the VL and VH domains of a single antibody.

As used herein, the term "IgG subclass modification" means an amino acid modification that converts one amino acid of one IgG isotype into the corresponding amino acid of another, aligned IgG isotype. For example, since IgG1 contains tyrosine and IgG2 contains phenylalanine at position EU 296, the F296Y substitution in IgG2 is considered an IgG subclass modification.

As used herein, the term "non-naturally occurring modification" means an amino acid modification that is not isotypic. For example, since none of the IgGs contain serine at position 434, a 434S substitution in IgG1, IgG2, IgG3, or IgG4 is considered a non-naturally occurring modification.

As used herein, the terms "amino acid" and "amino acid identity" mean one of the 20 naturally occurring amino acids or any non-natural analogs that may be present at a particular, defined position.

As used herein, the term "effector function" refers to a biochemical reaction resulting from the interaction of an Fc region of an antibody with a receptor or Fc ligand. Effector functions include, but are not limited to, ADCC, ADCP, and CDC.

As used herein, the term "IgG Fc ligand" refers to a molecule, preferably a polypeptide, from any organism that binds to the Fc region of an IgG antibody to form an Fc/Fc ligand complex. Fc ligands include, but are not limited to, FcγRs, FcγRs, FcγRs, FcRn, C1q, C3, mannan-binding lectin, mannose receptor, staphylococcal protein A, streptococcal protein G, and viral FcγR. Fc ligands also include Fc receptor homologs (FcRH), which are a family of Fc receptors homologous to FcγRs (Davis et al., 2002, Immunological Reviews 190:123-136, incorporated herein by reference in its entirety). Fc ligands may include uncharacterized Fc binding molecules. Particular Fc ligands of IgG are FcRn and Fc gamma receptors. As used herein, the term "Fc ligand" means a molecule, preferably a polypeptide, derived from any organism, that binds to the Fc region of an antibody to form an Fc/Fc ligand complex.

As used herein, the term "Fc gamma receptor," "FcγR," or "FcgammaR" refers to any member of a family of proteins that binds to the Fc region of an IgG antibody and is encoded by the FcγR gene. In humans, this family includes, but is not limited to, FcγRI (CD64), including the isoforms FcγRIa, FcγRIb, and FcγRIc; FcγRII (CD32), including the isoforms FcγRIIa (including the H131 and R131 allotypes), FcγRIIb (including FcγRIIb-1 and FcγRIIb-2), and FcγRIIc; and FcγRIII (CD16), including the FcγRIIIa (including the V158 and F158 allotypes) and FcγRIIIb (including the FcγRIIIb-NA1 and FcγRIIIb-NA2 allotypes) isoforms (Jefferis et al., 2002, Immunol Lett 82: 57-65, incorporated herein by reference in its entirety), as well as any uncharacterized human FcγRs or FcγR isoforms or allotypes. FcγR may be from any organism, including, but not limited to, humans, mice, rats, rabbits, and monkeys. Mouse FcγRs include, but are not limited to, FcγRI (CD64), FcγRII (CD32), FcγRIII (CD16), and FcγRIII-2 (CD16-2), as well as any uncharacterized mouse FcγRs or FcγR isoforms or allotypes.

As used herein, the term "FcRn" or "neonatal Fc receptor" refers to a protein that binds to the Fc region of an IgG antibody and is at least partially encoded by the FcRn gene. FcRn may be from any organism, including, but not limited to, humans, mice, rats, rabbits, and monkeys. As is known in the art, a functional FcRn protein comprises two polypeptides, often referred to as a "heavy chain" and a "light chain." The light chain is beta-2-microglobulin, and the heavy chain is encoded by the FcRn gene. Unless otherwise indicated herein, the term "FcRn" or "FcRn protein" refers to a complex of the FcRn heavy chain with beta-2-microglobulin. FcRn sequences, particularly those of humans, of particular interest are shown in the figures.

As used herein, the term "parent polypeptide" refers to an unmodified peptide that is subsequently modified to produce a variant. The parent polypeptide may be a naturally occurring polypeptide or a variant or engineered version of a naturally occurring polypeptide. The term "parent polypeptide" may refer to the polypeptide itself, compositions comprising said parent polypeptide, or an amino acid sequence encoding it. Accordingly, as used herein, the term "parent immunoglobulin" refers to an unmodified immunoglobulin peptide that is modified to produce a variant, and the term "parent antibody" refers to an unmodified antibody that is modified to produce a variant antibody. It should be noted that the term "parent antibody" includes known commercially available recombinantly produced antibodies, as described below.

As used herein, the term "position" refers to a location in a protein sequence. Positions may be numbered sequentially or according to an established scheme, such as the EU index for antibody numbering.

In the present description, the term "target antigen" means a molecule specifically bound by the variable region of the said antibody. The target antigen may be a protein, carbohydrate, lipid or other chemical compound.

As used herein, the term “target cell” means a cell expressing a target antigen.

As used herein, the term "variable region" means a region of an immunoglobulin that comprises one or more immunoglobulin domains substantially encoded by any of the Vκ, Vλ and/or VH genes that comprise the kappa, lambda and heavy chain immunoglobulin gene loci, respectively.

As used herein, the term "wild type or WT" means an amino acid sequence or nucleotide sequence that occurs in nature, including allelic variations. A wild type protein comprises an amino acid sequence or nucleotide sequence that has not been intentionally modified.

The present invention relates to antibodies that exhibit increased binding to FcRn compared to a wild-type antibody. For example, in some cases, the increased binding results in cellular recycling of the antibody, and thus an increased half-life. In addition, antibodies that exhibit increased binding to FcRn and altered binding to other Fc receptors, such as FcγRs, find use in the present invention.

Antibodies

The present invention relates to antibodies that contain amino acid modifications that modulate binding to FcRn. Of particular interest are antibodies that minimally contain an Fc region or a functional variant thereof that exhibit increased binding affinity to FcRn at a lower pH and substantially no altered binding at a higher pH.

Traditional antibody structural units typically comprise a tetramer. Each tetramer typically consists of two identical pairs of polypeptide chains, with each pair containing one "light" chain (typically having a molecular weight of approximately 25 kDa) and one "heavy" chain (typically having a molecular weight of approximately 50-70 kDa). Human light chains are divided into kappa and lambda light chains. Heavy chains are divided into mu, delta, gamma, alpha, or epsilon and determine the antibody's isotype, IgM, IgD, IgG, IgA, and IgE, respectively. IgG has several subclasses, including but not limited to IgG1, IgG2, IgG3, and IgG4. IgM has subclasses, including but not limited to IgM1 and IgM2. Thus, as used herein, the term "isotype" means any of the subclasses of immunoglobulins defined by the chemical and antigenic properties of their constant regions. Known human immunoglobulin isotypes are IgG1, IgG2, IgG3, IgG4, IgA1, IgA2, IgM1, IgM2, IgD, and IgE.

The N-terminus of each chain contains a variable region of approximately 100-110 or more amino acids, primarily responsible for antigen recognition. In the variable region, three loops are connected for each of the V domains of the heavy chain and light chain to form the antigen-binding site. Each of the loops is called a "complementarity-determining region" (hereinafter referred to as "CDR"), in which amino acid sequence variation is most significant.

The C-terminus of each chain defines a constant region primarily responsible for effector function. Kabat et al. obtained numerous primary sequences of the variable regions of the heavy chains and light chains. Based on the degree of conservation of these sequences, they divided the individual primary sequences into CDRs and framework regions and compiled a list of them (see SEQUENCES OF IMMUNOLOGICAL INTEREST, 5th edition, NIH publication, No. 91-3242, E.A. Kabat et al., which is incorporated herein by reference in its entirety).

In the IgG subclass of immunoglobulins, there are several immunoglobulin domains in the heavy chain. As used herein, the term "immunoglobulin (Ig) domain" refers to a region of an immunoglobulin having a distinct tertiary structure. Of interest in the present invention are heavy chain domains including constant heavy (CH) domains and hinge domains. In the context of IgG antibodies, each IgG isotype contains three CH regions. Accordingly, the "CH" domains in the context of IgG are as follows: "CH1" refers to positions 118-220 according to the EU index, as in Kabat. "CH2" refers to positions 237-340 according to the EU index, as in Kabat. and "CH3" refers to positions 341-447 according to the EU index, as in Kabat.

Another type of Ig heavy chain domain is the hinge region. As used herein, the term "hinge" or "hinge region" or "antibody hinge region" or "immunoglobulin hinge region" means a flexible polypeptide comprising amino acids between the first and second constant domains of an antibody. Structurally, the CH1 domain of IgG ends at EU position 220 and the CH2 domain of IgG begins at EU position 237. Thus, for IgG, the antibody hinge as used herein includes positions 221 (D221 in IgG1) to 236 (G236 in IgG1), with the numbering given according to the EU index as in Kabat. In some embodiments, for example in the context of an Fc region, a lower hinge is included, with the term "lower hinge" generally referring to positions 226 or 230.

Of particular interest in the present invention are the Fc regions. As used herein, the term "Fc" or "Fc region" refers to a polypeptide comprising the constant region of an antibody minus the first immunoglobulin domain of the constant region and, in some cases, a portion of the hinge. Thus, the term "Fc" refers to the last two immunoglobulin domains of IgA, IgD, and IgG of the constant region, the last three immunoglobulin domains of IgE and IgM of the constant region, and a flexible hinge N-terminal to these domains. For IgA and IgM, the Fc may comprise a J chain. For IgG, as depicted in Fig. 1, the Fc comprises the immunoglobulin domains Cgamma2 and Cgamma3 (Cg2 and Cg3) and a lower hinge region between Cgamma1 (Cg1) and Cgamma2 (Cg2). Although the boundaries of the Fc region may vary, the Fc region of a human IgG heavy chain typically includes residues C226 or P230 relative to the carboxyl terminal, with numbering given according to the EU index, as in Kabat. The term "Fc" may refer to this region alone or in the context of an Fc polypeptide, as described below. As used herein, the term "Fc polypeptide" refers to a polypeptide comprising all or a portion of an Fc region. Fc polypeptides include antibodies, Fc fusions, isolated Fc regions, and Fc fragments.

In some embodiments, the antibodies are full-length. As used herein, the term "full-length antibody" means a structure that forms the natural biological form of an antibody, including variable and constant regions containing one or more modifications as described herein.

Alternatively, antibodies can be a variety of structures, including, but not limited to, antibody fragments, monoclonal antibodies, bispecific antibodies, minibodies, domain antibodies, synthetic antibodies (sometimes referred to herein as "antibody mimetics"), chimeric antibodies, humanized antibodies, antibody hybrids (sometimes referred to herein as "antibody conjugates"), and fragments of each of these, respectively.

Antibody fragments

In one embodiment, the antibody is an antibody fragment. Of particular interest are antibodies comprising Fc regions, hybrids of Fc and a heavy chain constant region (CH1-hinge-CH2-CH3), also including hybrids of the heavy chain constant region.

Specific antibody fragments include, but are not limited to (i) a Fab fragment comprising the VL, VH, CL, and CH1 domains, (ii) an Fd fragment comprising the VH and CH1 domains, (iii) an Fv fragment comprising the VL and VH domains of a single antibody; (iv) a dAb fragment (Ward et al., 1989, Nature 341:544-546, incorporated herein by reference in its entirety) containing a single variable region, (v) isolated CDR regions, (vi) F(ab')2 fragments, a bivalent fragment containing two linked Fab fragments (vii) single-chain Fv molecules (scFv) in which the VH domain and the VL domain are linked by a peptide linker that allows the two domains to associate to form an antigen-binding site (Bird et al., 1988, Science 242:423-426, Huston et al., 1988, Proc. Natl. Acad. Sci. U.S.A. 85:5879-5883, incorporated herein by reference in its entirety), (viii) a bispecific single-chain Fv (WO 03/11161, hereby incorporated by reference in their entireties) and (ix) "diabodies" or "triabdomes," multivalent or multispecific fragments produced by gene fusion (Tomlinson et al., 2000, Methods Enzymol. 326: 461-479; WO94/13804; Holliger et al., 1993, Proc. Natl. Acad. Sci. U.S.A. 90: 6444-6448, all incorporated by reference in their entireties). The antibody fragments can be modified. For example, the molecules can be stabilized by introducing disulfide bridges linking the VH and VL domains (Reiter et al., 1996, Nature Biotech. 14: 1239-1245, incorporated by reference in their entireties).

Chimeric and humanized antibodies

In some embodiments, the framework components may be a mixture of different species. Thus, if the protein is an antibody, such an antibody may be a chimeric antibody and/or a humanized antibody. In general, both the term "chimeric antibodies" and "humanized antibodies" refer to antibodies that contain combined regions of more than one species. For example, "chimeric antibodies" conventionally contain variable region(s) from a mouse (or rat in some cases) and constant region(s) from a human. The term "humanized antibodies" generally refers to antibodies of non-human origin that contain framework regions of the variable domain that have been replaced with sequences found in human antibodies. In general, in a humanized antibody, the entire antibody, except for the CDRs, is encoded by a polynucleotide of human origin or is identical to such an antibody, except for the CDRs. CDR regions, some or all of which are encoded by nucleic acids originating from a non-human organism, are grafted onto the beta-pleated sheet framework of a human antibody variable region to produce an antibody whose specificity is determined by the introduced CDRs. The production of such antibodies is described, for example, in WO 92/11018, Jones, 1986, Nature 321: 522-525, Verhoeyen et al., 1988, Science 239: 1534-1536, all of which are incorporated herein by reference in their entireties. "Back mutation" of selected acceptor framework residues to the corresponding donor residues is often necessary to restore the affinity lost in the original grafted construct (US 5,530,101; US 5,585,089; US 5,693,761; US 5,693,762; US 6,180,370; US 5,859,205; US 5,821,337; US 6,054,297; US 6,407,213, all incorporated herein by reference in their entireties). Optimally, the humanized antibody will also comprise at least a portion of an immunoglobulin constant region, typically a human immunoglobulin, and will thus comprise a human Fc region. Humanized antibodies can also be produced using mice with genetically engineered immune systems. Roque et al., 2004, Biotechnol. Prog. 20: 639-654, incorporated herein by reference in its entirety. Numerous technologies and methods for humanizing and reengineering non-human antibodies are well known in the art (See Tsurushita & Vasquez, 2004, Humanization of Monoclonal Antibodies, Molecular Biology of B Cells, 533-545, Elsevier Science (USA) and references cited therein, all of which are incorporated herein by reference in their entireties). Humanization methods include, but are not limited to, those described in Jones et al., 1986, Nature 321: 522-525; Riechmann et al., 1988; Nature 332: 323-329; Verhoeyen et al., 1988, Science, 239: 1534-1536; Queen et al., 1989, Proc Natl Acad Sci, USA 86: 10029-33; He et al., 1998, J. Immunol. 160: 1029-1035; Carter et al., 1992, Proc Natl Acad Sci USA 89: 4285-9, Presta et al., 1997, Cancer Res. 57(20): 4593-9; Gorman et al., 1991, Proc. Natl. Acad. Sci. USA 88: 4181-4185; O'Connor et al., 1998, Protein Eng 11: 321-8, all of which are incorporated herein by reference in their entireties. Humanization or other methods for reducing the immunogenicity of variable regions of non-human antibodies can include resurfacing methods, such as described in Roguska et al., 1994, Proc. Natl. Acad. Sci. USA 91: 969-973, which is incorporated herein by reference in its entirety. In one embodiment, the parent antibody has been affinity matured as is known in the art. Structure-based methods can be used for humanization and affinity maturation, such as described in USSN 11/004,590. Selection-based methods can be used for humanization and/or affinity maturation of antibody variable regions, including, but not limited to, those described in Wu et al., 1999, J. Mol. Biol. 294: 151-162; Baca et al., 1997, J. Biol. Chem. 272(16): 10678-10684; Rosok et al., 1996, J. Biol. Chem. 271(37): 22611-22618; Rader et al., 1998, Proc. Natl. Acad. Sci. USA 95: 8910-8915; Krauss et al., 2003, Protein Engineering 16(10): 753-759, all of which are incorporated herein by reference in their entireties. Other methods of humanization may involve transplantation of only portions of the CDRs, including but not limited to the methods described in USSN 09/810,510; Tan et al., 2002, J. Immunol. 169: 1119-1125; De Pascalis et al., 2002, J. Immunol. 169: 3076-3084, all of which are incorporated herein by reference in their entireties.

