AU2004233493B2 - Polypeptide variants with altered effector function - Google Patents

Description

AUSTRALIA

Patents Act 1990 COMPLETE SPECIFICATION STANDARD PATENT Applicant: GENENTECH, INC.

Invention Title: POLYPEPTIDE VARIANTS WITH ALTERED EFFECTOR FUNCTION The following statement is a full description of this invention, including the best method of performing it known to us: 0 oPOLYPEPTIDE VARIANTS WITH ALTERED EFFECTOR FUNCTION BACKGROUND OF THE INVENTION Z Field of the Invention The present invention concerns polypeptides comprising a variant Fc region.

C More particularly, the present invention concerns Fc region-containing polypeptides that have altered effector function as a consequence of one or more amino acid en, modifications in the Fc region thereof.

N Description of Related Art Cfl 10 All references, including any patents or patent applications, cited in this C€3 Sspecification are hereby incorporated by reference. No admission is made that any reference constitutes prior art. The discussion of the references states what their authors oassert, and the applicants reserve the right to challenge the accuracy and pertinency of C the cited documents. It will be clearly understood that, although a number of prior art publications are referred to herein, this reference does not constitute an admission that any of these documents forms part of the common general knowledge in the art, in Australia or in any other country.

Antibodies are proteins which exhibit binding specificity to a specific antigen.

Native antibodies are usually heterotetrameric glycoproteins of about 150,000 daltons, composed of two identical light chains and two identical heavy chains. Each light chain is linked to a heavy chain by one covalent disulfide bond, while the number of disulfide linkages varies between the heavy chains of different immunoglobulin isotypes. Each heavy and light chain also has regularly spaced intrachain disulfide bridges. Each heavy chain has at one end a variable domain (V

H

followed by a number of constant domains. Each light chain has a variable domain at one end (VL) and a constant domain at its other end; the constant domain of the light chain is aligned with the first constant domain of the heavy chain, and the light chain variable domain is aligned with the variable domain of the heavy chain. Particular amino acid residues are believed to form an interface between the light and heavy chain variable domains.

The term "variable" refers to the fact that certain portions of the variable domains differ extensively in sequence among antibodies and are responsible for the binding specificity of each particular antibody for its particular antigen. However, the variability is not evenly distributed through the variable domains of antibodies. It is concentrated in three segments called complementarity determining regions (CDRs) both in the light chain and the heavy chain variable domains. The more highly conserved portions of the variable domains are called the framework regions (FRs). The variable domains of native heavy and light chains each comprise four FRs, largely adopting a n-sheet configuration, connected by three CDRs, which form loops connecting, and in some cases forming part of, the P-sheet structure. The CDRs in each chain are held together in close proximity by the FRs and, with the CDRs from the other chain, contribute to the formation of the antigen binding site of antibodies (see Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health H:\ciniae\Keep\s eci\28S05.0.doc 29/10104 -2o Service, National Institutes of Health, Bethesda, MD. (1991)).

c The constant domains are not involved directly in binding an antibody to an o antigen, but exhibit various effector functions. Depending on the amino acid sequence Z of the constant region of their heavy chains, antibodies or immunoglobulins can be I 5 assigned to different classes. There are five major classes of immunoglobulins: IgA, c IgD, IgE, IgG and IgM, and several of these may be further divided into subclasses (isotypes), e.g. IgGI, IgG2, IgG3, and IgG4; IgAl and IgA2. The heavy chain constant eC regions that correspond to the different classes of immunoglobulins are called ao, 8, e, y, ON and pt, respectively. Of the various human immunoglobulin classes, only human IgG1, en 10 IgG2, IgG3 and IgM are known to activate complement; and human IgG 1 and IgG3 Smediate ADCC more effectively than IgG2 and IgG4.

A schematic representation of the native IgGI structure is shown in FIG. 1, 0 where the various portions of the native antibody molecule are indicated. Papain digestion of antibodies produces two identical antigen binding fragments, called Fab fragments, each with a single antigen binding site, and a residual "Fc" fragment, whose name reflects its ability to crystallize readily. The crystal structure of the human IgG Fc region has been determined (Deisenhofer, Biochemistry 20:2361-2370 (1981)). In human IgG molecules, the Fc region is generated by papain cleavage N-terminal to Cys 226. The Fe region is central to the effector functions of antibodies.

The effector functions mediated by the antibody Fc region can be divided into two categories: effector functions that operate after the binding of antibody to an antigen (these functions involve the participation of the complement cascade or Fc receptor (FcR)-bearing cells); and effector functions that operate independently of antigen binding (these functions confer persistence in the circulation and the ability to be transferred across cellular barriers by transcytosis). Ward and Ghetie, Therapeutic Immunology 2:77-94 (1995).

While binding of an antibody to the requisite antigen has a neutralizing effect that might prevent the binding of a foreign antigen to its endogenous target (e.g.

receptor or ligand), binding alone may not remove the foreign antigen. To be efficient in removing and/or destroying foreign antigens, an antibody should be endowed with both high affinity binding to its antigen, and efficient effector functions.

Fc Receptor (FcR) Binding The interaction of antibodies and antibody-antigen complexes with cells of the immune system effects a variety of responses, including antibody-dependent cellmediated cytotoxicity (ADCC) and complement dependent cytotoxicity (CDC) (reviewed in Daeron, Annu. Rev. Immunol. 15:203-234 (1997); Ward and Ghetie, Therapeutic Immunol. 2:77-94 (1995); as well as Ravetch and Kinet, Annu. Rev.

Immunol. 9:457-492 (1991)).

