Heterodimeric proteins

Abstract

To provide heterodimeric antibodies that solve co-engagement of antigens in a multivalent manner which is a drawback of bispecific antibodies built with full length antibody-like formats.SOLUTION: In one aspect, the present invention discloses heterodimeric antibodies each comprising a first monomer comprising a first heavy chain constant domain comprising a first variant Fc domain and a first antigen-binding domain, and a second monomer comprising a second heavy chain constant domain comprising a second variant Fc domain and a second antigen-binding domain. In an additional aspect the heterodimeric antibody comprises a first monomer comprising a heavy chain comprising a first Fc domain and a single chain Fv region (scFv) that binds a first antigen, where the scFv comprises a charged scFv linker. The heterodimeric antibody further comprises a second monomer comprising a first heavy chain comprising a second Fc domain and a first variable heavy chain and a first light chain.SELECTED DRAWING: Figure 1-1

Description

Technical Field

[0001] Priority This application is a partial continuation application of International Patent Application PCT / US14 / 11549 filed on January 14, 2014, U.S. Patent Application Publication 14 / 155,334 filed on January 14, 2014, 14 / 205,248 filed on March 11, 2014, and 14 / 207,489 filed on March 12, 2014. Further, this application claims priority to U.S. Patent Application Publications 61 / 818,513 filed on May 1, 2013, 61 / 818,344 filed on May 1, 2013, 61 / 794,896 filed on March 15, 2013, 61 / 818,401 filed on May 1, 2013, 61 / 913,879 filed on December 9, 2013, 61 / 913,832 filed on December 9, 2013, 61 / 938,095 filed on February 10, 2014, and 61 / 913,870 filed on December 9, 2013, all of which are hereby expressly incorporated by reference in their entirety, particularly with respect to the drawings and related descriptions disclosed therein, and amino acid mutations.

Background Art

[0002] Success has been achieved in treating various diseases including cancer and autoimmune / inflammatory diseases using antibody-based therapeutic agents. However, there is still a need for further improvement in this type of agent, particularly with regard to enhancing clinical efficacy. One approach being studied is to engineer additional novel antigen-binding sites into antibody-based drugs such that one immunoglobulin molecule co-captures two different antigens. Such non-natural or alternative antibody formats that capture two different antigens are generally referred to as bispecific antibodies. Since the antibody variable region (Fv) has significant diversity, it is in fact possible to generate Fvs that recognize virtually any molecule, and a typical method for producing bispecific antibodies is the introduction of new variable regions into antibodies.

[0003] Regarding bispecific targeting, many alternative antibody formats have been developed (Chames & Baty, 2009, mAbs 1[6]:1-9; Holliger & Hudson, 2005, Nature Biotechnology 23[9]:1126-1136; Kontermann, mAbs 4(2):182 (2012). All of these are hereby incorporated by reference). Initially, bispecific antibodies were produced by fusing two cell lines, each producing one monoclonal antibody (Milstein et al., 1983, Nature 305:537-540). The resulting hybrid hybridomas or quadromas produced bispecific antibodies, but only a small number of groups, and extensive purification was required to isolate the desired antibody. A solution to this was the use of antibody fragments to produce bispecific antibodies. Since such fragments lack the complex quaternary structure of full-length antibodies, the variable light and heavy chains can be linked to one gene construct. Many different forms of antibody fragments have been produced, such as diabodies, single-chain diabodies, tandem scFvs, and Fab2 bispecific antibodies (Chames & Baty, 2009, mAbs 1[6]:1-9; Holliger & Hudson, 2005, Nature Biotechnology 23[9]:1126-1136. Explicitly incorporated by reference herein). These formats can be highly expressed in bacteria and are small in size, so they have the beneficial advantage of penetration, but are rapidly removed in vivo and have manufacturing difficulties related to production and stability. The main cause of these drawbacks is usually that the antibody fragments lack the constant region of the antibody related to its functional properties (increasing its size, enhancing its stability, and maintaining a long serum half-life (i.e., neonatal Fc receptor FcRn), or serving as a binding site for purification (i.e., protein A and protein G), binding to various Fc receptors and ligands).

[0004] More recently, attempts have been made to address the drawbacks of fragment-based bispecific antibodies by engineering double bonds into full-length antibody-like formats (Wu et al., 2007, Nature Biotechnology 25

[11] :1290-1297; USSN12 / 477,711; Michaelson et al., 2009, mAbs 1[2]:128-141; PCT / US2008 / 074693; Zuo et al., 2000, Protein Engineering 13[5]:361-367; USSN09 / 865,198; Shen et al., 2006, J Biol Chem 281

[16] :10706-10714; Lu et al., 2005, J Biol Chem 280

[20] :19665-19672; PCT / US2005 / 025472; all of which are hereby expressly incorporated by reference). Some of the drawbacks of antibody fragment bispecific antibodies have been overcome by these formats (mainly by containing an Fc region). However, a significant difficulty with these forms is that the binding to a new antigen is always bivalent in order to build a new antigen-binding site at the apex of the homodimeric constant chain.

[0005] For many antigens that are attractive as co-targets in therapeutic bispecific formats, the desired binding is monovalent rather than bivalent. For many immunoreceptors, cell activation is effected by cross-linking of monovalent binding interactions. The mechanism of cross-linking is often regulated by antibody / antigen immune complexes or through the binding of target cells to effector cells. For example, low-affinity Fcγ receptors (FcγR, e.g., FcγRIIa, FcγRIIb, and FcγRIIIa) bind monovalently to the antibody Fc region. Monovalent binding does not activate cells expressing these FcγRs, but when immune complexes are formed or when cell-to-cell contact occurs, the receptors are cross-linked and clustered on the cell surface, thereby leading to activation. For receptors involved in the regulation of cell killing, for example, FcγRIIIa on natural killer (NK) cells undergoes receptor cross-linking and cell activation when effector cells capture target cells more strongly (Bowles & Weiner, 2005, J Immunol Methods 304:88-99, which is expressly incorporated by reference). Similarly, on B cells, the inhibitory receptor FcγRIIb downregulates B cell activation only when forming immune complexes with the B cell receptor (BCR) on the cell surface, and the mechanism is regulated by immune complexes of soluble IgG and the same antigen recognized by the BCR (Heyman 2003, Immunol Lett 88[2]:157-161; Smith and Clatworthy, 2010, Nature Reviews Immunology 10:328-343, which are expressly incorporated by reference). As another example, CD3 activation of T cells occurs only when the associated T cell receptor (TCR) captures MHC carrying antigen on antigen-presenting cells in a very strong cell-to-cell synapse (Kuhns et al., 2006, Immunity 24:133-139).In fact, non-specific bivalent cross-linking of CD3 using anti-CD3 antibodies induces cytokine storm and toxicity (Perruche et al., 2009, J Immunol 183[2]:953-61; Chatenoud & Bluestone, 2007, Nature Reviews Immunology 7:622-632, which are hereby incorporated by reference). Therefore, for actual clinical applications, the preferred mode of co-capture of CD3 for target cell killing conversion is monovalent binding, whereby activation will only occur when the co-captured target is captured.

[0006] Therefore, while bispecific antibodies made from antibody fragments have biophysical and pharmacokinetic hurdles, the drawback of those made from full-length antibody-like forms is that in the absence of the major target antigen, they will co-capture antigens in a multivalent manner, thereby causing non-specific activation and the potential for toxicity. The present invention solves this problem by introducing a new set of bispecific forms that can co-capture different target antigens in a multivalent manner. Furthermore, the present invention discloses novel heterodimeric mutations that make the formation and purification of heterodimeric proteins containing antibodies more possible.

Summary of the Invention

[0007] In one aspect, the present invention discloses a heterodimeric antibody containing a first monomer containing a first heavy chain constant domain containing a first mutated Fc domain and a first antigen-binding domain, and a second monomer containing a second heavy chain constant domain containing a second mutated Fc domain and a second antigen-binding domain.

[0008] In a further aspect, the heterodimeric antibody contains a first monomer containing a heavy chain containing a single-chain Fv region (scFv) that binds to a first Fc domain and a first antigen, where the scFv contains a charged scFv linker. The heterodimeric antibody further contains a second monomer containing a second Fc domain and a second heavy chain containing a first variable heavy chain and a first light chain. In a further aspect, this charged linker has either 3 to 8 positive charges or 3 to 8 negative charges and is selected from the group consisting of the linkers shown in FIG. 9.

[0009] In a further aspect, according to the present invention, a first monomer containing a first mutated Fc domain compared to the human Fc domain; a first heavy chain sequence containing a first antigen-binding domain that binds to a first antigen; and a second mutated Fc domain compared to the human Fc domain; and a second heavy chain sequence containing a second antigen domain that binds to a second antigen, is provided, where the first and second Fc domains contain a set of amino acid substitutions selected from the group consisting of the amino acid sets shown in FIG. 3, and a heterodimeric antibody composition is disclosed.

[0010] In a further aspect, according to the present invention, a heterodimeric antibody composition is disclosed that contains a first monomer containing a first heavy chain sequence containing a first mutated Fc domain compared to the human Fc domain; a first antigen-binding domain that binds to a first antigen; a second heavy chain sequence containing a second mutated Fc domain compared to the human Fc domain; and a second antigen-binding domain that binds to CD19. The second antigen-binding domain contains a variable heavy chain domain containing the amino acid sequence of H1.227 (SEQ ID NO: X) and a variable light chain selected from the group consisting of the amino acid sequence of L1.198 (SEQ ID NO: X) and the amino acid sequence of 1.199 (SEQ ID NO: X) shown in FIG. 21.

[0011] In a further aspect, the present invention provides a first antigen-binding domain containing a hyper CD3 variable region having a sequence containing a first mutated Fc domain, a vhCDR1 having the sequence T-Y-A-M-Xaa1 (where Xaa1 is N, S, or H (SEQ ID NO: 435)), a vhCDR2 having the sequence R-I-R-S-K-Xaa1-N-Xaa2-Y-A-T-Xaa3-Y-Y-A-Xaa4-S-V-K-G (where Xaa1 is Y or A, Xaa2 is N or S, Xaa3 is Y or A, and Xaa4 is D or A (SEQ ID NO: 436)), a vhCDR3 having the sequence H-G-N-F-G-Xaa1-S-Y-V-S-W-F-Xaa2-Y (where Xaa1 is N, D, or Q, and Xaa2 is A or D (SEQ ID NO: 437)), a vlCDR1 having the sequence Xaa1-S-S-T-G-A-V-T-Xaa2-Xaa3-Xaa4-Y-A-N (where Xaa1 is G, R, or K, Xaa2 is T or S, Xaa3 is S or G, and Xaa4 is N or H (SEQ ID NO: 438)), a vlCDR2 having the sequence Xaa1-T-N-Xaa2-R-A-Xaa3 (where Xaa1 is G or D, Xaa2 is K or N, and Xaa3 is P or S (SEQ ID NO: 439)) and a vlCDR3 having the sequence Xaa1-L-W-Y-S-N-Xaa2-W-V (where Xaa1 is A or L, and Xaa2 is L or H (SEQ ID NO: 440)), a first monomer containing a first heavy chain sequence containing the first antigen-binding domain, and a heterodimeric antibody composition. The heterodimeric antibody further contains a second mutated Fc domain as compared to the human Fc domain; and an anti-C19 antigen-binding domain containing a variable heavy chain domain containing the amino acid sequence of H1.227 (SEQ ID NO: X), and a variable light chain selected from the group consisting of the amino acid sequence of L1.198 (SEQ ID NO: X) and the amino acid sequence of 1.199 (SEQ ID NO: X) shown in FIG. 21, and a second monomer containing a second heavy chain sequence containing the second antigen-binding domain.

[0012] In a further aspect, the present invention provides a heterodimeric antibody comprising: a heavy chain containing a first mutated Fc domain; and a first monomer containing a single-chain Fv region (scFv) that binds to a first antigen, wherein the scFv contains a charged linker; and a second monomer containing a second mutated Fc domain and a first heavy chain containing a first variable heavy chain, the second monomer also containing a first light chain, wherein the first and second mutated Fc domains contain amino acid substitutions (including pluralities) selected from the group consisting of those shown in FIG. 7.

[0013] In a further aspect, the present invention provides a heterodimeric antibody composition comprising a first monomer containing a first antigen-binding domain containing a vhCDR1 having the sequence T-Y-A-M-Xaa1 (where Xaa1 is N, S, or H (SEQ ID NO: 435)), a vhCDR2 having the sequence R-I-R-S-K-Xaa1-N-Xaa2-Y-A-T-Xaa3-Y-Y-A-Xaa4-S-V-K-G (where Xaa1 is Y or A, Xaa2 is N or S, Xaa3 is Y or A, and Xaa4 is D or A (SEQ ID NO: 436)), a vhCDR3 having the sequence H-G-N-F-G-Xaa1-S-Y-V-S-W-F-Xaa2-Y (where Xaa1 is N, D, or Q, and Xaa2 is A or D (SEQ ID NO: 437)), a vlCDR1 having the sequence Xaa1-S-S-T-G-A-V-T-Xaa2-Xaa3-Xaa4-Y-A-N (where Xaa1 is G, R, or K, Xaa2 is T or S, Xaa3 is S or G, and Xaa4 is N or H (SEQ ID NO: 438)), a vlCDR2 having the sequence Xaa1-T-N-Xaa2-R-A-Xaa3 (where Xaa1 is G or D, Xaa2 is K or N, and Xaa3 is P or S (SEQ ID NO: 439)) and a vlCDR3 having the sequence Xaa1-L-W-Y-S-N-Xaa2-W-V (where Xaa1 is A or L, and Xaa2 is L or H (SEQ ID NO: 440)). The first monomer also contains a first heavy chain sequence containing a first mutated Fc domain relative to the human Fc domain. The heterodimeric antibody also contains a second monomer containing a second antigen-binding domain; and a second heavy chain sequence containing a second mutated Fc domain relative to the human Fc domain, wherein the first and second mutated Fc domains have different amino acid sequences.In some embodiments, the anti-CD3 variable region comprises a vhCDR1 having the sequence T-Y-A-M-Xaa1 (where Xaa1 is N, S, or H (SEQ ID NO: 435)), a vhCDR2 having the sequence R-I-R-S-K-Xaa1-N-Xaa2-Y-A-T-Xaa3-Y-Y-A-Xaa4-S-V-K-G (where Xaa1 is Y or A, Xaa2 is N or S, Xaa3 is Y or A, and Xaa4 is D or A (SEQ ID NO: 436)), a vhCDR3 having the sequence H-G-N-F-G-Xaa1-S-Y-V-S-W-F-Xaa2-Y (where Xaa1 is N, D, or Q, and Xaa2 is A or D (SEQ ID NO: 437)), a vlCDR1 having the sequence Xaa1-S-S-T-G-A-V-T-Xaa2-Xaa3-Xaa4-Y-A-N (where Xaa1 is G, R, or K, Xaa2 is T or S, Xaa3 is S or G, and Xaa4 is N or H (SEQ ID NO: 438)), a vlCDR2 having the sequence Xaa1-T-N-Xaa2-R-A-Xaa3 (where Xaa1 is G or D, Xaa2 is K or N, and Xaa3 is P or S (SEQ ID NO: 439)), and a vlCDR3 having the sequence Xaa1-L-W-Y-S-N-Xaa2-W-V (where Xaa1 is A or L, and Xaa2 is L or H (SEQ ID NO: 440)).

[0014] In a further aspect of the invention, there is disclosed a heterodimeric protein comprising a first monomer containing a first mutated heavy chain constant region and a first fusion partner; and a second monomer containing a second mutated heavy chain constant region and a second fusion partner, wherein the Fc regions of the first and second constant regions contain a set of amino acid substitutions from FIGS. 3 and 12. In some examples, the first monomer contains a third fusion partner, and optionally, the second monomer contains a fourth fusion partner. The fusion partners are independently selected from the group consisting of immunoglobulin components, peptides, cytokines, chemokines, immune receptors, and blood factors. In some examples, the immunoglobulin component is selected from the group consisting of Fab, VH, VL, scFv, scFv2, dAb.

[0015] In many embodiments, one of the first and second mutated Fc domains contains an amino acid substitution(s) selected from the group consisting of those shown in FIGS. 6, 7 and / or 12. In some embodiments, the first antigen-binding domain is an scFv covalently attached to the first heavy chain constant domain. In further embodiments, the heterodimeric antibody has a structure selected from the structures of FIGS. 1B-1L and 2A-2M. In further embodiments, the first and / or second Fc domain of the heterodimeric antibody further contains an amino acid substitution(s) selected from the group consisting of 434A, 434S, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L, 252Y, 252Y / 254T / 256E, 259I / 308F / 428L, 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D, 332E / 330L, 236R, 328R, 236R / 328R, 236N / 267E, 243L, 298A and 299T. In some embodiments, one of the first and second mutated Fc domains contains the amino acid substitution 364K / E357Q and the other contains the amino acid substitution 368D / 370S. These antibodies may further contain an amino acid substitution(s) selected from the group listed in FIG. 7.

[0016] In a further aspect of the invention, nucleic acids, expression vectors and host cells for producing the heterodimeric proteins and antibodies of the invention are disclosed.

[0017] In a further aspect of the invention, a method for producing a heterodimeric protein of the invention is disclosed by culturing a host cell containing a nucleic acid encoding the heterodimeric protein and antibody of the invention under conditions such that the heterodimer is produced and recovered.

[0018] In a further aspect of the invention, there is provided a first nucleic acid encoding a first heavy chain containing a single-chain Fv region (scFv) that binds to a first antigen and a first Fc domain, wherein the scFv contains a charged linker; and a second nucleic acid encoding a second heavy chain containing a second Fc domain and a first variable heavy chain; and a third nucleic acid containing a light chain, a method for producing a heterodimeric antibody of the invention is disclosed. The method further comprises expressing the first, second, and third nucleic acids in a host cell, respectively, to produce the first, second, and third amino acid sequences, loading the first, second, and third amino acid sequences onto an ion exchange column; and collecting the heterodimeric fraction.

[0019] In a further aspect, the invention discloses a method of treating an individual in need thereof by administering the heterodimeric antibody or protein of the present specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Heterodimerization Form and Mutation

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

Figure 27

Figure 28

Figure 29

Figure 30

Figure 31

Figure 32

Figure 33

Figure 34

Figure 35

Figure 36

Figure 37

Figure 38

Figure 39

Figure 40

Figure 41

Figure 42

Figure 43

Figure 44

Figure 45

Figure 46

Figure 47

Figure 48

Figure 49

Figure 50

Figure 51

Figure 52

Figure 53

Figure 54

Figure 55

Figure 56

Figure 57

Modes for Carrying Out the Invention

[0021] Figures 66 - 80 of WO / 2013 / 055809 are sequences and associated descriptions, which are specifically incorporated herein by reference. Figures 2 - 111 of USSN 13 / 648,951 and their descriptions are specifically incorporated herein by reference.

[0022] I. Overview of Heterodimeric Proteins The present invention is directed to novel constructs that result in heterodimeric proteins capable of binding two or more antigens or ligands (e.g., capable of multivalent binding). The heterodimeric protein constructs are based on the property of two Fc domains of an antibody heavy chain to self - assemble (e.g., two "monomers" assemble into a "dimer"). The heterodimeric proteins are made by modifying the amino acid sequences of each monomer, as detailed below. Thus, the present invention generally aims at the production of heterodimeric proteins containing antibodies that can co - capture antigens in several ways, relying on amino acid mutations in the constant regions that are different for each chain, to promote heterodimer formation and / or to facilitate purification of the heterodimer over the homodimer. As detailed below, the heterodimeric proteins may be antibody variants or may be based on Fc fusion proteins. Generally, the focus of the present invention is on heterodimeric antibodies, but as will be apparent to those skilled in the art and as detailed below, the present invention is equally applicable to heterodimeric proteins.

[0023] Accordingly, the present invention discloses bispecific antibodies (or trispecific or tetra-specific antibodies, as will be discussed below, may also be produced). The current problems in antibody technology are the need for "bispecific (and / or multispecific)" antibodies that can bind simultaneously to two (or more) different antigens, thereby normally bringing different antigens into proximity, and as a result, enabling new functionality and new therapies. Generally, these antibodies are produced by introducing each of the genes for the heavy and light chains into a host cell. This typically results in the formation of the desired heterodimer (A-B), as well as two homodimers (A-A and B-B). However, the main obstacle to multispecific antibody formation is the difficulty of purifying the heterodimer antibody from the homodimer antibodies and / or of producing more heterodimers than homodimers.

[0024] There are many mechanisms that can be used to form the heterodimers of the present invention. Furthermore, as recognized by those skilled in the art, these mechanisms can be combined to ensure heterodimerization. Accordingly, amino acid mutations that result in the production of heterodimers are referred to as "heterodimerization mutations". As will be discussed below, heterodimerization mutations include conformational mutations (e.g., the "knobs and holes" or "twist" mutations, and "charge pair" mutations disclosed below), as well as "pI mutations" that enable the purification of homodimers from heterodimers.

[0025] One mechanism, commonly referred to in the art as "knobs and holes" (KIH), or in some cases, amino acid manipulations that have a conformational effect of increasing heterodimer formation and decreasing homodimer formation, may also optionally be referred to herein as "twist" mutations; this is, in some cases, referred to as "knobs and holes" (USSN 61 / 596,846, and USSN 12 / 875,0015, Ridgway et al., Protein Engineering 9(7):617 (1996); Atwell et al., J. Mol. Biol. 1997 270:26; U.S. Patent No. 8,216,805, US 2012 / 0149876, all of which are hereby incorporated by reference in their entirety). The drawings identify many "monomer A - monomer B" pairs containing "knobs and holes" amino acid substitutions. Further, as disclosed in Merchant et al., Nature Biotech. 16:677 (1998), these "knobs and holes" mutations can, in combination with disulfide bonds, twist the structure into a heterodimer.

[0026] Additional mechanisms that can be used to effect heterodimer formation are, in some cases, those referred to as "electrostatic steering" or "charge pairing" (disclosed in Gunasekaran et al., J. Biol. Chem. 285(25):19637 (2010), which is hereby incorporated by reference in its entirety). These are, in some cases, referred to herein as "charge pairing." In this embodiment, electrostatics is used to twist the structure towards heterodimer formation. Although known in the art, these may also have an effect on the pI and thus on purification, and in some cases may also be considered pI mutations. However, since they are engineered to promote heterodimerization and not used as tools for purification, they are classified as "conformational mutations." These include, but are not limited to, D221E / P228E / L368E paired with D221R / P228R / K409R (for example, these are a monomeric pair set) and C220E / P228E / 368E paired with C220R / E224R / P228R / K409R and others shown in the figures.

[0027] In some embodiments of the invention, pI mutations are used to modify the pI of one or both monomers, thereby enabling isoelectric purification of the A-A, A-B, and B-B dimer proteins.

[0028] In the present invention, there are several basic mechanisms that can facilitate the purification of heterodimeric proteins; one relies on the use of pI mutations, for example, each monomer has a different pI, thereby enabling the isoelectric purification of A-A, A-B, and B-B dimer proteins. Alternatively, size-based separation is also possible, for example, by some scaffold forms such as the "triple F" form. It is also possible to "twist" the structure towards heterodimers rather than homodimers (generally outlined below). Therefore, combinations of conformational heterodimer mutations and pI or charge pair mutations find particular applications in the present invention. Furthermore, as further outlined below, scaffolds utilizing scFv(s), such as the triple F form, can contain charged scFv linkers (either positive or negative), thereby providing further enhancement of pI for purification purposes. Although known to those skilled in the art, some triple F forms are useful for charged scFv linkers and do not have further pI regulation, but the present invention similarly presents the use of twisted mutations with charged scFv linkers (and combinations of Fc, FcRn, and KO mutations).

[0029] In the present invention, which utilizes pI as a separation mechanism to enable the production of heterodimeric proteins, amino acid mutations may be introduced into one or both of the monomer polypeptides; that is, the pI of one of the monomers (simply referred to herein as "monomer A") may be manipulated separately from monomer B, or both monomers A and B may be altered (the pI of monomer A increases and the pI of monomer B decreases). As further outlined below, the change in the pI of either or both monomers may be made by removing or adding charged residues (e.g., a neutral amino acid is substituted by a positively charged amino acid residue or a negatively charged amino acid residue (e.g., glycine is substituted by glutamic acid)), by changing a positive or negative charged residue to the opposite charge (aspartic acid to lysine), or by changing a charged residue to a neutral residue (e.g., for charge disappearance, lysine to serine). Many of these mutations are shown in the figures.

[0030] Accordingly, in this embodiment of the present invention, a sufficient change is brought about in the pI of at least one of the monomers, whereby the heterodimer is separated from the homodimer. As is known to those skilled in the art and will be further considered below, this can be accomplished by using the "wild type" heavy chain constant region and variable regions (wtA-+B, or wtA--B) whose pI has been manipulated to increase or decrease, or variable regions (A+-B-, or A-B+) that have been manipulated to increase one region and decrease the other.

[0031] Thus, generally, the components of some embodiments of the present invention involve modifying the isoelectric point (pI) of at least one (if not both) of the monomers of a dimeric protein to form a "pI heterodimer" (when the protein is an antibody, they are referred to as "pI antibodies") by incorporating amino acid substitutions ("pI mutations" or "pI substitutions") into one or both of the monomers. As shown herein, the separation of a heterodimer from two homodimers can be achieved if the pIs of the two monomers differ by as little as 0.1 pH unit (0.2, 0.2, 0.4 or 0.5 or more, which are used in the present invention).

[0032] As recognized by those skilled in the art, the number of pI mutations contained in each monomer or both monomers to achieve successful separation depends in part on the starting pIs of the scFv and Fab of interest. That is, to determine which monomer(s) to manipulate and its "direction" (e.g., more positive or more negative), the Fv sequences of the two target antigens are calculated and decisions are made therefrom. As is known in the art, different Fvs have different starting pIs, which are utilized in the present invention. Generally, as described herein, the pI is manipulated such that the total pI difference between the monomers is at least about 0.1 log (preferably 0.2 - 0.5).

[0033] Furthermore, as recognized by those skilled in the art and as described herein, heterodimers can also be separated from homodimers based on size. For example, as shown in FIGS. 1 and 2, a heterodimer having two scFvs can be separated by a "triple F" format and a bispecific mAb. This can be further utilized in more multivalent cases where additional antigen-binding sites are employed. For example, as additionally shown, one monomer has two Fab fragments and the other has one scFv, thereby differing in size and thus in molecular weight.

[0034] Furthermore, as recognized in the art and as described herein, the formats outlined herein can be extended to create trispecific and tetra-specific antibodies as well. In this embodiment, some of these variations are shown in FIG. 1A, and it will be recognized that some antigens can be bound bivalently (e.g., two antigen-binding sites for one antigen, e.g., A and B are part of a typical bivalent binding, and C and D may optionally be present and may optionally be the same or different). As recognized, any combination of Fabs and scFvs may be used to obtain the desired results and combinations.

[0035] When performing heterodimerization using pI mutations, a more modular method for designing and purifying multispecific proteins containing antibodies by using the constant region(s) of the heavy chain(s) is presented. Thus, in some embodiments, the heterodimerization mutations (including kink and purification heterodimerization mutations) are not included within the variable region, such that each individual antibody does not have to be manipulated. Furthermore, in some embodiments, the potential immunogenicity resulting from pI mutations is reduced by introducing pI mutations from different IgG isotypes, whereby the pI is changed without introducing obvious immunogenicity. Thus, an additional challenge to be solved is the elucidation of low pI constant domains that contain many human sequences, e.g., minimization or avoidance of non-human residues at any given position.

[0036] Another advantage obtained by pI manipulation is the extension of serum half-life and the increase in FcRn binding. That is, as described in USSN13 / 194,904, which is incorporated herein by reference in its entirety, the serum residence in vivo can be extended by lowering the pI of the antibody constant domain (including those present in antibodies and Fc fusions). These pI mutations for increasing serum half-life also facilitate pI changes for purification.

[0037] Furthermore, it should be noted that the pI mutations of the heterodimerization mutations provide further benefits with respect to the analysis and quality control processes of bispecific antibodies (particularly in the case of CD3 antibodies, where the ability to exclude, minimize, or distinguish in the presence of homodimers is prominent). Similarly, the ability to reliably verify the reproducibility of heterodimeric protein production is important.