Bispecific antibodies

In one embodiment, the antibodies of the present invention comprise a multispecific antibody, and in particular a bispecific antibody, also sometimes referred to as a "diabody". These antibodies are antibodies that bind to two (or more) different antigens. Diabodies can be produced by a variety of methods known in the art (Holliger and Winter, 1993, Current Opinion Biotechnol. 4: 446-449, incorporated herein by reference in its entirety), such as chemically produced or from interspecies hybridomas.

Miniantibodies

In one embodiment, the antibody is a miniantibody. Miniantibodies are minimized antibody-like proteins comprising an scFv linked to a CH3 domain. Hu et al., 1996, Cancer Res. 56:3055-3061, incorporated herein by reference in its entirety. In some cases, the scFv may be linked to an Fc region and may include some or all of the hinge region.

Antibody hybrids

In one embodiment, the antibodies of the present invention are antibody fusion proteins (sometimes referred to herein as "antibody conjugates"). One type of antibody fusion comprises Fc fusions, which join an Fc region to a conjugation partner. As used herein, the term "Fc fusion" refers to a protein in which one or more polypeptides are operably linked to an Fc region. As used herein, the term "Fc fusion" is synonymous with the terms "immunoadhesin," "Ig fusion," "Ig chimera," and "receptor globulin" (sometimes with impurities) used in the art (Chamow et al., 1996, Trends Biotechnol 14:52-60; Ashkenazi et al., 1997, Curr Opin Immunol 9:195-200, both of which are incorporated herein by reference in their entireties). An Fc fusion combines an Fc region of an immunoglobulin with a fusion partner, which can generally be any protein or small molecule. Indeed, any protein or small molecule can be linked to an Fc to form an Fc fusion. Protein fusion partners can include, but are not limited to, a variable region of any antibody, a target-binding receptor region, an adhesion molecule, a ligand, an enzyme, a cytokine, a chemokine, or some other protein or protein domain. Small molecule fusion partners can include any therapeutic agent that directs the Fc fusion to a target for therapeutic effect. Such targets can be any molecule, preferably an extracellular receptor, that is involved in a disease. Thus, IgG variants can be linked to one or more fusion partners. In an alternative embodiment, an IgG variant is conjugated or operably linked to another therapeutic compound. The therapeutic compound may be a cytotoxic agent, a chemotherapeutic agent, a toxin, a radioisotope, a cytokine, or other therapeutically active agent. IgG may be linked to one of a variety of non-protein polymers, such as polyethylene glycol, polypropylene glycol, polyoxyalkylenes, or copolymers of polyethylene glycol and polypropylene glycol.

In addition to Fc hybrids, antibody hybrids include a hybrid of a heavy chain constant region with one or more fusion partners (including the variable region of any antibody), while other antibody hybrids are essentially or completely full-length antibodies with fusion partners. In one embodiment, the role of the fusion partner is to mediate binding to the target and thus is (and in fact can be) functionally similar to the variable regions of an antibody. In fact, any protein or small molecule can be fused to an Fc hybrid to form an Fc hybrid (or antibody hybrid). Protein fusion partners can include, but are not limited to, a target-binding receptor region, an adhesion molecule, a ligand, an enzyme, a cytokine, a chemokine, or some other protein or protein domain. Small molecule fusion partners may include any therapeutic agent that directs the Fc fusion to a target for therapeutic effect. Such targets may be any molecule, preferably an extracellular receptor, that is involved in the disease.

The conjugation "partner" may be proteinaceous or non-proteinaceous; the latter is typically prepared using functional groups of the antibody and the conjugation "partner". For example, linkers are known in the art; for example, homo- or heterobifunctional linkers are well known (see the 1994 Pierce Chemical Company catalog, section on linkers, pages 155-200, which is incorporated herein by reference).

Suitable conjugates include, but are not limited to, labels as described below, drugs, and cytotoxic agents, including, but not limited to, cytotoxic drugs (e.g., chemotherapeutic agents) or toxins or active fragments of such toxins. Suitable toxins and corresponding fragments thereof include diphtheria A chain, exotoxin A chain, ricin A chain, abrin A chain, curcin, crotin, phenomycin, enomycin, and the like. Cytotoxic agents also include radiochemicals produced by conjugating radioisotopes to antibodies or by linking a radionuclide to a chelating agent that has been covalently attached to an antibody. In further embodiments, calicheamicin, auristatin, geldanamycin, maytansine, and duocarmycin and analogs are used; for the latter, see U.S. Patent No. 1,388,100. 2003/0050331A1, which is incorporated herein by reference in its entirety.

Covalent modifications of antibodies

Covalent modifications of antibodies are included within the scope of the present invention and are usually, but not always, performed post-translationally. For example, some types of covalent modifications of an antibody are introduced into the molecule by reacting specific amino acid residues of the antibody with an organic derivatizing agent that can react with selected side chains or N- or C-terminal residues.

Cysteinyl residues are most often reacted with α-haloacetates (and corresponding amines) such as chloroacetic acid or chloroacetamide to give carboxymethyl or carboxyamidomethyl derivatives. Cysteinyl residues can also be derivatized by reaction with bromotrifluoroacetone, α-bromo-β-(5-imidozyl)propionic acid, chloroacetyl phosphate, N-alkylmaleimides, 3-nitro-2-pyridyl disulfide, methyl 2-pyridyl disulfide, p-chloromercuric benzoate, 2-chloromercuric-4-nitrophenol or chloro-7-nitrobenzo-2-oxa-1,3-diazole, etc.

Histidyl residues are derivatized by reaction with diethyl pyrocarbonate at pH 5.5-7.0, as this agent has relative specificity for the histidyl side chain. Para-bromophenacyl bromide is also suitable; this reaction is preferably carried out in 0.1 M sodium cacodylate at pH 6.0.

Lysinyl and amino-terminal residues are reacted with succinic acid or other carboxylic acid anhydrides. Derivatization with these agents results in a charge change on the lysinyl residues. Other suitable reagents for derivatization of alpha-amino group-containing residues include imidoesters such as methyl picolinimidate; pyridoxal phosphate; pyridoxal; chloroborohydride; trinitrobenzenesulfonic acid; O-methylisourea; 2,4-pentanedione, and transaminase-catalyzed reaction with glyoxylate.

Arginine residues are modified by reaction with one or more standard reagents, including phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin. Derivatization of arginine residues requires an alkaline reaction due to the high pKa of the guanidine functional group. These reagents can also react with lysine groups as well as the epsilon-amino group of arginine.

Specific modification of tyrosyl residues can be achieved with particular interest by introducing spectral labels into tyrosyl residues by reaction with aromatic diazonium compounds or tetranitromethane. N-acetylimidazole and tetranitromethane are most commonly used to form O-acetyltyrosyl compounds and 3-nitro derivatives, respectively. Tyrosyl residues are iodinated using 125I or 131I to produce labeled proteins for use in radioimmunoassay, the chloramine method described above being suitable.

The side carboxyl groups (aspartyl or glutamyl) are selectively modified by reaction with carbodiimides (R'--N=C=N--R'), where R and R' are optionally different alkyl groups, such as 1-cyclohexyl-3-(2-morpholinyl-(4-ethyl) carbodiimide or 1-ethyl-3-(4-azonia-4,4-dimethylpentyl) carbodiimide. In addition, aspartyl and glutamyl residues are converted to asparaginyl and glutaminyl residues by reaction with ammonium ions.

Derivatization with bifunctional agents is useful for crosslinking antibodies to a water-insoluble support matrix or surface for use in a variety of methods in addition to those described below. Commonly used crosslinking agents include, for example, 1,1-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters such as 4-azidosalicylic acid esters, homobifunctional imidoesters including disuccinimidyl esters such as 3,3'-dithiobis(succinimidyl propionate), and bifunctional maleimides such as bis-N-maleimido-1,8-octane. Derivatizing agents such as methyl 3-[(p-azidophenyl)dithio]propioimidate form photoactivatable intermediates that can form crosslinks in the presence of light. Alternatively, reactive water-insoluble matrices such as cyanogen bromide-activated carbohydrates and reactive substrates described in U.S. Patents 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537 and 4,330,440, all of which are incorporated herein by reference in their entireties, are used to immobilize proteins.

Glutaminyl and asparaginyl residues are repeatedly deamidated to form the corresponding glutamyl and aspartyl residues, respectively. Alternatively, these residues are deamidated in a moderately acidic medium. Any form of these residues is within the scope of the present invention. Other modifications include hydroxylation of proline and lysine, phosphorylation of the hydroxyl groups of seryl or threonyl residues, methylation of the α-amino groups of the side chains of lysine, arginine, and histidine (T. E. Creighton, Proteins: Structure and Molecular Properties, W. H. Freeman & Co., San Francisco, pp. 79-86 [1983], incorporated herein by reference in its entirety), acetylation of the N-terminal amine, and amidation of any C-terminal carboxyl group.

Glycosylation

Another type of covalent modification is glycosylation. In another embodiment, the IgG variants described herein may be modified to comprise one or more engineered glycoforms. As used herein, the term "engineered glycoform" refers to a carbohydrate composition covalently attached to an IgG, wherein the carbohydrate composition is chemically different from the composition of the original IgG. Engineered glycoforms may be useful for a variety of purposes, including, but not limited to, enhancing or reducing effector function. The engineered glycoforms can be prepared by a variety of methods known in the art (Umaña et al., 1999, Nat Biotechnol 17: 176-180; Davies et al., 2001, Biotechnol Bioeng 74: 288-294; Shields et al., 2002, J Biol Chem 277: 26733-26740; Shinkawa et al., 2003, J Biol Chem 278: 3466-3473; US 6,602,684; USSN 10/277,370; USSN 10/113,929; PCT WO 00/61739A1; PCT WO 01/29246A1; PCT WO 02/31140A1; PCT WO 02/30954A1, all of which are incorporated herein by reference in their entireties; (Potelligent® technology [Biowa, Inc., Princeton, NJ]; GlycoMAb® glycosylation engineering technology [Glycart Biotechnology AG, Zurich, Switzerland]). Many of these technologies are based on the control of the level of fucosylated and/or fissile oligosaccharides covalently attached to the Fc region, such as by expressing IgG in various organisms or cell lines engineered or otherwise obtained (e.g., CHO Lec-13 cells or rat YB2/0 hybridoma cells), by regulating enzymes involved in the glycosylation pathway (e.g., FUT8 [α1,6-fucosyltransferase] and/or β1-4-N-acetylglucosaminyltransferase III [GnTIII]), or by modifying the carbohydrate(s) after expression of the IgG. The term "engineered glycoform" generally refers to another carbohydrate or oligosaccharide; thus, an IgG variant, such as an antibody or Fc hybrid, may comprise an engineered glycoform. Alternatively, the term "engineered glycoform" may refer to an IgG variant comprising another carbohydrate or oligosaccharide. As is known in the art, glycosylation patterns may depend on both the protein sequence (e.g., the presence or absence of particular glycosylation amino acid residues, as noted below) and the host cell or organism in which the protein is produced. Specific expression systems are noted below.

Glycosylation of peptides is typically either N-linked or O-linked. N-linked refers to the attachment of a carbohydrate moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine-X-threonine, where X is any amino acid except proline, are the recognition sequences for the enzymatic attachment of a carbohydrate moiety to the asparagine side chain. Thus, the presence of either of these tripeptide sequences in a polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of one of the sugars, N-acetylgalactosamine, galactose, or xylose, to a hydroxyamino acid, most commonly serine or threonine, although 5-hydroxyproline or 5-hydroxylysine may also be used.

The addition of glycosylation sites to the antibody is conveniently accomplished by altering the amino acid sequence so that it contains one or more of the above-described tripeptide sequences (for N-linked glycosylation sites). Said alteration can also be accomplished by adding or replacing one or more serine or threonine residues to the original sequence (for O-linked glycosylation sites). For convenience, the amino acid sequence of the antibody is preferably altered by changes at the DNA level, in particular by mutating the DNA encoding the target polypeptide at preselected bases so that codons are formed that will translate into the desired amino acids.

Another method for increasing the number of carbohydrate moieties of an antibody is the chemical or enzymatic coupling of glycosides to the protein. These procedures have the advantage that they do not require the formation of the protein in a host cell that is capable of glycosylation with respect to N- and O-linked glycosylation. Depending on the coupling method used, the sugar(s) can be attached to (a) arginine and histidine, (b) free carboxyl groups, (c) free sulfhydryl groups such as those of cysteine, (d) free hydroxyl groups such as those of serine, threonine, or hydroxyproline, (e) aromatic residues such as those of phenylalanine, tyrosine, or tryptophan, or (f) the amide group of glutamine. Said methods are described in WO 87/05330 and in Aplin and Wriston, 1981, CRC Crit. Rev. Biochem., pp. 259-306, both of which are incorporated herein by reference in their entirety.

Removal of carbohydrate moieties present in the parent antibody can be accomplished chemically or enzymatically. Chemical deglycosylation requires exposure of the protein to the compound trifluoromethanesulfonic acid or an equivalent compound. This exposure results in the degradation of most or all of the sugars except the linking sugar (N-acetyl glucosamine or N-acetyl galactosamine), while the polypeptide remains intact. Chemical deglycosylation is described by Hakimuddin et al., 1987, Arch. Biochem. Biophys. 259:52 and Edge et al., 1981, Anal. Biochem. 118:131, both of which are incorporated herein by reference in their entireties. Enzymatic cleavage of carbohydrate moieties in polypeptides can be accomplished using a variety of endo- and exoglycosidases, as described by Thotakura et al., 1987, Meth. Enzymol. 138:350, which is incorporated herein by reference in its entirety. Glycosylation at potential glycosylation sites can be prevented by using the compound, tunicamycin, as described by Duskin et al., 1982, J. Biol. Chem. 257:3105, which is incorporated herein by reference in its entirety. Tunicamycin blocks the formation of protein-N-glycosidic bonds.

Another type of covalent modification of an antibody involves linking the antibody to various non-proteinaceous polymers, including but not limited to various polyols such as polyethylene glycol, polypropylene glycol, or polyoxyalkylenes, in a manner described, for example, in the 2005-2006 Nektar Therapeutics PEG Catalog (available on the Nektar website), U.S. Patents 4,640,835; 4,496,689; 4,301,144; 4,670,417; 4,791,192, or 4,179,337, all of which are incorporated herein by reference in their entireties. Additionally, as is known in the art, amino acid substitutions can be made at various positions within the antibody to facilitate the addition of polymers such as PEG. See, for example, U.S. Publication No. 2005/0114037A1, which is incorporated herein by reference in its entirety.

Labeled antibodies

In some embodiments, covalent modification of the antibodies of the present invention includes the addition of one or more labels. In some cases, these are considered hybrids of antibodies. The term "labeling group" means any detectable label. In some embodiments, the labeling group is attached to the antibody via spacer groups of varying lengths to reduce potential steric hindrance. Various methods for labeling proteins are known in the art and can be used in the practice of the present invention.

In general, labels fall into a variety of classes depending on the assay in which they are to be detected: a) isotopic labels, which may be radioactive or heavy isotopes; b) magnetic labels (e.g., magnetic particles); c) redox active moieties; d) optical dyes; enzymatic groups (e.g., horseradish peroxidase, β-galactosidase, luciferase, alkaline phosphatase); e) biotinylated groups; and f) predetermined polypeptide epitopes recognized by a secondary reporter (e.g., leucine zipper paired sequences, binding sites for secondary antibodies, metal binding domains, epitope tags, etc.). In some embodiments, the labeling group is linked to the antibody via spacer groups of varying lengths to reduce potential steric hindrance. Various methods for labeling proteins are known in the art and may be used in the practice of the present invention.

Specific labels include optical dyes including, but not limited to, chromophores, phosphors, and fluorophores, the latter being specific in many cases. Fluorophores may be either "small molecule" fluorescent agents or protein fluorescent agents.