Several antibody effector functions are mediated by Fc receptors (FcRs), which bind the Fc region of an antibody. FcRs are defined by their specificity for immunoglobulin isotypes; Fc receptors for IgG antibodies are referred to as FcyR, for IgE as FcR, for IgA as FcoR and so on. Three subclasses of FcyR have been H:\cintae\Keep\speci\2B505.00.doc 29/10/04 -3- Sidentified: FcyRI (CD64), FcyRI (CD32) and FcyRIm (CD16). Because each FcyR subclass is encoded by two or three genes, and alternative RNA spicing leads to O multiple transcripts, a broad diversity in FcyR isoforms exists. The three genes Z encoding the FcyRI subclass (FcyRIA, FcyRIB and FcyRIC) are clustered in region I) 5 lq21.1 of the long arm of chromosome 1; the genes encoding FcyRII isoforms c (FcyRIIA, FcyRIIB and FcyRIIC) and the two genes encoding FcyRIII (FcyRIIIA and FcyRIIIB) are all clustered in region lq22. These different FcR subtypes are expressed Ce on different cell types (reviewed in Ravetch and Kinet, Annu. Rev. Immunol. 9:457-492 O (1991)). For example, in humans, FcyRIlB is found only on neutrophils, whereas e c n 10 FcyRIIIA is found on macrophages, monocytes, natural killer (NK) cells, and a C subpopulation of T-cells. Notably, FcyRIIIA is the only FcR present on NK cells, one Sof the cell types implicated in ADCC.

SFcyRI, FcyRII and FcyRIII are immunoglobulin superfamily (IgSF) receptors; FcyRI has three IgSF domains in its extracellular domain, while FcyRII and FcyRIII have only two IgSF domains in their extracellular domains.

Another type of.Fc receptor is the neonatal Fe receptor (FcRn). FcRn is structurally similar to major histocompatibility complex (MHC) and consists of an otchain noncovalently bound to P2-microglobulin.

The binding sites on human and murine antibodies for FcyR have been previously mapped to the so-called "lower hinge region" consisting of residues 233-239 (EU index numbering as in Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, Md.

(1991)). Woof et al. Molec. Immunol. 23:319-330 (1986); Duncan et al. Nature 332:563 (1988); Canfield and Morrison, J. Exp. Med. 173:1483-1491 (1991); Chappel et al., Proc. Natl. Acad. Sci USA 88:9036-9040 (1991). Of residues 233-239, P238 and S239 have been cited as possibly being involved in binding, but these two residues have never been evaluated by substitution or deletion.

Other previously cited areas possibly involved in binding to FcyR are: G316- K338 (human IgG) for human FcyRI (by sequence comparison only; no substitution mutants were evaluated) (Woof et al. Molec Immunol. 23:319-330 (1986)); K274-R301 (human IgGI) for human FcyRIII (based on peptides) (Sarmay et al. Molec. Immunol.

21:43-51 (1984)); Y407-R416 (human IgG) for human FcyRIII (based on peptides) (Gergely et al. Biochem. Soc. Trans. 12:739-743 (1984)); as well as N297 and E318 (murine IgG2b) for murine FcyRII (Lund et al., Molec. Immunol., 29:53-59 (1992)).

Pro331 in IgG3 was changed to Ser, and the affinity of this variant to target cells analyzed. The affinity was found to be six-fold lower than that of unmutated IgG3, indicating the involvement of Pro331 in FcyRI binding. Morrison et al., Immunologist, 2:119-124 (1994); and Canfield and Morrison, J. Exp. Med. 173:1483-91 (1991).

H\cintae\Keep\speci\28505.00.doc 29/10/04

O

O

C1q binding O Clq and two serine proteases, C1r and Cls, form the complex C1, the first component Z of the complement dependent cytotoxicity (CDC) pathway. Clq is a hexavalent molecule with a C molecular weight of approximately 460,000 and a structure likened to a bouquet of tulips in which six collagenous "stalks" are connected to six globular head regions. Burton and Woof, Advances n In Immunol. 51:1-84 (1992). To activate the complement cascade, it is necessary for Clq to bind to at least two molecules of IgG1, IgG2, or lgG3 (the consensus is that lgG4 does not activate cn complement), but only one molecule of IgM, attached to the antigenic target. Ward and Ghetie, Therapeutic Immunology 2:77-94 (1995) at page o 10 Based upon the results of chemical modifications and crystallographic studies, Burton et al. (Nature, 288:338-344 (1980)) proposed that the binding site for the complement subcomponent Clq on IgG involves the last two (C-terminal) p-strands of the CH2 domain. Burton later suggested (Molec. Immunol., 22(3):161-206 (1985)) that the region comprising amino acid residues 318 to 337 might be involved in complement fixation.

Duncan and Winter (Nature 332:738-40 (1988)), using site directed mutageniesis, reported that Glu318, Lys320 and Lys322 form the binding site to Clq. The data of Duncan and Winter were generated by testing the binding of a mouse lgG2b isotype to guinea pig Clq. The role of Glu318, Lys320 and Lys322 residues in the binding of Clq was confirmed by the ability of a short synthetic peptide containing these residues to inhibit complement mediated lysis. Similar results are disclosed in U.S. Patent No. 5,648,260 issued on July 15, 1997, and U.S. Patent No.

5,624,821 issued on April 29, 1997.

The residue Pro331 has been implicated in Clq binding by analysis of the ability of human IgG subclasses to carry out complement mediated cell lysis. Mutation of Ser331 to Pro331 in lgG4 conferred the ability to activate complement. (Tao et al., J. Exp. Med., 178:661-667 (1993); Brekke et al., Eur. J. Immunol., 24:2542-47 (1994)).