[0038] In addition to all or part of the mutated heavy chain constant domain, one or both monomers may contain one or two fusion partners such that the heterodimer forms a multivalent protein. As shown in the figures (particularly Figure 1A), the fusion partners are shown as A, B, C, and D, and all combinations are possible. Generally, A, B, C, and D are selected such that the heterodimer is at least bispecific or bivalent in its ability to interact with additional proteins.

[0039] As will be recognized by those skilled in the art and as discussed below, the heterodimeric fusion proteins of the present invention can assume a wide variety of structures (outlined in Figures 1 and 2). Some of the figures show a "one-terminal" structure, where one type of specificity is present on one "arm" of the molecule and a different specificity is present on the other "arm". In other figures, a "two-terminal" structure is shown, where at least one type of specificity is present on the "top" of the molecule and one or more different specificities are present on the "bottom" of the molecule. Furthermore, these two structures can be combined to have triple or quadruple specificities based on specific combinations. Thus, "multispecific" binding proteins (including multispecific antibodies) are disclosed by the present invention. Thus, the present invention is directed to novel immunoglobulin compositions that co-capture at least a first and a second antigen. The first and second antigens of the present invention are referred to as Antigen 1 and Antigen 2, respectively.

[0040] One heterodimeric scaffold particularly useful in the present invention is the "triple F" or "pull - out type" scaffold format. In this embodiment, one heavy chain of the antibody contains a single - chain Fc (hereinafter defined as "scFv"), and the other heavy chain is in the "standard" Fab format (containing a variable heavy chain and a light chain). This structure is sometimes referred to herein as the "triple F" format (scFv - FAb - Fc) or the "pull - out type" format from its resemblance to a pull - out appearance (see Figure 1B). The two chains are covalently linked together by using amino acid mutations in the constant regions (e.g., Fc domain and / or hinge region) that promote the formation of the heterodimeric antibody (detailed below).

[0041] Regarding the "triple F" format of the present invention, several different advantages exist. As is known in the art, antibody analogs relying on two scFv constructs often have problems with stability and aggregation, which can be alleviated in the present invention by adding a pair of "standard" heavy and light chains. Further, in contrast to formats relying on two heavy chains and two light chains, there is also no problem of incorrect pairing of heavy and light chains (e.g., pairing of heavy chain 1 with light chain 2, etc.).

[0042] In addition to all or part of the mutated heavy - chain constant domain, one or both of the monomers may contain one or two fusion partners, whereby the heterodimer forms a multivalent protein. As outlined in Figure 64 of USSN13 / 648,951 (incorporated herein by reference along with the accompanying specification), the fusion partners are designated as A, B, C, and D, and all combinations are possible. Generally, A, B, C, and D are selected such that the heterodimer is at least bispecific or divalent in its ability to interact with additional proteins. In the context of the "triple F" format of the present invention, typically A and B are scFv and Fv (as recognized, either monomer may contain scFv and the other may contain Fv / Fab), and then optionally one or two additional fusion partners.

[0043] Furthermore, as outlined herein, additional amino acid mutations may be introduced into the bispecific antibodies of the invention to add additional functionality. For example, amino acid changes within the Fc region may be added (to either one monomer or both monomers) to promote enhanced ADCC or CDC (e.g., modify binding to Fcγ receptors), may be added to enable the production of additional toxins and drugs (e.g., for ADCs) or to increase production levels, and may also be added to increase binding to FcRn and / or increase the serum half-life of the resulting molecule. As further detailed herein and as recognized by those skilled in the art, any and all mutations outlined herein may optionally and independently be combined with other mutations.

[0044] Similarly, another category of functional mutations is "Fcγ deletion mutations" or "Fc knockout (FcKO or KO) mutations". In these embodiments, for some therapeutic applications, it is desirable to reduce or eliminate the normal binding of the Fc domain to one or more or all Fcγ receptors (e.g., FcγR1, FcγRIIa, FcγRIIb, FcγRIIIa, etc.) to avoid additional effector mechanisms. That is, for example, in many embodiments, particularly in the use of bispecific antibodies that bind monovalently to CD3 and other tumor antigens (e.g., CD19, her2 / neu, etc.), it is typically desirable to remove FcγRIIIa binding to eliminate or significantly reduce ADCC activity.

[0045] Definition To more fully understand this use, several definitions are provided below. Such definitions are intended to include those that are grammatically equivalent.

[0046] As used herein, "loss" means a decrease or removal of activity. Thus, for example, "loss of FcγR binding" means that an amino acid mutation in the Fc region has less than 50% starting binding as compared to an Fc region that does not contain the particular mutation, preferably less than 70 - 80 - 90 - 95 - 98% loss of activity, and is typically an activity below the binding level detectable in a Biacore assay. What is particularly useful for loss of FcγR binding is shown in FIG. 7.

[0047] As used herein, "ADCC" or "antibody - dependent cell - mediated cytotoxicity" means a cell - mediated reaction, where non - specific cytotoxic effector cells expressing FcγR recognize an antibody bound on a target cell and then cause lysis of the said target cell. ADCC is associated with binding to FcγRIIIa, and an increase in binding to FcγRIIIa results in an increase in ADCC activity.

[0048] As used herein, "ADCP" or antibody - dependent cell - mediated phagocytosis means a cell - mediated reaction, where non - specific cytotoxic cells expressing FcγR recognize an antibody bound on a target cell and then cause phagocytosis of the target cell.

[0049] As used herein, "modification" means an amino acid substitution, insertion, and / or deletion in a polypeptide sequence, or a modification to a moiety chemically linked to a protein. For example, the modification may be a modified carbohydrate or a modified PEG structure attached to the protein. As used herein, "amino acid substitution" means an amino acid substitution, insertion, and / or deletion in a polypeptide sequence. For clarity, unless otherwise defined, amino acid modifications are always with respect to the amino acids encoded by DNA (e.g., the 20 amino acids having codons in DNA and RNA).

[0050] As used herein, "amino acid substitution" or "substitution" means substituting an amino acid with a different amino acid at a specific position in the original polypeptide sequence. In particular, in some embodiments, the substitution is for a non-natural amino acid at a specific position that does not occur naturally in the organism or in any organism. For example, substitution E272Y refers to a mutant polypeptide, and in the case of an Fc mutation, the glutamic acid at position 272 is substituted with tyrosine. For the sake of clarity, a protein in which the nucleic acid coding sequence has been changed but the start amino acid (e.g., the exchange from CGG (encoding arginine) to CGA (still encoding arginine) to increase the expression level in the host organism) has been manipulated so as not to change is not an "amino acid substitution". That is, even though it is the creation of a new gene encoding the same protein, if the protein has the same amino acid at the specific position where it starts, it is not an amino acid substitution.

[0051] As used herein, "amino acid insertion" or "insertion" means adding an amino acid sequence at a specific position in the original polypeptide sequence. For example, -233E or 233E means inserting glutamic acid after position 233 and before position 234. Further, -233AED or A233ADE means inserting AlaAspGlu after position 233 and before position 234.

[0052] As used herein, "amino acid deletion" or "deletion" means removing an amino acid at a specific position in the original polypeptide sequence. For example, E233- or E233# or E233() means deleting glutamic acid at position 233. Further, EDA233- or EDA233# means deleting the sequence GluAspAla starting at position 233.

[0053] As used herein, "mutant protein" or "protein variant" or "variant" means a protein that is different from the original protein due to at least one amino acid modification. A protein variant may refer to its own protein, a composition containing the protein, or the amino acid sequence encoding it. Preferably, the protein variant has at least one amino acid modification compared to the original protein (e.g., about 1 to about 70 amino acid modifications, and preferably about 1 to about 5 amino acid modifications compared to the original). As described below, in some embodiments, the original polypeptide (e.g., the original polypeptide such as Fc) is a human wild-type sequence such as an Fc region derived from IgG1, IgG2, IgG3, or IgG4, but a human sequence having a mutation may also function as the "original polypeptide" (e.g., the IgG1 / 2 hybrid of FIG. 13). As used herein, a protein variant sequence preferably has at least about 80% identity with the sequence of the original protein, and most preferably at least about 90% identity, even more preferably at least about 95 - 98 - 99% identity. A mutant protein may refer to its own mutant protein, a composition containing the protein variant, or the DNA sequence encoding it. Thus, as used herein, "antibody variant" or "mutant antibody" means an antibody that is different from the original antibody due to at least one amino acid modification, and as used herein, "IgG variant" or "mutant IgG" means an antibody that is different from the original IgG (again, often different from the human IgG sequence) due to at least one amino acid modification, and as used herein, "immunoglobulin variant" or "mutant immunoglobulin" means an immunoglobulin sequence that is different from the original immunoglobulin sequence due to at least one amino acid modification. As used herein, "Fc variant" or "Fc mutation" means a protein containing an amino acid modification in the Fc domain. The Fc variants of the present invention are defined according to the amino acid modifications that constitute them. For example, N434S or 434S means an Fc variant having a serine substitution at position 434 compared to the original Fc polypeptide (the numbering follows the EU index).Similarly, M428L / N434S defines an Fc variant having the substitutions of M428L and N434S as compared to the original Fc polypeptide. The identity of the WT amino acids need not be specified, in which case the variant described above is designated 428L / 434S. The order in which the substitutions are made is arbitrary; for example, it should be noted that 428L / 434S is the same Fc variant as M428L / N434S, etc. For all positions considered in the context of the antibodies of the present invention, unless otherwise specified, the amino acid position numbering follows the EU index. The EU index or EU numbering scheme such as in the EU index or Kabat refers to the numbering of EU antibodies (Edelman et al., 1969, Proc Natl Acad Sci USA 63:78-85, which is hereby incorporated by reference in its entirety). The modification may be an addition, deletion or substitution. The substitution may contain natural amino acids and, in some cases, may contain synthetic amino acids. Examples include U.S. Patent No. 6,586,207; WO98 / 48032; WO03 / 073238; US2004-0214988A1; WO05 / 35727A2; WO05 / 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 hereby incorporated by reference in their entirety).

[0054] As used herein, "protein" means at least two covalently linked amino acids and includes proteins, polypeptides, oligopeptides, and peptides. The peptidyl group may contain a natural amino acid and a peptide bond, or may contain a synthetic peptide mimetic structure (i.e., an "analog" such as, for example, peptoid (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 natural or synthetic (e.g., not an amino acid encoded by DNA), as recognized by those skilled in the art. For example, homo-phenylalanine, citrulline, ornithine, and norleucine are considered synthetic amino acids that meet the objectives of the present invention, and both D- and L-(R or S)-configured amino acids may be utilized. Variants of the present invention may contain modifications including, for example, the use of synthetic amino acids incorporated using techniques developed by Schultz et al. (including, but not limited to, the methods described in 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 entirety). Further, the polypeptide may contain one or more side chain or terminal synthetic derivatizations, glycosylations, PEGylations, cyclizations, linkers to other molecules, fusions to proteins or protein domains, and the addition of peptide tags or labels.

[0055] As used herein, "residue" means the position in a protein and the identity of its associated amino acid. For example, asparagine 297 (also referred to as Asn297 or N297) is the residue at position 297 in the human antibody IgG1.

[0056] As used herein, "Fab" or "Fab region" means a polypeptide containing the VH, CH1, VL, and Cl immunoglobulin domains. Fab may refer to this region in isolation, or may refer to this region in the context of a full-length antibody, antibody fragment, or Fab fusion protein. As used herein, "Fv" or "Fv fragment" or "Fv region" means a polypeptide containing the VL and VH domains of one antibody.

[0057] As used herein, "IgG subclass modification" or "isotype modification" means an amino acid modification that converts one IgG isotype to the corresponding amino acid in a different, aligned IgG isotype. For example, since IgG1 contains tyrosine at position 296 in the EU position and IgG2 contains phenylalanine, the F296Y substitution in IgG2 is considered an IgG subclass modification.

[0058] As used herein, "non-natural modification" means an amino acid modification that is not of the isotype type. For example, since IgG containing serine at position 434 does not exist, the 434S substitution in IgG1, IgG2, IgG3, or IgG4 (or hybrids thereof) is considered a non-natural modification.

[0059] As used herein, "amino acid" and "amino acid identity" mean one of the 20 amino acids encoded by DNA and RNA.

[0060] As used herein, "effector function" means a biochemical event resulting from the interaction of the Fc region of an antibody with an Fc receptor or ligand. Effector functions include, but are not limited to, ADCC, ADCP, and CDC.

[0061] As used herein, "IgG Fc ligand" means any organism-derived molecule, preferably a polypeptide, that binds to the Fc region of an IgG antibody and forms an Fc / Fc ligand complex. Fc ligands include, but are not limited to, FcγRI, FcγRII, FcγRIII, FcRn, C1q, C3, mannose-binding lectin, mannose receptor, staphylococcal protein A, streptococcal protein G, and viral FcγR. Fc ligands also include Fc receptor homologs (FcRH), a family of Fc receptors homologous to FcγR (Davis et al., 2002, Immunological Reviews 190:123-136, incorporated herein by reference in its entirety). Fc ligands may contain undiscovered molecules that bind to Fc. Specific IgG Fc ligands are FcRn and Fcγ receptors. As used herein, "Fc ligand" means any organism-derived molecule, preferably a polypeptide, that binds to the Fc region of an antibody and forms an Fc / Fc ligand complex.

[0062] As used herein, "Fcγ receptor", "FcγR", or "Fc gamma R" means any one of the protein families that bind to the Fc region of IgG antibodies and is encoded by the FcγR gene. In humans, this family includes, but is not limited to, FcγRI (CD64) (including isoforms of FcγRIa, FcγRIb, and FcγRIc); FcγRII (CD32) (including isoforms of FcγRIIa (including allotypes of H131 and R131), FcγRIIb (including FcγRIIb01 and FcγRIIb-2), and FcγRIIc); and FcγRIII (CD16) (including isoforms of FcγRIIIa (including allotypes of V158 and F158) and FcγRIIIb (including allotypes of FcγRIIb-NA1 and FcγRIIb-NA2)) (see Jefferis et al., 2002, Immunol Lett 82:57-65, which is incorporated herein by reference in its entirety), as well as any undiscovered human FcγR or FcγR isoform or allotype. FcγR may be derived from any organism, including, but not limited to, humans, mice, rats, rabbits, and monkeys. Mouse Fcγ includes, but is not limited to, FcγRI (CD64), FcγRII (CD32), FcγRIII (CD16), and FcγRIII-2 (CD16-2), as well as any undiscovered mouse FcγR or FcγR isoform or allotype.

[0063] As used herein, "FcRn" or "neonatal Fc receptor" means 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 derived from any organism (including, but not limited to, human, mouse, rat, rabbit, and monkey). As is known in the art, a functional FcRn protein is composed of two polypeptides (often referred to as the heavy chain and the light chain). The light chain is beta-2-microglobulin, and the heavy chain is encoded by the FcRn gene. Unless otherwise noted herein, FcRn or FcRn protein refers to a complex of the FcRn heavy chain and beta-2-microglobulin. Various FcRn variants used to enhance binding to the FcRn receptor and, in some cases, to extend serum half-life are shown in the description of the figure in FIG. 83.

[0064] As used herein, "original polypeptide" means the starting polypeptide, which is then modified to create variants. The original polypeptide may be a native polypeptide or a variant or engineered version of a native polypeptide. The original polypeptide may refer to the polypeptide itself, a composition containing the original polypeptide, or the amino acid sequence encoding it. Thus, as used herein, "original immunoglobulin" means an unmodified immunoglobulin polypeptide that is modified to create variants, and as used herein, "original antibody" means an unmodified antibody that is modified to create variant antibodies. Note that "original antibodies" include known commercially available, recombinantly produced antibodies (outlined below).

[0065] As used herein, the term "Fc fusion protein" or "immunoadhesin" generally refers to a protein containing an Fc region linked to a different protein (e.g., the binding moiety to the target protein described herein), optionally via a linker moiety described herein. In some cases, one monomer of a heterodimeric protein contains an antibody heavy chain (either containing a scFv or further containing a light chain), and the other monomer is an Fc fusion containing a mutant Fc domain and a ligand. In some embodiments, these "half-antibody - half-fusion proteins" are also referred to as "Fusionbodies".

[0066] As used herein, the term "position" refers to a position in the sequence of a protein. The positions may be numbered consecutively or may follow an established format (e.g., for antibody numbering, such as the EU index, etc.).

[0067] As used herein, the term "target antigen" refers to a molecule specifically bound by the variable region of a given antibody. The target antigen may be a protein, carbohydrate, lipid, or other chemical compound. Many suitable target antigens are described below.

[0068] In the context of the monomers of the heterodimeric proteins of the present invention, "chain - like nature" means that dimerization mutations are incorporated into each monomer such that they retain the ability to "match" to form a heterodimer, similar to the two "matching" strands of "complementary" DNA. For example, if some pI mutations are engineered and incorporated into monomer A (e.g., to increase the pI), then, if the conformational mutations, which are "charge pairs" that can also be used in a similar manner, do not interfere with that pI mutation, e.g., the charge mutation that increases the pI is placed in the same "chain" or "monomer" to retain both functionalities.

[0069] As used herein, the term "target cell" refers to a cell expressing a target antigen.

[0070] As used herein, the "variable region" means a region of an immunoglobulin containing one or more Ig domains substantially encoded by any of V.kappa., V.lamda., and / or VH genes (each constituting the kappa, lambda, and heavy chain immunoglobulin loci).

[0071] As used herein, "wild type" or "WT" means an amino acid sequence or nucleotide sequence that exists in nature and includes allelic variations. A WT protein has an amino acid sequence or nucleotide sequence that has not been intentionally modified.

[0072] The antibodies of the present invention may typically be isolated or recombinant. When used to describe the various polypeptides disclosed herein, "isolated" means a polypeptide that has been identified from the cell or cell culture in which it was expressed and separated and / or recovered. Usually, an isolated polypeptide is prepared by at least one purification step. An "isolated antibody" refers to an antibody that is substantially separated from other antibodies having different antigen specificities.

[0073] "Specific binding" or "binds specifically" or "is specific for" a particular antigen or epitope means binding that is measurably different from non-specific interactions. Specific binding can be measured, for example, by measuring the binding of a molecule as compared to the binding of a control molecule (usually a molecule of a similar structure that does not have binding activity). For example, specific binding can be measured by competition with a control molecule similar to the target.

[0074] Specific binding to a particular antigen or epitope may be demonstrated, for example, by an antibody having a KD for the antigen or epitope of at least about 10-4 M, at least about 10-5 M, at least about 10-6 M, at least about 10-7 M, at least about 10-8 M, at least about 10-9 M, or at least about 10-10 M, at least about 10-11 M, at least about 10-12 M or more, where KD refers to the dissociation rate of a particular antibody-antigen interaction. Typically, an antibody that specifically binds an antigen has a KD that is 20-, 50-, 100-, 500-, 1000-, 5,000-, 10,000-fold or more greater for the antigen or epitope than for a control molecule.

[0075] Alternatively, specific binding to a particular antigen or epitope may be demonstrated, for example, by an antibody having a KA or Ka for the antigen or epitope that is 20-, 50-, 100-, 500-, 1000-, 5,000-, 10,000-fold or more greater than that for a control, when compared to the control, where KA or Ka refers to the association rate of a particular antibody-antigen interaction.

[0076] Heterodimeric Protein The present invention is directed to the generation of multispecific, particularly bispecific, binding proteins, particularly multispecific antibodies. The invention generally relates to the use of engineered Fc domains or mutant Fc domains that are capable of self-assembly in production cells for producing heterodimeric proteins and methods of making and purifying such heterodimeric proteins.

[0077] Antibody The present invention relates to the generation of multispecific antibodies (usually therapeutic antibodies). As will be discussed below, the term "antibody" is commonly used. The antibodies for which uses are found in the present invention can take many forms described herein, including the antibodies and antibody derivatives, fragments and mimetics described below. Generally, the term "antibody" includes any polypeptide containing at least one constant domain (including but not limited to CH1, CH2, CH3 and CL).

[0078] Units of conventional antibody structures usually contain a tetramer. Each tetramer is usually composed of a pair of two identical polypeptide chains, and each pair has one "light" chain (usually having a molecular weight of about 25 kDa) and one "heavy" chain (usually having a molecular weight of about 50-70 kDa). Human light chains are classified as κ light chains and λ light chains. The present invention is directed to the IgG class having several subclasses (including but not limited to IgG1, IgG2, IgG3 and IgG4). Thus, as used herein, "isotype" means any subclass of immunoglobulin defined by the chemical and antigenic characteristics of its constant region. It should be understood that therapeutic antibodies can also contain hybrids of isotypes and / or subclasses. For example, as shown in US Publication 2009 / 0163699 (incorporated by reference), the present invention involves the pI manipulation of IgG1 / G2 hybrids.

[0079] The amino-terminal portion of each chain contains a variable region of approximately 100 to 110 or more amino acids that is primarily responsible for antigen recognition and is commonly referred to in the art and herein as the "Fv domain" or "Fv region". In the variable region, three loops converge for each of the V domains of the heavy and light chains to form the antigen-binding site. Each loop is referred to as a complementarity-determining region (hereinafter referred to as CDR), where the amino acid sequence variation is most prominent. "Variable" refers to the fact that certain segments of the variable region have significant sequence differences between antibodies. The variability within the variable region is not equally distributed. Rather, the V regions consist of relatively conserved continuous sequences called framework regions (FRs) of 15 to 30 amino acids, separated by short regions of relatively high variability called "hypervariable regions", each of which is 9 to 15 or more amino acids in length.

[0080] Each of VH and VL is composed of three hypervariable regions ("complementarity-determining regions", "CDRs") and four FRs, arranged in the following order from the amino terminus to the carboxy terminus: FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4.

[0081] The hypervariable regions generally include amino acid residues approximately 24-34 (LCDR1; "L" indicates the light chain), 50-56 (LCDR2), and 89-97 (LCDR3) in the light chain variable region, and approximately 31-35B (HCDR1; "H" indicates the heavy chain), 50-65 (HCDR2), and 95-102 (HCDR3) in the heavy chain variable region; Kabat et al., SEQUENCES OF PROTEINS OF IMMUNOLOGICAL INTEREST, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md. (1991), and / or residues that form hypervariable loops (e.g., 26-32 residues (LCDR1), 50-52 residues (LCDR2), and 91-96 residues (LCDR3) in the light chain variable region, and 26-32 residues (HCDR1), 53-55 residues (HCDR2), and 96-101 residues (HCDR3) in the heavy chain variable region); Chothia and Lesk (1987) J. Mol. Biol. 196:901-917. Specific CDRs of the present invention are described below.

[0082] Throughout this specification, when referring to residues in the variable domain (approximately residues 1-107 of the light chain variable region and residues 1-113 of the heavy chain variable region), the Kabat numbering system is used, and for the Fc region, the EU numbering system is commonly used (e.g., Kabat et al., (1991) above).

[0083] CDRs contribute to the formation of antigen binding, or more specifically, to the epitope binding site of an antibody. An "epitope" refers to a determinant that interacts with a specific antigen binding site in the variable region of an antibody molecule known as a paratope. An epitope is a population of molecules such as, for example, amino acids or sugar side chains, and usually has specific structural and charge characteristics. One antigen may have two or more epitopes.

[0084] An epitope may contain amino acid residues that are directly involved in binding (also referred to as the major antigenic components of the epitope), and may also contain other amino acid residues that are not directly involved in binding (for example, amino acid residues that are efficiently blocked by a specific antigen-binding peptide. That is, the amino acid residues are within the footprint of the specific antigen-binding peptide).

[0085] An epitope may be either conformational or linear. A conformational epitope is produced by amino acids that are spatially juxtaposed from different segments of a linear polypeptide chain. A linear epitope is produced by adjacent amino acid residues in a polypeptide chain. Conformational and non-conformational epitopes may be distinguished in that the former binds even in the presence of a denaturing solvent, and the latter loses binding.

[0086] An epitope usually contains at least 3, more commonly at least 5 or 8 - 10 amino acids in a unique spatial structure. Antibodies that recognize the same epitope can be demonstrated in a simple immunoassay that shows the activity of an antibody that inhibits the binding of other antibodies to the target antigen, such as "binning".

[0087] The carboxy-terminal portion of each chain defines a constant region that is mainly involved in effector functions. Kabat et al. collected many of the major sequences of the variable regions of the heavy and light chains. Based on the degree of sequence conservation, the individual major sequences were classified into CDRs and frameworks, and a list was made (see SEQUENCES OF IMMUNOLOGICAL INTEREST, 5th edition, NIH publication, No. 91 - 3242, E.A. Kabat et al., which is incorporated by reference in its entirety).

[0088] In the IgG subclasses of immunoglobulins, several immunoglobulin domains are present in the heavy chain. As used herein, the term "immunoglobulin (Ig) domain" means a region of an immunoglobulin having a distinct three-dimensional structure. The heavy chain domains are of interest in the present invention and include the constant heavy (CH) domains and the hinge domain. In the description of IgG antibodies, each of the IgG isotypes has three CH regions. Thus, for IgG, the "CH" domains are as follows: "CH1" refers to positions 118 to 220 according to the Kabat EU index. "CH2" refers to positions 237 to 340 according to the Kabat EU index, and "CH3" refers to positions 341 to 447 according to the Kabat EU index. As shown herein and as described below, the pI mutations may be in one or more of the CH regions and may also be in the hinge region as considered below.

[0089] It should be noted that the sequences disclosed herein begin at the CH1 region (position 118) and the variable regions are not included except as described. For example, the first amino acid of SEQ ID NO: 2 is designated as position "1" in the sequence listing but corresponds to position 118 of the CH1 region (according to EU numbering).

[0090] Another type of Ig domain of the heavy chain is the hinge region. As used herein, "hinge" or "hinge region" or "antibody hinge region" or "immunoglobulin hinge region" means a flexible polypeptide containing amino acids between the first and second constant domains of an antibody. Structurally, the IgG CH1 domain ends at position 220 of EU, and the IgG CH2 domain begins at residue 237 of EU. Thus, as used herein, the hinge of an antibody with respect to IgG is defined to include positions 221 (D221 in IgG1) to 236 (G236 in IgG1) (numbering follows the EU index of Kabat et al.). In some embodiments, for example, in the context of the Fc region, a lower hinge is included, and "lower hinge" typically refers to position 226 or 230. As described herein, pI mutations may be made in the hinge region as well.

[0091] The light chain usually contains two domains: a variable light chain domain (containing light chain CDRs and forming an Fv region together with the variable heavy chain domain) and a constant light chain region (often referred to as CL or C κ ).

[0092] As another area of interest for additional replacement, there is the Fc region, as outlined below. As used herein, "Fc" or "Fc region" or "Fc domain" means a polypeptide containing the constant region of an antibody excluding the first constant region immunoglobulin domain (and in some cases a portion of the hinge). Thus, Fc refers to the last two constant region immunoglobulin domains of IgA, IgD and IgG, the last three constant region immunoglobulin domains of IgE and IgM, and the flexible hinge at the N-terminus of these domains. For IgA and IgM, Fc may contain a J chain. For IgG, the Fc domain contains the immunoglobulin domains Cγ2 and Cγ3 (Cγ2 and Cγ3), and the lower hinge region between Cγ1 (Cγ1) and Cγ2 (Cγ2). The boundaries of the Fc region can vary, but the human IgG heavy chain Fc region is typically defined as containing residues from C226 or P230 to its carboxyl terminus (numbering follows the Kabat EU index). In some embodiments, as described more fully below, amino acid modifications are made to the Fc region, for example, altering binding to one or more FcγR or FcRn receptors.

[0093] Accordingly, in some embodiments of the invention, heterodimeric antibodies are disclosed that use two different heavy chain mutant Fc domains that self-assemble to form heterodimeric antibodies.

[0094] In some embodiments, the antibody is full-length. As used herein, "full-length antibody" means the structure that constitutes the native biological form of the antibody, contains a variable region and a constant region, and in particular in the Fc domain contains one or more modifications outlined herein to enable either heterodimer formation or purification of heterodimers from homodimers. Full-length heterodimeric antibodies are two heavy chains that have different Fc domains and either two light chains or a normal light chain.