The term "fluorescent label" refers to any molecule that can be detected by its inherent fluorescent properties. Suitable fluorescent labels include, but are not limited to, fluorescein, rhodamine, tetramethylrhodamine, eosin, erythrosine, coumarin, methylcoumarins, pyrene, malachite green, stilbene, Lucifer yellow, Cascade Blue, Texas red, IAEDANS, EDANS, BODIPY FL, LC Red 640, Cy 5, Cy 5.5, LC Red 705, Oregon green, Alexa-Fluor dyes (Alexa Fluor 350, Alexa Fluor 430, Alexa Fluor 488, Alexa Fluor 546, Alexa Fluor 568, Alexa Fluor 594, Alexa Fluor 633, Alexa Fluor 660, Alexa Fluor 680), Cascade Blue, Cascade Yellow, and R-phycoerythrin (PE) (Molecular Probes, Eugene, OR). FITC, Rhodamine, and Texas Red (Pierce, Rockford, IL), Cy5, Cy5.5, Cy7 (Amersham Life Science, Pittsburgh, PA). Suitable optical dyes, including fluorophores, are described in Molecular Probes Handbook by Richard P. Haugland, which is incorporated herein by reference in its entirety.

Suitable protein fluorescent labels also include, but are not limited to, green fluorescent protein, including the Renilla, Ptilosarcus, or Aequorea varieties of green fluorescent protein (Chalfie et al., 1994, Science 263: 802-805), EGFP (Clontech Laboratories, Inc., Genbank accession number U55762), blue fluorescent protein (BFP, Quantum Biotechnologies, Inc. 1801 de Maisonneuve Blvd. West, 8th Floor, Montreal, Quebec, Canada H3H 1J9; Stauber, 1998, Biotechniques 24: 462-471; Heim et al., 1996, Curr. Biol. 6: 178-182), “enhanced” yellow fluorescent protein (EYFP, Clontech Laboratories, Inc.), luciferase (Ichiki et al., 1993, J. Immunol. 150: 5408-5417), β-galactosidase (Nolan et al., 1988, Proc. Natl. Acad. Sci. U.S.A. 85: 2603-2607) and Renilla (WO92/15673, WO95/07463, WO98/14605, WO98/26277, WO99/49019, U.S. Patents 5,292,658, 5,418,155, 5,683,888, 5,741,668, 5,777,079, 5,804,387, 5,874,304, 5,876,995, 5925558). All of the above references in this paragraph are expressly incorporated into this description by reference.

IgG variants

In one embodiment, the present invention provides variant IgG proteins. The IgG variants at least comprise an antibody fragment comprising a CH2-CH3 region of a heavy chain. In addition, suitable IgG variants include Fc domains (e.g., including a lower hinge region), as well as IgG variants comprising a heavy chain constant region (CH1-hinge-CH2-CH3), which are also suitable according to the present invention, all of which can be fused to fusion partners.

The IgG variant comprises one or more amino acid modifications compared to the parent IgG polypeptide, in some cases compared to wild-type IgG. The IgG variant may have one or more optimized properties. The IgG variant differs in its amino acid sequence from the parent IgG by having at least one amino acid modification. Thus, the IgG variants comprise at least one amino acid modification compared to the parent IgG. Alternatively, the IgG variants may comprise several amino acid modifications compared to the parent IgG, such as from about one to fifty amino acid modifications, preferably from about one to ten amino acid modifications, and most preferably from about one to about five amino acid modifications compared to the parent IgG.

Thus, the IgG variant sequences and the parent Fc polypeptide sequences are substantially homologous. For example, the variant sequences of the Fc variant herein will have about 80% homology with the variant sequence of the parent IgG, preferably at least about 90% homology, and most preferably at least about 95% homology. The modifications can be genetically engineered using molecular biology, or can be enzymatically or chemically produced.

Antigens as targets for antibodies

In fact, any antigen can be targeted by IgG variants, including but not limited to proteins, subunits, domains, motifs, and/or epitopes belonging to the following list of target antigens, which includes both soluble factors such as cytokines and membrane-bound factors including transmembrane receptors: 17-IA, 4-1BB, 4Dc, 6-keto-PGF1a, 8-iso-PGF2a, 8-oxo-dG, Adenosine Receptor A1, A33, ACE, ACE-2, Activin, Activin A, Activin AB, Activin B, Activin C, Activin RIA, Activin RIA ALK-2, Activin RIB ALK-4, Activin RIIA, Activin RIIB, ADAM, ADAM10, ADAM12, ADAM15, ADAM17/TACE, ADAM8, ADAM9, ADAMTS, ADAMTS4, ADAMTS5, Addressins, aFGF, ALCAM, ALK, ALK-1, ALK-7, alpha-1-antitrypsin, alpha-V/beta-1 antagonist, ANG, Ang, APAF-1, APE, APJ, APP, APRIL, AR, ARC, ART, Artemin, anti-Id, ASPARTIC, atrial natriuretic factor, integrin av/b3, Axl, b2M, B7-1, B7-2, B7-H, B-lymphocyte stimulator (BlyS), BACE, BACE-1, Bad, BAFF, BAFF-R, Bag-1, BAK, Bax, BCA-1, BCAM, Bcl, BCMA, BDNF, b-ECGF, bFGF, BID, Bik, BIM, BLC, BL-CAM, BLK, BMP, BMP-2 BMP-2a, osteogenin BMP-3, BMP-4 BMP-2b, BMP-5, BMP-6 Vgr-1, BMP-7 (OP-1), BMP-8 (BMP-8a, OP-2), BMPR, BMPR-IA (ALK-3), BMPR-IB (ALK-6), BRK-2, RPK-1, BMPR-II (BRK-3), BMPs, b-NGF, BOK, Bombesin, bone-derived neurotrophic factor, BPDE, BPDE-DNA, BTC, complementary factor 3 (C3), C3a, C4, C5, C5a, C10, CA125, CAD-8, Calcitonin, cAMP, carcinoembryonic antigen (CEA), carcinoma antigen, Cathepsin A, Cathepsin B, Cathepsin C/DPPI, Cathepsin D, Cathepsin E, Cathepsin H, Cathepsin L, Cathepsin O, Cathepsin S, Cathepsin V, Cathepsin X/Z/P, CBL, CCI, CCK2, CCL, CCL1, CCL11, CCL12, CCL13, CCL14, CCL15, CCL16, CCL17, CCL18, CCL19, CCL2, CCL20, CCL21, CCL22, CCL23, CCL24, CCL25, CCL26, CCL27, CCL28, CCL3, CCL4, CCL5, CCL6, CCL7, CCL8, CCL9/10, CCR, CCR1, CCR10, CCR10, CCR2, CCR3, CCR4, CCR5, CCR6, CCR7, CCR8, CCR9, CD1, CD2, CD3, CD3E, CD4, CD5, CD6, CD7, CD8, CD10, CD11a, CD11b, CD11c, CD13, CD14, CD15, CD16, CD18, CD19, CD20, CD21, CD22, CD23, CD25, CD27L, CD28, CD29, CD30, CD30L, CD32, CD33 (p67 proteins), CD34, CD38, CD40, CD40L, CD44, CD45, CD46, CD49a, CD52, CD54, CD55, CD56, CD61, CD64, CD66e, CD74, CD80 (B7-1), CD89, CD95, CD123, CD137, CD138, CD140a, CD146, CD147, CD148, CD152, CD164, CEACAM5, CFTR, cGMP, CINC, Clostridium botulinum toxin, Clostridium perfringens toxin, CKb8-1, CLC, CMV, CMV UL, CNTF, CNTN-1, COX, C-Ret, CRG-2, CT-1, CTACK, CTGF, CTLA-4, CX3CL1, CX3CR1, CXCL, CXCL1, CXCL2, CXCL3, CXCL4, CXCL5, CXCL6, CXCL7, CXCL8, CXCL9, CXCL10, CXCL11, CXCL12, CXCL13, CXCL14, CXCL15, CXCL16, CXCR, CXCR1, CXCR2, CXCR3, CXCR4, CXCR5, CXCR6, cytokeratin tumor-associated antigen, DAN, DCC, DcR3, DC-SIGN, decay-accelerating factor, des(1-3)-IGF-I (brain IGF-1), Dhh, digoxin, DNAM-1, DNase, Dpp, DPPIV/CD26, Dtk, ECAD, EDA, EDA-A1, EDA-A2, EDAR, EGF, EGFR (ErbB-1), EMA, EMMPRIN, ENA, endothelin receptor, Enkephalinase, eNOS, Eot, eotaxin1, EpCAM, Ephrin B2/EphB4, EPO, ERCC, E-selectin, ET-1, Factor IIa, Factor VII, Factor VIIIc, Factor IX, fibroblast activator protein (FAP), Fas, FcR1, FEN-1, Ferritin, FGF, FGF-19, FGF-2, FGF3, FGF-8, FGFR, FGFR-3, Fibrin, FL, FLIP, Flt-3, Flt-4, Follicle-stimulating hormone, Fractalkine, FZD1, FZD2, FZD3, FZD4, FZD5, FZD6, FZD7, FZD8, FZD9, FZD10, G250, Gas 6, GCP-2, GCSF, GD2, GD3, GDF, GDF-1, GDF-3 (Vgr-2), GDF-5 (BMP-14, CDMP-1), GDF-6 (BMP-13, CDMP-2), GDF-7 (BMP-12, CDMP-3), GDF-8 (Myostatin), GDF-9, GDF-15 (MIC-1), GDNF, GDNF, GFAP, GFRa-1, GFR-alpha1, GFR-alpha2, GFR-alpha3, GITR, Glucagon, Glut 4, glycoprotein IIb/IIIa (GP IIb/IIIa), GM-CSF, gp130, gp72, GRO, growth hormone-releasing factor, Hapten (NP-cap or NIP-cap), HB-EGF, HCC, HCMV envelope glycoprotein gB, HCMV envelope glycoprotein gH, HCMV UL, hematopoietic growth factor (HGF), Hep B gp120, heparanase, Her2, Her2/neu (ErbB-2), Her3 (ErbB-3), Her4 (ErbB-4), herpes simplex virus (HSV) glycoprotein gB, HSV glycoprotein gD, HGFA, high molecular weight melanoma-associated antigen (HMW-MAA), HIV gp120, V3 loop of gp 120 HIV IIIB, HLA, HLA-DR, HM1.24, HMFG PEM, HRG, Hrk, human cardiac myosin, human cytomegalovirus (HCMV), human growth hormone (HGH), HVEM, I-309, IAP, ICAM, ICAM-1, ICAM-3, ICE, ICOS, IFNg, Ig, IgA receptor, IgE, IGF, IGF binding proteins, IGF-1R, IGFBP, IGF-I, IGF-II, IL, IL-1, IL-1R, IL-2, IL-2R, IL-4, IL-4R, IL-5, IL-5R, IL-6, IL-6R, IL-8, IL-9, IL-10, IL-12, IL-13, IL-15, IL-18, IL-18R, IL-23, interferon (INF)-alpha, INF-beta, INF-gamma, Inhibin, iNOS, insulin A chain, insulin B chain, insulin-like growth factor 1, integrin alpha2, integrin alpha3, integrin alpha4, integrin alpha4/beta, integrin alpha4/beta7, integrin alpha5 (alphaV), integrin alpha5/beta1, integrin alpha5/beta3, integrin alpha6, integrin beta1, integrin beta2, interferon gamma, IP-10, I-TAC, JE, Kallikrein 2, Kallikrein 5, Kallikrein 6, Kallikrein 11, Kallikrein 12, Kallikrein 14, Kallikrein 15, Kallikrein L1, Kallikrein L2, Kallikrein L3, Kallikrein L4, KC, KDR, keratinocyte growth factor (KGF), laminin 5, LAMP, LAP, LAP (TGF-1), Latent TGF-1, Latent TGF-1 bp1, LBP, LDGF, LECT2, Lefty, Lewis antigen-Y, Lewis-related antigen-Y, LFA-1, LFA-3, Lfo, LIF, LIGHT, lipoproteins, LIX, LKN, Lptn, L-Selectin, LT-a, LT-b, LTB4, LTBP-1, lung surfactant, luteinizing hormone, lymphotoxin-beta receptor, Mac-1, MAdCAM, MAG, MAP2, MARC, MCAM, MCAM, MCK-2, MCP, M-CSF, MDC, Mer, metalloproteases, MGDF receptor, MGMT, MHC (HLA-DR), MIF, MIG, MIP, MIP-1-alpha, MK, MMAC1, MMP, MMP-1, MMP-10, MMP-11, MMP-12, MMP-13, MMP-14, MMP-15, MMP-2, MMP-24, MMP-3, MMP-7, MMP-8, MMP-9, MPIF, Mpo, MSK, MSP, mucin (Muc1), MUC18, Müllerian inhibitory substance, Mug, MuSK, NAIP, NAP, NCAD, N-Cadherin, NCA 90, NCAM, NCAM, Neprilysin, Neurotrophin-3,-4, or -6, Neuroturin, Nerve growth factor (NGF), NGFR, NGF-beta, nNOS, NO, NOS, Npn, NRG-3, NT, NTN, OB, OGG1, OPG, OPN, OSM, OX40L, OX40R, p150, p95, PADPr, Parathyroid hormone, PARC, PARP, PBR, PBSF, PCAD, P-Cadherin, PCNA, PDGF, PDGF, PDK-1, PECAM, PEM, PF4, PGE, PGF, PGI2, PGJ2, PIN, PLA2, Placental alkaline phosphatase (PLAP), PlGF, PLP, PP14, Proinsulin, Prorelaxin, Protein C, PS, PSA, PSCA, Prostate-specific membrane antigen (PSMA), PTEN, PTHrp, Ptk, PTN, R51, RANK, RANKL, RANTES, RANTES, Relaxin A-chain, Relaxin B-chain Relaxin, Renin, Respiratory Syncytial Virus (RSV) F, RSV Fgp, Ret, Rheumatoid Factors, RLIP76, RPA2, RSK, S100, SCF/KL, SDF-1, SERINE, Serum Albumin, sFRP-3, Shh, SIGIRR, SK-1, SLAM, SLPI, SMAC, SMDF, SMOH, SOD, SPARC, Stat, STEAP, STEAP-II, TACE, TACI, TAG-72 (tumor-associated glycoprotein-72), TARC, TCA-3, T-Cell Receptors (e.g., T-Cell Receptor Alpha/Beta), TdT, TECK, TEM1, TEM5, TEM7, TEM8, TERT, Testicular PLAP-like Alkaline Phosphatase, TfR, TGF, TGF-alpha, TGF-beta, polyspecific TGF-beta, TGF-beta RI (ALK-5), TGF-beta RII, TGF-beta RIIb, TGF-beta RIII, TGF-beta, TGF-beta 2, TGF-beta3, TGF-beta4, TGF-beta5, Thrombin, Thymus Ck-1, Thyroid-stimulating hormone, Tie, TIMP, TIQ, Tissue factor, TMEFF2, Tmpo, TMPRSS2, TNF, TNF-alpha, TNF-alpha beta, TNF-beta2, TNFc, TNF-RI, TNF-RII, TNFRSF10A (TRAIL R1 Apo-2, DR4), TNFRSF10B (TRAIL R2 DR5, KILLER, TRICK-2A, TRICK-B), TNFRSF10C (TRAIL R3 DcR1, LIT, TRID), TNFRSF10D (TRAIL R4 DcR2, TRUNDD), TNFRSF11A (RANK ODF R, TRANCE R), TNFRSF11B (OPG OCIF, TR1), TNFRSF12 (TWEAK R FN14), TNFRSF13B (TACI), TNFRSF13C (BAFF R), TNFRSF14 (HVEM ATAR, HveA, LIGHT R, TR2), TNFRSF16 (NGFR p75NTR), TNFRSF17 (BCMA), TNFRSF18 (GITR AITR), TNFRSF19 (TROY TAJ, TRADE), TNFRSF19L (RELT), TNFRSF1A (TNF RI CD120a, p55-60), TNFRSF1B (TNF RII CD120b, p75-80), TNFRSF26 (TNFRH3), TNFRSF3 (LTbR TNF RIII, TNFC R), TNFRSF4 (OX40 ACT35, TXGP1 R), TNFRSF5 (CD40 p50), TNFRSF6 (Fas Apo-1, APT1, CD95), TNFRSF6B (DcR3 M68, TR6), TNFRSF7 (CD27), TNFRSF8 (CD30), TNFRSF9 (4-1BB CD137, ILA), TNFRSF21 (DR6), TNFRSF22 (DcTRAIL R2 TNFRH2), TNFRST23 (DcTRAIL R1 TNFRH1), TNFRSF25 (DR3 Apo-3, LARD, TR-3, TRAMP, WSL-1), TNFSF10 (TRAIL Apo-2 Ligand, TL2), TNFSF11 (TRANCE/RANK Ligand ODF, OPG Ligand), TNFSF12 (TWEAK Apo-3 Ligand, DR3 Ligand), TNFSF13 (APRIL TALL2), TNFSF13B (BAFF BLYS, TALL1, THANK, TNFSF20), TNFSF14 (LIGHT HVEM Ligand, LTg), TNFSF15 (TL1A/VEGI), TNFSF18 (GITR Ligand AITR Ligand, TL6), TNFSF1A (TNF-a Conectin, DIF, TNFSF2), TNFSF1B (TNF-b LTa, TNFSF1), TNFSF3 (LTb TNFC, p33), TNFSF4 (OX40 Ligand gp34, TXGP1), TNFSF5 (CD40 Ligand CD154, gp39, HIGM1, IMD3, TRAP), TNFSF6 (Fas Ligand Apo-1 Ligand, APT1 Ligand), TNFSF7 (CD27 Ligand CD70), TNFSF8 (CD30 Ligand CD153), TNFSF9 (4-1BB Ligand CD137 Ligand), TP-1, t-PA, Tpo, TRAIL, TRAIL R, TRAIL-R1, TRAIL-R2, TRANCE, transferrin receptor, TRF, Trk, TROP-2, TSG, TSLP, tumor-associated antigen CA 125, tumor-associated antigen expressing Lews-Y-related carbohydrate, TWEAK, TXB2, Ung, uPAR, uPAR-1, Urokinase, VCAM, VCAM-1, VECAD, VE-Cadherin, VE-cadherin-2, VEFGR-1 (flt-1), VEGF, VEGFR, VEGFR-3 (flt-4), VEGI, VIM, Viral antigens, VLA, VLA-1, VLA-4, VNR integrin, von Willebrand factor, WIF-1, WNT1, WNT2, WNT2B/13, WNT3, WNT3A, WNT4, WNT5A, WNT5B, WNT6, WNT7A, WNT7B, WNT8A, WNT8B, WNT9A, WNT9A, WNT9B, WNT10A, WNT10B, WNT11, WNT16, XCL1, XCL2, XCR1, XCR1, XEDAR, XIAP, XPD and hormone and growth factor receptors.