From the comparison of the data of the Winter group, and the Tao et al. and Brekke et al papers, Ward and Ghetie concluded in their review article that there are at least two different regions involved in the binding of Clq: one on the p-strand of the CH2 domain bearing the Glu318, Lys320 and Lys322 residues, and the other on a turn located in close proximity to the same p-strand, and containing a key amino acid residue at position 331.

Other reports suggested that human IgG1 residues Leu235, and Gly237, located in the lower hinge region, play a critical role in complement fixation and activation. Xu et al., Immuno.

150:152A (Abstract) (1993). W094/29351 published December 22,.1994 reports that amino acid residues necessary for Clq and FcR binding of human IgG1 are located in the N-terminal region of the CH2 domain, residues 231 to 238.

O

O

It has further been proposed that the ability of IgG to bind Clq and activate the O complement cascade also depends on the presence, absence, or modification of the carbohydrate moiety positioned between the two CH2 domains (which is normally anchored at c Asn297). Ward and Ghetie, Therapeutic Immunology 2:77-94 (1995) at page 81.

SUMMARY OF THE INVENTION Ce The present invention provides a variant of a parent polypeptide comprising an Fc region, en which variant mediates antibody-dependent cell-mediated cytotoxicity (ADCC) in the presence C- of human effector cells more effectively, or binds an Fc gamma receptor (FcyR) with better affinity, S than the parent polypeptide and comprises at least one amino acid modification in the Fc region.

The polypeptide variant may, for example, comprise an antibody or an immunoadhesin. The Fc region of the parent polypeptide preferably comprises a human Fc region; a human IgG1, IgG2, lgG3 or lgG4 Fc region. The polypeptide variant preferably comprises an amino acid modification a substitution) at any one or more of amino acid positions 256, 290, 298, 312, 326, 330, 333, 334, 360, 378 or 430 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

In addition, the invention provides a polypeptide comprising a variant Fc region with altered Fc gamma receptor (FcyR) binding affinity, which polypeptide comprises an amino acid modification at any one or more of amino acid positions 238, 239, 248, 249, 252, 254, 255, 256, 258, 265, 267, 268, 269, 270, 272, 276, 278, 280, 283, 285, 286, 289, 290, 292, 293, 294, 295, 296, 298, 301, 303, 305, 307, 309, 312, 315, 320, 322, 324, 326, 327, 329, 330, 331, 333, 334, 335, 337, 338, 340, 360, 373, 376, 378, 382, 388, 389, 398, 414, 416, 419, 430, 434, 435, 437, 438 or 439 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat The variant Fc region preferably comprises a variant human IgG Fc region, a variant human IgG1, lgG2, lgG3 or lgG4 Fc region. In this respect, it is noted that, in the work in the above-cited art where the parent polypeptide had a non-human murine Fc region, different residues from those identified herein were thought to impact FcR binding. For example, in the murine IgG2b/murine FcyRII system, IgG E318 was found to be important for binding (Lund et a. Molec. Immunol. 27(1):53-59 (1992)), whereas E318A had no effect in the human IgG/human FcyRII system (Table 6 below).

In one embodiment, the polypeptide variant with altered FcyR binding activity displays reduced binding to an FcyR and comprises an amino acid modification at any one or more of amino acid positions 238, 239, 248, 249, 252, 254, 265, 268, 269, 270, 272, 278, 289, 292, 293, 294, 295, 296, 298, 301, 303, 322, 324,327, 329, 333, 335, 338, 340, 373, 376, 382, 388, 389, 414, 416, 419, 434, 435, 437, 438 or 439 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

(t For example, the polypeptide variant may display reduced binding to an Fc-yRI and 0 comprise an amino acid modification at any one or more of amino acid positions 238, 265, 269, Z 270, 327 or 329 of the Fc region, wherein the numbering of the residues in the Fc region is that N of the EU index as in Kabat.

The polypeptide variant may display reduced binding to an FcyRIl and comprise an amino S acid modification at any one or more of amino acid positions 238, 265, 269, 270, 292, 294, 295, 298, 303, 324, 327, 329, 333, 335, 338, 373, 376, 414,416, 419, 435, 438 or 439 of the Fc region, en wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

1 The polypeptide variant of interest may display reduced binding to an FcyRIll and comprise an amino acid modification at one or more of amino acid positions 238, 239, 248, 249, 252, 254, 265, 268, 269, 270,272,278, 289, 293, 294, 295, 296, 301,303,322, 327, 329, 338, 340, 373, 376, 382, 388, 389, 416, 434, 435 or 437 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

In another embodiment, the polypeptide variant with altered FcyR binding affinity displays improved binding to the FcyR and comprises an amino acid modification at any one or more of amino acid positions 255, 256, 258, 267, 268, 272, 276, 280, 283, 285, 286, 290, 298, 301, 305, 307, 309, 312, 315, 320, 322, 326, 330, 331, 333, 334, 337, 340, 360, 378, 398 or 430 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

For example, the polypeptide variant may display increased binding to an FcyRlli and, optionally, may further display decreased binding to an FcyRII. An exemplary such variant comprises amino acid modification(s) at position(s) 298 and/or 333 of the Fc region, wherein the numbering of the residues in the Pc region is that of the EU index as in Kabat.