[0095] Alternatively, the antibody may contain various structures (outlined in the figure. Without limitation, antibody fragments, monoclonal antibodies, bispecific antibodies, minibodies, domain antibodies, synthetic antibodies (sometimes referred to herein as "antibody mimetics"), chimeric antibodies, humanized antibodies, antibody fusions (sometimes referred to as "antibody conjugates") and their respective fragments are included).

[0096] In one embodiment, the antibody is an antibody fragment as long as it contains at least one constant domain capable of performing an operation for producing a heterodimer, such as pI manipulation. Other antibody fragments that can be used include fragments containing one or more of the CH1, CH2, CH3, hinge and CL domains of the present invention that have been pI-manipulated. For example, an Fc fusion is a fusion of an Fc region (CH2 and CH3, optionally with a hinge region) fused to another protein. Many Fc fusions are known in the art and can be improved by the addition of the heterodimerization mutations of the present invention. In this case, the antibody fusion can be made containing CH1; CH1, CH2 and CH3; CH2; CH3; CH2 and CH3; CH1 and CH3, utilizing any combination of the heterodimerization mutations described herein, any or all of which can be made optionally with a hinge region.

[0097] scFv Embodiment In some embodiments of the present invention, one monomer contains a heavy chain containing an scFv linked to an Fc domain, and the other monomer contains a heavy chain containing a Fab linked to an Fc domain, for example, there are "typical" heavy and light chains. As used herein, "Fab" or "Fab region" means a polypeptide containing the VH, CH1, VL, and CL immunoglobulin domains. Fab refers to this region when isolated or in the context of a full-length antibody, antibody fragment or Fab fusion protein. As used herein, "Fv" or "Fv fragment" or "Fv region" means a polypeptide containing the VL and VH domains of one antibody.

[0098] Some embodiments of the heterodimeric antibodies of this specification rely on the use of one or more scFv domains that are covalently linked using a linker and contain variable heavy and variable light chains that form antigen-binding domains. In some embodiments of the invention, "standard" linkers (usually serine and glycine linkers known in the art) are used.

[0099] The invention further discloses a charged scFv linker for facilitating separation at the pI between the first and second monomers. That is, by incorporating a charged scFv linker (either a positive or negative charge (or both in the case of a scaffold using scFv on different monomers)), the monomer containing the charged linker can change its pI without undergoing further changes in the Fc domain. These charged linkers may be substituted for any scFv containing a standard linker. As recognized by those skilled in the art, charged scFv linkers are used on the correct "chain" or monomer according to the desired change in pI. For example, as discussed herein, to create a triple F format heterodimeric antibody, the original pI of the Fv region for each of the desired antigen-binding domains is calculated, one is selected to create the scFv, and either a positive or negative linker is selected depending on the pI.

[0100] Furthermore, in the case of the anti-CD3 scFv region, disulfide bonds may be incorporated into the variable heavy and variable light chains to provide additional stability. The sequence of appropriate disulfide bonds for the anti-CD3 scFv is shown in Figure 8.

[0101] Chimeric Antibody and Humanized Antibody In some embodiments, the antibody may be a mixture from different species (e.g., chimeric antibodies and / or humanized antibodies). Generally, both “chimeric antibodies” and “humanized antibodies” refer to antibodies in which regions from two or more species are mixed. For example, a “chimeric antibody” has traditionally contained variable region(s) from a mouse (or in some cases a rat) and constant region(s) from a human. A “humanized antibody” generally refers to a non-human antibody in which the sequences present in a human antibody and the framework regions of the variable domain have been exchanged. Generally, in a humanized antibody, all of the antibody except for the CDRs is encoded by a human-derived polynucleotide or is identical to the antibody except for its CDRs. The CDRs are encoded by nucleic acids having their origin in a non-human organism, at least in part or in whole, and are transplanted into the β-sheet framework of the human antibody variable region to produce an antibody, the specificity of which is determined by the transplanted CDRs. The production of such antibodies is described, for example, in WO92 / 11018, Jones, 1986, Nature 321:522-525, Verhoeyen et al., 1988, Science 239:1534-1536 (all incorporated by reference in their entirety). To restore the affinity lost in the initially transplanted construct, “backmutations” of selected acceptor framework residues to the corresponding donor residues are often required (US5530101; US5585089; US5693761; US5693762; US6180370; US5859205; US5821337; US6054297; US6407213, all incorporated by reference in their entirety). Also, a humanized antibody preferably contains at least a portion of an immunoglobulin constant region (usually that of a human immunoglobulin) and thereby usually contains the human Fc region. A humanized antibody may also be produced using a mouse having a genetically engineered immune system. Roque et al., 2004, Biotechnol. Prog. 20:639-654, incorporated by reference in its entirety.A variety of techniques and methods for humanizing and reforming into non-human antibodies are known in the art (see Tsurushita & Vasquez, 2004, Humanization of Monoclonal Antibodies, Molecular Biology of B Cells, 533-545, Elsevier Science (USA) and the references cited therein. All are incorporated by reference in their entirety). Methods of humanization 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 incorporated by reference in their entirety). Other methods for reducing the immunogenicity of humanized or non-human antibody variable regions include, for example, the methods described in Roguska et al., 1994, Proc. Natl. Acad. Sci. USA 91:969-973 (incorporated by reference in its entirety). In one embodiment, as is known in the art, the original antibody has its affinity matured. Structure-based methods may be used for humanization and affinity maturation (e.g., as described in USSN 11 / 004,590).Antibody variable regions may be humanized and / or affinity matured using selection-based methods, including but not limited to, the methods 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 incorporated by reference in their entirety). Other humanization methods transplant only portions of the CDRs and include but are 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 incorporated by reference in their entirety).

[0102] Heterodimeric Heavy Chain Constant Region Accordingly, the present invention discloses a heterodimeric protein based on the use of a monomer containing a mutant heavy chain constant region, and in particular an Fc domain as a first domain. As used herein, "monomer" means half of a heterodimeric protein. Note that conventional antibodies are actually tetramers (two heavy chains and two light chains). In the context of the present invention, one pair of heavy and light chains (if applicable, for example if the monomer contains a Fab) is considered a "monomer". Similarly, a heavy chain region containing an scFv is considered a monomer. In the case where the Fv region is one fusion partner (e.g., the variable domains of the heavy and light chains) and a protein that is not an antibody is the other fusion partner, each "half" is considered a monomer. Essentially, each monomer contains a sufficient heavy chain constant region for heterodimerization, whether it is all of the constant region (e.g., Ch1-hinge-CH2-CH3), the Fc region (CH2-CH3), or just the CH3 domain only.

[0103] The mutant heavy chain constant region may contain all or part of the heavy chain constant region, including full-length constructs, CH1-hinge-CH2-CH3, or a portion thereof (e.g., CH2-CH3 or CH3 only). Furthermore, the heavy chain regions of each monomer may be the same backbone (CH1-hinge-CH2-CH3, or CH2-CH3) or different. N-terminal truncations, C-terminal truncations, and enrichments are also included in the definition; for example, some pI mutations involve the addition of charged amino acids to the C-terminus of the heavy chain domain.

[0104] Thus, generally, one monomer of the "triple F" construct of the present invention is an scFv region-hinge-Fc domain, and the other is VH-CH1-hinge-CH2-CH3 + associated light chain, with heterodimerization mutations (including conformational mutations, isotype mutations, charge steering, as well as pI mutations, Fc and FcRn mutations, deletion mutations), and additional antigen-binding domains (along with any linkers) contained within these regions.

[0105] In addition to the heterodimerization mutations (e.g., conformational mutations and pI mutations) outlined herein, the heavy chain region may also contain additional amino acid substitutions (including changes to modify FcγR and FcRn binding as discussed below).

[0106] In addition, some monomers may use a linker between the mutant heavy chain constant region and its fusion partner. For the "pull - out type" scFv portion, standard linkers known in the art or the charged scFv linkers disclosed herein can be used. When creating additional fusion partners (e.g., FIGS. 1 and 2), conventional peptide linkers (such as the flexible linkers of glycine and serine, or the charged linker of FIG. 9) may be used. In some cases, the linker for use as a monomer component is different from that defined below for ADC constructs and, in many embodiments, is not a cleavable linker (e.g., protease - sensitive), although cleavable linkers may also find use in some embodiments.

[0107] Heterodimerization mutations include many different types of mutations, including but not limited to conformational mutations (including charge mutations) and pI mutations, and can be optionally and independently combined with any other mutations. In these embodiments, it is important to match "monomer A" and "monomer B"; that is, if the heterodimeric protein depends on both conformational and pI mutations, these must precisely match each other's monomers: for example, the set of conformational mutations that act (one set for monomer A, one set for monomer B) is combined with the set of pI mutations (one set for monomer A, one set for monomer B) such that the conformational mutations that can change the pI are placed on the appropriate monomer, taking into account the "chain - like nature" of the pI, and the mutations on each monomer are designed to obtain the desired function.

[0108] The heterodimerization mutations outlined herein (e.g., but not limited to, the mutations shown in Figures 3 and 12) are noted to be optionally and independently combinable on any other monomer and with any other mutation. That is, what is important for heterodimerization is that there is a "set" of mutations, and one set for one monomer is also one set for the other monomer. Whether these are combined one-to-one from form 1 (e.g., the list of monomer 1 can be compatible) or exchanged (exchanging the pI mutation of monomer 1 with the conformational mutation of monomer 2) is not important. However, as noted herein, when the above combinations are formed, the "chain-like structure likeness" must be preserved. Further, for additional Fc mutations (e.g., for FcγR binding, FcRn binding, etc.), either one monomer, or both monomers, can independently and optionally contain any of the listed mutations. In some cases, both monomers have additional mutations, and in some, only one monomer has an additional mutation, or they may be combined.

[0109] Heterodimerization Mutation The present invention discloses heterodimeric proteins containing various forms of heterodimeric antibodies that utilize heterodimerization mutations that enable the formation of heterodimers and / or the purification of heterodimers from homodimers.

[0110] Conformational Mutation In some embodiments, the formation of heterodimers can be facilitated by the addition of conformational mutations. That is, by changing the amino acids of each heavy chain, different heavy chains may associate, and the formation of heterodimeric structures may be more likely than the formation of homodimeric structures having the same Fc amino acid sequence. Suitable conformational mutations are shown in Figure 3 and Figures 12A, 12B, 12C, 12D, 12F, and 12G.

[0111] One mechanism, often referred to in the art as "knobs and holes", refers to amino acid manipulations that provide steric effects to act unfavorably on homodimer formation and favorably on heterodimer formation, and this can also be used optionally; this is sometimes referred to as "knobs and holes" and is described in USSN 61 / 596,846, Ridgway et al., Protein Engineering 9(7):617 (1996); Atwell et al., J. Mol. Biol. 1997 270:26; U.S. Patent No. 8,216,805 (all incorporated herein by reference in their entirety). In the figure, many "monomer A - monomer B" pairs that rely on "knobs and holes" are identified. Further, as described in Merchant et al., Nature Biotech. 16:677 (1998), these "knobs and holes" mutations may be combined with disulfide bonds to twist the structure towards heterodimerization.

[0112] Additional mechanisms that can be used to generate heterodimer formation are, in some cases, those referred to as "electrostatic steering" (disclosed in Gunasekaran et al., J. Biol. Chem. 285(25):19637 (2010), which is hereby incorporated by reference in its entirety). This is sometimes referred to herein as a "charge pair". In this embodiment, electrostatics is used to twist the structure towards heterodimer formation. Although known in the art, these may also have an effect on the pI and thus on purification and in some cases may also be considered pI mutations. However, since these were created to promote heterodimerization and not used as tools for purification, they are classified as "conformational mutations". These include, but are not limited to, for example, D221E / P228E / L368E paired with D221R / P228R / K409R (for example, these are a "monomer pair set") and C220E / P228E / 368E paired with C220R / E224R / P228R / K409R.

[0113] Additional monomer A and monomer B mutations that can be combined with other mutations can be optionally and independently incorporated into one or both monomers in any amount, such as pI mutations outlined herein or other conformational mutations shown in FIG. 37 of US2012 / 0149876, the figure, its description, and sequence numbers of which are hereby expressly incorporated by reference herein.

[0114] In some embodiments, the conformational mutations outlined herein can be optionally and independently incorporated into one or both monomers together with any pI mutation (or other mutations such as Fc mutations, FcRn mutations, etc.), and can optionally and independently be included in or excluded from the proteins of the invention.

[0115] pI (Isoelectric Point) Mutation for Heterodimer Generally, as recognized by those skilled in the art, there are two general categories of pI mutations: those that increase the pI of the protein (basic changes), and those that decrease the pI of the protein (acidic changes). As disclosed herein, all combinations of these mutations can be made: one monomer may be wild-type or a mutation that does not exhibit a pI significantly different from wild-type, and the other may be more basic or more acidic. Alternatively, each monomer can be modified such that one is more basic and one is more acidic.

[0116] Preferred combinations of pI mutations are shown in Figures 3 and 12E.

[0117] pI Change of Heavy Chain Many pI mutations are shown in Figures 54 and 55. As outlined herein and shown in the drawings, these changes are shown relative to IgG1, but all isotypes as well as isotype hybrids can be modified in this manner.

[0118] When the heavy chain constant domain is derived from IgG2-4, R133E and R133Q may be used.

[0119] Mutation of Antibody Heterodimeric Light Chain In the case of antibody-based heterodimers, for example, when at least one of the monomers contains a light chain in addition to the heavy chain domain, pI mutations may be made to the light chain. Amino acid substitutions for decreasing the pI of the light chain include, but are not limited to, K126E, K126Q, K145E, K145Q, N152D, S156E, K169E, S202E, K207E, and the addition of the peptide DEDE to the C-terminus of the light chain. Changes in this category based on the constant lambda light chain contain one or more substitutions with R108Q, Q124E, K126Q, N138D, K145T, and Q199E. Additionally, an increase in the pI of the light chain may be made.

[0120] Isotype Mutation Furthermore, many embodiments of the present invention rely on "introducing" pI amino acids at specific positions from one IgG isotype to another, thereby reducing or eliminating the possibility that unwanted immunogenicity is introduced into the variant. Many of these are shown in FIGS. 10A and 10B. That is, IgG1 is a universal isotype as a therapeutic antibody for various reasons (including high effector function). However, the heavy chain constant region of IgG1 has a higher pI (8.10 versus 7.31) than IgG2. By introducing IgG2 residues into the IgG1 backbone at specific positions, the pI of the resulting monomer is decreased (or increased), and furthermore, the serum half-life is extended. For example, IgG1 has glycine (pI 5.97) at position 137, and IgG2 has glutamic acid (pI 3.22); the introduction of glutamic acid affects the pI of the resulting protein. As described below, usually, many amino acid substitutions are required to significantly affect the pI of the mutant antibody. However, as discussed below, by bringing about an equal change in the IgG2 molecule, the serum half-life can be increased.

[0121] In other embodiments, non-isotype amino acid changes are made (detailed below) to reduce the overall charge state of the resulting protein (e.g., by changing high pI amino acids to low pI amino acids), or to allow for structural modulation for stability purposes.

[0122] Furthermore, by pI engineering both the constant domains of the heavy and light chains, significant changes are seen in each monomer of the heterodimer. As discussed herein, by changing the pI of the two monomers by at least 0.5, separation by ion exchange chromatography, isoelectric focusing electrophoresis, or other methods capable of sensing isoelectric points becomes possible.

[0123] Calculation of pI The pI of each monomer may depend on the pI of the mutant heavy chain constant domain and the total monomer (including the mutant heavy chain constant domain and the fusion partner). Thus, in some embodiments, the change in pI is calculated based on the mutant heavy chain constant domain using the chart of FIG. 53. As discussed herein, which monomer should be engineered is typically determined by the native pI of the Fv and the scaffold region. Alternatively, the pI of each monomer may be compared.

[0124] Heterodimeric Fc Fusion Protein In addition to heterodimeric antibodies, the present invention discloses a heterodimeric protein containing a first monomer containing a mutant Fc region and a first fusion partner and a second fusion partner (or alternatively, containing a mutant Fc region and a second fusion partner). The mutant Fc region is engineered relative to the antibody as described herein and is thus different, and generally, the first fusion partner and the second fusion partner are also different. In some examples, when one monomer is antibody-based (e.g., either contains a standard heavy and light chain or an Fc domain with an scFv) and the other is an Fc fusion protein, the resulting heterodimeric protein is referred to as a "fusion body".

[0125] pI Mutation Contributing to Improvement of in vivo Binding of FcRn When a pI mutation decreases the pI of a monomer, the advantage of improving serum retention in vivo may be added.

[0126] Although still under verification, the Fc region is thought to extend the in vivo half-life because Fc is sequestered by binding to FcRn at pH 6 within endosomes (see Ghetie and Ward, 1997 Immunol Today. 18(12): 592-598, which is incorporated herein by reference in its entirety). The endosomal fraction then recycles the Fc to the cell surface. When the fraction opens to the extracellular region (where the pH is high, about 7.4), the Fc is released into the blood. In mice, Dall’ Acqua et al. have shown that Fc mutants with enhanced FcRn binding at pH 6 and pH 7.4 actually decreased serum concentration and had a half-life comparable to that of wild-type Fc (Dall’ Acqua et al. 2002, J. Immunol. 169:5171-5180, which is incorporated herein by reference in its entirety). It is thought that increasing the affinity of Fc for FcRn at pH 7.4 inhibits the release of Fc into the blood. Therefore, Fc mutants that increase the in vivo half-life of Fc ideally enhance FcRn binding at low pH while releasing Fc at high pH. The amino acid histidine changes its charge state in the pH range of 6.0 - 7.4. Therefore, it is not surprising that His residues are present at important positions in the Fc / FcRn complex.

[0127] In recent years, it has been speculated that antibodies with variable regions having a low isoelectric point have a long serum half-life (Igawa et al., 2010 PEDS. 23(5): 385-392, which is incorporated herein by reference in its entirety). However, the mechanism has not yet been elucidated. Furthermore, the variable regions vary from antibody to antibody. Constant region mutants with a decreased pI and an extended half-life present a more modular approach for improving the pharmacokinetic properties of antibodies, as disclosed herein.

[0128] The pI mutations found for this embodiment and for use in purification optimization are disclosed in FIG. 20.

[0129] Combination of Heterodimeric Mutations As will be appreciated by those skilled in the art, all of the enumerated heterodimerization mutations can be optionally and independently combined in any way, as long as their "chain-like structure-ness" or "monomer separation" is retained. Further, all of these mutations can be combined into any heterodimeric form.

[0130] In the case of pI mutations, while specific uses found in the embodiments are shown in the figures, other combinations may be made according to the basic rules regarding the modification of the pI difference between the two monomers to facilitate purification.

[0131] Nucleic Acid of the Present Invention The present invention further discloses a nucleic acid composition encoding the heterodimeric protein of the present invention. As will be appreciated by those skilled in the art, the nucleic acid composition depends on the form and scaffold of the heterodimeric protein. Thus, for example, if the form requires three amino acid sequences, for example, for a triple F form (e.g., the first amino acid monomer contains an Fc domain and an scFv, and the second amino acid monomer contains a heavy chain and a light chain), the three nucleic acid sequences may be incorporated into one or more expression vectors for expression. Similarly, in some forms (e.g., the dual scFv form as disclosed in Figure 1M), only two nucleic acids are required (again, they can be inserted into one or two expression vectors).

[0132] Target Antigen The heterodimeric proteins of the present invention can in fact target virtually any antigen. The "triple F" format is particularly beneficial when targeting 2 (or more) different antigens. (As outlined herein, this targeting can be any combination of monovalent and bivalent binding, depending on the format). Thus, herein, the immunoglobulin preferably co-captures two target antigens, although in some cases 3 or 4 antigens may be captured monovalently. The specificity of each monomer may be selected from the following list. While the triple F immunoglobulins disclosed herein are particularly beneficial when targeting different antigens, in some cases it may be beneficial to target only one antigen. That is, each monomer may have specificity for the same antigen.

[0133] As a particular suitable application of the heterodimeric proteins of the present specification, it is a co-target pair where it is beneficial or important to capture each target antigen monovalently. Such an antigen may be, for example, an immunoreceptor activated in the formation of an immune complex. The cellular activation of many immunoreceptors usually occurs only by antibody / antigen immune complexes or by cross-linking obtained through target cell capture by effector cells. For some immunoreceptors, for example, with respect to the example of the CD3 signaling receptor on T cells, activation that occurs only in the capture of co-captured targets is important, and when non-specific cross-linking occurs in a clinical situation, cytokine storms and toxicity can occur. Therapeutically, by using the immunoglobulins of the present specification to capture such antigens monovalently rather than polyvalently, it responds only to cross-linking in the microenvironment of the main target antigen, and such activation occurs. The ability to target two different antigens with different valences is novel and is also a utility of the present invention. Examples of target antigens that are therapeutically beneficial and require monovalent co-capture include, but are not limited to, for example, CD3, FcγR, toll-like receptors (TLRs) (e.g., TLR4 and TLR9), cytokines, chemokines, cytokine receptors, and chemokine receptors, such as immunoreceptor activation receptors. In many embodiments, one of the antigen-binding sites binds to CD3, and in some embodiments, it is a scFv-containing monomer.

[0134] In fact, all antigens can be targets of the immunoglobulins herein, including, but not limited to, proteins, subunits, domains, motifs, and / or epitopes in the following list of target antigens, which are, for example, soluble factors such as cytokines and membrane-bound factors (transmembrane receptors (17-IA, 4-1BB, 4Dc, 6-keto-PGF1a, 8-iso-PGF2a, 8-oxo-dG, A1 adenosine receptor, 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, addressin, aFGF, ALCAM, ALK, ALK-1, ALK-7, α-1-antitrypsin, α-V / β-1 antagonist, ANG, Ang, APAF-1, APE, APJ, APP, APRIL, AR, ARC, ART, Artemin, anti-Id, ASPARTIC, atrial natriuretic factor, av / b3 integrin, 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, BMP-3 Osteogenin, BMP-4 BMP-2b, BMP-5, BMP-6Vgr-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, complement factor 3 (C3), C3a, C4, C5, C5a, C10, CA125, CAD-8, calcitonin, cAMP, carcinoembryonic antigen (CEA), cancer-associated 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, botulinum toxin, welchii toxin, Ckb8-1, CLC, CMV, CMVUL, 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-related antigen, DAN, DCC, DcR3, DC-SIGN, Disintegrin and metalloproteinase with thrombospondin motifs, 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, Eotaxin 1, EpCAM, Ephrin B2 / EphB4, EPO, ERCC, E-selectin, ET-1, Factor IIa, Factor VII, Factor VIIIc, Factor IX, Fibroblast activation 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-α1, GFR-α2, GFR-α3, GITR, Glucagon, Glut4, Glycoprotein IIb / IIIa (GPIIb / IIIa), GM-CSF, gp130, gp72, GRO, Growth hormone releasing factor, Hapten (NP-cap or NIP-cap), HB-EGF, HCC, HCMVgB envelope glycoprotein, HCMV; gH envelope glycoprotein, HCMV UL; hematopoietic growth factor (HGF); HepB gp120; heparanase; Her2; Her2 / neu (ErbB-2); Her3 (ErbB-3); Her4 (ErbB-4); herpes simplex virus (HSV) gB glycoprotein; HSV gD glycoprotein; HGFA; high molecular weight melanoma associated antigen (HMW-MAA); HIV gp120; HIV IIIB gp120 V3 loop; 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 protein; 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)-α; INF-β; INF-γ; inhibin; iNOS; insulin A chain; insulin B chain; insulin-like growth factor 1; integrin α2; integrin α3; integrin α4; integrin α4 / β1; integrin α4 / β7; integrin α5 (αV); integrin α5 / β1; integrin α5 / β3; integrin α6; integrin β1; integrin β2; interferon γ; 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-1bp1, LBP, LDGF, LECT2, Lefty, Lewis Y antigen, Lewis Y-related antigen, LFA-1, LFA-3, Lfo, LIF, LIGHT, lipoprotein, LIX, LKN, Lptn, L-selectin, LT-a, LT-b, LTB4, LTBP-1, lung surfactant, luteinizing hormone, lymphotoxin Β 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-α, 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, Muellerian inhibiting substance, Mug, MuSK, NAIP, NAP, NCAD, N-Cadherin, NCA 90, NCAM, NCAM, neprilysin, neurotrophin-3, -4, or 6, neuritin, nerve growth factor (NGF), NGFR, NGF-β, 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, prolactin, 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, renin, respiratory syncytial virus (RSV) F, RSVFgp, Ret, rheumatoid factor, 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 receptor (e.g., T cell receptor α / β), TdT, TECK, TEM1, TEM5, TEM7, TEM8, TERT, testicular PLAP-like alkaline phosphatase, TfR, TGF, TGF-α, TGF-β, TGF-β Pan Specific, TGF-β RI (ALK-5), TGF-β RII, TGF-β RIIb, TGF-β RIII, TGF-β1, TGF-β2, TGF-β3, TGF-β4, TGF-β5, thrombin, thymic Ck-1, thyroid-stimulating hormone, Tie, TIMP, TIQ, tissue factor, TMEFF2, Tmpo, TMPRSS2, TNF, TNF-α, TNF-α β, TNF-β2, 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), TNFR SF13C (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 connectin, 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 ligandCD153), TNFSF9 (4-1BB ligand, CD137 ligand), TP-1, t-PA, Tpo, TRAIL, TRAIL R, TRAIL-R1, TRAIL-R2, TRANCE, transferring receptor, TRF, Trk, TROP-2, TSG, TSLP, tumor-associated antigen CA 125, tumor-associated antigen-expressing Lewis 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 antigen, VLA, VLA-1, VLA-4, VNR integrin, von Willebrands 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 the like), as well as receptors for hormones and growth factors (and the like), both of which are included. To form the bispecific or trispecific antibodies of the present invention, antibodies can be prepared against any combination of these antigens; that is, each of these antigens can optionally and independently be excluded from, or included in, the multispecific antibodies according to the present invention.

[0135] Examples of antigens that can be specifically targeted by the immunoglobulins of the present invention include, but are not limited to, the following: CD20, CD19, Her2, EGFR, EpCAM, CD3, FcγRIIIa (CD16), FcγRIIa (CD32a), FcγRIIb (CD32b), FcγRI (CD64), Toll-like receptors (TLRs) (e.g., TLR4 and TLR9), cytokines (e.g., IL-2, IL-5, IL-13, IL-12, IL-23, and TNFα), cytokine receptors (e.g., IL-2R), chemokines, chemokine receptors, growth factors (e.g., VEGF and HGF). To form the multispecific antibodies of the present invention, antibodies against any combination of these antigens can be made; that is, each of these antigens may optionally and independently be excluded from, or included in, the multispecific antibodies according to the present invention.

[0136] Particularly preferred combinations of bispecific antibodies are the antigen-binding domain against CD3 and the antigen-binding domain against CD19; the antigen-binding domain against CD3 and the antigen-binding domain against CD33; the antigen-binding domain against CD3 and the antigen-binding domain against CD38. Again, in many embodiments, the CD3-binding domain is a scFv with an exemplary sequence shown in the figure and / or the CD3 CDRs outlined.

[0137] The selection of appropriate target antigens and co-targets depends on the desired therapeutic application. Some of the targets that have proven to be particularly suitable for antibody therapy are those with signal transduction functions. Other therapeutic antibodies exert their effects by interfering with receptor signal transduction by inhibiting the binding between the receptor and its associated ligand. Other mechanisms of action of therapeutic antibodies result in downregulation of the receptor. Other antibodies do not act via signal transduction through their target antigen. The selection of co-targets depends on the detailed biology underlying the condition being treated.