It will be apparent to one skilled in the art that the above list of targets refers not only to specific proteins and biomolecules, but also to a biochemical pathway or pathways that include them. For example, when referring to CTLA-4 as a target antigen, it is intended that the ligands and receptors that comprise the T cell co-stimulatory pathway, including CTLA-4, B7-1, B7-2, CD28, and any other uncharacterized ligands or receptors that bind to these proteins, also represent targets. Thus, as used herein, the term "target" refers not only to a specific biomolecule, but also to a set of proteins that interact with said target and to members of a biochemical pathway to which said target belongs. It will also be apparent to those skilled in the art that any of the above-mentioned target antigens, ligands or receptors that bind thereto, or other participants in the relevant biochemical pathway, can be operably linked to the Fc variants of the present invention to form an Fc hybrid. Thus, for example, an Fc hybrid that targets EGFR (epidermal growth factor receptor) can be created by operably linking an Fc variant to EGF, TGF-b, or any other ligand, tested or untested, that binds EGFR. Accordingly, an Fc variant of the present invention can be operably linked to EGFR to form an Fc hybrid that binds EGF, TGF-b, or any other ligand, tested or untested, that binds EGFR. Thus, in reality, any polypeptide, whether a ligand, a receptor, or some other protein or protein domain, including but not limited to the above-mentioned targets and proteins comprising the corresponding biochemical pathways, can be operably linked to the Fc variants of the present invention to form an Fc hybrid.

The choice of an appropriate antigen depends on the desired application. Anticancer therapy requires a target whose expression is restricted to cancer cells. Some targets that have proven particularly amenable to antibody therapy are those that have signaling functions. Other therapeutic antibodies act by blocking receptor signaling by inhibiting binding between the receptor and its cognate ligand. Another mechanism of action for therapeutic antibodies is to cause downregulation of the receptor. Other antibodies do not act by signaling through their target antigen. In some cases, antibodies directed against infectious agents are used.

In one embodiment, the Fc variants of the present invention are included in an antibody to a cytokine. Alternatively, the Fc variants are fused or conjugated to a cytokine. As used herein, the term "cytokine" is a general term for proteins released by one cell population that act on another cell as intercellular mediators. For example, as described in Penichet et al., 2001, J Immunol Methods 248:91-101, which is expressly incorporated herein by reference, cytokines can be coupled to an antibody to provide a variety of desired properties. Examples of such cytokines include lymphokines, monokines, and general polypeptide hormones. Cytokines include growth hormone such as human growth hormone; N-methionyl human growth hormone and bovine growth hormone; parathyroid hormone; thyroxine, insulin; proinsulin; relaxin; prorelaxin; Glycoprotein hormones such as follicle-stimulating hormone (FSH), thyroid-stimulating hormone (TSH), and luteinizing hormone (LH); liver growth factor; fibroblast growth factor; prolactin; placental lactogen; tumor necrosis factor-alpha and -beta; Müller inhibitory substance; mouse gonadotropin-associated peptide; inhibin; activin; vascular endothelial growth factor; integrin; thrombopoietin (TPO); nerve growth factors such as NGF-beta; platelet-derived growth factor; transforming growth factors (TGFs) such as TGF-alpha and TGF-beta; insulin-like growth factor-I and -II; erythropoietin (EPO); osteoinductive factors; interferons such as interferon-alpha, -beta, and -gamma; Colony stimulating factors (CSF) such as macrophage-CSF (M-CSF); granulocyte-macrophage-CSF (GM-CSF) and granulocyte-CSF (G-CSF); interleukins (IL) such as IL-1, IL-1alpha, IL-2, IL-g, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-11, IL-12, IL-15; tumor necrosis factor such as TNF-alpha or TNF-beta; C5a; and other polypeptide factors including LIF and kit ligand (KL). As used herein, the term "cytokine" includes proteins from natural sources or from recombinant cell culture and biologically active equivalents of cytokines containing native sequences.

Cytokines and soluble targets such as members of the TNF superfamily are preferred targets for use with the variants of the present invention. For example, anti-VEGF, anti-CTLA-4 and anti-TNF antibodies or fragments of these antibodies are antibodies that are particularly suitable for use with Fc variants that enhance FcRn binding. Drugs against these targets are often used in the treatment of autoimmune diseases, requiring multiple injections over long periods of time. Therefore, the longer serum half-life and less frequent treatment courses afforded by the variants of the present invention are most advantageous.

The Fc variants of the present invention may be useful in a variety of antibodies and Fc fusions approved for use in clinical trials or development. These antibodies and Fc fusions are referred to herein as "clinical products and candidates." Thus, in a preferred embodiment, the Fc polypeptides of the present invention may find use in a variety of clinical products and candidates. For example, the Fc polypeptides of the present invention may find use in a variety of antibodies that target CD20. For example, the Fc polypeptides of the present invention may find use in an antibody that is substantially similar to rituximab (Rituxan®, IDEC/Genentech/Roche) (see, e.g., U.S. Patent No. 5,736,137), a chimeric anti-CD20 antibody approved for the treatment of non-Hodgkin's lymphoma; HuMax-CD20, an anti-CD20 currently under development by Genmab, an anti-CD20 antibody described in US 5,500,362, AME-133 (Applied Molecular Evolution), hA20 (Immunomedics, Inc.), HumaLYM (Intracel), and PRO70769 (PCT/US2003/040426, entitled “Immunoglobulin Variants and Uses Thereof”). The Fc polypeptides of the present invention may be useful for a variety of antibodies targeting members of the epidermal growth factor receptor family, including EGFR (ErbB-1), Her2/neu (ErbB-2), Her3 (ErbB-3), Her4 (ErbB-4). For example, the Fc polypeptides of the present invention may find use in an antibody that is substantially similar to trastuzumab (Herceptin®, Genentech) (see, e.g., U.S. Patent No. 5,677,171), a humanized anti-Her2/neu antibody approved for the treatment of breast cancer; pertuzumab (rhuMab-2C4, Omnitarg™), currently under development by Genmab; the anti-Her2 antibody described in U.S. Patent No. 4,753,894; cetuximab (Erbitux®, Imclone) (US 4943533; PCT WO 96/40210), a chimeric anti-EGFR antibody in clinical trials for multiple cancers, ABX-EGF (US 6235883), currently being developed by Abgenix-Immunex-Amgen; HuMax-EGFr (USSN 10/172317), currently being developed by Genmab; 425, EMD55900, EMD62000 and EMD72000 (Merck KGaA) (US 5558864; Murthy et al. 1987, Arch Biochem Biophys. 252(2):549-60; Rodeck et al., 1987, J Cell Biochem. 35(4):315-20; Kettleborough et al., 1991, Protein Eng. 4(7):773-83); ICR62 (Institute of Cancer Research) (PCT WO 95/20045; Modjtahedi et al., 1993, J. Cell Biophys. 1993, 22(1-3): 129-46; Modjtahedi et al., 1993, Br J Cancer. 1993, 67(2): 247-53; Modjtahedi et al, 1996, Br J Cancer, 73(2): 228-35; Modjtahedi et al, 2003, Int J Cancer, 105(2): 273-80; TheraCIM hR3 (YM Biosciences, Canada and Centro de Immunologia Molecular, Cuba (US 5891996; US 6506883; Mateo et al, 1997, Immunotechnology, 3(1): 71-81); mAb-806 (Ludwig Institute for Cancer Research, Memorial Sloan-Kettering) (Jungbluth et al. 2003, Proc Natl Acad Sci U S A. 100(2): 639-44); KSB-102 (KS Biomedix); MR1-1 (IVAX, National Cancer Institute) (PCT WO 0162931A2); and SC100 (Scancell) (PCT WO 01/88138). In another preferred embodiment, Fc polypeptides of the present invention may find use in alemtuzumab (Campath®, Millenium), a humanized monoclonal antibody currently approved for the treatment of B-cell chronic lymphocytic leukemia. Fc polypeptides of the present invention may find use in a variety of antibodies or Fc fusions that are substantially similar to other clinical products and candidates, including, but not limited to, muromonab-CD3 (Orthoclone OKT3®), an anti-CD3 antibody developed by Ortho Biotech/Johnson & Johnson, ibritumomab tiuxetan (Zevalin®), an anti-CD20 antibody developed by IDEC/Schering AG, gemtuzumab ozogamicin (Mylotarg®), anti-CD33 antibody (p67 protein) developed by Celltech/Wyeth, alefacept (Ameviv®), anti-LFA-3 Fc fusion developed by Biogen, abciximab (ReoPro®) developed by Centocor/Lilly, basiliximab (Simulect®) developed by Novartis, palivizumab (Synagis®) developed by MedImmune, infliximab (Remicade®), anti-TNFalpha antibody developed by Centocor, adalimumab (Humira®), anti-TNFalpha antibody developed by Abbott, Humicade™, anti-TNFalpha antibody developed by Celltech, etanercept (Enbrel®), anti-TNFalpha Fc fusion developed by Immunex/Amgen, ABX-CBL, anti-CD147 antibody developed by Abgenix, ABX-IL8, anti-IL8 antibody developed by Abgenix, ABX-MA1, anti-MUC18 antibody developed by Abgenix, Pemtumomab (R1549, 90Y-muHMFG1), anti-MUC1, developed by Antisoma, Therex (R1550), anti-MUC1 antibody developed by Antisoma, AngioMab (AS1405), developed by Antisoma, HuBC-1, developed by Antisoma, Thioplatin (AS1407), developed by Antisoma, Antegren® (natalizumab) (Antegren®, natalizumab), an anti-alpha-4-beta-1 (VLA-4) and alpha-4-beta-7 antibody in development by Biogen, VLA-1 mAb, an anti-VLA-1 integrin antibody in development by Biogen, LTBR mAb, an anti-lymphotoxin beta receptor (LTBR) antibody in development by Biogen, CAT-152, an anti-TGF-β2 antibody in development by Cambridge Antibody Technology, J695, an anti-IL-12 antibody in development by Cambridge Antibody Technology and Abbott, CAT-192, an anti-TGFβ1 antibody in development by Cambridge Antibody Technology and Genzyme, CAT-213, an anti-Eotaxin1 antibody in development by Cambridge Antibody Technology, an anti-Blys LymphoStat-B™ antibody in development Cambridge Antibody Technology and Human Genome Sciences Inc., TRAIL-R1mAb, an anti-TRAIL-R1 antibody being developed by Cambridge Antibody Technology and Human Genome Sciences, Inc., Avastin™ (bevacizumab, rhuMAb-VEGF), an anti-VEGF antibody being developed by Genentech, an anti-HER receptor family antibody being developed by Genentech, Anti-tissue factor (ATF), an anti-tissue factor antibody being developed by Genentech, Xolair™ (Omalizumab), an anti-IgE antibody being developed by Genentech, Raptiva™ (Efalizumab), an anti-CD11a antibody being developed by Genentech and Xoma, MLN-02 antibody (formerly LDP-02), being developed by Genentech and Millenium Pharmaceuticals, HuMax CD4, an anti-CD4 antibody being developed by Genmab, HuMax-IL15, an anti-IL15 antibody being developed by Genmab and Amgen, HuMax-Inflam, being developed by Genmab and Medarex, HuMax-Cancer, an anti-heparanase I antibody being developed by Genmab and Medarex and Oxford GcoSciences, HuMax-Lymphoma, being developed by Genmab and Amgen, HuMax-TAC, being developed by Genmab, IDEC-131 and anti-CD40L antibody being developed by IDEC Pharmaceuticals, IDEC-151 (Clenoliximab), an anti-CD4 antibody being developed by IDEC Pharmaceuticals, IDEC-114, an anti-CD80 antibody being developed by IDEC Pharmaceuticals, IDEC-152, an anti-CD23 antibody in development by IDEC Pharmaceuticals, an anti-macrophage migration factor (MIF) antibody in development by IDEC Pharmaceuticals, BEC2, an anti-idiotype antibody in development by Imclone, IMC-1C11, an anti-KDR antibody in development by Imclone, DC101, an anti-flk-1 antibody in development by Imclone, an anti-VE cadherin antibody in development by Imclone, CEA-Cide™ (labetuzumab), an antibody to carcinoembryonic antigen (CEA), in development by Immunomedics, LymphoCide™ (Epratuzumab), an anti-CD22 antibody in development by Immunomedics, AFP-Cide, in development by Immunomedics, MyelomaCide, in development by Immunomedics, LkoCide, developed by Immunomedics, ProstaCide, developed by Immunomedics, MDX-010, an anti-CTLA4 antibody in development by Medarex, MDX-060, an anti-CD30 antibody in development by Medarex, MDX-070, in development by Medarex, MDX-018, in development by Medarex, Osidem™ (IDM-1) and anti-Her2 antibody in development by Medarex and Immuno-Designed Molecules, HuMax™-CD4, an anti-CD4 antibody in development by Medarex and Genmab, HuMax-IL15, an anti-IL15 antibody in development by Medarex and Genmab, CNTO 148, an anti-TNFα antibody in development by Medarex and Centocor/J&J, CNTO 1275, an antibody anti-cytokine developed by Centocor/J&J, MOR101 and MOR102, intercellular adhesion molecule-1 (ICAM-1) antibodies (CD54) developed by MorphoSys, MOR201, an anti-fibroblast growth factor receptor 3 (FGFR-3) antibody developed by MorphoSys, Nuvion® (visilizumab), an anti-CD3 antibody developed by Protein Design Labs, HuZAF™, an anti-gamma interferon antibody developed by Protein Design Labs, anti-α5β1 Integrin developed by Protein Design Labs, anti-IL-12 developed by Protein Design Labs, ING-1, an anti-Ep-CAM antibody developed by Xoma, and MLN01, an anti-Beta2 integrin antibody developed by Xoma; All references cited in this paragraph are expressly incorporated into this description by reference.