The polypeptide variant may display increased binding to an FcyRII and comprise an amino acid modification at any one or more of amino acid positions 255, 256, 258, 267, 268, 272, 276,280, 283, 285, 286,290, 301,305,307, 309, 312, 315, 320, 322,326,330,331,337,340, 378, 398 or 430 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat. Such polypeptide variants with increased binding to an FcyRlI may optionally further display decreased binding to an FcyRIII and may, for example, comprise an amino acid modification at any one or more of amino acid positions 268, 272, 298, 301, 322 or 340 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

The invention further provides a polypeptide comprising a variant Fc region with altered neonatal Fc receptor (FcRn) binding affinity, which polypeptide comprises an amino acid modification at any one or more of amino acid positions 238, 252, 253, 254, 255, 256, 265, 272, 286,288.303,305,307,309,311,312,317,340,356,360,362,376,378, 380,382, 386, 388,

O

O

400, 413, 415, 424, 433, 434, 435, 436, 439 or 447 of the Fc region, wherein the numbering of O the residues in the Fc region is that of the EU index as in Kabat. Such polypeptide variants with reduced binding to an FcRn may comprise an amino acid modification at any one or more of Cl amino acid positions 252, 253, 254, 255, 288, 309, 386, 388, 400, 415, 433, 435, 436, 439 or 447 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat. The above-mentioned polypeptide variants may, alternatively, display increased e binding to FcRn and comprise an amino acid modification at any one or more of amino acid S positions 238, 256, 265, 272, 286, 303, 305, 307, 311, 312, 317, 340, 356, 360, 362, 376, 378, 380, 382, 413, 424 or 434 of the Fc region, wherein-the numbering of the residues in the Fc region is that of the EU index as in Kabat.

The invention also provides a composition comprising the polypeptide variant and a physiologically or pharmaceutically acceptable carrier or diluent. This composition for potential therapeutic use is sterile and may be lyophilized.

Diagnostic and therapeutic uses for the polypeptide variants disclosed herein are contemplated. In one diagnostic application, the invention provides a method for determining the presence of an antigen of interest comprising exposing a sample suspected of containing the antigen to the polypeptide variant and determining binding of the polypeptide variant to the sample. In one therapeutic application, the invention provides a method of treating a mammal suffering from or predisposed to a disease or disorder, comprising administering to the mammal a therapeutically effective amount of a polypeptide variant as disclosed herein, or of a composition comprising the polypeptide variant and a pharmaceutically acceptable carrier.

The invention further provides: isolated nucleic acid encoding the polypeptide variant; a vector comprising the nucleic acid, optionally, operably linked to control sequences recognized by a host cell transformed with the vector; a host cell containing the vector; a method for producing the polypeptide variant comprising culturing this host cell so that the nucleic acid is expressed and, optionally, recovering the polypeptide variant from the host cell culture from the host cell culture medium).

The inventien further provides a method for making a variant Fc region with altfe-te-F receptor (FcR) binding affinity, or altered antibody-dependent cell-mediated cyfotoxicity (ADCC) activity, comprising: introducing one or more amino acid-modifications into an Fc region of a parent polypeptide in order to generate a variant Focregion; determining binding of the variant Fc region to an FcR, or determining ADCC activity of the variant Fo region.

O

O

Step of the method may comprise determining binding of the variant Fe region to one O or more FcRs in vitro. Moreover, the method may result in the identification of a variant Fc region with improved FcR binding affinity, or with improved ADCC activity, in step thereof. Where Cl step comprises determining binding of the Fc region to an FcR, the FcR may, for example, be human Fc gamma receptor III (FcyRll). Where step comprises determining binding of the variant Fc region to at least two different FcRs, the FcRs tested preferably include human Fc en gamma receptor II (FcyRII) and human Fc gamma receptor III (FcyRIII).

e Brief Description of the Drawings o. Figure 1 is a schematic representation of a native IgG. Disulfide bonds are represented by heavy lines between CH1 and CL domains and the two CH2 domains. V is variable domain; C is constant domain; L stands for light chain and H stands for heavy chain.

Figure 2 shows Clq binding of wild type (wt) C2B8 antibody; C2B8 antibody with a human lgG2 constant region (IgG2); and variants K322A, K320A and E318A.

Figure 3 depicts C1q binding of variants P331A, P329A and K322A.

Figures 4A and 4B depict the amino acid sequences of E27 anti-lgE antibody light chain (Fig. 4A; SEQ ID NO:1) and heavy chain (Fig. 4B; SEQ ID NO:2).

Figure 5 is a schematic diagram of the "immune complex" prepared for use in the FcR assay described in Example 1. The hexamer comprising three anti-igE antibody molecules (the "Fc region-containing polypeptide") and three IgE molecules (the "first target molecule") is shown.

IgE has two "binding sites" for the anti-lgE antibody (E27) in the Fc region thereof. Each IgE molecule in the complex is further able to bind two VEGF molecules ("the second target polypeptide"). VEGF has two "binding sites" for IgE.

Figure 6 shows C1q binding results obtained for variants D270K and D270V compared to wild type C2B8.

Figure 7 depicts complement dependent cytotoxicity (CDC) of variants D270K and D270V, compared to wild type C2B8.

Figure 8 shows Clq binding ELISA results for 293 cell-produced wild type C2B8 antibody (293-Wt-C2B8), CHO-produced wild type C2B8 antibody (CHO-Wt-C2B8) and various variant antibodies.

Figure 9 shows Clq binding ELISA results obtained for wild type (wt) C288 and various variant antibodies as determined in Example 3.

Figure 10 depicts the three-dimensional structure of a human IgG Fc region, highlighting residues: Asp270, Lys326, Pro329, Pro331, Lys322 and Glu333.

Figure 11 shows C1q binding ELISA results obtained for wild type C2B8 and various variant antibodies as determined in Example 3.