[0138] Monoclonal antibody therapy has been integrated as an important treatment modality for cancer (Weiner et al., 2010, Nature Reviews Immunology 10:317-327; Reichert et al., 2005, Nature Biotechnology 23[9]:1073-1078, which are hereby expressly incorporated by reference). For anti-cancer treatment, it is desirable to target one antigen (antigen 1), the expression of which is limited to cancerous cells, while co-targeting a second antigen (antigen 2) that modulates some immunological killing activity. For other treatments, it may be beneficial to co-target two antigens, for example, co-targeting two angiogenesis factors, or two growth factors, etc., that are known to play some role in tumor growth. Examples of co-targets for tumors include, but are not limited to, HGF and VEGF, IGF-1R and VEGF, Her2 and VEGF, CD19 and CD3, CD20 and CD3, Her2 and CD3, CD19 and FcγRIIIa, CD20 and FcγRIIIa, Her2 and FcγRIIIa. The immunoglobulins of the present invention are VEGF and phosphatidylserine; VEGF and ErbB3; VEGF and PLGF; VEGF and ROBO4; VEGF and BSG2; VEGF and CDCP1; VEGF and ANPEP; VEGF and c-MET; HER-2 and ERB3; HER-2 and BSG2; HER-2 and CDCP1; HER-2 and ANPEP; EGFR and CD64; EGFR and BSG2; EGFR and CDCP1; EGFR and ANPEP; IGF1R and PDGFR; IGF1R and VEGF; IGF1R and CD20; CD20 and CD74; CD20 and CD30; CD20 and DR4; CD20 and VEGFR2; CD20 and CD52; CD20 and CD4; HGF and c-MET; HGF and NRP1; HGF and phosphatidylserine; ErbB3 and IGF1R; ErbB3 and IGF1,2; c-Met and Her-2; c-Met and NRP1; c-Met and IGF1R; IGF1,2 and PDGFR; IGF1,2 and CD20; IGF1,2 and IGF1R; IGF2 and EGFR; IGF2 and HER2; IGF2 and CD20; IGF2 and VEGF;IGF2 and IGF1R; IGF1 and IGF2; PDGFRa and VEGFR2; PDGFRa and PLGF; PDGFRa and VEGF; PDGFRa and c-Met; PDGFRa and EGFR; PDGFRb and VEGFR2; PDGFRb and c-Met; PDGFRb and EGFR; RON and c-Met; RON and MTSP1; RON and MSP; RON and CDCP1; VGFR1 and PLGF; VGFR1 and RON; VGFR1 and EGFR; VEGFR2 and PLGF; VEGFR2 and NRP1; VEGFR2 and RON; VEGFR2 and DLL4; VEGFR2 and EGFR; VEGFR2 and ROBO4; VEGFR2 and CD55; LPA and S1P; EPHB2 and RON; CTLA4 and VEGF; CD3 and EPCAM; CD40 and IL6; CD40 and IGF; CD40 and CD56; CD40 and CD70; CD40 and VEGFR1; CD40 and DR5; CD40 and DR4; CD40 and APRIL; CD40 and BCMA; CD40 and RANKL; CD28 and MAPG; CD80 and CD40; CD80 and CD30; CD80 and CD33; CD80 and CD74; CD80 and CD2; CD80 and CD3; CD80 and CD19; CD80 and CD4; CD80 and CD52; CD80 and VEGF; CD80 and DR5; CD80 and VEGFR2; CD22 and CD20; CD22 and CD80; CD22 and CD40; CD22 and CD23; CD22 and CD33; CD22 and CD74; CD22 and CD19; CD22 and DR5; CD22 and DR4; CD22 and VEGF; CD22 and CD52; CD30 and CD20; CD30 and CD22; CD30 and CD23; CD30 and CD40; CD30 and VEGF; CD30 and CD74; CD30 and CD19; CD30 and DR5; CD30 and DR4; CD30 and VEGFR2; CD30 and CD52; CD30 and CD4; CD138 and RANKL; CD33 and FTL3; CD33 and VEGF; CD33 and VEGFR2; CD33 and CD44; CD33 and DR4; CD33 and DR5; DR4 and CD137; DR4 and IGF1,2; DR4 and IGF1R; DR4 and DR5; DR5 and CD40; DR5 and CD137; DR5 and CD20;It can bind to DR5 and EGFR; DR5 and IGF1,2; DR5 and IGFR, DR5 and HER-2, and EGFR and DLL4. Other combinations of targets include one or more factors of the EGF / efb-2 / erb-3 family.;

[0139] Other target(s) (one or more) involved in cancerous diseases to which the immunoglobulins of the present specification can bind include, but are not limited to, CD52, CD20, CD19, CD3, CD4, CD8, BMP6, IL12A, IL1A, IL1B, IL2, IL24, INHA, TNF, TNFSF10, BMP6, EGF, FGF1, FGF10, FGF11, FGF12, FGF13, FGF14, FGF16, FGF17, FGF18, FGF19, FGF2, FGF20, FGF21, FGF22, FGF23, FGF3, FGF4, FGF5, FGF6, FGF7, FGF8, FGF9, GRP, IGF1, IGF2, IL12A, IL1A, IL1B, IL2, INHA, TGFA, TGFB1, TGFB2, TGFB3, VEGF, CDK2, FGF10, FGF18, FGF2, FGF4, FGF7, IGF1R, IL2, BCL2, CD164, CDKN1A, CDKN1B, CDKN1C, CDKN2A, CDKN2B, CDKN2C, CDKN3, GNRH1, IGFBP6, IL1A, IL1B, ODZ1, PAWR, PLG, TGFB1I1, AR, BRCA1, CDK3, CDK4, CDK5, CDK6, CDK7, CDK9, E2F1, EGFR, ENO1, ERBB2, ESR1, ESR2, IGFBP3, IGFBP6, IL2, INSL4, MYC, NOX5, NR6A1, PAP, PCNA, PRKCQ, PRKD1, PRL, TP53, FGF22, FGF23, FGF9, IGFBP3, IL2, INHA, KLK6, TP53, CHGB, GNRH1, IGF1, IGF2, INHA, INSL3, INSL4, PRL, KLK6, SHBG, NR1D1, NR1H3, NR1I3, NR2F6, NR4A3, ESR1, ESR2, NR0B1, NR0B2, NR1D2, NR1H2, NR1H4, NR1I2, NR2C1, NR2C2, NR2E1, NR2E3, NR2F1, NR2F2, NR3C1, NR3C2, NR4A1, NR4A2, NR5A1, NR5A2, NR6A1, PGR, RARB, FGF1, FGF2, FGF6, KLK3, KRT1, APOC1, BRCA1, CHGA, CHGB, CLU, COL1A1, COL6A1, EGF, ERBB2, ERK8, FGF1, FGF10, FGF11, FGF13, FGF14, FGF16, FGF17, FGF18, FGF2FGF20, FGF21, FGF22, FGF23, FGF3, FGF4, FGF5, FGF6, FGF7, FGF8, FGF9, GNRH1, IGF1, IGF2, IGFBP3, IGFBP6, IL12A, IL1A, IL1B, IL2, IL24, INHA, INSL3, INSL4, KLK10, KLK12, KLK13, KLK14, KLK15, KLK3, KLK4, KLK5, KLK6, KLK9, MMP2, MMP9, MSMB, NTN4, ODZ1, PAP, PLAU, PRL, PSAP, SERPINA3, SHBG, TGFA, TIMP3, CD44, CDH1, CDH10, CDH19, CDH20, CDH7, CDH9, CDH1, CDH10, CDH13, CDH18, CDH19, CDH20, CDH7, CDH8, CDH9, ROBO2, CD44, ILK, ITGA1, APC, CD164, COL6A1, MTSS1, PAP, TGFB1I1, AGR2, AIG1, AKAP1, AKAP2, CANT1, CAV1, CDH12, CLDN3, CLN3, CYB5, CYC1, DAB21P, DES, DNCL1, ELAC2, ENO2, ENO3, FASN, FLJ12584, FLJ25530, GAGEB1, GAGEC1, GGT1, GSTP1, HIP1, HUMCYT2A, IL29, K6HF, KAI1, KRT2A, MIB1, PART1, PATE, PCA3, PIAS2, PIK3CG, PPID, PR1, PSCA, SLC2A2, SLC33μl, SLC43μl, STEAP, STEAP2, TPM1, TPM2, TRPC6, ANGPT1, ANGPT2, ANPEP, ECGF1, EREG, FGF1, FGF2, FIGF, FLT1, JAG1, KDR, LAMA5, NRP1, NRP2, PGF, PLXDC1, STAB1, VEGF, VEGFC, ANGPTL3, BAI1, COL4A3, IL8, LAMA5, NRP1, NRP2, STAB1, ANGPTL4, PECAM1, PF4, PROK2, SERPINF1, TNFAIP2, CCL11, CCL2, CXCL1, CXCL10, CXCL3, CXCL5, CXCL6, CXCL9, IFNA1, IFNB1, IFNG, IL1B, IL6, MDK, EDG1, EFNA1, EFNA3, EFNB2, EGF, EPHB4, FGFR3, HGF, IGF1, ITGB3,PDGFA, TEK, TGFA, TGFB1, TGFB2, TGFBR1, CCL2, CDH5, COL1A1, EDG1, ENG, ITGAV, ITGB3, THBS1, THBS2, BAD, BAG1, BCL2, CCNA1, CCNA2, CCND1, CCNE1, CCNE2, CDH1 (E-cadherin), CDKN1B (p27Kip1), CDKN2A (p16INK4a), COL6A1, CTNNB1 (β-catenin), CTSB (cathepsin B), ERBB2 (Her-2), ESR1, ESR2, F3 (TF), FOSL1 (FRA-1), GATA3, GSN (gelsolin), IGFBP2, IL2RA, IL6, IL6R, IL6ST (glycoprotein 130), ITGA6 (α6 integrin), JUN, KLK5, KRT19, MAP2K7 (c-Jun), MKI67 (Ki-67), NGFB (NGF), NGFR, NME1 (M23A), PGR, PLAU (uPA), PTEN, SERPINB5 (maspin), SERPINE1 (PAI-1), TGFA, THBS1 (thrombospondin-1), TIE (Tie-1), TNFRSF6 (Fas), TNFSF6 (FasL), TOP2A (topoisomerase IIα), TP53, AZGP1 (zinc-α-glycoprotein), BPAG1 (placentin), CDKN1A (p21Waf1 / Cip1), CLDN7 (claudin-7), CLU (clusterin), ERBB2 (Her-2), FGF1, FLRT1 (fibronectin), GABRP (GABAa), GNAS1, ID2, ITGA6 (α6 integrin), ITGB4 (β4 integrin), KLF5 (GC Box BP), KRT19 (keratin 19), KRTHB6 (hair-specific type II keratin), MACMARCKS, MT3 (metallothionein-III), MUC1 (mucin), PTGS2 (COX-2), RAC2 (p21Rac2), S100A2, SCGB1D2 (lipophilin B), SCGB2A1 (mammaglobin 2), SCGB2A2 (mammaglobin 1), SPRR1B (Spr1), THBS1, THBS2, THBS4 and TNFAIP2 (B94), RON, c-Met, CD64, DLL4, PLGF, CTLA4, phosphatidylserine, ROBO4, CD80, CD22, CD40, CD23, CD28, CD80,Those selected from the group consisting of CD55, CD38, CD70, CD74, CD30, CD138, CD56, CD33, CD2, CD137, DR4, DR5, RANKL, VEGFR2, PDGFR, VEGFR1, MTSP1, MSP, EPHB2, EPHA1, EPHA2, EpCAM, PGE2, NKG2D, LPA, SIP, APRIL, BCMA, MAPG, FLT3, PDGFR alpha, PDGFR beta, ROR1, PSMA, PSCA, SCD1, and CD59 are included. To form the bispecific or trispecific antibodies of the present invention, antibodies against any combination of these antigens can be prepared; that is, each of these antigens may optionally and independently be excluded from, or contained in, the multispecific antibodies according to the present invention.,

[0140] Monoclonal antibody therapy has become an important therapeutic modality for the treatment of autoimmune and inflammatory diseases (Chan & Carter, 2010, Nature Reviews Immunology 10:301-316; Reichert et al., 2005, Nature Biotechnology 23[9]:1073-1078, which are hereby expressly incorporated by reference). Many proteins are involved in common autoimmune and inflammatory responses and can thus be targets of the immunoglobulins of the present invention. Targets for autoimmunity and inflammation include, but are not limited to, C5, CCL1 (I-309), CCL11 (eotaxin), CCL13 (mcp-4), CCL15 (MIP-1d), CCL16 (HCC-4), CCL17 (TARC), CCL18 (PARC), CCL19, CCL2 (mcp-1), CCL20 (MIP-3a), CCL21 (MIP-2), CCL23 (MPIF-1), CCL24 (MPIF-2 / eotaxin-2), CCL25 (TECT), CCL26, CCL3 (MIP-1a), CCL4 (MIP-1b), CCL5 (RANTES), CCL7 (mcp-3), CCL8 (mcp-2), CXCL1, CXCL10 (IP-10), CXCL11 (1-TAC / IP-9), CXCL12 (SDF1), CXCL13, CXCL14, CXCL2, CXCL3, CXCL5 (ENA-78 / LIX), CXCL6 (GCP-2), CXCL9, IL13, IL8, CCL13 (mcp-4), CCR1, CCR2, CCR3, CCR4, CCR5, CCR6, CCR7, CCR8, CCR9, CX3CR1, IL8RA, XCR1 (CCXCR1), IFNA2, IL10, IL13, IL17C, IL1A, IL1B, IL1F10, IL1F5, IL1F6, IL1F7, IL1F8, IL1F9, IL22, IL5, IL8, IL9, LTA, LTB, MIF, SCYE1 (endothelial monocyte activating cytokine), SPP1, TNF, TNFSF5, IFNA2, IL10RA, IL10RB, IL13, IL13RA1, IL5RA, IL9, IL9R, ABCF1, BCL6, C3, C4A, CEBPB, CRP, ICEBERG, IL1R1, IL1RN, IL8RB, LTB4R, TOLLIP, FADD, IRAK1,IRAK2, MYD88, NCK2, TNFAIP3, TRADD, TRAF1, TRAF2, TRAF3, TRAF4, TRAF5, TRAF6, ACVR1, ACVR1B, ACVR2, ACVR2B, ACVRL1, CD28, CD3E, CD3G, CD3Z, CD69, CD80, CD86, CNR1, CTLA4, CYSLTR1, FCER1A, FCER2, FCGR3A, GPR44, HAVCR2, OPRD1, P2RX7, TLR2, TLR3, TLR4, TLR5, TLR6, TLR7, TLR8, TLR9, TLR10, BLR1, CCL1, CCL2, CCL3, CCL4, CCL5, CCL7, CCL8, CCL11, CCL13, CCL15, CCL16, CCL17, CCL18, CCL19, CCL20, CCL21, CCL22, CCL23, CCL24, CCL25, CCR1, CCR2, CCR3, CCR4, CCR5, CCR6, CCR7, CCR8, CCR9, CX3CL1, CX3CR1, CXCL1, CXCL2, CXCL3, CXCL5, CXCL6, CXCL10, CXCL11, CXCL12, CXCL13, CXCR4, GPR2, SCYE1, SDF2, XCL1, XCL2, XCR1, AMH, AMHR2, BMPR1A, BMPR1B, BMPR2, C19orf10(IL27w), CER1, CSF1, CSF2, CSF3, DKFZp451J0118, FGF2, GFI1, IFNA1, IFNB1, IFNG, IGF1, IL1A, IL1B, IL1R1, IL1R2, IL2, IL2RA, IL2RB, IL2RG, IL3, IL4, IL4R, IL5, IL5RA, IL6, IL6R, IL6ST, IL7, IL8, IL8RA, IL8RB, IL9, IL9R, IL10, IL10RA, IL10RB, IL11, IL12RA, IL12A, IL12B, IL12RB1, IL12RB2, IL13, IL13RA1, IL13RA2, IL15, IL15RA, IL16, IL17, IL17R, IL18, IL18R1, IL19, IL20, KITLG, LEP, LTA, LTB, LTB4R, LTB4R2, LTBR, MIF, NPPB, PDGFB, TBX21, TDGF1, TGFA, TGFB1, TGFB1I1, TGFB2, TGFB3, TGFB1, TGFBR1, TGFBR2, TGFBR3, TH1L, TNFTNFRSF1A, TNFRSF1B, TNFRSF7, TNFRSF8, TNFRSF9, TNFRSF11A, TNFRSF21, TNFSF4, TNFSF5, TNFSF6, TNFSF11, VEGF, ZFPM2, and RNF110 (ZNF144) are included. Antibodies against any combination of these antigens can be made to form the bispecific or trispecific antibodies of the present invention; that is, each of these antigens may optionally and independently be excluded from, or included in, the multispecific antibodies according to the present invention.,

[0141] Examples of co-targets for autoimmune and inflammatory diseases include, but are not limited to, IL-1 and TNFalpha, IL-6 and TNFalpha, IL-6 and IL-1, IgE and IL-13, IL-1 and IL-13, IL-4 and IL-13, IL-5 and IL-13, IL-9 and IL-13, CD19 and FcγRIIb, and CD79 and FcγRIIb.

[0142] The immunoglobulins of the present invention having specificity for the following target pairs for the treatment of inflammatory diseases are expected: TNF and IL-17A; TNF and RANKL; TNF and VEGF; TNF and SOST; TNF and DKK; TNF and αVβ3; TNF and NGF; TNF and IL-23p19; TNF and IL-6; TNF and SOST; TNF and IL-6R; TNF and CD-20; IgE and IL-13; IL-13 and IL23p19; IgE and IL-4; IgE and IL-9; IgE and IL-9; IgE and IL-13; IL-13 and IL-9; IL-13 and IL-4; IL-13 and IL-9; IL-13 and IL-9; IL-13 and IL-4; IL-13 and IL-23p19; IL-13 and IL-9; IL-6R and VEGF; IL-6R and IL-17A; IL-6R and RANKL; IL-17A and IL-1β; IL-1β and RANKL; IL-1β and VEGF; RANKL and CD-20; IL-1α and IL-1β; IL-1α and IL-1β.

[0143] Targets to which the immunoglobulins disclosed herein can bind and which may be useful in the treatment of asthma may be determined. In one embodiment, such targets include, but are not limited to, IL-13 and IL-1β (since IL-1β is involved in the inflammatory response of asthma); IL-13 and cytokines and chemokines involved in inflammation (e.g., IL-13 and IL-9; IL-13 and IL-4; IL-13 and IL-5; IL-13 and IL-25; IL-13 and TARC; IL-13 and MDC; IL-13 and MIF; IL-13 and TGF-β; IL-13 and LHR agonist; IL-13 and CL25; IL-13 and SPRR2a; IL-13 and SPRR2b; and IL-13 and ADAM8). The immunoglobulins herein may have specificity for one or more of the targets involved in asthma selected from the group consisting of CSF1 (MCSF), CSF2 (GM-CSF), CSF3 (GCSF), FGF2, IFNA1, IFNB1, IFNG, histamine and histamine receptors, IL1A, IL1B, IL2, IL3, IL4, IL5, IL6, IL7, IL8, IL9, IL10, IL11, IL12A, IL12B, IL13, IL14, IL15, IL16, IL17, IL18, IL19, KITLG, PDGFB, IL2RA, IL4R, IL5RA, IL8RA, IL8RB, IL12RB1, IL12RB2, IL13RA1, IL13RA2, IL18R1, TSLP, CCLi, CCL2, CCL3, CCL4, CCL5, CCL7, CCL8, CCL13, CCL17, CCL18, CCL19, CCL20, CCL22, CCL24, CX3CL1, CXCL1, CXCL2, CXCL3, XCLi, CCR2, CCR3, CCR4, CCR5, CCR6, CCR7, CCR8, CX3CR1, GPR2, XCR1, FOS, GATA3, JAK1, JAK3, STAT6, TBX21, TGFB1, TNF, TNFSF6, YY1, CYSLTR1, FCER1A, FCER2, LTB4R, TB4R2, LTBR, and chitinase.To form the bispecific or trispecific antibodies of the invention, antibodies can be made against any combination of these antigens; i.e., each of these antigens may optionally and independently be excluded from, or included in, the multispecific antibodies according to the invention.

[0144] Target pairs involved in rheumatoid arthritis (RA) may be co-captured according to the invention, including but not limited to, TNF and IL-18; TNF and IL-12; TNF and IL-23; TNF and IL-1beta; TNF and MIF; TNF and IL-17; and, TNF and IL-15.

[0145] Antigens that can be targeted by the immunoglobulins herein for the treatment of systemic lupus erythematosus include, but are not limited to, CD-20, CD-22, CD-19, CD28, CD4, CD80, HLA-DRA, IL10, IL2, IL4, TNFRSF5, TNFRSF6, TNFSF5, TNFSF6, BLR1, HDAC4, HDAC5, HDAC7A, HDAC9, ICOS-L, IGBP1, MS4A1, RGS1, SLA2, CD81, IFNB1, IL10, TNFRSF5, TNFRSF7, TNFSF5, AICDA, BLNK, GALNAC4S-6ST, HDAC4, HDAC5, HDAC7A, HDAC9, IL10, IL11, IL4, INHA, INHBA, KLF6, TNFRSF7, CD28, CD38, CD69, CD80, CD83, CD86, DPP4, FCER2, IL2RA, TNFRSF8, TNFSF7, CD24, CD37, CD40, CD72, CD74, CD79A, CD79B, CR2, IL1R2, ITGA2, ITGA3, MS4A1, ST6GAL1, CD1C, CHST10, HLA-A, HLA-DRA, and NT5E.; CTLA4, B7.1, B7.2, BlyS, BAFF, C5, IL-4, IL-6, IL-10, IFN-α, and TNF-α. To form the bispecific or trispecific antibodies of the present invention, antibodies against any combination of these antigens can be made; that is, each of these antigens may optionally and independently be excluded from, or included in, the multispecific antibodies according to the present invention.

[0146] The immunoglobulins herein may target antigens for the treatment of multiple sclerosis (MS), including, but not limited to, IL-12, TWEAK, IL-23, CXCL13, CD40, CD40L, IL-18, VEGF, VLA-4, TNF, CD45RB, CD200, IFNgamma, GM-CSF, FGF, C5, CD52, and CCR2. Embodiments include co-capture of anti-IL-12 and TWEAK for the treatment of MS.

[0147] One aspect of the present invention relates to an immunoglobulin capable of binding to one or more targets involved in sepsis. In one embodiment, the targets are two, and are selected from the group consisting of TNF, IL-1, MIF, IL-6, IL-8, IL-18, IL-12, IL-23, FasL, LPS, Toll-like receptor, TLR-4, tissue factor, MIP-2, ADORA2A, CASP1, CASP4, IL-10, IL-1B, NFκB1, PROC, TNFRSFIA, CSF3, CCR3, ILIRN, MIF, NFκB1, PTAFR, TLR2, TLR4, GPR44, HMOX1, midkine, IRAK1, NFκB2, SERPINA1, SERPINE1, and TREM1. Antibodies against any combination of these antigens can be made to form bispecific or trispecific antibodies of the present invention; that is, each of these antigens may optionally and independently be excluded from, or included in, the multispecific antibody according to the present invention.

[0148] In some examples, the immunoglobulins herein may be directed against antigens for the treatment of infectious diseases.

[0149] Antigen-Binding Domain As recognized by those skilled in the art, there are two basic types of antigen-binding domains, those similar to the antigen-binding domain of an antibody (e.g., containing a set of six CDRs), and those that are ligands or receptors (e.g., bind to a target without using CDRs).

[0150] Modified Antibody In addition to the above modifications, other modifications may be made. For example, the molecule may be stabilized by incorporation of disulfide bonds that link the VH and VL domains (see Reiter et al., 1996, Nature Biotech. 14:1239-1245, which is incorporated herein by reference in its entirety). Further, various covalent modifications can be made to the antibody as outlined below.

[0151] Covalent modification of antibodies is included within the scope of the present invention and is typically (but not always) performed post-translationally. For example, covalent modification of antibodies of several types is introduced into the molecule by reacting specific amino acid residues of the antibody with an organic derivatizing agent capable of reacting with selected side chains, or the N-terminal or C-terminal residues.

[0152] Cysteinyl residues are the most common residues that react with α-haloacetates (and corresponding amines) such as chloroacetic acid or chloroacetamide, for example, to produce carboxymethyl or carboxamidomethyl derivatives. Cysteinyl residues may also be derivatized by reaction with bromotrifluoroacetone, α-bromo-β-(5-imidazolyl)propionic acid, phosphoryl chloroacetyl, N-alkylmaleimide, 3-nitro-2-pyridyl disulfide, methyl 2-pyridyl disulfide, mercury p-chlorobenzoate, 2-chloromercury-4-nitrophenol, or chloro-7-nitrobenzo-2-oxa-1,3-diazole, among others.

[0153] Furthermore, modification at cysteine is particularly useful in antibody-drug conjugate (ADC) applications (detailed below). In some embodiments, the constant region of the antibody can be engineered to contain one or more cysteines that are specifically "thiol-reactive", thereby making the placement of the drug moiety more specific and controllable. See, for example, U.S. Patent No. 7,521,541, which is incorporated herein by reference in its entirety.

[0154] Histidyl residues are derivatized by reaction with diethyl pyrocarbonate at pH 5.5 - 7.0 (since this agent is relatively specific for histidyl side chains). Parabromophenacyl bromide is also useful; the reaction is preferably carried out at pH 6.0 in 0.1 M sodium cacodylate.

[0155] The lysinyl residue and the amino-terminal residue react with succinic acid or other carboxylic anhydrides. Derivatization with these reagents is effective in restoring the charge of the lysine residue. Other reagents suitable for derivatization of α-amino-containing residues include, for example, imidoesters such as methyl picolinimidate; pyridoxal phosphate; pyridoxal; chloroformic acid; trinitrobenzenesulfonic acid; O-methylisourea; 2,4-pentanedione; and transaminase-catalyzed reactions with glyoxylic acid.

[0156] Arginyl residues are modified by reaction with one or several standard reagents, especially phenylglyoxal, 2,3-butanedione, 1,2-cyclohexanedione, and ninhydrin. For the derivatization of arginine residues, due to the high pKa of the guanidine functional group, the reaction needs to be carried out under alkaline conditions. Furthermore, these reagents can react with lysine groups as well as the arginine epsilon-amino group.

[0157] Specific modification of tyrosyl residues may be carried out with particular interest in introducing spectral labels into the tyrosyl residues by reaction with aromatic diazonium or tetranitromethane. Most commonly, N-acetylimidazole and tetranitromethane are used to form O-acetyltyrosyl species and 3-nitro derivatives, respectively. Tyrosyl residues are iodinated using 125I or 131I to prepare labeled proteins for use in radioimmunoassays (the chloramine T method described above is appropriate).

[0158] Carboxyl side chains (aspartyl or glutamyl) are selectively modified by reaction with carbodiimide (R´-N=C=N--R´), where R and R´ are optionally different alkyl groups (e.g., 1-cyclohexyl-3-(2-morpholinyl-4-ethyl)carbodiimide, or 1-ethyl-3-(4-azonia-4,4-dimethylpentyl)carbodiimide, etc.). Furthermore, aspartyl residues and glutamyl residues are converted to asparaginyl residues and glutaminyl residues by reaction with ammonium ions.

[0159] Derivatization of the bifunctional agent is useful for cross-linking to water-insoluble support matrices or surfaces of antibodies for use in a variety of methods, in addition to the methods described below. Commonly used cross-linking agents include, for example, 1-bis(diazoacetyl)-2-phenylethane, glutaraldehyde, N-hydroxysuccinimide esters (such as esters with 4-azidosalicylic acid), homobifunctional imide esters (such as disuccinimidyl esters (e.g., 3,3´-dithiobis(succinimidylpropionate), etc.)), and bifunctional maleimides (such as bis-N-maleimide-1,8-octane). For example, derivatizing agents such as methyl-3-[(p-azidophenyl)dithio]propionimidate produce photoactivatable intermediates that can form cross-links in the presence of light. Alternatively, reactive water-insoluble matrices (such as cynomolgusogen bromide-activated carbohydrates) and reactive substrates (described in U.S. Patent Nos. 3,969,287; 3,691,016; 4,195,128; 4,247,642; 4,229,537; and 4,330,440; all incorporated by reference in their entirety) are used for protein immobilization.

[0160] Glutaminyl and asparaginyl residues are often deamidated to the corresponding glutamyl and aspartyl residues, respectively. Alternatively, these residues are deamidated under weakly acidic conditions. Any form of these residues is within the scope of the present invention.