The Fc polypeptides of the present invention may be included in the above clinical candidates and products, or antibodies and Fc hybrids substantially similar thereto. The Fc polypeptides of the present invention may be included in variants of the above clinical candidates and products, humanized, affinity matured, engineered or otherwise modified.

In one embodiment, the Fc polypeptides of the present invention are used to treat autoimmune symptoms, inflammatory symptoms, or transplantation. Target antigens and clinical products and candidates suitable for the treatment of the said diseases include, but are not limited to, anti-α4β7 integrin antibodies such as LDP-02, anti-beta2 integrin antibodies such as LDP-01, anti-complement (C5) antibodies such as 5G1.1, anti-CD2 antibodies such as BTI-322, MEDI-507, anti-CD3 antibodies such as OKT3, SMART anti-CD3, anti-CD4 antibodies such as IDEC-151, MDX-CD4, OKT4A, anti-CD11a antibodies, anti-CD14 antibodies such as IC14, anti-CD18 antibodies, anti-CD23 antibodies such as IDEC 152, anti-CD25 antibodies such as Zenapax, anti-CD40L antibodies such as 5c8, Antova, IDEC-131, anti-CD64 antibodies such as MDX-33, anti-CD80 antibodies such as IDEC-114, anti-CD147 antibodies such as ABX-CBL, anti-E-selectin antibodies such as CDP850, anti-gpIIb/IIIa antibodies such as ReoPro/Abciximab, anti-ICAM-3 antibodies such as ICM3, anti-ICE antibodies such as VX-740, anti-FcR1 antibodies such as MDX-33, anti-IgE antibodies such as rhuMab-E25, anti-IL-4 antibodies such as SB-240683, anti-IL-5 antibodies such as SB-240563, SCH55700, anti-IL-8 antibodies such as ABX-IL8, anti-interferon gamma antibodies, anti-TNF (TNF, TNFa, TNFa, TNF-alpha) such as CDP571, CDP870, D2E7, Infliximab, MAK-195F and anti-VLA-4 antibodies such as Antegren.

Fc variants of the present invention, such as those having enhanced binding to FcRn, can be used in TNF inhibitor molecules to achieve improved properties. Suitable TNF inhibitor molecules include any molecule that inhibits the action of TNF-alpha in a mammal. Suitable variants include the Fc fusion Enbrel® (etanercept) and the antibodies Humira® (adalimumab) and Remicade® (infliximab). Monoclonal antibodies (such as Remicade and Humira) created using Fc variants of the present invention with enhanced FcFn binding can result in greater potency through increased half-life.

In some embodiments, antibodies to infectious diseases are used. Antibodies to eukaryotic cells include antibodies that target yeast cells, including but not limited to Saccharomyces cerevisiae, Hansenula polymorpha, Kluyveromyces fragilis and K. lactis, Pichia guillerimondii and P. pastoris, Schizosaccharomyces pombe, plasmodium falciparium, and Yarrowia lipolytica.

Also suitable are antibodies to fungal cells, including target antigens associated with Candida strains including, among others, Candida glabrata, Candida albicans, C. krusei, C. lusitaniae, and C. maltosa, as well as species of Aspergillus, Cryptococcus, Histoplasma, Coccidioides, Blastomyces, and Penicillium.

Antibodies directed against target antigens associated with protozoa include, but are not limited to, antibodies associated with Trypanosoma, Leishmania species including Leishmania donovanii;, Plasmodium spp., Pneumocystis carinii, Cryptosporidium parvum, Giardia lamblia, Entamoeba histolytica, and Cyclospora cayetanensis.

Also suitable are antibodies to prokaryotic antigens, including antibodies to suitable bacteria, such as pathogenic and non-pathogenic prokaryotes, including but not limited to Bacillus, including Bacillus anthracis; Vibrio, such as V. cholerae; Escherichia, such as Enterotoxigenic E. coli, Shigella, such as S. dysenteriae; Salmonella, such as S. typhi; Mycobacteria, such as M. tuberculosis, M. leprae; Clostridia, such as C. botulinum, C. tetani, C. difficile, C. perfringens; Corynebacteria, such as C. diphtheriae; Streptococcus, S. pyogenes, S. pneumoniae; Staphylococcus, such as S. aureus; Haemophilus influenzae; Neisseria, such as N. meningitidis, N. gonorrhoeae; Yersinia, such as Y. lamblia, Y. pestis, Pseudomonas bacilli, such as P. aeruginosa, P. putida; Chlamydia, such as C. trachomatis; Bordetella, such as B. pertussis; Treponema, such as T. palladium; B. anthracis, Y. pestis, Brucella spp., F. tularensis, B. mallei, B. pseudomallei, B. mallei, B. pseudomallei, C. botulinum, Salmonella spp., SEB V. cholerae toxin B, E. coli O157:H7, Listeria spp., Trichosporon beigelii, Rhodotorula species, Hansenula anomala, Enterobacter sp., Klebsiella sp., Listeria sp., Mycoplasma sp. etc.

According to some aspects, antibodies are directed against viral infections; These viruses include, but are not limited to, orthomyxoviruses (e.g., influenza virus), paramyxoviruses (e.g., respiratory syncytial virus, mumps virus, measles virus), adenoviruses, rhinoviruses, coronaviruses, reoviruses, togaviruses (e.g., rubella virus), parvoviruses, poxviruses (e.g., smallpox virus, vaccinia virus), enteroviruses (e.g., poliovirus, coxsackievirus), hepatitis viruses (including A, B, and C), herpesviruses (e.g., herpes simplex virus, varicella-zoster virus, cytomegalovirus, Epstein-Barr virus), rotaviruses, noroviruses, hantaviruses, arenaviruses, rhabdoviruses (e.g., rabies virus), retroviruses (including HIV, HTLV-I and -II), papovaviruses (e.g., human papillomaviruses), polyomaviruses, and picornaviruses, etc.

Optimized properties of IgG variants

The present application also provides IgG variants optimized for a variety of therapeutically important properties. As used herein, an IgG variant that is created or intended to exhibit one or more optimized properties is referred to as an "optimized IgG variant." Particularly preferred properties that may be optimized include, but are not limited to, increased or decreased affinity for FcRn and increased or decreased half-life in vivo. Suitable embodiments include antibodies that exhibit increased binding affinity for FcRn at a reduced pH, such as the pH associated with endosomes, e.g., pH 6.0, while maintaining decreased affinity at a higher pH, such as 7.4, to allow for increased endosomal uptake but normal release rates. Similarly, these antibodies with modulated FcRn binding may have other desirable properties, such as modulated FcγR binding, such as those disclosed in U.S.S.N.s 11/174,287, 11/124,640, 10/822,231, 10/672,280, 10/379,392, and patent application Ser. No. 11/256,060, entitled "IgG Immunoglobulin Variants Having Optimized Effector Function," filed October 21, 2005. That is, the optimized properties also include, but are not limited to, increased or decreased affinity for FcγR. In one possible embodiment, the IgG variants are optimized such that they have increased affinity for an activating human FcγR, preferably FcγRIIIa, in addition to the FcRn binding characteristic. In yet another possible alternative embodiment, the IgG variants are optimized to have reduced affinity for the human FcγRIIb inhibitory receptor. That is, particular embodiments include using antibodies that exhibit increased binding to FcRn and increased binding to FcγRIIIa. Other embodiments use antibodies that exhibit increased binding to FcRn and increased binding to FcγRIIIa. These embodiments are believed to provide IgG polypeptides that have improved therapeutic properties in humans, such as enhanced effector function and increased anti-cancer activity. In an alternative embodiment, the IgG variants are optimized to have increased or decreased affinity for FcRn and increased or decreased affinity for human FcγRs, including but not limited to FcγRI, FcγRIIa, FcγRIIb, FcγRIIc, FcγRIIIa, and FcγRIIIb, including allelic variants thereof. These embodiments are believed to provide IgG polypeptides that have improved therapeutic properties in humans, such as increased half-life and decreased effector function. In other embodiments, the IgG variants provide increased affinity for FcRn and increased affinity for one or more FcγRs, but decreased affinity for one or more other FcγRs. For example, an IgG variant may have increased binding to FcRn and FcγRIIIa, but decreased binding to FcγRIIb. Alternatively, an IgG variant may have decreased binding to FcRn and FcγR's. In another embodiment, an IgG variant may have decreased affinity for FcRn and increased affinity for FcγRIIb, but decreased affinity for one or more FcγR. In yet another embodiment, an IgG variant may have increased half-life and decreased effector functions.

Preferred embodiments include optimization for binding to human FcRn and FcγR, but in alternative embodiments, the IgG variants have increased or decreased affinity for non-human FcRn and FcγR, including but not limited to rodents and non-human primates. IgG variants that are optimized for binding to non-human FcRn may find use in research. For example, mouse models are available for a variety of diseases that allow testing of properties such as efficacy, toxicity, and pharmacokinetic characteristics for a given drug candidate. As is known in the art, cancer cells can be transplanted or injected into mice to mimic human cancer, a technique called "xenografting." Testing IgG variants, including IgG variants optimized for FcRn, may provide useful information regarding the clearance characteristics of the protein, its mechanism of clearance, etc. IgG variants may also be optimized for increased functionality and/or solution properties in a deglycosylated form. Fc ligands include, but are not limited to, FcRn, FcγRs, C1q, and proteins A and G, and may be from any source, including, but not limited to, human, mouse, rat, rabbit, or monkey, preferably human. In an alternative preferred embodiment, IgG variants are optimized to be more stable and/or more soluble than the deglycosylated form of the parent IgG variant.

IgG variants may contain modifications that modulate interactions with Fc ligands other than FcRn and FcγRs, including but not limited to complement proteins and Fc receptor homologs (FcRHs). FcRHs include but are not limited to FcRH1, FcRH2, FcRH3, FcRH4, FcRH5, and FcRH6 (Davis et al., 2002, Immunol. Reviews 190:123-136, incorporated herein by reference in its entirety).

Preferably, the specificity of the IgG variant for the Fc ligand will determine its therapeutic utility. The utility of a given IgG variant for therapeutic purposes will depend on the epitope and form of the target antigen and the disease or symptom being treated. For most targets and symptoms, increased FcRn binding may be advantageous because increased FcRn binding may result in increased serum half-life. A longer serum half-life allows the drug to be administered less frequently or at a lower dose. This is most advantageous when the therapeutic agent is administered in response to a symptom requiring repeated administration. For some targets and symptoms, decreased FcRn binding may be advantageous. This may be most advantageous when an Fc variant with increased clearance and decreased serum half-life is needed, such as in Fc polypeptides used as radiopharmaceuticals or radiotherapeutics.

IgG variants may be used, including IgG variants that provide increased affinity for FcRn with increased activating FcγRs, and/or decreased affinity for inhibitory FcγRs. For some targets and indications, it may also be useful to use IgG variants that provide differential selectivity for different activating FcγRs; for example, in some cases, increased binding to FcγRIIa and FcγRIIIa but not FcγRI may be needed, whereas in other cases, increased binding to FcγRIIa alone may be preferred. For some targets and symptoms, it may be preferable to use IgG variants that alter FcRn binding and enhance both FcγR-mediated and complement-mediated effector functions, whereas in other cases it may be beneficial to use IgG variants that increase FcRn binding or serum half-life and either FcγR-mediated or complement-mediated effector functions. For some cancer targets or symptoms, it may be beneficial to reduce or remove one or more effector functions, such as by ‘knocking out’ binding to C1q, one or more FcγRs, FcRn, or one or more other Fc ligands. For other targets and symptoms, it may be preferable to use IgG variants that provide increased binding to inhibitory FcγRIIb but wild-type levels, reduced or deleted binding to activating FcγRs. In particular, this may be suitable, for example, in the case where the task of the IgG variant is to inhibit inflammation or an autoimmune disease or to modulate the immune system in some way. Since autoimmune diseases are generally long-term and are treated with repeated doses, their treatment with Fc variants having an increased half-life due to an increased FcRn is most preferred.

Modifications may be made to improve the stability, solubility, function, or clinical utility of the IgG. In a preferred embodiment, the IgG variants may contain modifications to reduce immunogenicity in humans. In a most preferred embodiment, the immunogenicity of the IgG variant is reduced using the method described in USSN 11/004,590, which is incorporated herein by reference in its entirety. In alternative embodiments, the IgG variants are humanized (Clark, 2000, Immunol Today 21:397-402, which is incorporated herein by reference in its entirety.

IgG variants may contain modifications that reduce immunogenicity. Modifications to reduce immunogenicity may include modifications that reduce the binding of processed polypeptides derived from the parent sequence to MHC proteins. For example, amino acid modifications may be made such that there are no or a minimal number of immune epitopes that are predicted to bind with high affinity to any of the predominant MHC alleles. Several methods for identifying MHC binding epitopes in protein sequences are known in the art and can be used to enumerate epitopes in an IgG variant. See, for example, WO 98/52976; WO 02/079232; WO 00/3317; USSN 09/903,378; USSN 10/039,170; USSN 60/222,697; USSN 10/754,296; PCT WO 01/21823; and PCT WO 02/00165; Mallios, 1999, Bioinformatics 15: 432-439; Mallios, 2001, Bioinformatics 17: 942-948; Sturniolo et al., 1999, Nature Biotech. 17: 555-561; WO 98/59244; WO 02/069232; WO 02/77187; Marshall et al., 1995, J. Immunol. 154: 5927-5933; and Hammer et al., 1994, J. Exp. Med. 180: 2353-2358, all of which are incorporated herein by reference in their entireties. Sequence-based information can be used to determine the binding outcome for a given peptide-MHC interaction (see, e.g., Mallios, 1999, Bioinformatics 15:432-439; Mallios, 2001, Bioinformatics 17:p942-948; Sturniolo et al., 1999, Nature Biotech. 17:555-561, all of which are incorporated herein by reference in their entireties).

Creation of IgG variants

Variants of the present invention can be made in a variety of ways. Variants described herein can be insertions, deletions, substitutions, other modifications, or combinations of these and other changes. In particular, a novel embodiment of the present invention is to create insertions and deletions that either increase or decrease binding of an Fc polypeptide to an Fc ligand. As described herein, insertions or deletions can be made that increase or decrease the affinity of an Fc polypeptide for FcRn. Insertions and deletions can be made using rational approaches or approaches involving the use of random components, such as random and semi-random library generation or screening. In an alternative embodiment, substitutions are described that increase or decrease the affinity of an Fc polypeptide for FcRn.

Backbone modifications: insertions and deletions

Variant Fc polypeptides can be prepared by substituting a variant amino acid in place of a parent amino acid at a position in the Fc polypeptide. By substituting one or more amino acids for the variant amino acids in the Fc polypeptide, the side chains at these positions are altered. The most suitable substitutions modify the properties of the Fc by altering the side chains of the Fc. The substituted side chains can interact directly or indirectly with an Fc binding partner that is associated with the function or property of the Fc. At least one substitution alters the covalent structure of one or more side chains of the parent Fc polypeptide.

Alternatively, variant Fc polypeptides can be produced that alter the covalent structure of the backbone of a parent Fc polypeptide. In proteins, backbone atoms are a peptide nitrogen, an alpha carbon, a carbonyl or peptide carbon, and a carbonyl oxygen. Altering the covalent structure of the backbone provides additional ways to alter the properties of Fc polypeptides. The covalent structure of the Fc backbone can be altered by adding atoms to the backbone, such as by inserting one or more amino acids, or by removing atoms from the backbone, such as by deleting one or more amino acids. The covalent structure of the backbone can also be altered by replacing individual backbone atoms with other atoms (Deechongkit et al., J Am Chem Soc. 2004. 126(51):16762-71, incorporated herein by reference in its entirety). As is known in the art and illustrated herein, amino acid insertions or deletions in Fc polypeptides can be accomplished by inserting or deleting the corresponding nucleotides in DNA encoding said Fc polypeptide. Alternatively, as is known in the art, amino acid insertions or deletions can be accomplished during synthesis of Fc polypeptides.

The creation of amino acid insertions or deletions that alter the interaction of an Fc polypeptide with one or more binding partners (e.g., FcgammaR's, FcRn, C1q) can be accomplished by considering the structure of a complex of the Fc polypeptide and its binding partner. In a less preferred embodiment, the creation can be accomplished by considering the structure of the Fc polypeptide and information about the region of the Fc involved in binding to the binding partner. This information can be obtained by performing mutagenesis experiments, phage display experiments, by comparing homology, computer modeling, or other methods.