O

O

c Figure 12 shows Clq binding ELISA results obtained for wild type C2B8 and double O variants, K326M-E333S and K326A-E333A.

Z Figure 13 shows CDC of wild type C2B8 and double variants, K326M-E333S and K326AiE333A.

Figure 14 depicts Clq binding ELISA results obtained for C2B8 with a human lgG4 (lgG4), tn wild type C2B8 (Wt-C2B8), C2B8 with a human lgG2 constant region (lgG2), and variant antibodies as described in Example 3.

tC Figures 15A and 15B show binding patterns for parent antibody (E27) to FcyRIIB and FcyRIIA. Figure 15A shows the binding pattern for the humanized anti-lgE E27 IgG1 as a O 10 monomer (open circles), hexamer (closed squares), and immune complex consisting of multiple hexamers (closed triangles) to a recombinant GST fusion protein of the human FcyRIIB (CD32) receptor a subunit The hexameric complex (closed squares) was formed by the mixture of equal molar concentrations of E27 (which binds to the Fc region of human IgE) and a human myeloma IgE. The hexamer is a stable 1.1 kD complex consisting of 3 IgG molecules (150 kD each) and 3 IgE molecules (200 kD each). The immune complex (closed triangles) was formed sequentially by first mixing equal molar concentrations of E27 and recombinant anti-VEGF IgE (human IgE with Fab variable domains that bind human VEGF) to form the hexamer. Hexamers were then linked to form an immune complex by the addition of 2x molar concentration of human VEGF, a 44 kD homodimer which has two binding sites for the anti-VEGF IgE per mole of VEGF. Figure 15B shows the binding pattern to a recombinant GST fusion protein of the human FcyRIIIA (CD16) receptor a subunit.

Figure 16A shows the binding of immune complexes using different antigen-antibody pairs to recombinant GST fusion protein of the FcyRIIA receptor a subunit. Figure 16B shows the binding of the same antigen-antibody pairs to the GST fusion protein of the FcyRIIIA receptor a subunit. Closed circles represent binding of human IgE:anti-lgE E27 IgG; open circles represent binding of human VEGF:humanized anti-VEGF IgG1.

Figure 17 summarizes differences in binding selectivity of some alanine variants between the different FcyRs. Binding of alanine variants at residues in the CH2 domain of anti-lgE E27 IgG1 are shown to FcyRIIA, FcyRIIB, and FcyRIIIA. Type 1 abrogates binding to all three receptors: D278A (265 in EU numbering). Type 2 improves'binding to FcyRIIA and FcyRIIB, while binding to FcyRllA is unaffected: S280A (267 in EU numbering). Type 3 improves binding to FcyRIlA and FcyRIIB, but reduces binding to FcyRIIIA: H281A (268 in EU numbering). Type 4 reduces binding to FcyRIIA and FcyRIIB, while improving binding to FcyRIIIA: S317A (298 in EU numbering). Type 5 improves binding to FcyRllIA, but does not affect binding to FcyRIIA and FcyRIIB: E352A, K353A (333 and 334 in EU numbering).

O

O

C Figures 18A and 18B compare the FcyRIIIA protein/protein assay and CHO GPI-FcyRIIIA O cell based assay, respectively. Figure 18A illustrates binding of selected alanine variants to Z FcyRIIIA-GST fusion protein. S317A (298 in EU numbering) and S317A/K353A (298 and 334 in c EU numbering) bind better than E27 wildtype, while D278A (265 in EU numbering) almost completely abrogates binding. Figure 18B illustrates that a similar pattern of binding is found on en CHO cells expressing a recombinant GPI-linked form of FcyRIIlA.

Figures 19A and 19B compare the FcyRIIB protein/protein assay and CHO GPI-FcyRIIB cn cell based assay, respectively. Figure 19A illustrates binding of selected alanine variants to Cl FcyRIIB-GST fusion protein. H281A (268 in EU numbering) binds better than E27 wildtype while S317A (298 in EU numbering) shows reduced binding. Figure 19B illustrates that a similar pattern of binding is found on CHO cells expressing a recombinant membrane bound form of FcyRIIB.

Figure 20 shows single alanine substitutions in the CH2 domain of anti-HER2 IgG1 (HERCEPTIN®) that influence FcyRIIA binding in both the protein-protein and cell-based assays alter the ability to bind to FcyRIltA on peripheral blood mononuclear cell (PBMC) effector cells.

Recombinant humanized anti-HER2 (HERCEPTIN®), which binds to HER2-expressing SK-BR-3 breast tumor cells, was preincubated with 5 1Cr-labeled SK-BR-3 cells for 30 minutes (opsonization) at 100 ng/ml (filled circles) and 1.25 ng/ml (filled squares). Keeping the SK-BR-3 tumor target cell concentration constant, the ratio of effector cells was increased from 0 to 100.

The spontaneous cytotoxicity in the absence of antibody (hatched squares) was 20% at an effector:target ratio of 100:1. A single alanine mutation that did not affect FcyRIIlA binding, variant G31 R309A (292 in EU numbering), did not effect ADCC (filled triangles). A single alanine mutation that only slightly increased binding to FcyRIIIA, variant G30 K307A (290 in EU numbering), also showed slightly improved ADCC a 1.1 fold improvement in ADCC activity, calculated as area under the curve) at 1.25 ng/ml at all E:T ratios (filled diamonds) compared to wildtype antibody at 1.25 ng/mI (filled square). A single alanine mutation that decreased binding to FcyRIIIA, variant G34 Q312A (295 in EU numbering), also showed decreased ADCC activity (filled inverted triangles).