[0161] Other modifications include hydroxylation of proline and lysine, phosphorylation of the hydroxyl group of a seryl or threonyl residue, methylation of the α-amino group of lysine, arginine, and histidine side chains (see T. E. Creighton, Proteins: Structure and Molecular Properties, W. H. Freeman & Co., San Francisco, pp. 79-86

[1983] , which is incorporated herein by reference in its entirety), acetylation of the N-terminal amine, and amidation of the C-terminal carboxyl group.

[0162] Furthermore, as will be recognized by those skilled in the art, any label (including fluorescent, enzymatic, magnetic, radioactive labels, etc.) may be added to the antibody (as well as other compositions of the present invention).

[0163] Glycosylation Another type of covalent modification is modification in glycosylation. In other embodiments, the antibodies disclosed herein may be modified to contain one or more engineered glycoforms. As used herein, "engineered glycoform" means a carbohydrate composition covalently attached to an antibody, wherein the carbohydrate composition is chemically different from the native antibody. Engineered glycoforms can be useful for various purposes, including but not limited to enhancing or reducing effector function. A preferred form of engineered glycoform is defucosylation, which has been shown to correlate with increased ADCC function and is presumed to be through stronger binding to the FcγRIIIa receptor. In this context, "defucosylation" means that the majority of the antibodies produced in the host cell are substantially devoid of fucose, and 90-95-98% of the antibodies produced do not have a detectable amount of fucose as a component of the carbohydrate moiety of the antibody (usually attached to N297 of the Fc region). By virtue of a given function, defucosylated antibodies typically exhibit at least 50% or more affinity for FcγRIIIa.

[0164] The engineered glycoform may be produced by various methods known in the art (Umana 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; US6,602,684; USSN10 / 277,370; USSN10 / 113,929; PCT WO00 / 61739A1; PCT WO01 / 29246A1; PCT WO02 / 31140A1; PCT WO02 / 30954A1. All are incorporated by reference in their entirety); (Potelligent (registered trademark) technology [Biowa, Inc., Princeton, NJ]; GlycoMab (registered trademark) glycosylation engineering technology [Glycart Biotechnology AG, Zurich, Switzerland]). Many of these technologies are based on, for example, the expression of IgG in engineered, various organisms or cell lines (e.g., Lec-13 CHO cells or rat hybridoma YB2 / 0 cells), or by controlling enzymes involved in the glycosylation pathway (e.g., FUT8 [α1,6-fucosyltransferase] and / or β1-4-N-acetylglucosaminyltransferase III [GnIII]), or by modifying the carbohydrate(s) after IgG has been expressed, based on controlling the level of fucosylation and / or bisecting of the oligosaccharide covalently attached to the Fc region. For example, the "sugar-engineered antibody" or the Seattle Genetics "SEA technology" by adding modified monosaccharides that inhibit fucosylation during production (see, for example, 20090317869 (incorporated herein by reference in its entirety)). The engineered glycoform typically refers to an unusual carbohydrate or oligosaccharide, and thus, the antibody may contain the engineered glycoform.

[0165] Alternatively, the engineered glycoform may refer to an IgG variant containing an unusual carbohydrate or oligosaccharide. As is known in the art, glycosylation patterns depend on both the protein sequence (e.g., the presence or absence of specific glycosylated amino acid residues discussed below) and the host cell or organism in which the protein is produced. Specific expression systems are discussed below.

[0166] Glycosylation of polypeptides is usually 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 other than proline, are recognition sequences for the enzymatic attachment of carbohydrate moieties to the asparagine side chain. Thus, the presence of any of these tripeptide sequences in a polypeptide creates potential glycosylation sites. 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, but 5-hydroxyproline or 5-hydroxyserine may also be used).

[0167] Addition of glycosylation sites to antibodies can be readily accomplished by modifying the amino acid sequence to contain one or more of the above-described tripeptide sequences (for N-linked glycosylation sites). Modification may also be done by adding or substituting one or more serine or threonine residues to the starting sequence (for O-linked glycosylation sites). For convenience, the amino acid sequence of the antibody is preferably modified via changes at the DNA level such that codons are created that translate to the desired amino acids, and in particular, mutagenesis is preferably performed with preselected bases in the DNA encoding the target polypeptide to effect the modification.

[0168] Other means of increasing the number of carbohydrate moieties on an antibody are by enzymatically or chemically coupling glycosides to the protein. These procedures are advantageous when the acidity of the protein in a host cell having the ability to glycosylate for N-linked and O-linked glycosylation is not required. Depending on the mode of coupling used, the sugar(s) may be attached to (a) arginine and histidine, (b) free carboxyl groups, (c) free sulfhydryl groups (e.g., those of cysteine), (d) free hydroxyl groups (e.g., those of serine, threonine, or hydroxyproline), (e) aromatic residues (e.g., those of phenylalanine, tyrosine or tryptophan), or (f) the amide group of glutamine. These methods are described in WO87 / 05330, and Aplin and Wriston, 1981, CRC Crit. Rev. Biochem., pp. 259-306 (both incorporated by reference in their entirety).

[0169] Removal of sugar moieties present on starting antibodies (e.g., post-translationally) may be performed chemically or enzymatically. Chemical deglycosylation requires that the protein be exposed to trifluoromethanesulfonic acid compounds or equivalent compounds. This treatment cleaves most or all sugars other than the linked sugars (N-acetylglucosamine or N-acetylgalactosamine), while leaving the polypeptide intact. Chemical deglycosylation is described in Hakimuddin et al., 1987, Arch. Biochem. Biophys. 259:52, and Edge et al., 1981, Anal. Biochem. 118:131 (both incorporated herein by reference in their entirety). Enzymatic cleavage of sugar moieties on polypeptides can be accomplished by using various endoglycosidases and exoglycosidases (described in Thotakura et al., 1987, Meth. Enzymol. 138:350; incorporated herein by reference in its entirety). Glycosylation at potential glycosylation sites can be inhibited by use of tunicamycin compounds (described in Duskin et al., 1982, J. Biol. Chem. 257:3105; incorporated herein by reference in its entirety). Tunicamycin inhibits the formation of protein-N-glycoside bonds.

[0170] Conjugation modifications of other types of antibodies include conjugating the antibody to various non-proteinaceous polymers (including, but not limited to, various polyols such as polyethylene glycol, polypropylene glycol, or polyoxyalkylene) by methods described in, for example, the 2005-2006 PEG Catalog (Nektar Therapeutics) (available from Nektar's website), U.S. Patent Nos. 4,640,835; 4,496,689; 4,301,144; 4,670,417; 4,791,192, or 4,179,337 (all incorporated by reference in their entirety). Further, as is known in the art, amino acid substitutions may be made at various positions within the antibody to facilitate the addition of polymers such as PEG. See, for example, U.S. Patent Publication 2005 / 0114037A1 (incorporated by reference in its entirety).

[0171] Additional Fc Mutation for Functional Addition In addition to pI amino acid mutations, there are many useful Fc amino acid modifications that can be made for various reasons (including, but not limited to, modification of binding to one or more FcγR receptors, modification of binding to the FcRn receptor, etc.).

[0172] Accordingly, the proteins of the present invention can contain amino acid modifications including the heterodimerization mutations outlined herein, including pI mutations and conformational mutations. Each set of mutations can be independently and optionally included or excluded in any particular heterodimeric protein.

[0173] FcγR Mutation Thus, there are numerous useful Fc substitutions that can be made to modify the binding to one or more FcγR receptors. Substitutions that result in increased binding, as well as decreased binding, can be useful. For example, an increase in binding to FcγRIIIa typically results in an increase in ADCC (antibody-dependent cell-mediated cytotoxicity; a cell-mediated reaction in which non-specific cytotoxic cells expressing FcγR recognize an antibody bound to a target cell and subsequently cause lysis of the target cell). Similarly, a decrease in binding to FcγRIIb (an inhibitory receptor) can also be beneficial in some circumstances. Amino acid substitutions for which uses have been found in the present invention include those listed in USSN11 / 124,620 (especially FIG. 41), 11 / 174,287, 11 / 396,495, 11 / 538,406 (all of which are hereby expressly incorporated by reference in their entirety for the mutations disclosed therein). Specific mutations for which uses have been found include, but are not limited to, 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D, 332E / 330L, 243A, 243L, 264A, 264V, and 299T.

[0174] Furthermore, these can also be additional Fc substitutions for which uses have been found in increasing the binding to the FcRn receptor and in the extension of the effector arm (specifically disclosed in USSN12 / 341,769, which is hereby incorporated by reference in its entirety herein), including, but not limited to, 434S, 434A, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L, and 259I / 308F / 428L.

[0175] Linker The present invention optionally comprises a linker, if necessary, for example when adding further antigen-binding sites, and there is, for example, that shown in Figure 2, where the "other end" of the molecule contains an additional antigen-binding component. Further, as outlined below, the linker is optionally also used in an antibody-drug conjugate (ADC) system. When used to link the components of the central mAb-Fv construct, the linker is usually a polypeptide containing two or more amino acid residues linked by peptide bonds and is used to link one or more components of the present invention. Such linker polypeptides are known in the art (see, for example, Holliger, P., et al. (1993) Proc. Natl. Acad. Sci. USA 90:6444-6448; Poljak, R. J., et al. (1994) Structure 2:1121-1123). Various linkers find use in some embodiments herein. As will be appreciated by those skilled in the art, there are at least three different types of linkers in the present invention.

[0176] As used herein, "linker" refers to "linker sequence", "spacer", "tethering sequence", or grammatical equivalents thereof. Homogeneous or heterobifunctional linkers are also known in the art (see the technical section on crosslinking agents in the 1994 Pierce Chemical Company catalog, pages 155 - 200, which is incorporated herein by reference in its entirety). Many strategies can be used to covalently attach molecules, including but not limited to polypeptide bonds between the N - and C - termini of a protein or protein domain, bonds via disulfide linkages, and bonds via chemical crosslinking agents. In one aspect of this embodiment, the linker is a peptide bond made by recombinant techniques or peptide synthesis. The linker peptide may mainly contain the following amino acid residues: Gly, Ser, Ala, or Thr. The linker peptide must have an appropriate length to link two molecules in such a way that they adopt the correct structure relative to each other so that the desired activity is retained. In one embodiment, the linker has a length of about 1 - 50 amino acids, preferably about 1 - 30 amino acids. In one embodiment, a linker having a length of 1 - 20 amino acids may be used. Useful linkers include polymers of glycine - serine (e.g., (GS)n, (GSGGS)n, (GGGGS)n, and (GGGS)n, where n is at least one integer), polymers of glycine - alanine, polymers of alanine - serine, and other flexible linkers. Alternatively, various non - proteinaceous polymers (including but not limited to polyethylene glycol (PEG), polypropylene glycol, polyoxyalkylene, or copolymers of polyethylene glycol and polypropylene glycol) find use as linkers.

[0177] Other linker sequences include any sequence of the CL / CH1 domain (excluding the full - length CL / CH1) of any length; for example, the first 5 - 12 amino acid residues of the CL / CH1 domain. The linker may be from an immunoglobulin light chain (e.g., C κOr it may be derived from Cλ). The linker may be derived from the immunoglobulin heavy chain of any isotype (e.g., Cγ1, Cγ2, Cγ3, Cγ4, Cα1, Cα2, Cδ, Cε, and Cμ). The linker sequence may also be derived from other proteins such as, for example, Ig-like proteins (e.g., TCR, FcR, KIR, etc.), sequences derived from hinge regions, and native sequences of other proteins.

[0178] Antibody-Drug Conjugate In some embodiments, the multispecific antibodies of the invention are conjugated to a drug to form an antibody-drug conjugate (ADC). Generally, ADCs are used in cancer applications, and by using the antibody-drug conjugate for local delivery of a cytotoxic agent or a cytostatic agent, it is possible to target and deliver the drug moiety to the tumor, thereby resulting in low toxicity and high efficacy, etc. The general outline of this technology is disclosed in Ducry et al., Bioconjugate Chem., 21:5-13 (2010), Carter et al., Cancer J. 14(3):154 (2008) and Current Opin. Chem. Biol. 13:235-244 (2009) (all of which are incorporated herein by reference in their entirety).

[0179] Therefore, the present invention discloses a multispecific antibody conjugated to a drug. Generally, the conjugation is made by covalent bonding to the antibody as detailed below and usually relies on a linker (in many cases, a peptide linker) (designed to be sensitive or not sensitive to cleavage by proteases at the target site as described below). Further, as described above, the linkage of the linker-drug unit (LU-D) may be made by addition to a cysteine in the antibody. As recognized by those skilled in the art, the number of drugs per antibody may be varied depending on the reaction conditions and may vary in a drug:antibody ratio of 1:1 to 10:1. As recognized by those skilled in the art, the real numbers are averages.

[0180] Accordingly, the present invention discloses a multispecific antibody conjugated to a drug. As described below, the drug of the ADC may be any number of agents and is not limited to, but includes cytotoxic agents (e.g., chemotherapeutic agents, growth inhibitors, toxins (e.g., enzymatically active bacterial toxins, mycotoxins, phytotoxins, or toxins of animal origin, or fragments thereof, etc.) or radioisotopes (i.e., radiolabeled compounds), etc.). In other embodiments, the present invention discloses a method of using an ADC in a dish.

[0181] Drugs for use in the present invention include cytotoxic agents, particularly those used in cancer treatment. Such drugs typically include DNA-damaging agents, antimetabolites, natural products, and their analogs. Exemplary types of cytotoxic agents include enzyme inhibitors such as dihydrofolate reductase inhibitors and thymidylate synthase inhibitors, DNA intercalators, DNA cleaving agents, topoisomerase inhibitors, anthracycline drug families, vinca agents, mitomycin, bleomycin, cytotoxic nucleosides, pteridine drug families, diynenes, podophyllotoxin, dolastatin, maytansinoid, differentiation inducers, and taxol.

[0182] Examples of drugs in these classifications include, for example, methotrexate, methopterin, dichloromethotrexate, 5-fluorouracil, 6-mercaptopurine, cytosine arabinoside, melphalan, leurosine, leurosideine, actinomycin, daunorubicin, doxorubicin, mitomycin C, mitomycin A, caminomycin, aminopterin, tallysomycin, podophyllotoxin and podophyllotoxin derivatives (e.g., etoposide or etoposide phosphate), vinblastine, vincristine, vindesine, taxanes (including taxol), taxotere, tretinoin, butyric acid, N8-acetylspermidine, camptothecin, calicheamicin, esperamicin, enediyne, zuocamycin A, zuocamycin SA, calicheamicin, camptothecin, maytansinoids (including DM1), monomethyl auristatin E (MMAE), monomethyl auristatin F (MMAF), and maytansinoids (DM4) and their analogs.

[0183] The toxin may be used as an antibody-toxin conjugate, and examples include bacterial toxins such as diphtheria toxin, plant toxins such as ricin, small molecule toxins such as geldanamycin (Mandler et al (2000) J. Nat. Cancer Inst. 92(19):1573-1581; Mandler et al (2000) Bioorganic & Med. Chem. Letters 10:1025-1028; Mandler et al (2002) Bioconjugate Chem. 13:786-791), maytansinoids (EP 1391213; Liu et al., (1996) Proc. Natl. Acad. Sci. USA 93:8618-8623), and calicheamicin (Lode et al (1998) Cancer Res. 58:2928; Hinman et al (1993) Cancer Res. 53:3336-3342). The toxin can exert its cytotoxic effect and cell growth inhibitory effect through mechanisms including tubulin binding, DNA binding, or topoisomerase inhibition.

[0184] Multispecific antibodies are expected to be conjugated to one or more small molecule toxins (e.g., maytansinoids, dolastatin, auristatin, trichothecene, calicheamicin, and CC1065, and derivatives of these toxins having toxin activity, etc.).

[0185] Mytansine Maytansinoid compounds suitable for use as the maytansinoid drug moiety are known in the art and can be isolated from natural substances according to known methods, can be produced using genetic engineering techniques (see Yu et al (2002) PNAS 99:7968 - 7973), or maytansinol and maytansinol analogs can be prepared synthetically according to known methods. As described below, the drug may be modified by incorporating a group that is functionally active towards the binding of the antibody, such as, for example, a thiol or amine group, etc.

[0186] Examples of exemplary maytansinoid drug moieties include, for example, C - 19 - dechloro (U.S. Patent No. 4,256,746) (prepared by lithium aluminum hydride reduction of ansamitocin P2); C - 20 - hydroxy (or C - 20 - demethyl) + / - C - 19 - dechloro (U.S. Patents Nos. 4,361,650 and 4,307,016) (prepared by demethylation using Streptomyces or Micromonospora, or dechlorination using LAH); and those having a modified aromatic ring such as C - 20 - de-methoxy, C - 20 - acyloxy (--OCOR), + / - dechloro (U.S. Patent No. 4,294,757) (prepared by acylation using acyl chloride), and those having modifications at other positions.

[0187] Examples of mitansinoid drug moieties also include, for example, C-9-SH (U.S. Patent No. 4,424,219) (prepared by reaction of mitansinol with H2S or P2S5); C-14-alkoxymethyl (demethoxy / CH2OR) (U.S. Patent No. 4,331,598); C-14-hydroxymethyl or acyloxymethyl (CH2OH or CH2OAc) (U.S. Patent No. 4,450,254) (prepared from Nocardia); C-15-hydroxy / acyloxy (U.S. Patent No. 4,364,866) (prepared by conversion of mitansinol by Streptomyces); C-15-methoxy (U.S. Patents Nos. 4,313,946 and 4,315,929) (isolated from Trewia nudiflora); C-18-N-demethyl (U.S. Patents Nos. 4,362,663 and 4,322,348) (prepared by demethylation of mitansinol by Streptomyces); and those having modifications such as 4,5-deoxy (U.S. Patent No. 4,371,533) (prepared by LAH reduction of titanium trichloride / mitansinol), etc.

[0188] DM1 (disclosed in U.S. Patent No. 5,208,020, incorporated by reference) and DM4 (disclosed in U.S. Patent No. 7,276,497, incorporated by reference) are particularly used. See also many further mitansinoid derivatives and methods of 5,416,064, WO / 01 / 24763, 7,303,749, 7,601,354, USSN 12 / 631,508, WO02 / 098883, 6,441,163, 7,368,565, WO02 / 16368 and WO04 / 1033272 (all of which are hereby expressly incorporated by reference in their entirety).

[0189] ADCs containing maytansinoids, methods for their preparation, and their therapeutic uses are disclosed, for example, in U.S. Patent Nos. 5,208,020; 5,416,064; 6,441,163 and European Patent No. 0 425 235 B1 (the disclosures of which are hereby expressly incorporated by reference). In Liu et al., Proc. Natl. Acad. Sci. USA 93:8618-8623 (1996), an ADC containing a maytansinoid designated DM1 conjugated to the monoclonal antibody C242 against human colorectal cancer was described. The conjugate was found to be highly cytotoxic to cultured colon cancer cells and showed antitumor activity in in vivo tumor growth assays.

[0190] In Chari et al., Cancer Research 52:127-131 (1992), an ADC in which a maytansinoid is conjugated via a disulfide bond to the mouse antibody A7 that binds to an antigen on a human colon cancer cell line or to another mouse monoclonal antibody TA.1 that binds to the HER-2 / neu oncogene was disclosed. The cytotoxicity of the TA.1 maytansinoid conjugate was verified in vitro against the human breast cancer cell line SK-BR-3 (expressing 3x105 HER-2 surface antigens / cell). The drug conjugate exerted cytotoxicity comparable to that of the free maytansinoid drug, and increasing the number of maytansinoid molecules per antibody molecule also increased the degree of cytotoxicity. The A7-maytansinoid conjugate had low systemic cytotoxicity in mice.

[0191] Auristatin and Dolastatin In some embodiments, the ADC contains a multispecific antibody conjugated to dolastatin or a dolastatin peptide analog and derivative, auristatin (U.S. Patent Nos. 5,635,483; 5,780,588). Dolastatin and auristatin have been shown to interfere with microtubule dynamics, GTP hydrolysis, and nuclear and cell division (Woyke et al (2001) Antimicrob. Agents and Chemother. 45(12):3580-3584), and have anticancer activity (U.S. Patent No. 5,663,149) and antifungal activity (Pettit et al (1998) Antimicrob. Agents Chemother. 42:2961-2965). The dolastatin or auristatin drug moiety may be attached to the antibody via the N (amino) terminus or C (carboxyl) terminus of the peptide drug moiety (WO 02 / 088172).

[0192] Exemplary embodiments of auristatin include the N-terminal conjugated monomethyl auristatin drug moieties DE and DF, as disclosed in Senter et al, Proceedings of the American Association for Cancer Research, Volume 45, Abstract Number 623 (published March 28, 2004), and described in U.S. Patent Publication 2005 / 0238648 (the disclosures of which are hereby expressly incorporated by reference in their entireties).

[0193] An exemplary embodiment of auristatin is MMAE (see U.S. Patent No. 6,884,869, which is hereby expressly incorporated by reference in its entirety).

[0194] Another exemplary embodiment of auristatin is MMAF (see US2005 / 0238649, 5,767,237 and 6,124,431, which are hereby expressly incorporated by reference in their entireties).

[0195] Further exemplary embodiments comprising MMAE or MMAF and various linker components (detailed herein) have the following structures and abbreviations (Ab means antibody and p is from 1 to about 8).

[0196] Typically, the peptide-based agent moiety can be prepared by forming peptide bonds between two or more amino acids and / or peptide fragments. Such peptide bonds may be prepared, for example, according to liquid phase synthesis methods known in the field of peptide chemistry (see E. Schroder and K. Lubke, “The Peptides”, volume 1, pp 76-136, 1965, Academic Press). The auristatin / dolastatin agent moiety may be prepared according to the methods of U.S. Patent No. 5,635,483; U.S. Patent No. 5,780,588; Pettit et al (1989) J. Am. Chem. Soc. 111:5463-5465; Pettit et al (1998) Anti-Cancer Drug Design 13:243-277; Pettit, G. R., et al. Synthesis, 1996, 719-725; Pettit et al (1996) J. Chem. Soc. Perkin Trans. 1 5:859-863; and Doronina (2003) Nat Biotechnol 21(7):778-784.

[0197] Calicheamicin In other embodiments, the ADC contains an antibody of the invention conjugated to one or more calicheamicin molecules. For example, Mylotarg is the first commercially available ADC drug and uses calicheamicin γ1 as a payload (see U.S. Patent No. 4,970,198, which is hereby incorporated by reference in its entirety). Further calicheamicin derivatives are disclosed in U.S. Patent Nos. 5,264,586, 5,384,412, 5,550,246, 5,739,116, 5,773,001, 5,767,285, and 5,877,296 (all of which are hereby expressly incorporated by reference). The antibiotic calicheamicin family can effect double-strand DNA breakage at sub-picomolar concentrations. For the preparation of calicheamicin family conjugates, see U.S. Patent Nos. 5,712,374, 5,714,586, 5,739,116, 5,767,285, 5,770,701, 5,770,710, 5,773,001, 5,877,296 (all to American Cyanamid Company). Structural analogs of calicheamicin that can be used include, but are not limited to, γ1I, α2I, α2I, N-acetyl-γ1I, PSAG, and θI1 (Hinman et al., Cancer Research 53:3336-3342 (1993), Lode et al., Cancer Research 58:2925-2928 (1998), and the aforementioned U.S. patents to American Cyanamid). Another antitumor agent to which an antibody can bind is the antifolate QFA. Both calicheamicin and QFA have an intracellular site of action and do not readily penetrate the cell membrane. Therefore, by incorporating these agents into cells by antibody-mediated internalization, the cytotoxic effect can be significantly enhanced.

[0198] Zucarminomycin CC-1065 (see U.S. Patent No. 4,169,888, incorporated by reference) and duocarmycin are one family of antitumor antibiotics used in ADCs. These compounds are thought to act by sequence-selectively alkylating DNA at the N3 of adenine in the minor groove, initiating a cascade of events that lead to apoptosis.

[0199] Important members of duocarmycin include duocarmycin A (U.S. Patent No. 4,923,990, incorporated by reference), and duocarmycin SA (U.S. Patent No. 5,101,038, incorporated by reference), and many analogs disclosed in U.S. Patent Nos. 7,517,903, 7,691,962, 5,101,038, 5,641,780, 5,187,186, 5,070,092, 5,070,092, 5,641,780, 5,101,038, 5,084,468, 5,475,092, 5,585,499, 5,846,545, WO2007 / 089149, WO2009 / 017394A1, 5,703,080, 6,989,452, 7,087,600, 7,129,261, 7,498,302, and 7,507,420 (all of which are expressly incorporated by reference).

[0200] Other Cytotoxic Agents Other antitumor agents that can bind to the antibodies of the present invention include BCNU, streptozocin, vincristine, and 5-fluorouracil, collectively the family of agents known as the LL-E33288 complex (described in U.S. Patent Nos. 5,053,394 and 5,770,710), and esperamicin (U.S. Patent No. 5,877,296).

[0201] Enzymatically active toxins and fragments thereof that can be used include diphtheria A chain, non-binding active fragments of diphtheria toxin, exotoxin A chain (from Pseudomonas aeruginosa), ricin A chain, abrin A chain, modeccin A chain, alpha-sarcin, Aleurites fordii protein, dianthin protein, Phytolaca americana proteins (PAPI, PAPII, and PAP-S), momordica charantia inhibitor, curcin, crotin, sapaonaria officinalis inhibitor, gelonin, mitogelin, restrictocin, phenomycin, enomycin, and trichothecin. See, for example, WO93 / 21232 (published October 1993).

[0202] Further expected from the present invention are ADCs formed between an antibody and a compound having nuclease activity (e.g., ribonuclease or DNA endonuclease (e.g., deoxyribonuclease; DNase)).

[0203] For the selective disintegration of tumors, the antibody may contain a high radioactive atom. Various radioisotopes are available for the production of radiolabeled antibodies. Examples include radioisotopes of At211, I131, I125, Y90, Re186, Re188, Sm153, Bi212, P32, Pb212, and Lu.

[0204] A radioactive label or other label may be incorporated into the conjugate by known methods. For example, a peptide may be biosynthesized or synthesized by chemical amino acid synthesis using a suitable amino acid precursor containing fluorine-19 instead of hydrogen, for example. For example, a label such as Tc99m or I123, Re186, Re188, and In111 may be added via a cysteine residue of the peptide. Yttrium-90 may be added via a lysine residue. The IODOGEN method (Fraker et al (1978) Biochem. Biophys. Res. Commun. 80: 49-57) can be used to incorporate iodine-123. “Monoclonal Antibodies in Immunoscintigraphy” (Chatal, CRC Press 1989) discloses other methods in detail.

[0205] For a composition containing multiple antibodies, the dosage is the average number of drug molecules per antibody, and the dosage is represented by p. It may be in the range of 1 to 20 drugs (D) / antibody. The average number of drugs per antibody in the reaction for preparing the conjugate may be characterized by conventional means such as mass spectrometry, ELISA assay, and HPLC. The quantitative distribution of the antibody-drug-conjugate may be measured with p as the unit.

[0206] In some examples, the separation, purification, and characterization of the homologous antibody-drug-conjugate (p is a predetermined value from an antibody-drug-conjugate with other drug carriers) may be performed by means such as reverse phase HPLC or electrophoresis. In an exemplary embodiment, p is 2, 3, 4, 5, 6, 7, or 8 or a fraction thereof.

[0207] The preparation of the antibody-drug-conjugate can be carried out by any method known to those skilled in the art. Briefly, the antibody-drug-conjugate may contain a multispecific antibody as an antibody unit, a drug, and optionally a linker and a binder that binds the drug.

[0208] Many different reactions are available for the formation of covalent bonds to the binder of the agent and / or linker. This can be done by reaction of the amino acid residues of the binder, and examples include the amine group of lysine, the free carboxylic acid groups of glutamic acid and aspartic acid, the sulfhydryl group of cysteine, and various moieties of aromatic amino acids, including antibody molecules. A commonly used non-specific covalent binding method is the carbodiimide reaction for linking the carboxy (or amino) group of a compound to the amino (or carboxy) group of an antibody. Additionally, a bifunctional agent such as a dialdehyde or imido ester is used to link the amino group of a compound to the amino group of an antibody molecule.