Preferred amino acid sequence positions for insertions or deletions that affect Fc binding interactions but do not affect the overall structure, stability, expression, or use of the Fc polypeptide are loops involved in Fc/Fc binding partner interactions. For altering FcRn binding to an Fc polypeptide, positions 244-257, 279-284, 307-317, 383-390, and 428-435 represent preferred loop locations for insertions or deletions (numbering according to the EU index of Kabat et al., Burmeister et al., 1994, Nature, 372:379-383; Martin et al., 2001, Mol Cell 7:867-877, all incorporated herein by reference in their entireties). To alter binding of the Fc gamma receptor to the Fc polypeptide, positions 229-239, 266-273, 294-299, and 324-331 are preferred loop locations for insertions or deletions (numbering according to EU index of Kabat et al., PDB code 1E4K.pdb Sondermann et al. Nature. 2000 406:267, all of which are incorporated herein by reference in their entireties). The loops are regions of the polypeptide that are not included in the alpha-helical structure or the beta-sheet structure. The loop positions are positions that are not included in either the alpha-helical structure or the beta-sheet structure (van Holde, Johnson and Ho. Principles of Physical Biochemistry. Prentice Hall, New Jersey 1998, Chapter 1 pp. 2-67, incorporated herein by reference in their entireties). Loop positions are preferred because backbone atoms are generally more flexible and less likely to engage in hydrogen bonding compared to backbone atoms of alpha helices and beta sheets. Therefore, loop lengthening or shortening by insertion or deletion of one or more amino acids is less likely to result in significant, disruptive changes to the Fc polypeptide, including stability, expression, or other difficulties.

Insertions and deletions can be used to alter the length of a polypeptide. For example, in loop regions, altering the length of the loop results in altered flexibility and conformational entropy of said loop. Insertions into a loop will generally increase the conformational entropy of the loop, which can be defined as Boltzmann's constant multiplied by the natural logarithm of the number of possible conformations (van Holde, Johnson and Ho. Principles of Physical Biochemistry. Prentice Hall, New Jersey 1998, pp. 78, incorporated herein by reference in its entirety). By inserting at least one amino acid into a polypeptide, the overall number of conformations of the polypeptide is increased. These additional conformations can be useful for forming suitable Fc/Fc binding partner interactions, since one of the additional conformations of the Fc polypeptide can be utilized in binding of the Fc binding protein. In this case, the insertion may result in stronger Fc/Fc binding partner interactions. In the case where the additional conformations are not used in the binding region, the insertion may result in weaker Fc/Fc binding partner interactions because the additional conformations compete with the binding conformation. Similarly, deletion of a polypeptide segment may also result in either stronger or weaker Fc/Fc binding partner interactions. In the case where the deletion of a segment that reduces the possible number of backbone conformations excludes a binding conformation, the deletion may result in weaker Fc/Fc binding partner interactions. In the case where the deletion does not exclude a binding conformation, the deletion may result in stronger Fc/Fc binding partner interactions because the deletion may exclude conformations competing with the binding conformation.

Insertions and deletions can be used to change the positions and orientations of amino acids in an Fc polypeptide. Because insertions and deletions cause a change in the covalent structure of the backbone, they inevitably cause a change in the positions of the backbone atoms. Figure 7 compares the backbone positions of several loop segments, labeled L1 through L4, in three different backbones. The reference backbone structure contains four loop segments, whereas the deletion backbone does not contain the L1 segment, and the insertion segment contains an additional segment before (i.e., N-terminal) the L1 segment. Deletions and insertions cause the greatest change in the backbone structure near the site of insertion or deletion. By deleting a segment located near the N-terminal region of the loop, such as the L1 segment, the loop is shortened and the remaining segments move their positions closer to the N-terminal of the loop. This causes a shift of the L2 segment toward the former location of the L1 segment and toward the N-terminal of the loop. This change in the position of the L2 segment toward the L1 segment may enhance binding of the Fc/Fc binding partner complex and is advantageous when there is prior information that the amino acid or amino acids located in L2 make favorable interactions with the Fc binding partner when located in L1. For example, if L2 contains an alanine and a tyrosine, and substitution of two amino acids of the L1 segment to an alanine and a tyrosine already results in an Fc variant with increased binding, then deletion of L1 may result in an Fc variant with increased affinity for the Fc binding partner.

Similarly, insertion of a polypeptide segment into an Fc polypeptide at the N-terminal region of the loop results in a shift in the positions of the loop segments toward the C-terminal region of the loop. In Fig. 7, insertion of one or more amino acids upstream (i.e., N-terminal) of the L1 segment alters the conformation of the backbone, including a shift of the L1 segment toward the C-terminal region of the loop. This type of insertion is preferred when the amino acids located in the L1 segment are known to make favorable interactions when located at the L2 positions, since this insertion can result in stronger Fc/Fc binding partner interactions. In cases where weaker Fc/Fc binding partner interactions are desired, insertion can be used to shift the inappropriate amino acid to a different position. The inserted, deleted, and reference segments (L1-L4 in Fig. 7) can represent one or more amino acids in an Fc polypeptide.

Alternatively, insertions or deletions may be used in the C-terminal region of the loops in a manner analogous to insertions or deletions in the N-terminal region of the loops. Insertions in the C-terminal region of the loop may result in a shift of positions N-terminal to the insertion toward the N-terminal of the loop. Deletions in the C-terminal region of the loop may result in a shift of positions N-terminal to the deletion toward the C-terminal of the loop. The choice of using an insertion or deletion in the N-terminal or C-terminal region of the loop is made based on the amino acids located in the loop, the need for increased or decreased Fc/Fc binding partner affinity, and the desired positional shift.

Insertions or deletions may be used in any region of the Fc polypeptide, including loops, alpha-helical regions, and beta-sheet regions. Preferred sites for insertions and deletions include loop regions, which are regions that are not alpha-helical or beta-sheet regions. Loops are preferred because they generally accept backbone changes better than alpha-helices or beta-sheet regions. The most preferred sites for insertions or deletions resulting in stronger protein/protein interactions are located at the N-terminal or C-terminal regions of the loop. If loop side chains are involved in Fc/Fc binding partner interactions, insertions or deletions at these sites are less likely to result in highly disruptive changes in binding interactions. Deletions within the precise center of the loop are more likely to remove important residues at the Fc/Fc binding partner interface, and insertions within the precise center of the loop are more likely to introduce inappropriate interactions at the Fc/Fc binding partner interface. The number of residues removed or inserted can be determined by the amount of backbone change required, with insertions or deletions of 15 or fewer residues being preferred, insertions or deletions of 10 or fewer residues being more preferred, and insertions or deletions of 5 or fewer residues being most preferred.

Once the position and size of the Fc deletion variant is determined, the entire polypeptide sequence is determined and the polypeptide can be obtained by methods known in the art.

However, Fc insertion variants involve the additional step of creating a sequence containing at least one inserted amino acid. Insertions of polar residues including Ser, Thr, Asn, Gln, Ala, Gly, His are preferred at positions expected to be exposed in the Fc polypeptide. Smaller amino acids including Ser, Thr, and Ala are most preferred because small size is less likely to sterically hinder Fc/Fc binding partner interactions. Ser and Thr can also form hydrogen bonds with atoms of the Fc binding partner.

Insertions also provide additional flexibility, whereby the inserted polypeptide can be designed to form suitable interactions with the Fc binding partner, as is necessary when stronger Fc/Fc binding partner binding is required. The length of the insertion in the backbone can be determined by modeling the variant backbone with a simple generic sequence inserted. For example, polyserine, polyglycine, or polyalanine insertions of varying lengths can be "designed" and modeled. Modeling can be accomplished in a variety of ways, including homology modeling based on known three-dimensional structures of homologues containing the said insertion, and by computer modeling including MODELLER (M.A. Marti-Renom et al. Annu. Rev. Biophys. Biomol. Struct. 29, 291-325, 2000) and ROSETTA (Kuhlman et al. (2003). Science 302, 1364-8), both of which are incorporated herein by reference in their entireties. Typically, various conformations of the backbone are first generated, and the final structure of the backbone can be determined after the identity of the side chain has been established. Side chains can be generated using PDA® algorithms (US 6188965; 6269312; 6403312; 6801861; 6804611; 6792356, 6950754 and USSN 09/782004; 09/927790; 10/101499; 10/666307; 10/666311; 10/218102, all incorporated herein by reference in their entireties).

Insertions or deletions may be made to alter the binding of Fc polypeptides to FcgammaR in a manner similar to that described for altering the binding properties of FcRn. The Fc domains bind to FcgammaR at the position shown in Fig. 1. Structures of the Fc/FcgammaR complex, including PDB codes 1T89 and 1IIS (Radaev S et al. J. Biol. Chem. v. 276, pp. 16469-16477, incorporated herein by reference in its entirety), show interacting residues and loops between the two structures. All mutagenesis results, such as those reported in US11/124620 and US6737056 (both of which are incorporated herein by reference in their entirety), are useful in determining appropriate shifts in backbone positions.

Insertions or deletions can be made in any polypeptide other than Fc polypeptides by the methods described herein. For example, insertions or deletions in the TNF superfamily member APRIL can be made using the three-dimensional structure (PDB accession number 1XU1.pdb, Hymowitz, et al. (2005) J. Biol. Chem. 280:7218, incorporated herein by reference in its entirety). Insertions or deletions can be made to enhance binding of APRIL to its receptor TACI. Loop residues preferred as sites of insertion or deletion are residues Ser118-Val124, Asp164-Phe167, Pro192-Ala198, Pro221-Lys226. These loops interact with TACI in the APRIL/TACI complex and mediate binding.

Polypeptides including variants

IgG variants may be based on human IgG sequences, and thus human IgG sequences are used as "base" sequences to which other sequences are compared, including but not limited to sequences from other organisms, such as rodent and primate sequences. IgG variants may also contain sequences from other immunoglobulin classes, such as IgA, IgE, IgD, IgM, etc. It is envisaged that, although IgG variants are created in the context of one parent IgG, variants may be created in the context of, or "transferred" to, another, second parent IgG. This is accomplished by determining "equivalent" or "corresponding" residues and substitutions between the first and second IgGs, typically based on sequence homology or structural homology between the IgG sequences. To establish homology, the amino acid sequence of the first specified IgG is directly compared to the sequence of the second IgG. After alignment of the sequences using one or more homology alignment programs known in the art (e.g. using conserved residues as intermediate species), providing for the necessary insertions and deletions to maintain the alignment (i.e. avoiding removal of conserved residues by arbitrary deletion and insertion), residues equivalent to specific amino acids in the primary sequence of the first IgG variant are determined. The alignment of conserved residues should preferably preserve 100% of such residues. However, alignment of more than 75% or up to 50% of the conserved residues is also suitable for determining equivalent residues. Equivalent residues can also be determined by establishing structural homology between the first and second IgG, i.e. at the tertiary structure level for the IgGs whose structures have been determined. In this case, equivalent residues are defined as residues for which the atomic coordinates of two or more backbone atoms of a particular amino acid residue of the parent or precursor (N to N, CA to CA, C to C, and O to O) are within 0.13 nm and preferably 0.1 nm after alignment. The alignment is performed after orientation and positioning of the best model to obtain the maximum coincidence of the atomic coordinates of the non-hydrogen protein atoms of the proteins. Regardless of how equivalent or corresponding residues are defined and regardless of the identity of the parent IgG in which the IgGs are generated, it remains the same that the investigated IgG variants can be generated in a second parent IgG that has significant sequence homology or structural homology to the IgG variant. Thus, for example, if a variant antibody is produced in which the parent antibody is human IgG1, by using the above-described methods or other methods for determining equivalent residues, the variant antibody can be generated in another parent IgG1 antibody that binds to a different antigen, a human IgG2 parent antibody, a human IgA parent antibody, a mouse IgG2a or IgG2b parent antibody, etc. Again, as described above, the context of the parent IgG variant does not affect the ability to transfer IgG variants to other parent IgGs.

Methods for creating, producing and screening IgG variants are provided. The disclosed methods do not limit any particular use or method of use. Preferably, the disclosed methods generally illustrate that one or more IgG variants can be created, produced and experimentally screened to obtain IgG variants having optimized effector function. A variety of methods for creating, producing and testing antibody and protein variants are described in USSN 10/754,296 and USSN 10/672,280, both of which are incorporated herein by reference in their entireties.

A variety of protein engineering methods can be used to create IgG variants having optimized effector function. In one embodiment, a structure-based design method can be used, in which available structural information is used to guide substitutions, insertions, or deletions. In a preferred embodiment, a computer-based screening method can be used, in which substitutions are made based on their energetic fit in computer calculations. See, for example, USSN 10/754,296 and USSN 10/672,280 and references cited therein, all of which are incorporated herein by reference in their entireties.

Sequence alignment can be used to guide substitutions at identified positions. It will be apparent to those skilled in the art that the use of sequence information can limit the introduction of substitutions that are potentially destructive to protein structure. The source of sequences can vary widely and include one or more known databases, including, but not limited to, the Kabat database (Northwestern University); Johnson & Wu, 2001, Nucleic Acids Res. 29:205-206; Johnson & Wu, 2000, Nucleic Acids Res. 28:214-218), the IMGT database (IMGT, the international ImMunoGeneTics information system®; Lefranc et al., 1999, Nucleic Acids Res. 27:209-212; Ruiz et al., 2000 Nucleic Acids Res. 28:219-221; Lefranc et al., 2001, Nucleic Acids Res. 29:207-209; Lefranc et al., 2003, Nucleic Acids Res. 31:307-310), and VBASE, all of which are incorporated herein by reference in their entireties. Antibody sequence information may be acquired, compiled, and/or derived from alignments of germline or naturally occurring antibody sequences from any organism, including, but not limited to, mammals. It will be apparent to those skilled in the art that the use of sequences of human or substantially human origin may also have the advantage of being less immunogenic when administered to humans. Other databases that are more general nucleic acid or protein databases, i.e., not specifically antibodies, include, but are not limited to, SwissProt, GenBank Entrez, and the EMBL nucleotide sequence database. The aligned sequences may include VH, VL, CH, and/or CL sequences. There are numerous sequence-based alignment programs and methods known in the art, all of which find use in generating sequence alignments.

Alternatively, random or semi-random mutagenesis methods can be used to create amino acid modifications at desired positions. In these cases, the positions are chosen randomly or the amino acid changes are made according to simplified rules. For example, all residues can be mutated to alanine, a mutagenesis termed "alanine scanning". These methods can be combined with more sophisticated engineering approaches that employ selection methods to screen for higher levels of sequence difference. As is well known in the art, there are a variety of selection technologies that can be used for these approaches, including, for example, display technologies such as phage display, ribosome display, cell surface display, etc., as described below.

Methods for producing and screening IgG variants are well known in the art. General methods for antibody molecular biology, expression, purification, and screening are described in Antibody Engineering, Duebel & Kontermann, Springer-Verlag, Heidelberg, 2001; and Hayhurst & Georgiou, 2001, Curr Opin Chem Biol 5: 683-689; Maynard & Georgiou, 2000, Annu Rev Biomed Eng 2:339-76. See also the methods described in USSN 10/754,296; USSN 10/672,280; and USSN 10/822,231; and 11/124,620, all of which are incorporated herein by reference in their entireties.

Preferred embodiments of the present invention include those shown in Fig. 8. Alternatively, preferred embodiments of the present invention include those shown in Fig. 9. Further alternatively, preferred embodiments of the present invention include those shown in Fig. 10. These variants have demonstrated increased binding to the Fc receptor, FcRn, as illustrated in the examples.