Figure 21 illustrates that a single alanine mutation which had the most improved binding to FcyRIIIA, variant G36 S317A (298 in EU numbering), in the protein-protein and cell-based assays also showed the most improvement in ADCC (filled triangles) among the variants compared to wildtype (closed squares) at 1.25 ng/ml. G36 displayed a 1.7 fold improvement in ADCC activity, calculated as area under the curve. Variants G17 E282A (269 In EU numbering) and G18 D283A (270 in EU numbering) both showed reduced binding to FcyRIIIA as well as reduced efficacy in ADCC. The effector cells were PBMCs.

-11- O FIG. 22A depicts alignments of native sequence IgG Fe regions. Native C sequence human IgG Fc region sequences, humIgGi (non-A and A allotypes) (SEQ ID NOs:3 and 4, respectively), humlgG2 (SEQ ID NO:5), humlgG3 (SEQ ID NO:6) and o humlgG4 (SEQ ID NO:7), are shown. The human IgGI sequence is the non-A allotype, Z 5 and differences between this sequence and the A allotype (at positions 356 and 358; EU tn numbering system) are shown below the human IgGI sequence. Native sequence murine IgG Fc region sequences, murIgGI (SEQ ID NO:8), murIgG2A (SEQ ID NO:9), murIgG2B (SEQ ID NO: 10) and murIgG3 (SEQ ID NO: 11), are also shown.

FIG. 22B shows percent identity among the Fc region sequences of FIG. 22A.

FIG. 23 depicts alignments of native sequence human IgG Fc region n sequences, humlgGI (non-A and A allotypes; SEQ ID NOs:3 and 4, respectively), C humlgG2 (SEQ ID NO:5), humlgG3 (SEQ ID NO:6) and humIgG4 (SEQ ID NO:7) with differences between the sequences marked with asterisks.

o FIG. 24 shows area under curve (AUC) for selected variants compared to antil 15 HER2 IgG 1 (HERCEPTIN®) in a 4 hour ADCC assay. The effector cells were PBMCs Variant G36 (S317A; 298 in Eu numbering) with improved binding to FcyRIIIA showed improved ADCC activity; variant G31 (R309A; 292 in Eu numbering) which did not display altered FcyRIIIA binding, also had unaltered ADCC activity; and G14 (D265A; 278 in Eu numbering) which had reduced FcyRIIIA binding, also had reduced ADCC activity.

Detailed Description of the Preferred Embodiments I. Definitions In the claims which follow and in the preceding description of the invention, except where the context requires otherwise due to express language or necessary implication, the word "comprise" or variations such as "comprises" or "comprising" is used in an inclusive sense, i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.

Throughout the present specification and claims, the numbering of the residues in an immunoglobulin heavy chain is that of the EU index as in Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD. (1991), expressly incorporated herein by reference. The "EU index as in Kabat" refers to the residue numbering of the human IgG 1 EU antibody.

A "parent polypeptide" is a polypeptide comprising an amino add sequence which lacks one or more of the Fc region modifications disclosed herein and which differs in effector function compared to a polypeptide variant as herein disclosed. The parent polypeptide may comprise a native sequence Fe region or an Fc region with preexisting amino acid sequence modifications (such as additions, deletions and/or substitutions).

The term "Fc region" is used to define a C-terminal region of an immunoglobulin heavy chain, as shown in FIG. 1. The "Fc region" may be a native sequence Fc region or a variant Fc region. Although the boundaries of the Fc region of an immunoglobulin heavy chain might vary, the human IgG heavy chain Fc region is usually defined to stretch from an amino acid H:\cintae\Koep\speci\28505.00.doc 29/10/04

O

O

residue at position Cys226, or from Pro230, to the carboxyl-terminus thereof. The Fc region of an O immunoglobulin generally comprises two constant domains, CH2 and CH3, as shown, for example, in Fig. 1.

C The "CH2 domain" of a human IgG Fc region (also referred to as "Cy2" domain) usually extends from about amino acid 231 to about amino acid 340. The CH2 domain is unique in that it is not closely paired with another domain. Rather, two N-linked branched carbohydrate chains et are interposed between the two CH2 domains of an intact native IgG molecule. It has been C speculated that the carbohydrate may provide a substitute for the domain-domain pairing and help S stabilize the CH2 domain. Burton, Molec. Immunol.22:161-206 (1985).

The "CH3 domain" comprises the stretch of residues C-terminal to a CH2 domain in an Fc region from about amino acid residue 341 to about amino acid residue 447 of an IgG) A "functional Fc region" possesses an "effector function" of a native sequence Fc region.

Exemplary "effector functions" include C1q binding; complement dependent cytotoxicity; Fc receptor binding; antibody-dependent cell-mediated cytotoxicity (ADCC); phagocytosis; down regulation of cell surface receptors B cell receptor BCR), etc. Such effector functions generally require the Fc region to be combined with a binding domain an antibody variable domain) and can be assessed using various assays as herein disclosed, for example.

A "native sequence Fc region" comprises an amino acid sequence identical to the amino acid sequence of an Fc region found in nature. Native sequence human Fc regions are shown in Fig. 23 and include a native sequence human IgG1 Fc region (non-A and A allotypes); native sequence human IgG2 Fc region; native sequence human IgG3 Fc region; and native sequence human lgG4 Fc region as well as naturally occurring variants thereof. Native sequence murine Fc regions are shown in Fig. 22A.

A "variant Fc region" comprises an amino acid sequence which differs from that of a native sequence Fc region by virtue of at least one "amino acid modification" as herein defined.