[0209] Also, the Schiff base reaction is available for attaching an agent to a binder. This method involves periodate oxidation of an agent containing a glycol or hydroxy group, thereby forming an aldehyde, which is then reacted with the binder. The addition occurs via the formation of a Schiff base with the amino group of the binder. Isothiocyanates can also be used as coupling agents for covalently binding an agent to a binder. Other techniques are known to those skilled in the art and are within the scope of the present invention.

[0210] In some embodiments, an intermediate, which is a precursor of the linker, is reacted with an agent under suitable conditions. In other embodiments, reactive groups are used on the agent and / or the intermediate. The product of the reaction between the agent and the intermediate, or the derivatized agent, is then reacted with the multispecific antibody of the present invention under suitable conditions.

[0211] The desired compound may be chemically modified to more readily effect its reaction for the purposes of preparing the conjugates of the present invention. For example, a functional group such as an amine, hydroxyl, or sulfhydryl may be added to the agent at a position that has a minimal acceptable impact on the activity or other properties of the agent.

[0212] ADC Linker Unit Typically, an antibody-drug conjugate contains a linker unit between the drug unit and the antibody unit. In some embodiments, the linker is cleavable under intracellular or extracellular conditions, and cleavage of the linker releases the drug unit from the antibody in an appropriate environment. For example, a solid tumor that secretes a certain protease is targeted by a cleavable linker, and in other embodiments, intracellular proteases are used. In yet other embodiments, the linker unit is not cleavable, and the drug is released, for example, by degradation of the antibody in lysosomes.

[0213] In some embodiments, the linker is cleavable by a cleavage agent present in the intracellular environment (e.g., within lysosomes or endosomes or caveolae). The linker may be, for example, a peptidyl linker that is cleaved by an intracellular peptidase enzyme or protease enzyme (including, but not limited to, lysosomal protease or endosomal protease). In some embodiments, the peptidyl linker is at least 2 amino acids in length, or at least 3 amino acids or more in length.

[0214] Cleavage agents include, but are not limited to, cathepsin B and D and plasmin (all of which are known to hydrolyze dipeptide drug derivatives and result in the release of the active drug inside the target cell) (see, for example, Dubowchik and Walker, 1999, Pharm. Therapeutics 83:67-123). A peptidyl linker cleavable by an enzyme present in CD38-expressing cells. For example, a peptidyl linker cleavable by the thiol-dependent protease cathepsin B, which is highly expressed in cancerous tissues, may be used (e.g., Phe-Leu or Gly-Phe-Leu-Gly linker (SEQ ID NO: X)). Other examples of such linkers are disclosed in U.S. Patent No. 6,214,345 (incorporated herein by reference in its entirety for all purposes).

[0215] In some embodiments, the peptidyl linker cleavable by an intracellular protease is a Val-Cit linker or a Phe-Lys linker (see, e.g., U.S. Patent No. 6,214,345, which discloses the synthesis of a val-cit linker and doxorubicin).

[0216] In other embodiments, the cleavable linker is pH-sensitive. That is, it is sensitive to hydrolysis at a certain pH value. Typically, the pH-sensitive linker is hydrolyzable under acidic conditions. For example, an acid-labile linker hydrolyzable in lysosomes (e.g., hydrazone, semicarbazone, thiosemicarbazone, cis-aconitamide, orthoester, acetal, ketal, etc.) may be used (see, e.g., U.S. Patent Nos. 5,122,368, 5,824,805, 5,622,929; Dubowchik and Walker, 1999, Pharm. Therapeutics 83:67-123; Neville et al., 1989, Biol. Chem. 264:14653-14661). Such linkers are relatively stable under neutral pH conditions, such as in blood, for example, but are unstable when below a pH of 5.5 or 5.0, which is the approximate pH of lysosomes. In certain embodiments, the hydrolyzable linker is a thioester linker (e.g., a thioester attached to a therapeutic agent via an acylhydrazone bond) (see U.S. Patent No. 5,622,929).

[0217] In yet other embodiments, the linker is cleavable under reducing conditions (e.g., a disulfide linker). A variety of disulfide linkers are known in the art, including, for example, SATA (N-succinimidyl-5-acetylthioacetate), SPDP (N-succinimidyl-3-(2-pyridyldithio)propionate), SPDB (N-succinimidyl-3-(2-pyridyldithio)butyrate), and SMPT (N-succinimidyl-oxycarbonyl-α-methyl-α-(2-pyridyl-dithio)toluene)-, SPDB, and SMPT. (See, for example, Thorpe et al., 1987, Cancer Res. 47:5924-5931; Wawrzynczak et al., In Immunoconjugates: Antibody Conjugates in Radioimagery and Therapy of Cancer (C. W. Vogel ed., Oxford U. Press, 1987. See also U.S. Patent No. 4,880,935).

[0218] In other embodiments, the linker is a malonic acid linker (Johnson et al., 1995, Anticancer Res. 15:1387-93), a maleimidobenzoyl linker (Lau et al., 1995, Bioorg-Med-Chem. 3(10):1299-1304), or a 3´-N-amide analog (Lau et al., 1995, Bioorg-Med-Chem. 3(10):1305-12).

[0219] In yet other embodiments, the linker unit is not cleavable and the agent is released by degradation of the antibody. See U.S. Patent Publication 2005 / 0238649 (which is hereby incorporated by reference in its entirety for all purposes).

[0220] In many embodiments, the linker is self - cleavable. As used herein, the term "self - cleaving spacer" refers to a bifunctional chemical moiety that can covalently link two distally - placed chemical moieties together to form a three - component molecule. If the bond to the first moiety is cleaved, it spontaneously separates from the second chemical moiety. See, for example, WO2007059404A2, WO06110476A2, WO05112919A2, WO2010 / 062171, WO09 / 017394, WO07 / 089149, WO07 / 018431, WO04 / 043493, and WO02 / 083180 (which are directed to conjugates of a drug and a cleavable substrate optionally linked via a self - cleaving linker, all of which are hereby expressly incorporated by reference).

[0221] In many cases, the linker is not substantially sensitive to the extracellular environment. As used herein, with respect to the linker, "not substantially sensitive to the extracellular environment" means that when the antibody - drug conjugate is present in the extracellular environment (e.g., in plasma), less than about 20%, 15%, 10%, 5%, 3% or less than about 1% of the linker in the antibody - drug conjugate sample is cleaved.

[0222] Whether the linker is not substantially sensitive to the extracellular environment can be measured, for example, by incubating the antibody - drug conjugate with plasma for a defined time (e.g., 2, 4, 8, 16 or 24 hours) and then quantifying the amount of free drug present in the plasma.

[0223] In other, non - mutually exclusive embodiments, the linker promotes endocytosis. In one embodiment, the linker promotes endocytosis when bound to a therapeutic agent (i.e., in the context of the linker - therapeutic agent portion of the antibody - drug conjugates described herein). In yet other embodiments, the linker promotes endocytosis when bound to both an auristatin compound and a multispecific antibody of the invention.

[0224] Examples of various linkers that can be used in the compositions and methods of the present invention are described in WO2004-010957, US Patent Publication 2006 / 0074008, US Patent Publication 20050238649, and US Patent Publication 2006 / 0024317 (each of which is hereby incorporated by reference in its entirety for all purposes).

[0225] Drug Dosage The drug load, represented by p, is the average number of drug moieties per antibody in the molecule. The drug load (p) may be 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20 or more moieties per antibody, although often the average is a fraction or a decimal. Generally, a drug load of 1-4 is often useful, and 1-2 is also useful. The ADCs of the present invention contain a collection of antibodies conjugated in the range of 1-20 drug moieties. The average number of drug moieties per antibody of an ADC prepared from a conjugation reaction may be analyzed by standard means such as mass spectrometry and ELISA assays.

[0226] The quantitative distribution of the ADC in units of p may also be measured. In some examples, the isolation, purification, and characterization of homogeneous ADCs (where p is a value from an ADC having other drug loads) can be performed by means such as electrophoresis.

[0227] For some antibody-drug conjugates, p can be limited by the number of attachment sites on the antibody. For example, where the attachment is a cysteine thiol, as in the exemplary embodiments above, the antibody may have only one or several cysteine thiol groups, or only one or several sufficiently reactive thiol groups, through which the linker is added. In certain embodiments, for example when the drug dose is high such as p>5, aggregation, insolubility, toxicity, or loss of cell permeability of certain antibody-drug conjugates may occur. In certain embodiments, the drug dose for the ADCs of the present invention ranges from 1 to about 8; about 2 to about 6; about 3 to about 5; about 3 to about 4; about 3.1 to about 3.9; about 3.2 to about 3.8; about 3.2 to about 3.7; about 3.2 to about 3.6; about 3.3 to about 3.8; or about 3.3 to about 3.7. It has been shown that for the implementation of certain ADCs, the optimal ratio of drug moieties per antibody can be less than 8 and about 2 to about 5. See 2005-0238649 A1 (incorporated herein by reference in its entirety).

[0228] In certain embodiments, an amount less than the theoretical maximum of the drug moiety is conjugated to the antibody during the conjugation reaction. The antibody may contain, for example, lysine residues that do not react with the drug-linker intermediate or linker reagent (described below). Generally speaking, the antibody does not contain many free and reactive cysteine thiol groups that can be conjugated to the drug moiety; in fact, most cysteine thiol residues in the antibody exist as disulfide bridges. In certain embodiments, the antibody may be reduced under partially or fully reducing conditions using a reducing agent such as dithiothreitol (DTT) or tricarbonyl ethyl phosphate (TCEP) to generate reactive cysteine thiol groups. In certain embodiments, the antibody is subjected to denaturing conditions to expose reactive nucleophilic groups such as lysine or cysteine.

[0229] The amount of ADC (drug / antibody ratio) may be controlled in different ways, for example, (i) limiting the molar excess of the drug-linker intermediate or the molar excess of the linker reagent relative to the antibody, (ii) limiting the binding reaction time or temperature, (iii) partial or limiting reduction conditions for cysteine thiol modification, (iv) manipulating the amino acid sequence of the antibody by recombinant techniques such that the number and position of cysteine residues are modified for control of the number and / or position of linker-drug adducts (e.g., thioMab or thioFab prepared as disclosed in this specification and in WO2006 / 034488, which is hereby incorporated by reference in its entirety).

[0230] It should be understood that when two or more nucleophilic groups react with the drug-linker intermediate or the linker reagent and then with the drug moiety reagent, the resulting product is a mixture of ADC compounds in which one or more drug moieties are attached to the antibody. The average number of drugs per antibody may be calculated from the mixture by a dual ELISA antibody assay that is specific for the antibody and specific for the drug. Individual ADC molecules may be identified in the mixture by mass spectrometry and HPLC separation (e.g., hydrophobic interaction chromatography).

[0231] In some embodiments, a hetero-ADC having a single payload value may be isolated from the conjugate mixture by electrophoresis or chromatography.

[0232] Method for Measuring Cytotoxic Effect of ADC Methods for measuring whether a drug or an antibody-drug conjugate exerts a cytostatic effect and / or a cytotoxic effect on cells are known. Generally, the cytotoxic or cytostatic activity of an antibody-drug conjugate can be measured as follows: exposing mammalian cells expressing the target protein of the antibody-drug conjugate in cell culture medium; culturing the cells for about 6 hours to about 5 days; and measuring the activity of the cells. Cell-based in vitro assays can be used to measure activity (proliferation), cytotoxicity, and induction of apoptosis (caspase activation) by the antibody-drug conjugate.

[0233] To measure whether an antibody-drug conjugate exerts a cell growth inhibitory effect, a thymidine incorporation assay may be used. For example, cancer cells expressing the target antigen may be seeded at a density of 5,000 cells / well, cultured for 72 hours, and exposed to 0.5 μCi of 3H-thymidine during the last 8 hours of the 72-hour period. Uptake of 3H-thymidine into the cultured cells is measured in the presence or absence of the antibody-drug conjugate.

[0234] For the measurement of cytotoxicity, necrosis or apoptosis (programmed cell death) may be measured. Necrosis usually occurs due to increased cell membrane permeability; cell swelling and rupture of the cell membrane. Apoptosis is usually characterized by membrane blebbing, cytoplasmic condensation, and activation of endogenous endonucleases. When these effects on cancer cells are measured, it indicates that the antibody-drug conjugate is useful for the treatment of cancer.

[0235] Cell viability can be determined, for example, by measuring the uptake of dyes such as neutral red, trypan blue, or ALAMAR® Blue (see, for example, Page et al., 1993, Intl. J. Oncology 3:473-476) into the cells. In such assays, the cells are incubated in a medium containing the dye, the cells are washed, and if the dye remains, it reflects the uptake of the dye by the cells and is measured by a spectrophotometer. Sulforhodamine B (SRB), a protein-binding dye, can also be used for the measurement of cytotoxicity (Skehan et al., 1990, J. Natl. Cancer Inst. 82:1107-12).

[0236] Alternatively, for example, tetrazolium salts such as MTT can be used in a quantitative colorimetric assay for the survival and proliferation of mammalian cells by detecting living cells but not dead cells (see, for example, Mosmann, 1983, J. Immunol. Methods 65:55-63).

[0237] Apoptosis can be quantified, for example, by measuring DNA fragmentation. Commercially available optical analysis methods for in vitro quantitative measurement of DNA fragmentation are available. Examples of such assays include TUNEL (detecting incorporation of labeled nucleotides in fragmented DNA) and ELISA-based assays, as described in Biochemica, 1999, no. 2, pp. 34-37 (Roche Molecular Biochemicals).

[0238] Apoptosis can also be determined by measuring morphological changes in cells. For example, with respect to necrosis, the loss of cell membrane integrity can be determined by measuring the uptake of a dye (e.g., a fluorescent dye such as acridine orange or ethidium bromide). Methods for measuring the number of apoptotic cells are disclosed in Duke and Cohen, Current Protocols in Immunology (Coligan et al. eds., 1992, pp. 3.17.1-3.17.16). Cells can also be labeled with a DNA dye (e.g., acridine orange, ethidium bromide, or propidium iodide, etc.), and chromatin condensation and marginalization along the inner nuclear membrane are observed in the cells. Other morphological changes that can be measured to determine apoptosis include, for example, cytoplasmic condensation, increased membrane blebbing, and cell atrophy.

[0239] The presence of apoptotic cells can be measured in both the adherent fraction and the "floating" fraction of the culture. For example, both fractions can be recovered by removing the supernatant, the adherent cells can be trypsinized, the preparations can be mixed after a centrifugation washing step (e.g., 10 minutes at 2000 rpm), and apoptosis can be detected (e.g., by measurement of DNA fragmentation) (see, e.g., Piazza et al., 1995, Cancer Research 55:3110-16).

[0240] The in vivo therapeutic effect of the therapeutic composition of the multispecific antibody of the present invention can be evaluated in a suitable animal model. For example, a heterograft cancer model can be used, where cancer grafts or passaged heterologous tissues are introduced into immunodeficient animals (e.g., nude mice or SCID mice) (Klein et al., 1997, Nature Medicine 3: 402-408). Efficacy can be measured using assays that measure inhibition of tumor formation, tumor regression, metastasis, etc.

[0241] The therapeutic composition used in the above-described implementation can be formulated into a pharmaceutical composition containing a carrier suitable for the desired delivery method. Suitable carriers include any substance that retains the anti-tumor function of the therapeutic composition when mixed with the therapeutic composition and is generally non-reactive with the patient's immune system. Examples include, but are not limited to, any of many standard pharmaceutical carriers such as sterile phosphate buffered saline, bacteriostatic water, etc. (see, for an overview, Remington’s Pharmaceutical Sciences 16th Edition, A. Osal., Ed., 1980).

[0242] Antibody Composition for in vivo Administration The formulations of the antibodies used in accordance with the present invention are prepared for storage in the form of lyophilized formulations or aqueous solutions by mixing an antibody having the desired degree of purity with any pharmaceutically acceptable carrier, excipient or stabilizer (Remington’s Pharmaceutical Sciences 16th edition, Osol, A. Ed.

[1980] ). Acceptable detectors, excipients or stabilizers are non-toxic to the recipient at the dosages and concentrations employed, and include, for example, buffers such as phosphoric acid, citric acid and other organic acids; antioxidants (including ascorbic acid and methionine); preservatives (such as octadecyl dimethyl benzyl ammonium chloride; hexamethonium chloride; benzalkonium chloride, benzethonium chloride; phenol, butyl, or benzyl alcohol; alkyl parabens (such as methyl paraben or propyl paraben); catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol, etc.); low molecular weight (less than about 10 residues) polypeptides; proteins such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers such as polyvinyl pyrrolidone; amino acids such as glycine, glutamine, asparagine, histidine, arginine or lysine; monosaccharides, disaccharides and other carbohydrates (including glucose, mannose or dextrin); chelating agents such as EDTA; sugars such as sucrose, mannitol, trehalose or sorbitol; salt-forming counterions (such as sodium); metal complexes (such as Zn-protein complexes); and / or nonionic surfactants (such as TWEEN®, PLURONICS® or polyethylene glycol (PEG)).

[0243] The formulations herein may also contain, if desired, two or more active compounds (preferably those having complementary activities that do not adversely affect each other) for a particular disease to be treated. For example, it may be desirable to add antibodies having other specificities. Alternatively, or in addition, the composition may contain a cytotoxic agent, a cytokine, a growth inhibitor, and / or a small molecule antagonist. Such molecules are present in appropriate combinations in effective amounts for the intended purpose.

[0244] The active ingredient may also be encapsulated in, for example, microcapsules prepared by droplet formation technology or by interfacial polymerization (such as hydroxy methyl cellulose or gelatin - microcapsules, and poly-(methyl methacrylate) microcapsules respectively), in a colloidal drug delivery system (such as liposomes, albumin microspheres, microemulsions, nanoparticles and nanocapsules), or in a macroemulsion. Such techniques are disclosed in Remington’s Pharmaceutical Sciences 16th edition, Osol, A. Ed. (1980).

[0245] Formulations for in vivo administration must be sterile or nearly so. This can be easily achieved by filtration through a sterile filtration membrane.

[0246] Sustained - release formulations may be prepared. Suitable examples of sustained - release formulations include semi - permeable matrices of hydrophobic, individual polymers containing an antibody, and the matrix is in the form of shaped articles such as films or microcapsules. Examples of sustained - release matrices include polyesters, hydrogels (such as poly(2 - hydroxyethyl - methacrylate), or poly(vinyl alcohol)), polylactide (U.S. Patent No. 3,773,919), copolymers of L - glutamic acid and γ - ethyl - L - glutamic acid, non - degradable ethylene vinyl acetate, degradable lactic acid - glycolic acid copolymers (such as LUPRON DEPOT® (injectable microspheres composed of lactic acid - glycolic acid copolymer and leuprolide acetate), and poly - D - (-)-3 - hydroxybutyric acid. Polymers such as ethylene vinyl acetate and lactic acid - glycolic acid can release molecules over a period of 100 days, while certain hydrogels release proteins over a much shorter period.

[0247] When the encapsulated antibody remains in the body for a long time, it may be degraded or aggregated by exposure to a moist environment at 37°C. As a result, it may lose its biological activity and its immunogenicity may change. Rational strategies for stabilization may be developed depending on the mechanisms involved. For example, if the mechanism of aggregation is found to be the formation of intermolecular S-S bonds via thiol-disulfide exchange, stabilization can be achieved by modifying sulfhydryl residues, lyophilizing from acidic solutions, controlling the water content using appropriate additives, and developing specific polymer matrix compositions.

[0248] Administration Modality The antibody and chemotherapeutic agent of the present invention are administered to a subject according to known methods such as, for example, bolus administration, or intravenous administration, intramuscular administration, intraperitoneal administration, intrathecal administration, subcutaneous administration, intraarterial administration, intra-articular administration, subarachnoid administration, oral administration, topical administration, or inhalation route over a period of time. Intravenous or subcutaneous administration of the antibody is preferred.

[0249] Therapeutic Modality In the method of the present invention, treatment is used to obtain a favorable response to treatment with respect to a disease or condition. "Favorable response to treatment" is intended to mean an improvement in the disease or condition and / or an improvement in the symptoms associated with the disease or condition. For example, a favorable response to treatment refers to one or more of the following improvements in a disease: (1) a decrease in the number of tumor cells; (2) an increase in tumor cell death; (3) inhibition of tumor cell survival; (5) inhibition of tumor growth (i.e., a slowdown to a certain extent, preferably a halt); (6) an increase in the survival rate of the patient; and (7) some relief from one or more of the symptoms associated with the disease or condition.

[0250] A favorable response to treatment in a given disease or condition can be measured by standardized response criteria specific to that disease or condition. Tumor response can be evaluated, for example, using screening techniques such as magnetic resonance imaging (MRI) scans, X-ray radiography, computed tomography (CT) scans, bone scan imaging, endoscopy, and measurement of the number of major cells in the circulation, including tumor biopsy sampling, for changes in tumor morphology (i.e., overall tumor volume, tumor size, etc.).

[0251] In addition to a favorable response to these treatments, subjects treated will experience a beneficial effect of improvement in symptoms associated with the disease.

[0252] Thus, for B cell tumors, the subject will experience a decrease in so-called B symptoms (i.e., night sweats, fever, weight loss, and / or pruritus). For pre-malignant conditions, treatment with a multispecific therapeutic agent may inhibit the development of related malignant diseases (e.g., the development of multiple myeloma in subjects suffering from monoclonal gammopathy of undetermined significance (MGUS)), and / or prolong the time to development.

[0253] Improvement of the disease may be characterized as a complete remission. "Complete remission" is intended to mean the disappearance of clinically detectable disease, accompanied by normalization of any previous abnormal X-ray examination (in the case of myeloma, in the bone marrow, and cerebrospinal fluid (CSF) or monoclonal protein above).

[0254] After treatment according to the method of the present invention, such a response may be maintained for at least 4 to 8 weeks, or in some cases 6 to 8 weeks. Alternatively, improvement in the disease may be categorized as a partial response. "Partial response" is intended to mean a reduction of at least about 50% in all measurable tumor tissue amounts (i.e., the number of malignant cells present in the subject, or the measured amount of tumor mass, or the amount of abnormal monoclonal protein) being maintained for 4 to 8 weeks, or 6 to 8 weeks, except for new lesion sites.

[0255] Treatment according to the present invention includes a "therapeutically effective amount" of the pharmaceutical used. A "therapeutically effective amount" refers to an effective amount for a required period and dosage to obtain a desired therapeutic effect.

[0256] The therapeutically effective amount can vary, for example, according to factors such as the individual disease state, age, gender, and weight, and the ability of the particular medicine to elicit a desired response. The therapeutically effective amount is also an amount where the therapeutically beneficial effect of the antibody or antibody portion exceeds the toxic or adverse effects.

[0257] The "therapeutically effective amount" for tumor therapy may also be measured by the ability to stabilize the progression of the disease. The ability of a compound to inhibit cancer may be evaluated in an animal model system that predicts the effect in human tumors.

[0258] Alternatively, this property of the composition may be evaluated by verifying the ability of the compound to inhibit cell proliferation and induce apoptosis, by an in vitro assay known to those skilled in the art. The therapeutically effective amount of the therapeutic compound may reduce the tumor size or may improve the symptoms in the subject. One skilled in the art can determine the amount based on factors such as the size of the subject, the severity of the subject's symptoms, and the particular composition or selected route of administration.

[0259] The dosing regimen is adjusted so as to provide an optimal desired response (e.g., a therapeutic response). For example, a single bolus administration may be performed, doses divided into several portions may be administered over time, or the dosage may be proportionally increased or decreased according to the requirements of the treatment situation. The parenteral administration composition may be formulated in unit dosage form for ease of administration and to uniformize the dosage. As used herein, the unit dosage form refers to physically distinct units suitable as a single dose for the subject to be treated; each unit contains a predetermined quantity of the active compound calculated to produce the desired therapeutic effect in association with the required pharmaceutical carrier.

[0260] The specifications for the dosage unit forms of the present invention are determined by, and directly dependent on, (a) the unique characteristics of the active compound and the particular therapeutic effect obtained, and (b) the limitations inherent in the art regarding the composition of the active compound, such as with respect to individual therapeutic susceptibilities.

[0261] The effective dosage and dosing regimen for the multispecific antibodies used in the present invention depend on the disease or condition being treated and may be determined by those skilled in the art.

[0262] Non-limiting examples of ranges for the therapeutically effective amount of the multispecific antibodies used in the present invention are from 0.1 to 100 mg / kg, for example, about 0.1 to 50 mg / kg, for example, about 0.1 to 20 mg / kg, for example about 0.1 to 10 mg / kg, for example about 0.5, for example about 0.3, about 1, or about 3 mg / kg. In other embodiments, the antibody is administered at a dosage of 1 mg / kg or more, for example, at a dosage of 1 to 20 mg / kg, for example, at a dosage of 5 to 20 mg / kg, for example, at a dosage of 8 mg / kg.

[0263] Medical experts in the art can readily determine and prescribe the effective amount of the pharmaceutical composition required. For example, a physician or veterinarian can initiate administration of the agent in the pharmaceutical composition at a level lower than the amount required to obtain the desired therapeutic effect and gradually increase the dosage until the desired effect is obtained.

[0264] In one embodiment, the multispecific antibody is administered by intravenous infusion once a week at a dosage of 10 to 500 mg / kg (for example, 200 to 400 mg / kg). Such administration may be repeated, for example, 1 to 8 times, for example 3 to 5 times. The administration may be by continuous infusion over a period of 2 to 24 hours (for example, 2 to 12 hours).

[0265] In one embodiment, if it is necessary to reduce side effects including toxicity, the multispecific antibody is administered by slow continuous infusion over a long period (e.g., 24 hours or more).

[0266] In one embodiment, the multispecific antibody is administered weekly at a dose of 250 mg to 2000 mg (e.g., 300 mg, 500 mg, 700 mg, 1000 mg, 1500 mg, or 2000 mg) up to 8 times (e.g., 4 to 6 times). The administration may be by continuous infusion over a period of 2 to 24 hours (e.g., 2 to 12 hours). The regimen may be repeated one or more times as needed, for example, after 6 to 12 months. The dose may be determined or adjusted, for example, by taking a biological sample and measuring the amount of the compound of the invention in the blood using an anti-idiotype antibody that targets the antigen-binding region of the multispecific antibody.

[0267] In a further embodiment, the multispecific antibody may be administered once every 2 to 12 weeks (e.g., 3 to 10 weeks, e.g., 4 to 8 weeks).

[0268] In one embodiment, the multispecific antibody is administered once a week for a period of, for example, 6 months or more, by maintenance therapy.

[0269] In one embodiment, the multispecific antibody is administered by a regimen that includes one infusion of the multispecific antibody followed by an infusion of the multispecific antibody conjugated to a radioisotope. The regimen may be repeated, for example, after 7 to 9 days.

[0270] As a non-limiting example, the treatment according to the present invention may be administered, after the start of treatment, at least one of the 1st, 2nd, 3rd, 4th, 5th, 6th, 7th, 8th, 9th, 10th, 11th, 12th, 13th, 14th, 15th, 16th, 17th, 18th, 19th, 20th, 21st, 22nd, 23rd, 24th, 25th, 26th, 27th, 28th, 29th, 30th, 31st, 32nd, 33rd, 34th, 35th, 36th, 37th, 38th, 39th, or 40th day, or at least one of the 1st, 2nd, 3rd, 4th, 5th, 6th, 7th, 8th, 9th, 10th, 11th, 12th, 13th, 14th, 15th, 16th, 17th, 18th, 19th, or 20th week, or any combination thereof, every 24, 12, 8, 6, 4, or 2 hours or any combination thereof, using a single administration or divided administration, at a daily dose of about 0.1 to 100 mg / kg (e.g., 0.5, 0.9, 1.0, 1.1, 1.5, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 40, 45, 50, 60, 70, 80, 90, or 100 mg / kg) of the antibody.