Creation of IgG variants

IgG variants can be generated by any method known in the art. In one embodiment, IgG variant sequences are used to generate nucleic acids that encode member sequences and which can then be cloned into host cells, expressed, and analyzed as needed. These techniques are performed using well-known procedures, and many methods that may find use therein are described in Molecular Cloning - A Laboratory Manual, ^3rd Ed. (Maniatis, Cold Spring Harbor Laboratory Press, New York, 2001) and Current Protocols in Molecular Biology (John Wiley & Sons), both of which are incorporated herein by reference in their entireties. Nucleic acids encoding IgG variants can be included in an expression vector for expression of the protein. Expression vectors typically contain the protein operably linked, i.e., in a functional relationship, to control or regulatory sequences, selectable markers, any fusion partners, and/or additional elements. IgG variants can be produced by growing a host cell transformed with a nucleic acid, preferably an expression vector, containing a nucleic acid encoding the IgG variants, under suitable conditions to induce or cause expression of the protein. A wide variety of suitable host cells can be used, including, but not limited to, mammalian cells, bacteria, insect cells, and yeast. For example, a variety of cell lines that may find use are described in the ATCC cell line catalog available from the American Type Culture Collection, which is incorporated herein by reference in its entirety. Methods for introducing exogenous nucleic acid into host cells are well known in the art and will vary depending on the host cell used.

In a preferred embodiment, the IgG variants are purified or isolated after expression. The antibodies can be isolated or purified by a variety of methods known to those skilled in the art. Standard purification methods include chromatographic methods, electrophoretic methods, immunological methods, precipitation methods, dialysis, filtration, concentration, and chromatofocusing methods. As is well known in the art, many natural proteins bind antibodies, such as bacterial proteins A, G, and L, and these proteins can find use in purification. Often, purification can be facilitated by a particular fusion partner. For example, proteins can be purified using glutathione resin in the case of a GST fusion, Ni ⁺² affinity chromatography in the case of a His tag, or immobilized anti-flag antibody in the case of a flag tag. For general guidance on suitable purification methods, see Antibody Purification: Principles and Practice, ^3rd Ed., Scopes, Springer-Verlag, NY, 1994, which is incorporated herein by reference in its entirety.

Screening for IgG variants

Fc variants can be screened using a variety of methods, including but not limited to methods that use in vitro assays, in vivo assays, and cell-based assays and selection methods. Automated and high-throughput screening technologies can be used in screening procedures. Screening can use a fusion partner or a label, such as an immunolabel, an isotopic label, or a label representing a small molecule, such as a fluorescent or colorimetric dye.

In a preferred embodiment, the functional and/or biophysical properties of the Fc variants are screened in an in vitro assay. In a preferred embodiment, the protein is screened for functionality, such as its ability to catalyze a reaction or its binding affinity for a target.

As is known in the art, a subset of screening methods are those that select suitable library members. These methods are referred to herein as "selection methods" and these methods find use in the present invention in screening for Fc variants. When protein libraries are screened using a selection method, only those library members that are suitable, i.e., that meet the selection criteria, are passed, isolated, and/or monitored. A variety of selection methods are known in the art and may find use in the present invention for screening protein libraries. Other selection methods that may find use in the present invention include methods that are not display-based, such as in vivo methods. A subset of selection methods, referred to as "directed evolution" methods, are methods that involve joining or "breading" suitable sequences during selection, sometimes including new mutations.

In a preferred embodiment, the Fc variants are screened using one or more cell-based or in vivo assays. For these assays, purified and unpurified proteins are typically added externally such that cells are exposed to individual variants or pools of variants belonging to a library. These assays are typically, but not always, based on the function of the Fc polypeptide; i.e., the ability of the Fc polypeptide to bind to a target and mediate some biochemical reaction, such as effector function, inhibition of ligand/receptor binding, apoptosis, etc. These assays often involve monitoring the response of cells to IgG, such as cell survival, cell death, changes in cellular morphology, or transcriptional activation, such as cellular expression of a natural gene or a reporter gene. For example, these assays can measure the ability of Fc variants to induce ADCC, ADCP, or CDC. Some assays may require the addition of additional cells or components, i.e., in addition to target cells, such as serum complement or effector cells such as peripheral blood monocytes (PBMCs), macrophages, natural killer cells, etc. Said additional cells may be from any organism, preferably humans, mice, rats, rabbits, and monkeys. The antibodies may induce apoptosis of certain cell lines expressing the target, or may mediate attack on target cells by immune cells added to the assay. Methods for monitoring cell death or viability are known in the art and include the use of dyes, immunochemicals, cytochemicals, and radioactive reagents. Transcriptional activation may serve as a method for analyzing function in cell-based assays. Alternatively, cell-based screens are performed using cells transformed or transfected with nucleic acids encoding the variants. That is, the Fc variants are not added externally to the cells.

The biological properties of IgG variants can be characterized based on data from experiments involving cells, tissues, and the whole organism. As is known in the art, drugs are often tested in animals, including but not limited to mice, rats, rabbits, dogs, cats, pigs, and monkeys, in order to measure the effectiveness of the drug in treating a disease or disease model or to measure the pharmacokinetic characteristics, toxicity, and other properties of the drug. These animals can be considered disease models. Drugs are often tested in mice, including but not limited to nude mice, SCID mice, xenograft mice, and transgenic mice (including knockins and knockouts). These studies can provide important data to determine the potential of a protein to be used as a drug. Any organism, preferably mammals, can be used for testing. For example, monkeys, due to their genetic similarity to humans, may be suitable models for treatment and thus can be used to test the efficacy, toxicity, pharmacokinetic properties, or other properties of IgGs. Ultimately, human testing is required for approval for use as drugs and, therefore, these experiments are necessarily contemplated. Thus, IgGs can be tested in humans to determine their therapeutic efficacy, toxicity, immunogenicity, pharmacokinetic properties, and/or other clinical properties.

Ways to use IgG variants

IgG variants may find use in a wide range of products. In one embodiment, the IgG variant is a therapeutic, diagnostic or experimental reagent, preferably therapeutic. The IgG variant may find use in an antibody composition that is monoclonal or polyclonal. In a preferred embodiment, the IgG variants are used to destroy target cells that produce a target antigen, such as cancer cells. In an alternative embodiment, the IgG variants are used to block, antagonize or agonize a target antigen, such as to antagonize a cytokine or a cytokine receptor. In an alternative preferred embodiment, the IgG variants are used to block, antagonize or agonize a target antigen and destroy target cells that produce the target antigen.

IgG variants can be used for various therapeutic purposes. In a preferred embodiment, an antibody comprising an IgG variant is administered to a patient for the treatment of an antibody-associated disorder. For these purposes, the term "patient" includes humans and other animals, preferably mammals and most preferably humans. As used herein, the term "antibody-associated disorder" or "antibody-responsive disorder" or "condition" or "disease" means a disorder that can be alleviated by administering a pharmaceutical composition comprising an IgG variant. Antibody-associated disorders include, but are not limited to, autoimmune diseases, immunological diseases, infectious diseases, inflammatory diseases, neurological diseases, and oncological and neoplastic diseases, including cancer. As used herein, the terms "cancer" and "cancerous" describe or refer to a physiological condition in mammals that is typically characterized by unregulated cell growth. Examples of cancer include, but are not limited to, carcinoma, lymphoma, blastoma, sarcoma (including liposarcoma), neuroendocrine tumors, mesothelioma, schwannoma, meningioma, adenocarcinoma, melanoma, leukemia, and lymphocytic leukemia.

In one embodiment, the IgG variant is the only therapeutically active agent administered to the patient. Alternatively, the IgG variant is administered in combination with one or more therapeutic agents, including, but not limited to, cytotoxic agents, chemotherapeutic agents, cytokines, growth inhibitors, antihormonal agents, kinase inhibitors, antiangiogenic agents, cardioprotectors, or other therapeutic agents. The IgG variants can be administered concurrently with one or more treatment regimens. For example, the IgG variant can be administered to the patient along with chemotherapy, radiation therapy, or both chemotherapy and radiation therapy. In one embodiment, the IgG variant can be co-administered with one or more antibodies, which may or may not be an IgG variant. According to another embodiment, the IgG variant and one or more other types of anti-cancer therapy are used to treat cancer cells ex vivo. It is suggested that the said ex vivo treatment may be suitable in bone marrow transplantation and in particular autologous bone marrow transplantation. It is certainly suggested that IgG variants may be used in combination with other therapeutic methods, such as surgery.

A variety of other therapeutic agents can be administered with IgG variants. In one embodiment, the IgG is co-administered with an anti-angiogenic agent. As used herein, the term "antiangiogenic agent" refers to a compound that blocks or interferes to some extent with the growth of blood vessels. The anti-angiogenic factor can, for example, be a small molecule or protein, such as an antibody, an Fc hybrid, or a cytokine that binds to a growth factor or growth factor receptor involved in stimulating angiogenesis. As used herein, a preferred anti-angiogenic factor is an antibody that binds to vascular endothelial growth factor (VEGF). In an alternative embodiment, the IgG is co-administered with a therapeutic agent that induces or enhances an adaptive immune response, such as an antibody that targets CTLA-4. In an alternative embodiment, the IgG is co-administered with a tyrosine kinase inhibitor. As used herein, the term "tyrosine kinase inhibitor" refers to a molecule that inhibits the tyrosine kinase activity of a tyrosine kinase to some extent. In an alternative embodiment, the IgG variants are co-administered with a cytokine.

Pharmaceutical compositions are envisaged when an IgG variant and one or more therapeutically active agents are prepared. Formulations of IgG variants are prepared for storage by mixing IgG of the required purity with optional pharmaceutically acceptable carriers, excipients or stabilizers (Remington's Pharmaceutical Sciences 16th edition, Osol, A. Ed., 1980, incorporated herein by reference in its entirety) in the form of lyophilized formulations or aqueous solutions. Said formulations for use in in vivo administration are preferably sterile. This is readily accomplished by filtration through sterilizing filtration membranes or other means. IgG variants and other therapeutically active agents described herein can also be prepared as immunoliposomes and/or enclosed in microcapsules.

The concentration of the therapeutically active IgG variant in the composition may vary from about 0.1 to 100% by weight. In a preferred embodiment, the IgG concentration is in the range of 0.003-1.0 mole. A therapeutically effective dose of the IgG variant may be administered to treat a patient. As used herein, the term "therapeutically effective dose" means a dose that provides the desired effects. The exact dose will depend on the purpose of treatment and will be determined by one skilled in the art using known methods. Doses may range from 0.01 to 100 mg/kg body weight or higher, such as 0.01, 0.1, 1.0, 10 or 50 mg/kg body weight, with a dose of 1-10 mg/kg being preferred. As is known in the art, adjustments may be necessary for protein degradation, systemic or localized delivery, and rate of synthesis of new proteases, as well as age, body weight, general health, gender, diet, duration of administration, drug interactions, and severity of the condition, which can be determined by one of skill in the art using routine testing.

The administration of a pharmaceutical composition comprising an IgG variant, preferably in the form of a sterile aqueous solution, can be carried out in a variety of ways, such as, but not limited to, orally, subcutaneously, intravenously, parenterally, intranasally, intra-aurally, intra-ocularly, rectally, vaginally, transdermally, topically (e.g., gels, ointments, lotions, creams, etc.), intraperitoneally, intramuscularly, intrapulmonarily (e.g., the AERx® inhalation system manufactured by Aradigm or the Inhance® pulmonary delivery system manufactured by Nektar Therapeutics, etc.). The drug described herein can be administered simultaneously with other drugs, i.e., the drugs described herein can be administered together with other therapies or drugs, including, for example, small molecules, other biological drugs, radiation therapy, surgery, etc.

Information confirming the possibility of implementing the invention

The following examples are provided to illustrate the present invention. These examples do not limit the present invention to any particular use or method of use. For all positions described in accordance with the present invention, numbering is given according to the EU index as in Kabat (Kabat et al., 1991, Sequences of Proteins of Immunological Interest, 5th Ed., United States Public Health Service, National Institutes of Health, Bethesda, incorporated herein by reference in its entirety). Those skilled in the art of antibodies will appreciate that this convention consists of non-sequential numbering in specific regions of the immunoglobulin sequence, providing standardized reference to conserved positions within immunoglobulin families. Accordingly, the positions of any given immunoglobulin defined by the EU index will not necessarily correspond to its ordinal sequence.

EXAMPLE 1. DNA generation, expression and purification of Fc variants

Amino acid modifications were created in the Fc region of IgG antibodies to enhance their affinity for the neonatal Fc receptor FcRn. Variants were screened against a number of different human IgG constant chains (Fig. 2), including IgG1, IgG2, and hybrid IgG sequences containing the CH1 and upper hinge of IgG1 and the Fc region of IgG2. It will be apparent to those skilled in the art that due to the different interactions of the IgG1 and IgG2 Fc region with FcγRs and complement, these different parental Fc regions will have different FcγR- and complement-mediated effector function properties. Representative Fc variant sequences against these parental IgG constant chains are shown in Fig. 3.

Fc variants were created for an antibody that targets vascular endothelial growth factor (VEGF). The variable regions (VH and VL) of the heavy and light chains are those of a humanized version of the A4.6.1 antibody, also called bevacizumab (Avastin®), which is approved for the treatment of multiple cancers. The amino acid sequences of the VH and VL regions of this antibody are shown in Fig. 4.

Genes encoding the heavy and light chains of anti-VEGF antibodies were generated in the mammalian expression vector pTT5. Human IgG1 and IgG2 constant chain genes were obtained from IMAGE clones and subcloned into the pTT5 vector. The IgG1/2 gene was generated using PCR-mediated mutagenesis. The VH and VL genes encoding anti-VEGF antibodies were synthesized commercially (Blue Heron Biotechnologies, Bothell WA) and subcloned into vectors encoding the appropriate CL, IgG1, IgG2, and IgG1/2 constant chains. Amino acid modifications were generated by site-directed mutagenesis using QuikChange® site-directed mutagenesis methods (Stratagene, La Jolla CA). All DNA sequences were sequenced and sequences were confirmed to be correct.

Plasmids containing the heavy chain gene (VH-Cγ1-Cγ2-Cγ3) were cotransfected with a plasmid containing the light chain gene (VL-Cκ) into 293E cells using lipofectamine (Invitrogen, Carlsbad CA) and grown in FreeStyle 293 media (Invitrogen, Carlsbad CA). After 5 days of growth, antibodies were purified from culture supernatants by protein A affinity chromatography using MabSelect resin (GE Healthcare). Antibody concentrations were determined by the bicinchoninic acid (BCA) assay (Pierce).

EXAMPLE 2. Fc variant antibodies retain binding to antigen

The accuracy of the expressed variant antibodies was confirmed by demonstrating that they retained specificity for the antigen. VEGF binding was monitored by surface plasmon resonance (SPR, Biacore) using a Biacore 3000 instrument. Recombinant VEGF (VEGF-165, PeproTech, Rocky Hill, NJ) was coupled to the surface of the CM5 chip by coupling with N-hydroxysuccinimide/N-ethyl-N'-(-3-dimethylaminopropyl)carbodiimide (NHS/EDC) using standard methods. Wild-type and variant antibodies were used as analytes, and a response measured in resonance units (RU) was obtained. The dissociation phase was too slow to measure true equilibrium constants and therefore relative binding was determined by measuring the RU at the end of the association phase, which should be proportional to the protein concentration (held constant in this experiment) and the association rate constant. The data obtained (Fig. 6) demonstrate that the variant anti-VEGF antibodies retain binding to the antigen, in contrast to the negative control anti-Her2 antibody, which does not bind VEGF.

EXAMPLE 3. Measurement of binding to human FcRn

Binding of variant antibodies to human FcRn was measured at pH 6.0, the pH at which it is naturally bound in endosomes. Vectors encoding beta-2-microglobulin and His-tagged FcRn alpha chain genes were generated, co-transfected into 293T cells, and purified by nickel chromatography. Affinity of antibody to human FcRn (hFcRn) at pH 6.0 was measured on a Biacore 3000 instrument by coupling human FcRn to the surface of a CM5 chip using standard NHS/EDC chemistry. Wild-type and variant antibodies were used in the mobile phase at 25-100 nM concentrations, and the response was measured in resonance units. Association and dissociation phases were achieved at pH 6.0 followed by the introduction of pH 7.4 buffer and measurement of antibody release from the receptor at the higher pH. The antibody and buffer cycle provided a baseline analysis that was subtracted from the sensorgram of each sample.

Figure 7 shows Biacore sensorgrams of binding of native IgG1 and selected Fc variant antibodies to human FcRn at two corresponding pH values. The data show that wild-type and variant antibodies readily bind to the FcRn chip at pH 6.0 and dissociate slowly at this pH, just as they would dissociate in an endosome, while they are rapidly released at pH 7.4, as if the endosome were recycled to the membrane and exposed to the higher pH of serum.