Preferably, the variant Fc region has at least one amino acid substitution compared to a native sequence Fc region or to the Fc region of a parent polypeptide, e.g. from about one to about ten amino acid substitutions, and preferably from about one to about five amino acid substitutions in a native sequence Fc region or in the Fc region of the parent polypeptide. The variant Fc region herein will preferably possess at least about 80% homology with a native sequence Fc region and/or with an Fc region of a parent polypeptide, and most preferably at least about homology therewith, more preferably at least about 95% homology therewith.

"Homology" is defined as the percentage of residues in the amino acid sequence variant that are identical after aligning the sequences and introducing gaps, if necessary, to achieve the maximum percent homology. Methods and computer programs for the alignment are well known

O

O

C

N in the art. One such computer program is "Align authored by Genentech, Inc., which was filed O with user documentation in the United States Copyright Office, Washington, DC 20559, on Z December 10, 1991.

SThe term "Fc region-containing polypeptide" refers to a polypeptide, such as an antibody or immunoadhesin (see definitions below), which comprises an Fc region.

n The terms "Fc receptor" or "FcR" are used to describe a receptor that binds to the Fc region of an antibody. The preferred FcR is a native sequence human FcR. Moreover, a n preferred FcR is one which binds an IgG antibody (a gamma receptor) and includes receptors of the FcyRI, FcyRIi, and FcyRIII subclasses, including allelic variants and alternatively spliced forms 0 10 of these receptors. FcyRIl receptors include FcyRIIA (an "activating receptor") and FcyRIIB (an "inhibiting receptor"), which have similar amino acid sequences that differ primarily in the cytoplasmic domains thereof. Activating receptor FcyRIIA contains an immunoreceptor tyrosinebased activation motif (ITAM) in its cytoplasmic domain. Inhibiting receptor FcyRIIB contains an immunoreceptor tyrosine-based inhibition motif (ITIM) in its cytoplasmic domain. (see review M.

in Dadron, Annu. Rev. Immunol 15:203-234 (1997)). FcRs are reviewed in Ravetch and Kinet, Annu. Rev. Immunol 9:457-92 (1991); Capel etal., Immunomethods 4:25-34 (1994); and de Haas etal., J. Lab. Clin. Med. 126:330-41 (1995). Other FcRs, including those to be identified in the future, are encompassed by the term "FcR" herein. The term also includes the neonatal receptor, FcRn, which is responsible for the transfer of maternal IgGs to the fetus (Guyer et al., J. Immunol. 117:587 (1976) and Kim et al., J. Immunol. 24:249 (1994)).

"Antibody-dependent cell-mediated cytotoxicity" and "ADCC" refer to a cell-mediated reaction in which nonspecific cytotoxic cells that express FcRs Natural Killer (NK) cells, neutrophils, and macrophages) recognize bound antibody on a target cell and subsequently cause lysis of the target cell. The primary cells for mediating ADCC, NK cells, express FcyRII only, whereas monocytes express FcyRI, FcyRII and FcyRIII. FcR expression on hematopoietic cells is summarized in Table 3 on page 464 of Ravetch and Kinet, Annu. Rev. Immunol 9:457-92 (1991).

"Human effector cells" are leukocytes which express one or more FcRs and perform effector'functions. Preferably, the cells express at least FcyRill and perform ADCC effector function. Examples of human leukocytes which mediate ADCC include peripheral blood mononuclear cells (PBMC), natural killer (NK) cells, monocytes, cytotoxic T cells and neutrophils; with PBMCs and NK cells being preferred. The effector cells may be isolated from a native source thereof, e.g. from blood or PBMCs as described herein.

A polypeptide variant with "altered" FcR binding affinity or ADCC activity is one which has either enhanced or diminished FcR binding activity and/or ADCC activity compared to a parent -13-

O

O

Spolypeptide or to a polypeptide comprising a native sequence Fc region. The polypeptide variant O which "displays increased binding" to an FcR binds at least one FcR with better affinity than the Z parent polypeptide. The polypeptide variant which "displays decreased binding" to an FcR, binds C at least one FcR with worse affinity than a parent polypeptide. Such variants which display decreased binding to an FcR may possess little or no appreciable binding to an FcR, 0-20% t' binding to the FcR compared to a native sequence IgG Fc region, e.g. as determined in the Examples herein.

M The polypeptide variant which binds an FcR with "better affinity" than a parent polypeptide, is one which binds any one or more of the above identified FcRs with substantially better binding o 10 affinity than the parent antibody, when the amounts of polypeptide variant and parent polypeptide in the binding assay are essentially the same. For example, the polypeptide variant with improved FcR binding affinity may display from about 1.15 fold to about 100 fold, e.g. from about 1.2 fold to about 50 fold improvement in FcR binding affinity compared to the parent polypeptide, where FcR binding affinity is determined, for example, as disclosed in the Examples herein.

The polypeptide variant which "mediates antibody-dependent cell-mediated cytotoxicity (ADCC) in the presence of humah effector cells more effectively" than a parent antibody is one which in vitro or in vivo is substantially more effective at mediating ADCC, when the amounts of polypeptide variant and parent antibody used in the assay are essentially the same. Generally, such variants will be identified using the in vitro ADCC assay as herein disclosed, but other assays or methods for determining ADCC activity, e.g. in an animal model etc, are contemplated.

The preferred variant is from about 1.5 fold to about 100 fold, e.g. from about two fold to about fifty fold, more effective at mediating ADCC than the parent, e.g. in the in vitro assay disclosed herein.

An "amino acid modification" refers to a change in the amino acid sequence of a predetermined amino acid sequence. Exemplary modifications include an amino acid substitution, insertion and/or deletion; The preferred amino acid modification herein is a substitution.