[0271] In some embodiments, the multispecific antibody molecule is used in combination with one or more additional therapeutic agents (e.g., chemotherapeutic agents). Non-limiting examples of DNA-damaging chemotherapeutic agents include topoisomerase I inhibitors (e.g., irinotecan, topotecan, camptothecin, and analogs or metabolites thereof, and doxorubicin); topoisomerase II inhibitors (e.g., etoposide, teniposide, and daunorubicin); alkylating agents (e.g., melphalan, chlorambucil, busulfan, thiotepa, ifosfamide, carmustine, lomustine, semustine, streptozocin, dacarbazine, methotrexate, mitomycin C, and cyclophosphamide); DNA intercalators (e.g., cisplatin, oxaliplatin, and carboplatin); DNA intercalators and free radical generating substances (e.g., bleomycin); and nucleoside analogs (e.g., 5-fluorouracil, capecitabine, gemcitabine, fludarabine, cytarabine, mercaptopurine, thioguanine, pentostatin, and hydroxyurea).

[0272] Chemotherapeutic agents that interfere with cell replication include: paclitaxel, docetaxel, and related analogs; vincristine, vinblastine, and related analogs; thalidomide, lenalidomide, and related analogs (e.g., CC-5013 and CC-4047); protein tyrosine kinase inhibitors (e.g., imatinib mesylate and gefitinib); proteasome inhibitors (e.g., bortezomib); NF-κB inhibitors (including inhibitors of IκB kinase); antibodies that bind to proteins overexpressed in cancer and thereby downregulate cell replication (e.g., trastuzumab, rituximab, cetuximab, and bevacizumab); and inhibitors of other proteins or enzymes that are known to be upregulated, overexpressed, or activated in cancer and whose inhibition results in downregulation of cell replication.

[0273] In some embodiments, the antibodies of the invention may be used before, during, or after treatment with Velcade® (bortezomib).

[0274] All cited references are hereby expressly incorporated by reference in their entirety.

[0275] While specific embodiments of the invention have been described for purposes of illustration, those skilled in the art will recognize that many changes in detail can be made without departing from the invention as described in the appended claims.

Examples

[0276] To illustrate the present invention, examples are presented below. These examples are not intended to limit the present invention to any particular application or theory of operation. With respect to the positions of all constant regions contemplated by the present invention, the numbering follows the EU index of Kabat (Kabat et al., 1991, Sequences of Proteins of Immunological Interest, 5th Ed., United States Public Health Service, National Institutes of Health, Bethesda, which is incorporated by reference in its entirety). One of ordinary skill in the art will recognize that this convention consists of non - contiguous numbering in specific regions of the immunoglobulin sequence, thereby making it possible to standardize the reference to conserved positions of the immunoglobulin family. Thus, the position of any given immunoglobulin defined by the EU index need not correspond to its contiguous sequence.

[0277] Example 1. Design of Non-Natural Charge Substitution for Lowering pI The antibody constant chain was modified to have a low pI by manipulating substitutions in the constant domain. The decrease in pI can be manipulated by substituting basic amino acids (K or R) with acidic amino acids (D or E), which results in the largest change in pI. A decrease in pI can also be brought about by mutating basic amino acids to neutral amino acids and mutating neutral amino acids to acidic amino acids. A list of amino acid pK values is in Table 1 of Bjellqvist et al., 1994, Electrophoresis 15:529 - 539.

[0278] We chose to study substitutions in the antibody CH1 (Cγ1) and CL (Cκ or CK) regions (sequences shown in Figure 13) (unlike the Fc region, they do not interact with natural ligands that affect the pharmacological properties of the antibody). In determining which positions to mutate, we considered the surrounding environment and the number of contact points between the WT amino acid and its neighbors, for example, to minimize the impact of substitutions or sets of substitutions on structure and / or function. The solvent accessibility and the exposed fraction for each position in CH1 and CK were calculated using the relevant crystal structures of the antibody Fab domain. The results are shown for Cγ1 and CK, respectively, in Figures 2 and 3 of USSN 13 / 648,951 (the figures and the accompanying descriptions are expressly incorporated herein by reference). Further design was performed by validating the CH1 and CL domains with respect to positions that are isotypic among immunoglobulin isotypes (IgG1, IgG2, IgG3, and IgG4). Since such variations occur naturally, these positions are predicted to be permissive to substitutions. Based on this analysis, many substitutions were identified that decrease the pI but are predicted to have a minimal impact on the biophysical properties of the domain.

[0279] For all heterodimeric proteins herein, genes encoding the heavy and light chains of the antibody were constructed in the mammalian expression vector pTT5. The human IgG1 constant chain gene was obtained from an IMAGE clone and subcloned into the pTT5 vector. The VH and VL genes encoding the anti-VEGF antibody were synthesized by a vendor (Blue Heron Biotechnologies, Bothell WA) and subcloned into vectors encoding the appropriate CL and IgG1 constant chains. Amino acid modifications were constructed by site-directed mutagenesis using the QuikChange® site-directed mutagenesis method (Stratagene, La Jolla CA). All DNA was sequenced and the sequence fidelity was confirmed.

[0280] A plasmid containing the heavy chain gene (VH-Cγ1-Cγ2-Cγ3) was co-transfected with a plasmid containing the light chain gene (VL-Cκ) into 293E cells using Lipofectamine (Invitrogen, Carlsbad CA) and grown in FreeStyle293 medium (Invitrogen, Carlsbad CA). After growing for 5 days, it was purified from the culture supernatant by protein A affinity using MabSelect resin (GE Healthcare). Antibody concentration was measured by the bicinchoninic acid (BCA) assay (Pierce).

[0281] The pI-engineered mAbs were grossly characterized by SDS PAGE, size exclusion chromatography (SEC), isoelectric focusing gel electrophoresis (IEF), binding to antigen by Biacore, and differential scanning calorimetry (DSC) on an Agilent Bioanalyzer. All mAbs showed high purity by SDS-PAGE and SEC. The IEF gels showed that each variant had the designed isoelectric point. Binding analysis by Biacore showed that the pI-engineered variants bound to the antigen with an affinity similar to that of the original antibody, suggesting that the designed substitutions did not inhibit the function of the mAb. DSC in the figure shows which variants had high thermal stability.

[0282] If necessary, pharmacokinetic experiments on serum half-life were performed in B6 mice (homozygous knockout for mouse FcRn and heterozygous knock-in for human FcRn (mFcRn− / −, hFcRn+; also referred to herein as hFcRn or hFcRn+ mice) (Petkova et al., 2006, Int Immunol 18(12):1759-69, incorporated herein by reference in its entirety)).

[0283] A single intravenous antibody injection (2 mg / kg) was administered to groups of 4 - 7 female mice (randomized by body weight in the range of 20 - 30 g). Blood (approx. 50 μl) was collected from the retro-orbital venous plexus at each time point, processed into serum, and stored at -80 °C until analysis. Antibody concentration was measured using an ELISA assay. Serum concentration of the antibody was measured using a recombinant antigen as the capture reagent, and detection was performed using biotinylated anti-human κ antibody and europium-labeled streptavidin. Time-resolved fluorescence signals were collected. PK parameters were measured for each individual mouse using a non-compartmental model with WinNonLin (Pharsight Inc, Mountain View CA). Nominal time and dose were used at a uniform measurement point.

[0284] Example 2. Method of operation for constant region pI manipulation Lowering the pI of a protein or antibody can be done using various methods. At the most basic level, residues with high pKa (lysine, arginine, and to some extent histidine) are replaced with neutral or negative residues, and / or neutral residues are replaced with residues with low pKa (asparagine and glutamic acid). Specific substitutions depend on various factors including position in the structure, role in function, and immunogenicity.

[0285] Since immunogenicity is a concern, one may take care to minimize the risk of inducing immunogenicity by substitutions that lower the pI. One way to minimize the risk is to minimize the amount of variation in the variant. That is, it is to reduce the number of mutations as much as possible while lowering the pI. Charge-exchange mutations (where K, R, or H is exchanged for D or E) have the greatest impact on lowering the pI, so these substitutions are preferred. Another way to minimize the risk of immunogenicity while lowering the pI is to utilize substitutions from homologous human proteins. Thus, for antibody constant chains, low-risk substitutions are achieved by taking advantage of the isotype differences between IgG subclasses (IgG1, IgG2, IgG3, and IgG4). Since immune recognition occurs at the local sequence level (i.e., MHC II and T cell receptors recognize an epitope, usually 9 residues in length), pI-modifying substitutions may be made by isotype substitutions that are proximal in the sequence. In this method, the epitope may be extended to match the native isotype. Such substitutions create epitopes present in other human IgG isotypes and tolerance is expected.

[0286] One way to manipulate the change in pI is the method using isotype exchange disclosed herein.

[0287] Another method of manipulating the pI reduction in proteins and antibodies is to fuse negatively charged residues to the N-terminus or C-terminus. Thus, for example, a peptide consisting mainly of aspartic acid and glutamic acid may be fused to the N-terminus or C-terminus of the antibody heavy chain, light chain, or both. Since the N-terminus is structurally close to the antigen-binding site, the C-terminus is preferred.

[0288] Based on the disclosed methods of manipulation, many variants are designed to modify the isoelectric point of the antibody heavy chain (usually the Fc region) and in some cases the light chain.

[0289] Example 3. Isotype light chain constant region mutations The homology between CK and Cλ is not as high as between IgG subclasses, but substitutions were introduced using the sequence homology and structural homology that exist to create an isotype low pI light chain constant region. In Figure 56, positions with residues contributing to high pI (K, R, and H) or low pI (D and E) are highlighted in bold. Gray indicates lysine, arginine, and histidine, which can preferably be substituted with aspartic acid or glutamic acid to lower the isoelectric point. These mutations can be combined, alone or in any combination, independently and optionally, with all other heavy chain mutations of a scaffold having at least one light chain.

[0290] Example 4. Purification of a mixture of antibody variants having a modified isoelectric point Substitutions that modify the antibody isoelectric point may be introduced into one or more chains of the antibody variant to facilitate analysis and purification. For example, a heterodimeric antibody such as that disclosed in US2011 / 0054151A1 can be purified by modifying the isoelectric point of one chain, thereby allowing the various species present after expression and Protein A purification to be purified by a protein separation method (e.g., ion exchange chromatography) based on differences in charge.

[0291] As an example, the heavy chain of bevacizumab was modified by introducing substitutions that lower the isoelectric point so that the difference in charge between three species produced when WT-IgG1-HC, low-pI-HC, and WT-LC were transfected into 293E cells was large enough to facilitate purification by anion exchange chromatography. Clones were generated, transfected, and the initial purification by protein A chromatography was performed as described above. The sequences of the three chains, "heavy chain 1 of XENP10653", "heavy chain 2 of XENP10653", and "light chain of XENP10653", are shown in the figure. After protein A purification, three species with almost the same molecular weight but different charges were obtained. These were the WT-IgG1-HC / WT-IgG1-HC homodimer (pI = 8.12), the WT-IgG1-HC / low pI-HC heterodimer (pI = 6.89), and the low pI-HC / low pI-HC homodimer (pI = 6.20). The mixture was loaded onto a GE HiTrap Q HP column in 20 mM Tris (pH 7.6) and eluted using a step gradient of NaCl consisting of 50 mM, 100 mM, and finally 200 mM in the same Tris buffer. Elution was monitored at A280 and each fraction was analyzed on an Invitrogen pH 3-10 IEF gel using Novex running buffer, and these results are shown in Figure 40. The WT-IgG1-HC / WT-IgG1-HC homodimer did not bind to the anion exchange column at pH 7.6 and was therefore present in the flow-through and wash (lanes 1-2). The desired heterodimer was eluted with 50 mM NaCl (lane 3), while the low pI-HC / low pI-HC homodimer bound most strongly to the column and was eluted with 100 mM (lane 4) and 200 mM (lane 5) NaCl. Thus, the desired heterodimer variant was easily purified by introducing a low pI substitution into one chain, despite being difficult to purify by other means due to its similar molecular weight to the other two species. The method of purifying antibodies by manipulating the isoelectric point of each chain can be applied to purification methods for various bispecific antibody constructs.This method is particularly useful when the desired substance in the mixture has similar molecular weights and other properties such that it cannot be separated in high yield by conventional purification techniques.

[0292] Example 5. Design of unnatural charge substitutions to modify pI The pI of the antibody constant chain was modified by performing substitutions in the constant domains. Low pI was engineered by substituting basic amino acids (K or R) with acidic amino acids (D or E) that most greatly decrease the pI. Mutations of basic amino acids to neutral amino acids, and of neutral amino acids to acidic amino acids also result in a decrease in pI. Conversely, an increase in pI can be engineered by substituting acidic amino acids (D or E) with basic amino acids (K or R) that result in the greatest increase in pI. Mutations of acidic amino acids to neutral amino acids, and of neutral amino acids to basic amino acids also result in an increase in pI. A list of the pK values of the amino acids can be found in Table 1 of Bjellqvist et al., 1994, Electrophoresis 15:529-539.

[0293] In determining where to make the mutations, the local environment and the number of contact points of the WT amino acid with its neighbors were considered, for example, to minimize the impact of the substitution or set of substitutions on structure and / or function. The accessibility and the exposed fraction of solvent at each constant region position were calculated using the relevant crystal structure. Based on this analysis, many substitutions were identified that were predicted to decrease or increase pI with a minimal impact on the biophysical properties of the domain.

[0294] The pI of the protein was calculated as follows. First, the total number of D, E, C, H, K, R, and Y amino acids present in the protein, as well as the number of N - and C - termini, was considered. Then, the pI was calculated by identifying the pH at which the protein would have an overall zero charge. This was done by calculating the net charge of the protein at a number of validation pH values. The validation pH values were set iteratively, increasing in steps of 0.001 from a low pH (0) to a high pH (14) until the charge of the protein reached or exceeded zero. The net charge of the protein at a given pH was calculated by the following formula:

Number

[0295] Example 6. Purification of a mixture of antibody variants with modified isoelectric points The variants were first purified by Protein A and then loaded onto a GE Healthcare HiTrap SP HP cation - exchange column in 50 mM MES (pH 6.0) and eluted using an NaCl gradient. After elution, fractions from each peak were loaded onto a Lonza IsoGel IEF plate (pH range 7 - 11) for analysis. Separation of the intermediate pI heterodimers was performed in each case and was better when the heterodimers had a larger pI difference compared to the homodimers.

[0296] Example 7. Stability of pI isoelectronic variants The stability of antibodies containing isoelectronic pI substitutions was evaluated using differential scanning fluorimetry (DSF). The DSF experiments were performed using a Bio-Rad CFX Connect Real-Time PCR Detection System. The protein was mixed with SYPRO Orange fluorescent dye and eluted in PBS at 0.25 - 0.50 mg / mL. The final concentration of SYPRO Orange was 10X. After an initial 10-minute incubation period (25 °C), the protein was heated from 25 °C to 95 °C (heating rate of 1 °C / min). Fluorescence measurements were taken every 30 seconds. The melting point was calculated using the software of the instrument. The results are shown in Figure 110. The results suggested that the isoelectronic (+) pI mutations were less stable. Therefore, we generated mutants to reduce the number of substitutions on the high pI side, but the results showed that only E269Q had a small effect on stability, while E272Q and E283Q had a large negative impact on stability.

[0297] Example 8. Design of charged scFv linkers to enable IEX purification of scFvs containing heterodimeric bispecific antibodies We have previously manipulated the antibody constant regions of heterodimeric antibodies to increase or decrease pI using both isotype and isoelectronic charge replacement. These methods allow for effective IEX purification of heterodimeric species, but the introduction of non-natural substitutions may affect antibody stability and immunogenicity. For scFvs containing heterodimeric bispecific antibodies (an example is shown in Figure 87), charge replacement is introduced into other regions (the scFv linker that connects the VH and VL of the scFv construct). The most commonly used linkers are (GGGGS)3 or (GGGGS)4, which have been shown to be plastic enough to allow for the formation of stable scFvs without diabody formation. These sequences are already non-natural and contain few sequences specific for epitopes that may be immunogenic. Therefore, we thought that introducing charge replacement into the scFv linker would allow for IEX purification of heterodimeric bispecific species containing scFvs. Various positively charged scFv linkers and negatively charged scFv linkers were designed and are shown in Figure 85. All linkers are novel constructs (except for the "Whitlow" linker reported by Whitlow et al., (Whitlow M, Protein Eng. 1993 (8), 989-995.)). The linkers designed as 6paxA_1(+A) and 3hsc_2(-A) were obtained from a database of unstructured regions in human proteins obtained from PDB files, and these linkers are approximately the same length as (GGGGS)3 and contain either a positive or negative charge. Other linkers are based on the introduction of repeated residues of Lys or Glu, as well as the introduction of a Lys-Pro motif designed to reduce the opportunity for proteolysis in the positively charged linkers.

[0298] The charged linker was first evaluated for its biophysical properties in the scFv-His format and then subsequently constructed into the anti-CD19xCD3 Fab-scFv-Fc bispecific format. Genes encoding engineered forms of the anti-CD3 antibody SP34 or the anti-CD19 4G7 antibody scFv were constructed in the mammalian expression vector pTT5. For the full-length constructs, the human IgG1 constant chain gene was obtained from an IMAGE clone and subcloned into the pTT5 vector. The scFv genes were synthesized by a vendor (Blue Heron Biotechnologies, Bothell WA). Amino acid modifications were constructed by site-directed mutagenesis using the QuikChange® site-directed mutagenesis method (Stratagene, La Jolla CA). All DNA was sequenced and the sequence fidelity was confirmed.

[0299] Plasmids containing the scFv or heavy and light chain genes were transfected into 293E cells (or co-transfected for the full-length format) using Lipofectamine (Invitrogen, Carlsbad CA) and grown in FreeStyle 293 medium (Invitrogen, Carlsbad CA). After growing for 5 days, purification from the culture supernatant was performed using Protein A (for full-length) with MabSelect resin (GE Healthcare) or Ni-NTA resin for His-tagged scFv. The heterodimers were further purified by ion exchange chromatography (IEX) to evaluate the ability to purify efficiently or the modified pI heavy chain. An example of IEX purification for an anti-CD19xCD3 bispecific antibody containing a positively charged linker in the CD3 scFv is shown in Figure 90. Antibody concentration was measured by the bicinchoninic acid (BCA) assay (Pierce).

[0300] The pI-engineered scFv or antibody was characterized by SDS-PAGE, size exclusion chromatography (SEC), isoelectric focusing (IEF) gel electrophoresis, and / or differential scanning fluorimetry (DSF).

[0301] Example 9. Stability and properties of scFv containing a charged linker Anti-CD3 scFv and anti-CD19 scFv containing a positively charged linker or a negatively charged linker, respectively, were evaluated for SEC properties and stability using DSF. The stability of scFv containing a charged linker was evaluated using differential scanning fluorimetry (DSF). The DSF experiment was performed using a Bio-Rad CFX Connect Real-Time PCR Detection System. The protein was mixed with SYPRO Orange fluorescent dye and diluted with PBS to 0.25 or 0.50 mg / mL. The final concentration of SYPRO Orange was 10X. After first incubating at 25 °C for 10 minutes, the protein was heated from 25 °C to 95 °C at a heating rate of 1 °C / min. Fluorescence measurements were taken every 30 seconds. The melting point was calculated using the software of the instrument. The values for scFv are shown in Figure 86. The charged linker had only a slight effect on the stability of the overall scFv, as indicated by its Tm value. The SEC chromatogram obtained from the purified scFv is shown in Figure 4. The highly charged linker had a longer elution time, and from the prominent peak tailing, it was shown that the scFv was stuck to the SEC resin longer than expected due to the excessive charge. From the results of the binding of the positively charged anti-CD3 scFv to CD4+ T cells (Figure 88), it was shown that the binding of most scFv except for the very highly charged (GKGKS)4 scFv (which showed weak binding) was similar. No unexpected binding was detected when gating against CD20+ cells in PBMC. However, when verifying unexpected binding using SP34 cells, some unexpected binding was observed at high concentrations of the most highly charged linker (Figure 89).

[0302] The positively charged scFv linker of the anti-CD3 scFv of the anti-CD19xCD3 Fab-scFv-Fc construct had the unexpected property of reducing the amount of high molecular weight aggregation (Figure 91). This phenomenon was confirmed by SEC chromatograms of two bispecific constructs (13121 - standard (GGGGS)4 linker, and 13124 - charged linker (GKPGS)4) incubated at various concentrations.

[0303] The activity of the anti-CD19xCD3 construct containing the charged scFv linker in the anti-CD3 scFv was evaluated by RTCC assay using PBMC and Fab-scFv-Fc format bispecific anti-CD19xCD3 antibodies containing different scFv linkers (Figure 92). The linker had little effect on RTCC activity, except for the highly charged linker (GKGKS)3 which had low activity.

[0304] The sequences for all constructs of the present invention are shown in Figure 93.

Claims

1. A heterodimeric antibody comprising: a) a first monomer containing: i) a first heavy chain constant domain containing a first mutated Fc domain; and ii) a first antigen-binding domain; and b) a second monomer containing: i) a second heavy chain constant domain containing a second mutated Fc domain; and ii) a second antigen-binding domain; wherein one of the first and second mutant Fc domains contains an amino acid substitution(s) selected from the group consisting of those shown in FIG.

2. The heterodimeric antibody of claim 1 , wherein the first antigen-binding domain is an scFv covalently linked to the first heavy chain constant domain.

3. The heterodimeric antibody of claim 1, wherein the heterodimeric antibody has a structure selected from the structures of Figures 1B-1L, and 2A-2M.

4. the first and / or second Fc domain is selected from the group consisting of 434A, 434S, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L, 252Y, 252Y / 254T / 256E, 259I / 308F / 428L, 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D, 332E / 330L, 236R, 328R, 236R / 328R, 236N / 267E, 243L, 298A and 299T. The heterodimeric antibody of any one of claims 1 to 3, further comprising an amino acid substitution (including a plurality of substitutions) selected from the group consisting of.

5. A heterodimeric antibody according to any one of claims 1 to 4, wherein one of the first and second mutant Fc domains contains the amino acid substitution 364K / E357Q and the other of the first and second mutant Fc domains contains the amino acid substitution 368D / 370S.

6. The heterodimeric antibody of any one of claims 1 to 5, wherein the first and / or second Fc domain further contains an amino acid substitution(s) selected from the group consisting of those listed in Figure 7.

7. The heterodimeric antibody of any one of claims 1 to 6, wherein the first monomer contains a heavy chain constant domain covalently linked to an scFv, and the second monomer contains a heavy chain and a light chain.

8. A heterodimeric antibody according to any one of claims 1 to 6, wherein one of the monomers contains N208D / Q295E / N384D / Q418E / N421D.

9. A nucleic acid composition comprising a nucleic acid encoding the first and second monomers of claim 1.

10. A host cell containing the nucleic acid composition of claim 9.

11. A method for producing the heterodimeric antibody of any one of claims 1 to 7, comprising culturing a host cell of claim K10 under conditions in which the heterodimeric antibody is produced, and recovering the heterodimeric antibody.

12. A method of treating an individual in need thereof by administering the heterodimeric antibody of any one of claims 1 to 8.

13. Heterodimeric antibodies comprising: a) a first monomer containing a heavy chain comprising: i) a first Fc domain; and ii) A single chain Fv region (scFv) that binds to a first antigen: wherein the scFv comprises a charged scFv linker; and b) a second monomer containing: I) A first heavy chain comprising: 1) a second Fc domain; 2) a first variable heavy chain; and ii) the first light chain.

14. The heterodimeric antibody of claim 13, wherein the charged scFv linker has 3 to 8 positive charges and is selected from the group consisting of those shown in Figure 9.

15. The heterodimeric antibody of claim 13, wherein the charged scFv linker has 3 to 8 negative charges and is selected from the group consisting of those shown in Figure 9.

16. The heterodimeric antibody of any one of claims 13 to 15, wherein the first and second Fc domains contain a set of amino acid substitutions selected from the group consisting of the sets shown in Figure 3.

17. the first and / or second Fc domain is selected from the group consisting of 434A, 434S, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L, 252Y, 252Y / 254T / 256E, 259I / 308F / 428L, 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D, 332E / 330L, 236R, 328R, 236R / 328R, 236N / 267E, 243L, 298A and 299T. The heterodimeric antibody of any one of claims 13 to 16, further comprising an amino acid substitution (including a plurality) selected from the group consisting of.

18. The heterodimeric antibody of any one of claims 13 to 17, wherein the first and / or second Fc domain further contains an amino acid substitution(s) selected from the group consisting of those listed in Figure 7.

19. A heterodimeric antibody composition comprising: a) a first monomer containing: i) a first heavy chain sequence comprising: A) a first mutated Fc domain relative to a human Fc domain; and B) a first antigen-binding domain that binds to a first antigen; and ii) a second heavy chain sequence comprising: A) a second mutated Fc domain relative to a human Fc domain; and B) a second antigen-binding domain that binds to a second antigen; wherein the first and second mutated Fc domains are selected from the set of amino acids shown in FIG. The amino acid substitutions are selected from the group consisting of:

20. A heterodimeric antibody composition comprising: a) a first monomer containing: i) a first heavy chain sequence comprising: A) a first mutated Fc domain relative to a human Fc domain; and B) a first antigen-binding domain that binds to a first antigen; and ii) a second heavy chain sequence comprising: A) a second mutated Fc domain relative to a human Fc domain; and B) A second antigen-binding domain that binds to CD19 and contains a variable heavy chain domain containing the amino acid sequence of H1.227 (SEQ ID NO:X), a variable light chain selected from the group consisting of the amino acid sequence of L1.198 (SEQ ID NO:X), and the amino acid sequence of 1.199 (SEQ ID NO:X) shown in FIG.

21. A heterodimeric antibody composition comprising: a) a first monomer containing: i) a first heavy chain sequence comprising: A) a first mutated Fc domain relative to a human Fc domain; and B) vhCDR1 having the sequence T-Y-A-M-Xaa1, where Xaa1 is N, S, or H (SEQ ID NO:435), vhCDR2 having the sequence R-I-R-S-K-Xaa1-N-Xaa2-Y-A-T-Xaa3-Y-Y-A-Xaa4-S-V-K-G, where Xaa1 is Y or A, Xaa2 is N or S, and aa3 is Y or A and Xaa4 is D or A (SEQ ID NO:436)), vhCDR3 having the sequence H-G-N-F-G-Xaa1-S-Y-V-S-W-F-Xaa2-Y, where Xaa1 is N, D, or Q and Xaa2 is A or D (SEQ ID NO:437), a first antigen binding domain containing an anti-CD3 variable region having a sequence containing: vlCDR1 having the sequence Xaa1-T-N-Xaa2-R-A-Xaa3, where Xaa1 is G, R, or K, Xaa2 is T, or S, Xaa3 is S, or G, and Xaa4 is N, or H (SEQ ID NO:438); vlCDR2 having the sequence Xaa1-T-N-Xaa2-R-A-Xaa3, where Xaa1 is G, or D, Xaa2 is K, or N, and Xaa3 is P, or S (SEQ ID NO:439); and vlCDR3 having the sequence Xaa1-L-W-Y-S-N-Xaa2-W-V, where Xaa1 is A, or L, and Xaa2 is L, or H (SEQ ID NO:440); and b) a second monomer containing: ii) a second heavy chain sequence comprising: A) a second mutated Fc domain relative to a human Fc domain; and B) An anti-CD19 antigen binding domain comprising a variable heavy chain domain comprising the amino acid sequence of H1.227 (SEQ ID NO:X), and a variable light chain selected from the group consisting of the amino acid sequence of L1.198 (SEQ ID NO:X), and the amino acid sequence of 1.199 (SEQ ID NO:X) shown in FIG.

22. The heterodimeric antibody composition of claim 21 , wherein the first and second mutant Fc domains contain a set of amino acid substitutions selected from the group consisting of those shown in FIG. 3 .

23. An antibody composition comprising an anti-CD19 variable domain comprising a variable heavy chain domain comprising the amino acid sequence of H1.227 (SEQ ID NO:X), and a variable light chain selected from the group consisting of the amino acid sequence of L1.198 (SEQ ID NO:X), and the amino acid sequence of 1.199 (SEQ ID NO:X) shown in Figure 21.