The FcRn association/dissociation curves did not fit a simple Langmuir model, possibly due to antibody and receptor polyvalency or chip heterogeneity. Pseudo-Ka values (denoted Ka*) were determined by fitting a conformational change model with a refractive index (RI) change fixed at 0 RU. These values for selected variant antibodies are plotted in Fig. 8. The relative affinity of each variant compared to parent IgG was calculated according to the equation Fold = (Ka* wild type / Ka* variant). Relative binding data for all Fc variants in the IgG1 Fc region are shown in Fig. 9, and binding data for variants in antibodies with the IgG2 Fc region (IgG1 and IgG1/2 constant chains) are shown in Fig. 10. For many variants, the binding experiment was repeated multiple times (n-fold), for the experiment, the multiplicity was calculated relative to the initial wild-type IgG within each specific binding experiment. By averaging these data, the mean and standard deviation were obtained, as shown in Figs. 9 and 10.

Figures 9 and 10 show that a number of the generated variants bind with greater affinity for human FcRn binding at pH 6.0 compared to wild-type IgG1. Improvements were largely dependent on the identity of the substitution at a given position. For example, using 2-fold as the criteria for improved binding, a number of mutations at position 434 in IgG2 increased affinity (A, S, Y, F, and W), some were neutral (within 2-fold of wild-type IgG2) (G, H, M, and T), and a number decreased affinity (fold < 0.5) (D, E, K, P, R, and V). Greater binding in IgG1 did not necessarily translate into greater binding in IgG2 (e.g., 434T had improved binding in IgG1 but not IgG2). Furthermore, the improvements due to individual variants were not always additive when combined. This is demonstrated graphically in Fig. 11a by plotting the experimental fold binding of FcRn by selected double substitution variants against the fold binding of FcRn by the individual single variants that comprise them. The straight line represents the absolute additivity, i.e., the value that is expected or inferred from the results of the individual substitutions. A number of double variants fall on or near this line (259I/319I, 259I/428L, 319I/428L, and 308F/428L). Some variants have lower additivity (319I/308F, 252Y/428L, and 428L/434M). For these variants, particularly the last two (252Y/428L and 428L/434M), the affinity increases of the single substitutions could be incompatible with each other when combined. Surprisingly, the affinity increases of the 259I/308F and 428L/434S variants for FcRn were larger than expected from the affinity values of their respective single substitutions. These particular single substitutions had unexpected synergistic increases when combined. The differences between the experimental affinities and the affinities predicted from the affinities of the single variants are plotted in Fig. 11b, with the variants grouped according to the constituent single variants (259I, 308F, and 319I on the left side, and combinations containing 482L on the right side). Synergy can be quantified by calculating the fold of the experimental value relative to the predicted value, then standardizing to 1 and converting to a percentage (% synergy = 100×[(experimental fold/predicted fold) – 1)]. This analysis is plotted in Fig. 11b, with the variants grouped according to the single variants they comprise. The plot again indicates synergy among some variants, particularly 259I/308F and 428L/434S. Figs. 11b and 11c also highlight the unpredictable nature of combining many of the best single substitutions in the screen. For example, while the combination of 428L with 434S and 259I resulted in synergistic improvements in binding, either 252Y or 434M had a negative result when combined with 428L. The significant difference between the combination of 428L with 434S and the combination with 434M also indicates the importance of the specific amino acid identity of the substitution at a particular position.

EXAMPLE 4. Testing variants in cases of other antibodies

Selected variants were generated for antibodies targeting other antigens, including TNF (TNFα), CD25 (TAC), EGFR, and IgE. Figure 4 provides the amino acid sequences of the VH and VL regions of antibodies targeting these antigens that were used in accordance with the present invention. The wild-type anti-TNF antibodies and the Fc variant anti-TNF antibodies comprise the variable region of the fully human antibody adalimumab (Humira®), currently approved for the treatment of rheumatoid arthritis (RA), juvenile idiopathic arthritis (JIA), psoriatic arthritis (PsA), ankylosing spondylitis (AS), and Crohn's disease (CD). Wild-type anti-CD25 and Fc variant anti-CD25 antibodies are humanized versions of the anti-TAC antibody (Junghans et al., 1990, Cancer Research 50:1495-1502), designated H1.8/L1 anti-TAC. Wild-type anti-EGFR and Fc variant anti-EGFR antibodies are humanized versions of the mouse antibody, C225, designated H4.42/L3.32 C225. Finally, wild-type anti-IgE and Fc variant anti-IgE antibodies contain the variable region of the humanized antibody omalizumab (Xolair®), approved for the treatment of allergic asthma.

Wild-type and variant antibodies were generated, expressed, and purified as described above. Antibodies were tested for binding to human FcRn at pH 6.0 using Biacore as described above. Relative binding data for the anti-TNF, -CD25, -EGFR, and -IgE variant antibodies to human FcRn are shown in Fig. 12. As can be seen, these variants increase the affinity for FcRn for antibodies targeting multiple antigens.

EXAMPLE 5. Pharmacokinetic experiments in mice with human FcRn knocked in

To test the ability of the selected variants to increase half-life in vivo, pharmacokinetic experiments were performed in B6 mice, which are homozygous knock-outs for mouse FcRn and heterozygous knock-ins for human FcRn (mFcRn ^-/- , hFcRn ⁺ ) (Petkova et al., 2006, Int Immunol 18(12): 1759-69, incorporated by reference in its entirety), herein referred to as "hFcRn or hFcRn ⁺ mice"). Groups of 4-7 female mice, randomized by body weight (range 20-30 g), received a single intravenous injection of anti-VEGF antibody (2 mg/kg) in the tail vein. Blood (~50 µl) was collected from the orbital plexus at specific time points, converted to serum, and stored at -80°C until analysis. The duration of the studies was 28 or 49 days. No animals were harmed during these studies.

Antibody concentrations were determined using two enzyme-linked immunosorbent assays (ELISA). In the first two studies (referred to as Study 1 and Study 2), goat anti-human Fc antibody (Jackson Immuno research) was attached to the plate, the wells were washed with PBST (phosphate buffered saline with 0.05% Tween) and blocked with 3% BSA in PBST. Serum or calibration standards were then added, followed by a wash with PBST, addition of europium-labeled anti-human IgG (Perkin Elmer) and a subsequent wash with PBST. Time-resolved fluorescence signal was obtained. For Studies 3–5, serum concentrations were detected using a similar enzyme-linked immunosorbent assay (ELISA), but recombinant VEGF (VEGF-165, PeproTech, Rocky Hill, NJ) was used as the binding reagent and detection was performed with biotinylated anti-human kappa antibody and europium-labeled streptavidin. PK parameters were determined for individual mice in a non-compartmental model using WinNonLin (Pharsight Inc, Mountain View CA). Nominal time and dose were used with the same time point comparisons. Time points (lambda Z bands) used ranged from 4 days to the end of the study, although all time points were used to allow for more rapid clearance of the mutants, P257N and P257L.

Five PK studies of antibodies were conducted in mFcRn ^-/- hFcRn ⁺ mice. Figure 13 shows the serum concentration data for the wild-type and IgG1 (Study 3) and IgG2 (Study 5) variant antibodies, respectively. The fitted PK parameters of all in vivo PK studies conducted in mFcRn ^-/- hFcRn ⁺ mice are shown in Figure 14. The PK data include half-life, which is the beta phase characterizing the elimination of antibody from serum, Cmax, which is the maximum observed serum concentration, AUC, which is the area under the concentration-time curve, and clearance, which is the clearance of antibody from serum. Also given for each variant is the calculated fold increase or decrease in half-life relative to the parent IgG1 or IgG2 antibody [fold half-life = half-life (variant) / half-life (wild type)].

These data show that a series of engineered Fc variant antibodies with enhanced affinity for FcRn at pH 6.0 have increased in vivo half-lives. Figure 15a shows a plot of in vivo half-life versus fold FcRn binding for IgG1 antibodies to selected labeled variants. Results from replicate experiments (circled in the figure) show that these in vivo models are reproducible. The best single variants include 308F and 434S, the best double variants include 259I/308F, 308F/428L, 308F/434S, and 428L/434S, and the best triple variant is 259I/308F/428L. There is a general correlation between affinity for FcRn and in vivo half-life, but it is not completely predictable. In particular, the 257L and 257N variants, which increased FcRn binding by 3.4- and 3.5-fold, respectively, reduced the in vivo half-life by 0.6 and 0.3, respectively. The graph also again highlights the importance of amino acid identity at a particular position – while 308F/434S resulted in a significant increase in half-life, 308F/434M was only slightly better compared to wild-type IgG1.

Figure 15b shows a plot of in vivo half-life versus fold FcRn binding for the variant IgG2 antibodies with labeled variants. When the IgG2 in vivo data were compared with the IgG1 in vivo data (Figure 15c), an unexpected result was observed. The variants resulted in substantially greater increases in in vivo half-life in the IgG2 Fc region than in the IgG1 Fc region. The longest half-lives of the single and double variants of all antibodies across all 5 studies were 12.2 and 16.5, attributed to 434S IgG2 and 428L/434S IgG2, respectively. The dramatic increases in half-life for IgG2 variants compared to IgG1 occurred despite the fact that the fold increases by the variants in IgG2 were comparable to or even lower than in IgG1 (434S IgG1 fold = 3.8, 434S IgG2 fold = 4.9, 428L/434S IgG1 fold = 17.3, 428L/434S IgG2 fold = 14.8). Thus, unexpectedly, IgG2 antibody may represent a better application for Fc variants to increase half-life in vivo in mammals.

EXAMPLE 6. Variant immunoadhesins

The Fc variants of the present invention were also evaluated for their ability to increase the half-life of immunoadhesins (also referred to as "Fc fusions"). Selected Fc variants were generated in the anti-TNF immunoadhesin, etanercept (Enbrel®). Etanercept is a fusion of human TNF receptor 2 (TNF RII) and a human IgG1 Fc region, and is clinically approved for the treatment of rheumatoid arthritis, juvenile idiopathic arthritis, ankylosing spondylitis, psoriatic arthritis, and psoriasis. An IgG2 Fc region variant of this Fc fusion was also generated, and accordingly, selected Fc variants were also generated. The amino acid sequences of the anti-TNF immunoadhesins characterized in accordance with the present invention are shown in Fig. 16. Genes were generated using recursive PCR and subcloned into the pTT5 vector, and Fc variants were generated using QuikChange® mutagenesis methods. Immunoashesins were expressed in 293E cells and purified as described above.

The binding specificity of the purified immunoadhesins was confirmed by testing binding to recombinant TNF using Biacore. Immunoadhesins were immobilized on a CM5 biosensor chip (Biacore) with immobilized Protein A/G (Pierce) prepared using a standard combination of primary amines. Immunoadhesins were immobilized on the Protein A/G surface and serial dilutions of recombinant TNF were introduced onto the antibody-bound surface, followed by a dissociation phase. After each cycle, the surface was regenerated with buffer. Data were processed by zeroing the time and reaction before receptor introduction and by subtracting from the baseline data to account for changes due to introductions. Kinetic data were fitted to a 1:1 binding model (Langmuir). The equilibrium association constants (Ka) obtained from these fits are shown in Fig. 17. These results indicate that the variant immunoadhesins retained affinity for TNF comparable to the commercially available Enbrel.

The variant immunoadhesins were tested for binding to human FcRn at pH 6.0 using Biacore as described above. The results (Fig. 18) show that, as with the antibodies, the variants improve binding to FcRn compared to their parental IgG1 and IgG2 immunoadhesin proteins.

The half-lives of the variant immunoadhesins were tested in mFcRn ^−/− hFcRn ⁺ mice as described above. Twelve mice per group were injected with 2 mg/kg of the variant and parental IgG1 immunoadhesin. Serum concentration was determined using an enzyme-linked immunosorbent assay (ELISA) similar to that described above, except that goat anti-human TNF RII antibody was used as the binding reagent; detection was accomplished with biotinylated anti-human kappa antibody and europium-labeled streptavidin. Figure 19 shows the serum concentration data for wild-type Fc IgG1 and variant Fc immunoadhesins. The adjusted PK parameters as described above obtained from the PK study are shown in Figure 19. 20. Also given for each variant is the calculated % increase in half-life, calculated as the half-life of the variant Fc hybrid relative to the half-life of the parental wild-type IgG1 Fc multiplied by 100. These results indicate that the variants increase the half-life in vivo for the immunoadhesin.

EXAMPLE 7. Pharmacokinetic experiment on lower primates

The PK properties of biologics in nonhuman primates are well established, making them predictable in humans. A PK study was performed in cynomolgus macaques (macaca fascicularis) to assess the ability of variant anti-VEGF antibodies to increase serum half-life in nonhuman primates.

In preparation for the PK study, binding of variant antibodies to cynomolgus monkey FcRn (cFcRn) was measured in cynomolgus monkeys at pH 6.0. cFcRn was generated, expressed, and purified as described above for human FcRn. Binding of the anti-VEGF variant antibodies to cFcRn was measured using Biacore as described above. The data are shown in Fig. 21. These results indicate that the variants increase affinity for cyno FcRn in a manner similar to that for human FcRn. Dissociation at higher pH (7.4) was also very rapid (data not shown), similar to that observed for binding to human FcRn. These results are not unexpected, given the high sequence homology between the human and cyno receptors (FcRn alpha chain 96%, beta-2-microglobulin 91%).

The pharmacokinetic characteristics of the variants were studied in nonhuman primates in vivo. Male cynomolgus macaques (macaca fascicularis, also called crab-eating macaques) weighing 2.3–5.1 kg were randomized by weight into 5 groups of 3 monkeys per group. The monkeys received a single 1-hour infusion of 4 mg/kg antibody into a peripheral vein. Blood samples (1 ml) were collected from another vein 5 minutes to 90 days after completion of the infusion, converted to serum, and stored at -70C. No animals were harmed during these studies.

Antibody concentrations were determined using the VEGF capture assay as described above. PK parameters were determined by fitting concentration-time relationships to a non-compartmental model as performed in the mouse PK studies. However, the day 10–day 90 time points were used for PK parameter determinations. The PK results are plotted in Fig. 22 and the fitted parameters are shown in Fig. 23. These results show that the variants increased the in vivo half-life of the antibody by up to 3.2-fold. In the best case (428L/434S variant), the half-life was increased from 9.7 days to 31.1 days. The PK results obtained in cynomolgus monkeys are comparable to those obtained in mFcRn ^−/− hFcRn ⁺ mice, validating the hFcRn mouse model as a system for determining the PK properties of variants in vivo and corroborating the results of these studies.

While specific embodiments of the present invention have been described above for illustrative purposes, it will be apparent to those skilled in the art that numerous variations may be made within the scope of the present invention, which is defined by the appended claims. The contents of all references cited are herein incorporated in their entirety.

Claims (13)

1. Anti-TNF antibody comprising: - two identical heavy chain polypeptides, each having the amino acid sequence of SEQ ID NO: 29; and - two identical light chain polypeptides, each of which has the amino acid sequence of SEQ ID NO: 23. 2. A nucleic acid for producing an anti-TNF antibody according to claim 1, comprising: - a first polynucleotide encoding a heavy chain polypeptide having the amino acid sequence of SEQ ID NO: 29; and/or - a second polynucleotide encoding a light chain polypeptide having the amino acid sequence of SEQ ID NO: 23. 3. An expression vector containing the nucleic acid according to item 2. 4. An expression vector system comprising two expression vectors according to claim 3, where: - the first expression vector comprises a first polynucleotide encoding a heavy chain polypeptide having the amino acid sequence of SEQ ID NO: 29; and - the second expression vector comprises a second polynucleotide encoding a light chain polypeptide having the amino acid sequence of SEQ ID NO: 23. 5. A host cell for producing an anti-TNF antibody according to claim 1, comprising a nucleic acid according to claim 2, wherein said nucleic acid contains: - a first polynucleotide encoding a heavy chain polypeptide having the amino acid sequence of SEQ ID NO: 29; and - a second polynucleotide encoding a light chain polypeptide having the amino acid sequence of SEQ ID NO: 23.