24. The antibody of claim Q1, wherein the antibody is a heterodimeric antibody.

25. Heterodimeric antibodies comprising: a) a first monomer containing a heavy chain comprising: i) a first mutated Fc domain; and ii) A single chain Fv region (scFv) that binds to a first antigen: wherein the scFv comprises a charged scFv linker; and b) a second monomer containing: I) A first heavy chain comprising: 1) a second variant Fc domain; 2) a first variable heavy chain; and ii) a first light chain; wherein the first and second mutant Fc domains contain an amino acid substitution(s) selected from the group consisting of those shown in FIG.

26. A heterodimeric antibody composition comprising: a) a first monomer containing: i) a vhCDR1 having the sequence T-Y-A-M-Xaa1, where Xaa1 is N, S, or H (SEQ ID NO:435), a vhCDR2 having the sequence R-I-R-S-K-Xaa1-N-Xaa2-Y-A-T-Xaa3-Y-Y-A-Xaa4-S-V-K-G, where Xaa1 is Y or A, Xaa2 is N or S, and aa3 is Y or A and Xaa4 is D or A (SEQ ID NO:436)), vhCDR3 having the sequence H-G-N-F-G-Xaa1-S-Y-V-S-W-F-Xaa2-Y, where Xaa1 is N, D, or Q and Xaa2 is A or D (SEQ ID NO:437), a first antigen binding domain containing an anti-CD3 variable region having a sequence containing: vlCDR1 having the sequence Xaa1-T-N-Xaa2-R-A-Xaa3, where Xaa1 is G, R, or K, Xaa2 is T, or S, Xaa3 is S, or G, and Xaa4 is N, or H (SEQ ID NO:438); vlCDR2 having the sequence Xaa1-T-N-Xaa2-R-A-Xaa3, where Xaa1 is G, or D, Xaa2 is K, or N, and Xaa3 is P, or S (SEQ ID NO:439); and vlCDR3 having the sequence Xaa1-L-W-Y-S-N-Xaa2-W-V, where Xaa1 is A, or L, and Xaa2 is L, or H (SEQ ID NO:440); and ii) a first heavy chain sequence containing a first variant Fc domain relative to a human Fc domain; and b) a second monomer containing: i) a second antigen-binding domain; and ii) a second heavy chain sequence containing a second mutated Fc domain to human; wherein said first and second variant Fc domains have different amino acid sequences.

27. The heterodimeric antibody composition of claim 26, wherein the first and second mutant Fc domains contain a set of amino acid substitutions selected from the group consisting of the sets of amino acid substitutions shown in Figure 3.

28. The heterodimeric antibody of claim C2, wherein the set is selected from the group consisting of: L368D / K370S and S364K; L368D / K370S and S364K / E357L; L368D / K370S and S364K / E357Q; T411E / K360E / Q362E and D401K; L368E / K370S and S364K; K370S and S364K / E357Q; and K370S and S364K / E357Q.

29. The heterodimeric antibody of any one of claims 26 to 28, wherein the anti-CD3 variable region is an anti-CD3 scFv sequence covalently linked to the first heavy chain sequence.

30. The heterodimeric antibody of claim 29, wherein the anti-CD3 scFv contains a charged scFv linker.

31. The charged scFv linker is IRPRAIGGSKPRVA (SEQ ID NO:X), GKGGSGKGGSGKGGS (SEQ ID NO:X), GGKGSGGKGSGGKGS (SEQ ID NO:X), GGGKSGGGKSGGGKGS (SEQ ID NO:X), GKGKSGGKSGKSGKGKS (SEQ ID NO:X), (SEQ ID NO:X), GGGKSGGKGSGKGGS (SEQ ID NO:X), (SEQ ID NO:X), GKPGSGKPGSGKPGS (SEQ ID NO:X), GKPGSGKPGSGKPGS (SEQ ID NO:X), GKPGSGKPGSGKPGSGKPGS (SEQ ID NO:X), (SEQ ID NO:X), 32. The heterodimeric antibody of claim 31 , selected from the group consisting of: GKGKSGKGKSGKSGKSGKGS (SEQ ID NO: X), STAGDTHLGGEDFD (SEQ ID NO: X), GEGGSGEGGSGEGGS (SEQ ID NO: X), GGEGSGGEGSGGEGS (SEQ ID NO: X), GGGESGGGESGGGES (SEQ ID NO: X), GEGESGEGESGEGES (SEQ ID NO: X), GGGESGGEGSGEGGS (SEQ ID NO: X) and GEGESGEGESGEGESGEGES (SEQ ID NO: X).

32. The heterodimeric antibody of any one of claims 26 to 28, wherein the first monomer comprises: a) the first heavy chain sequence comprising a heavy chain variable domain comprising the vhCDR1, vhCDR2 and vhCDR3 sequences; and b) a first light chain sequence comprising a light chain variable domain comprising said vLCDR1, said vLCDR2 and said vLCDR3 sequences.

33. The heterodimeric antibody of any one of claims 26 to 32, wherein the second antigen-binding domain comprises an scFv.

34. The heterodimeric antibody of any one of claims 26 to 32, wherein the second monomer comprises: a) the heavy chain sequence further comprising a second heavy chain variable domain; and b) light chain sequence; wherein said heavy chain variable domain and said light chain sequence form said second antigen-binding domain.

35. The heterodimeric antibody of any one of claims 26 to 34, wherein the pI of the first monomer and the pI of the second monomer are at least 0.5 log apart.

36. The heterodimeric antibody of claim 35, wherein the first and second mutant Fc domains contain a set of amino acid substitutions (including multiple substitutions) selected from the group consisting of those shown in Figure 3.

37. one of the variant Fc domains is 434A, 434S, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I, or V / 434S, 436V / 428L, 252Y, 252Y / 254T / 256E, 259I / 308F / 428L, 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D, 332E / 330L, 236R, 328R, 236R / 328R, 236N / 267E, 243L, 298A, and 299T. The heterodimeric antibody of any one of claims 26 to 36, further comprising an amino acid substitution (including a plurality of substitutions) selected from the group consisting of.

38. A composition comprising an anti-CD3 variable region having sequences containing: a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:

430.

39. vhCDR1 having the sequence T-Y-A-M-Xaa1, where Xaa1 is N, S, or H (SEQ ID NO:435), vhCDR2 having the sequence R-I-R-S-K-Xaa1-N-Xaa2-Y-A-T-Xaa3-Y-Y-A-Xaa4-S-V-K-G, where Xaa1 is Y or A and Xaa2 is N or S; Xaa3 is Y or A and Xaa4 is D or A (SEQ ID NO:436)), vhCDR3 having the sequence H-G-N-F-G-Xaa1-S-Y-V-S-W-F-Xaa2-Y, where Xaa1 is N, D, or Q and Xaa2 is A or D (SEQ ID NO:437), a vlCDR1 having the sequence Xaa1-T-N-Xaa2-R-A-Xaa3, where Xaa1 is G or D, Xaa2 is K or N, and Xaa3 is P or S (SEQ ID NO:439), and a vlCDR3 having the sequence Xaa1-L-W-Y-S-N-Xaa2-W-V, where Xaa1 is A or L, and Xaa2 is L or H (SEQ ID NO:440).

40. 40. The composition of claim 39, wherein the anti-CD3 variable region has a sequence selected from the group consisting of: a) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; b) a sequence containing a vhCDR1 having SEQ ID NO:412, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; c) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:414, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; d) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:417, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; e) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:418, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; f) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:421, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; g) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:422, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; h) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:427, and a vlCDR3 having SEQ ID NO:430; i) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:428, and a vlCDR3 having SEQ ID NO:430; j) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:431; k) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; l) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:423, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:432; m) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:424, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:432; n) a sequence containing a vhCDR1 having SEQ ID NO:412, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:417, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; o) a sequence containing a vhCDR1 having SEQ ID NO:412, a vhCDR2 having SEQ ID NO:414, a vhCDR3 having SEQ ID NO:419, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; p) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:415, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; q) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:415, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; r) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:417, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; s) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:419, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:430; t) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:413, a vhCDR3 having SEQ ID NO:417, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:433; u) a sequence containing a vhCDR1 having SEQ ID NO: 411, a vhCDR2 having SEQ ID NO: 413, a vhCDR3 having SEQ ID NO: 416, a vlCDR1 having SEQ ID NO: 420, a vlCDR2 having SEQ ID NO: 425, and a vlCDR3 having SEQ ID NO: 433, and v) a sequence containing a vhCDR1 having SEQ ID NO:411, a vhCDR2 having SEQ ID NO:434, a vhCDR3 having SEQ ID NO:416, a vlCDR1 having SEQ ID NO:420, a vlCDR2 having SEQ ID NO:425, and a vlCDR3 having SEQ ID NO:

430.

41. 41. The composition of any one of claims 39 or 40, wherein the composition comprises a first amino acid sequence containing a variable heavy chain CDR and a second amino acid sequence containing a variable light chain CDR.

42. 41. The composition of any one of claims 39 or 40, wherein the composition comprises an scFv.

43. 41. The composition of any one of claims 39 or 40, wherein the anti-CD3 variable region comprises a variable heavy chain region and a variable light chain region selected from the group consisting of: SEQ ID NOs:5 and 6, SEQ ID NOs:9 and 10, SEQ ID NOs:13 and 14, SEQ ID NOs:17 and 18, SEQ ID NOs:21 and 22, SEQ ID NOs:25 and 26, SEQ ID NOs:29 and 30, SEQ ID NOs:33 and 34, SEQ ID NOs:37 and 38, SEQ ID NOs:41 and 42, SEQ ID NOs:45 and 46, SEQ ID NOs:49 and 50, SEQ ID NOs:53 and 54, SEQ ID NOs:57 and 58, SEQ ID NOs:61 and 62, SEQ ID NOs:65 and 66, SEQ ID NOs:69 and 70, SEQ ID NOs:73 and 74, SEQ ID NOs:77 and 78, SEQ ID NOs:81 and 82, SEQ ID NOs:85 and 86, SEQ ID NOs:89 and 90, SEQ ID NOs:93 and 94, SEQ ID NO:97 and 98, SEQ ID NO:101 and 102, SEQ ID NO:105 and 106, SEQ ID NO:109 and 110, SEQ ID NO:113 and 114, SEQ ID NO:117 and 118, SEQ ID NO:121 and 122, SEQ ID NO:125 and 126, SEQ ID NO:129 and 130, SEQ ID NO:133 and 134, SEQ ID NO:137 and 138, SEQ ID NO:141 and 142, SEQ ID NO:145 and 146, SEQ ID NO:149 and 150, SEQ ID NO:153 and 154, SEQ ID NO:157 and 158, SEQ ID NO:161 and 162, SEQ ID NO:165 and 166, SEQ ID NO:169 and 170 , SEQ ID NOs: 173 and 174, SEQ ID NOs: 177 and 178, SEQ ID NOs: 181 and 182, SEQ ID NOs: 185 and 186, SEQ ID NOs: 189 and 190, SEQ ID NOs: 193 and 194, SEQ ID NOs: 197 and 198, SEQ ID NOs: 201 and 202, SEQ ID NOs: 205 and 206, SEQ ID NOs: 209 and 210, SEQ ID NOs: 213 and 214, SEQ ID NOs: 217 and 218, SEQ ID NOs: 221 and 222, SEQ ID NOs: 225 and 226, SEQ ID NOs: 229 and 230, SEQ ID NOs: 233 and 234, SEQ ID NOs: 237 and 238, SEQ ID NOs: 241 and 242, SEQ ID NOs: 245 and 246 , SEQ ID NO:249 and 250, SEQ ID NO:253 and 254, SEQ ID NO:257 and 258, SEQ ID NO:261 and 262, SEQ ID NO:265 and 266, SEQ ID NO:269 and 270, SEQ ID NO:273 and 274, SEQ ID NO:277 and 278, SEQ ID NO:281 and 282, SEQ ID NO:285 and 286, SEQ ID NO:289 and 290, SEQ ID NO:293 and 294, SEQ ID NO:297 and 298, SEQ ID NO:301 and 302, SEQ ID NO:305 and 306, SEQ ID NO:309 and 310, SEQ ID NO:313 and 314, SEQ ID NO:317 and 318, SEQ ID NO:321 and 322,SEQ ID NOs: 325 and 326, 329 and 330, 333 and 334, 337 and 338, 341 and 342, 345 and 346, 349 and 350, 353 and 354, 357 and 358, 361 and 362, 365 and 366, 369 and 370, 373 and 374, 377 and 378, 381 and 382, ​​385 and 386, 389 and 390, 393 and 394, 397 and 398, 401 and 402, 405 and 406, 409 and 410. ,

44. 44. The composition of any one of claims 42 or 43, wherein the scFv contains a charged scFv linker.

45. The composition of claim 44, wherein the charged scFv linker is selected from the group consisting of those shown in Figure 9.

46. The composition of claim 40, wherein the scFv has a sequence selected from the group consisting of: SEQ ID NO:4, SEQ ID NO:8, SEQ ID NO:12, SEQ ID NO:16, SEQ ID NO:20, SEQ ID NO:24, SEQ ID NO:28, SEQ ID NO:32, SEQ ID NO:36, SEQ ID NO:40, SEQ ID NO:44, SEQ ID NO:48, SEQ ID NO:52, SEQ ID NO:56, SEQ ID NO:60, SEQ ID NO:64, SEQ ID NO:68, SEQ ID NO:72, SEQ ID NO:76, SEQ ID NO:80, SEQ ID NO:84, SEQ ID NO:88, SEQ ID NO:92, SEQ ID NO:96, SEQ ID NO:100, SEQ ID NO:104, SEQ ID NO:108, SEQ ID NO:112, SEQ ID NO:116, SEQ ID NO:120, SEQ ID NO:124, SEQ ID NO:128, SEQ ID NO:132, SEQ ID NO:136, SEQ ID NO:140, SEQ ID NO:144, SEQ ID NO:148, SEQ ID NO:152, SEQ ID NO:156, SEQ ID NO:160, SEQ ID NO:164, SEQ ID NO:168, SEQ ID NO:172, SEQ ID NO:176, SEQ ID NO:180, SEQ ID NO:184, SEQ ID NO:188, SEQ ID NO:192, SEQ ID NO:196, SEQ ID NO:200, SEQ ID NO:204, SEQ ID NO:208, SEQ ID NO:2 12, SEQ ID NO:216, SEQ ID NO:220, SEQ ID NO:224, SEQ ID NO:228, SEQ ID NO:232, SEQ ID NO:236, SEQ ID NO:240, SEQ ID NO:244, SEQ ID NO:248, SEQ ID NO:252, SEQ ID NO:256, SEQ ID NO:260, SEQ ID NO:264, SEQ ID NO:268, SEQ ID NO:272, SEQ ID NO:276, SEQ ID NO:280, SEQ ID NO:284, SEQ ID NO:288, SEQ ID NO:292, SEQ ID NO:296, SEQ ID NO:300, SEQ ID NO:304, SEQ ID NO:308, SEQ ID NO: Number 312, SEQ ID NO:316, SEQ ID NO:320, SEQ ID NO:324, SEQ ID NO:328, SEQ ID NO:332, SEQ ID NO:336, SEQ ID NO:340, SEQ ID NO:344, SEQ ID NO:348, SEQ ID NO:352, SEQ ID NO:356, SEQ ID NO:360, SEQ ID NO:364, SEQ ID NO:368, SEQ ID NO:372, SEQ ID NO:376, SEQ ID NO:380, SEQ ID NO:384, SEQ ID NO:388, SEQ ID NO:392, SEQ ID NO:396, SEQ ID NO:400, SEQ ID NO:404, SEQ ID NO:

408.

47. A nucleic acid composition encoding an anti-CD3 variable region comprising a variable heavy chain region and a variable light chain region selected from the group consisting of: SEQ ID NOs:5 and 6, SEQ ID NOs:9 and 10, SEQ ID NOs:13 and 14, SEQ ID NOs:17 and 18, SEQ ID NOs:21 and 22, SEQ ID NOs:25 and 26, SEQ ID NOs:29 and 30, SEQ ID NOs:33 and 34, SEQ ID NOs:37 and 38, SEQ ID NOs:41 and 42, SEQ ID NOs:45 and 46, SEQ ID NOs:49 and 50, SEQ ID NOs:53 and 54, SEQ ID NOs:57 and 58, SEQ ID NOs:61 and 62, SEQ ID NOs:65 and 66, SEQ ID NOs:69 and 70, SEQ ID NOs:73 and 74, SEQ ID NOs:77 and 78, SEQ ID NOs:81 and 82, SEQ ID NOs:85 and 86, SEQ ID NOs:89 and 90, SEQ ID NOs:93 and 94, SEQ ID NO:97 and 98, SEQ ID NO:101 and 102, SEQ ID NO:105 and 106, SEQ ID NO:109 and 110, SEQ ID NO:113 and 114, SEQ ID NO:117 and 118, SEQ ID NO:121 and 122, SEQ ID NO:125 and 126, SEQ ID NO:129 and 130, SEQ ID NO:133 and 134, SEQ ID NO:137 and 138, SEQ ID NO:141 and 142, SEQ ID NO:145 and 146, SEQ ID NO:149 and 150, SEQ ID NO:153 and 154, SEQ ID NO:157 and 158, SEQ ID NO:161 and 162, SEQ ID NO:165 and 166, SEQ ID NO:169 and 170 , SEQ ID NOs: 173 and 174, SEQ ID NOs: 177 and 178, SEQ ID NOs: 181 and 182, SEQ ID NOs: 185 and 186, SEQ ID NOs: 189 and 190, SEQ ID NOs: 193 and 194, SEQ ID NOs: 197 and 198, SEQ ID NOs: 201 and 202, SEQ ID NOs: 205 and 206, SEQ ID NOs: 209 and 210, SEQ ID NOs: 213 and 214, SEQ ID NOs: 217 and 218, SEQ ID NOs: 221 and 222, SEQ ID NOs: 225 and 226, SEQ ID NOs: 229 and 230, SEQ ID NOs: 233 and 234, SEQ ID NOs: 237 and 238, SEQ ID NOs: 241 and 242, SEQ ID NOs: 245 and 246 , SEQ ID NO:249 and 250, SEQ ID NO:253 and 254, SEQ ID NO:257 and 258, SEQ ID NO:261 and 262, SEQ ID NO:265 and 266, SEQ ID NO:269 and 270, SEQ ID NO:273 and 274, SEQ ID NO:277 and 278, SEQ ID NO:281 and 282, SEQ ID NO:285 and 286, SEQ ID NO:289 and 290, SEQ ID NO:293 and 294, SEQ ID NO:297 and 298, SEQ ID NO:301 and 302, SEQ ID NO:305 and 306, SEQ ID NO:309 and 310, SEQ ID NO:313 and 314, SEQ ID NO:317 and 318, SEQ ID NO:321 and 322,SEQ ID NOs: 325 and 326, 329 and 330, 333 and 334, 337 and 338, 341 and 342, 345 and 346, 349 and 350, 353 and 354, 357 and 358, 361 and 362, 365 and 366, 369 and 370, 373 and 374, 377 and 378, 381 and 382, ​​385 and 386, 389 and 390, 393 and 394, 397 and 398, 401 and 402, 405 and 406, 409 and 410. ,

48. 48. The nucleic acid composition of claim 47, wherein the nucleic acid composition contains a first nucleic acid encoding a variant heavy chain region and a second nucleic acid encoding a variable light chain region.

49. 48. The nucleic acid composition of claim 47, wherein the anti-CD3 variable region comprises an scFv and the nucleic acid encodes an scFv amino acid sequence selected from the group consisting of: SEQ ID NO:4, SEQ ID NO:8, SEQ ID NO:12, SEQ ID NO:16, SEQ ID NO:20, SEQ ID NO:24, SEQ ID NO:28, SEQ ID NO:32, SEQ ID NO:36, SEQ ID NO:40, SEQ ID NO:44, SEQ ID NO:48, SEQ ID NO:52, SEQ ID NO:56, SEQ ID NO:60, SEQ ID NO:64, SEQ ID NO:68, SEQ ID NO:72, SEQ ID NO:76, SEQ ID NO:80, SEQ ID NO:84, SEQ ID NO:88, SEQ ID NO:92, SEQ ID NO:96, SEQ ID NO:100, SEQ ID NO:104, SEQ ID NO:108, SEQ ID NO:112, SEQ ID NO:116, SEQ ID NO:120, SEQ ID NO:124, SEQ ID NO:128, SEQ ID NO:132, SEQ ID NO:136, SEQ ID NO:140, SEQ ID NO:144, SEQ ID NO:148, SEQ ID NO:152, SEQ ID NO:156, SEQ ID NO:160, SEQ ID NO:164, SEQ ID NO:168, SEQ ID NO:172, SEQ ID NO:176, SEQ ID NO:180, SEQ ID NO:184, SEQ ID NO:188, SEQ ID NO:192, SEQ ID NO:196, SEQ ID NO:200, SEQ ID NO:204, SEQ ID NO:208, SEQ ID NO:2 12, SEQ ID NO:216, SEQ ID NO:220, SEQ ID NO:224, SEQ ID NO:228, SEQ ID NO:232, SEQ ID NO:236, SEQ ID NO:240, SEQ ID NO:244, SEQ ID NO:248, SEQ ID NO:252, SEQ ID NO:256, SEQ ID NO:260, SEQ ID NO:264, SEQ ID NO:268, SEQ ID NO:272, SEQ ID NO:276, SEQ ID NO:280, SEQ ID NO:284, SEQ ID NO:288, SEQ ID NO:292, SEQ ID NO:296, SEQ ID NO:300, SEQ ID NO:304, SEQ ID NO:308, SEQ ID NO: Number 312, SEQ ID NO:316, SEQ ID NO:320, SEQ ID NO:324, SEQ ID NO:328, SEQ ID NO:332, SEQ ID NO:336, SEQ ID NO:340, SEQ ID NO:344, SEQ ID NO:348, SEQ ID NO:352, SEQ ID NO:356, SEQ ID NO:360, SEQ ID NO:364, SEQ ID NO:368, SEQ ID NO:372, SEQ ID NO:376, SEQ ID NO:380, SEQ ID NO:384, SEQ ID NO:388, SEQ ID NO:392, SEQ ID NO:396, SEQ ID NO:400, SEQ ID NO:404, SEQ ID NO:

408.

50. 50. The composition of any one of claims 48 or 49, wherein the nucleic acid composition comprises a first expression vector containing the first nucleic acid and a second expression vector containing the second nucleic acid.

51. 51. The composition of claim 50, wherein the nucleic acid composition comprises an expression vector.

52. A host cell comprising the nucleic acid composition of any one of claims 47 to 51.

53. 53. A method of making a composition containing an anti-CD3 variable region, comprising culturing the host cell of claim 52 under conditions in which the composition is expressed.

54. 54. A method of treating a patient in need thereof by administering to the patient a composition according to any one of claims 1 to 53.

55. A heterodimeric antibody comprising: a) a first heavy chain comprising: i) a first Fc domain; and ii) a vhCDR1 having the sequence T-Y-A-M-Xaa1, where Xaa1 is N, S, or H (SEQ ID NO:435), and a vhCDR2 having the sequence R-I-R-S-K-Xaa1-N-Xaa2-Y-A-T-Xaa3-Y-Y-A-Xaa4-S-V-K-G, where Xaa1 is Y or A and Xaa2 is N or S; , Xaa3 is Y or A, and Xaa4 is D or A (SEQ ID NO:436)), vhCDR3 having the sequence H-G-N-F-G-Xaa1-S-Y-V-S-W-F-Xaa2-Y, where Xaa1 is N, D, or Q, and Xaa2 is A or D (SEQ ID NO:437), a single chain Fv region (scFv) that binds CD3, comprising a vlCDR1 having the sequence Xaa3-Xaa4-Y-A-N, where Xaa1 is G, R, or K, Xaa2 is T or S, Xaa3 is S or G, and Xaa4 is N or H (SEQ ID NO:438), a vlCDR2 having the sequence Xaa1-T-N-Xaa2-R-A-Xaa3, where Xaa1 is G or D, Xaa2 is K or N, and Xaa3 is P or S (SEQ ID NO:439), and a vlCDR3 having the sequence Xaa1-L-W-Y-S-N-Xaa2-W-V, where Xaa1 is A or L, and Xaa2 is L or H (SEQ ID NO:440); and b) a second heavy chain comprising: i) a second Fc domain; ii) a first variable heavy chain; and ii) a first variable light chain; wherein said first and second Fc domains are different.

56. Methods for making a heterodimeric antibody containing: a) providing a first nucleic acid encoding a first heavy chain comprising: i) a first heavy chain comprising:

1. a first Fc domain; and 2 a single chain Fv region (scFv) that binds to a first antigen; wherein the scFv contains a charged linker; and b) providing a second nucleic acid encoding a second heavy chain comprising: i) a second Fc domain; ii) a first variable heavy chain; and c) providing a third nucleic acid containing a light chain; d) expressing the first, second, and third nucleic acids in a host cell to produce the first, second, and third amino acid sequences, respectively; e) loading the first, second and third amino acid sequences onto an ion exchange column; and f) harvesting the heterodimer fraction.

57. A composition comprising a heterodimeric protein comprising: a) a first monomer containing: i) a first mutant heavy chain constant region; ii) a first fusion partner; and b) a second monomer containing: i) a second mutant heavy chain constant region; ii) a second fusion partner; wherein the pI of said first and second mutant heavy chain constant regions are at least 0.5 log apart.

58. A composition comprising a heterodimeric protein comprising: a) a first monomer containing: i) a first mutant heavy chain constant region; ii) a first fusion partner; and b) a second monomer containing: i) a second mutant heavy chain constant region; ii) a second fusion partner; wherein the Fc region of said first and second constant regions contains the set of amino acid substitutions in Figures 3 and 12.

59. 59. The composition of claim 58, wherein the first fusion partner is an scFv.

60. 59. The composition of claim 58, wherein the first fusion partner comprises a variable heavy and light chain.

61. 61. The composition of any one of claims 57-60, wherein the heterodimeric protein has a structure selected from the group consisting of the structures of Figures 1B-1M and Figures 2A-2N, and 2P-2S.

62. 62. The composition of claim 61 , wherein the heterodimeric protein has the structure depicted in FIG. 1B.

63. 62. The composition of claim 61 , wherein the heterodimeric protein has the structure depicted in FIG. 1M.

64. The composition of any one of claims 1 to 63, wherein the first monomer comprises a third fusion partner.

65. 65. The composition of claim 64, wherein the second monomer contains a fourth fusion partner.

66. 66. The composition of any one of claims 1 to 65, wherein the fusion partners are independently selected from the group consisting of immunoglobulin components, peptides, cytokines, chemokines, immune receptors, and blood factors.

67. 67. The composition of claim 66, wherein the immunoglobulin component is selected from the group consisting of Fab, VH, VL, scFv, scFv2, and dAb.

68. The composition of any one of claims 1 to 67, wherein both of the fusion partners are immunoglobulin moieties.

69. The composition of any one of claims 1 to 68, wherein one of the fusion partners is an scFv.

70. The composition of any one of claims 1 to 69, wherein one of the fusion partners is a Fab.

71. 71. The composition of any one of claims 1 to 70, wherein at least one Fc domain of a heavy chain of one monomer contains an amino acid mutation selected from the group consisting of 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D, 332E / 330L, 236R, 328R, 236R / 328R, 243L, 298A and 299T.

72. 72. The composition of any one of claims 1 to 71, wherein the Fc domain of each monomer comprises a chain containing an amino acid mutation selected from the group consisting of 236A, 239D, 239E, 332E, 332D, 239D / 332E, 267D, 267E, 328F, 267E / 328F, 236A / 332E, 239D / 332E / 330Y, 239D, 332E / 330L, 236R, 328R, 236R / 328R, 243L, 298A and 299T.

73. 73. The composition of any one of claims 1 to 72, wherein at least one Fc domain of one monomeric heavy chain contains an amino acid mutation selected from the group consisting of 434A, 434S, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L, 252Y, 252Y / 254T / 256E and 259I / 308F / 428L.

74. 74. The composition of any one of claims 1 to 73, wherein the Fc domain of each monomer comprises a chain containing an amino acid mutation selected from the group consisting of 434A, 434S, 428L, 308F, 259I, 428L / 434S, 259I / 308F, 436I / 428L, 436I or V / 434S, 436V / 428L, 252Y, 252Y / 254T / 256E and 259I / 308F / 428L.

Images