An "amino acid modification at" a specified position, e.g. of the Fc region, refers to the substitution or deletion of the specified residue, or the insertion of at least one amino acid residue adjacent the specified residue. By insertion "adjacent" a specified residue is meant insertion within one to two residues thereof. The insertion may be N-terminal or C-terminal to the specified residue.

An "amino acid substitution" refers to the replacement of at least one existing amino acid residue in a predetermined amino acid sequence with another different "replacement" amino acid residue. The replacement residue or residues may be "naturally occurring amino acid residues" (Le. encoded by the genetic code) and selected from the group consisting of: alanine (AJa);

O

O

arginine (Arg); asparagine (Asn); aspartic acid (Asp); cysteine (Cys); glutamine (Gin); glutamic O acid (Glu); glycine (Gly); histidine (His); isoleucine (Ile): leuclne (Leu); lysine (Lys); methionine z (Met); phenylalanine (Phe); proline (Pro); serine (Ser); threonine (Thr); tryptophan (Trp); tyrosine Ci (Tyr); and valine (Val). Preferably, the replacement residue is not cysteine. Substitution with one or more non-naturally occurring amino acid residues is also encompassed by the definition of an amino acid substitution herein. A "non-naturally occurring amino acid residue" refers to a residue, other than those naturally occurring amino acid residues listed above, which is able to covalently tn bind adjacent amino acid residues(s) in a polypeptide chain. Examples of non-naturally occurring amino acid residues include norleucine, ornithine, norvaline, homoserine and other amino acid 0 10 residue analogues such as those described in Ellman et al. Meth. Enzym. 202:301-336 (1991).

To generate such non-naturally occurring amino acid residues, the procedures of Noren et al Science 244:182 (1989) and Ellman et al., supra, can be used. Briefly, these procedures involve chemically activating a suppressor tRNA with a non-naturally occurring amino acid residue followed by in vitro transcription and translation of the RNA.

An "amino acid insertion" refers to the incorporation of at least one amino acid into a predetermined amino acid sequence. While the insertion will usually consist of the insertion of one or two amino acid residues, the present application contemplates larger "peptide insertions", e.g. insertion of about three to about five or even up to about ten amino acid residues. The inserted residue(s) may be naturally occurring or non-naturally occurring as disclosed above.

An "amino acid deletion" refers to the removal of at least one amino acid residue from a predetermined amino acid sequence.

"Hinge region" is generally defined as stretching from Glu216 to Pro230 of human lgG1 (Burton, Molec. ImmunoL.22:161-206 (1985)). Hinge regions of other IgG isotypes may be aligned with the IgG1 sequence by placing the first and last cysteine residues forming inter-heavy chain S-S bonds in the same positions.

The "lower hinge region" of an Fc region is normally defined as the stretch of residues immediately C-terminal to the hinge region, I.e. residues 233 to 239 of the Fc region. Prior to the present invention, FcyR binding was generally attributed to amino acid residues in the lower hinge region of an IgG Fc region.

"Clq" is a polypeptide that includes a binding site for the Fc region of an immunoglobulin.

Clq together with two serine proteases, C1r and Cls, forms the complex C1, the first component of the complement dependent cytotoxicity (CDC) pathway. Human Clq can be purchased commercially from, e.g. Quidel, San Diego, CA.

The term "binding domain" refers to the region of a polypeptide that binds to another molecule. In the case of an FcR, the binding domain can comprise a portion of a polypeptide 0 0 chain thereof the a chain thereof) which is responsible for binding an Fc region. One useful o binding domain is the extracellular domain of an FcR a chain.

SThe term "antibody" is used in the broadest sense and specifically covers monoclonal Cl antibodies (including full length monoclonal antibodies), polyclonal antibodies, multispecific antibodies bispecific antibodies), and antibody fragments so long as they exhibit the desired biological activity.

ten "Antibody fragments", as defined for the purpose of the present invention, comprise a Sportion of an intact antibody, generally including the antigen binding or variable region of the intact o antibody or the Fc region of an antibody which retains FcR binding capability. Examples of antibody fragments include linear antibodies; single-chain antibody molecules; and multispecific antibodies formed from antibody fragments. The antibody fragments preferably retain at least part of the hinge and optionally the CH1 region of an IgG heavy chain. More preferably, the antibody fragments retain the entire constant region of an IgG heavy chain, and include an IgG light chain.

The term "monoclonal antibody" as used herein refers to an antibody obtained from a population of substantially homogeneous antibodies, Le., the individual antibodies comprising the population are identical except for possible naturally occurring mutations that may be present in minor amounts. Monoclonal antibodies are highly specific, being directed against a single antigenic site. Furthermore, in contrast to conventional (polyclonal) antibody preparations that typically include different antibodies directed against different determinants (epitopes), each monoclonal antibody is directed against a single determinant on the antigen. The modifier "monoclonal" indicates the character of the antibody as being obtained from a substantially homogeneous population of antibodies, and is not to be construed as requiring production of the antibody by any particular method. For example, the monoclonal antibodies to be used in accordance with the present invention may be made by the hybridoma method first described by Kohler et al., Nature 256:495 (1975), or may be made by recombinant DNA methods (see, e.g., U.S. Patent No. 4,816,567). The "monoclonal antibodies" may also be isolated from phage antibody libraries using the techniques described in Clackson et at, Nature 352:624-628 (1991) and Marks et al, J. Mol. Biol. 222:581-597 (1991), for example.

The monoclonal antibodies herein specifically include "chimeric" antibodies (immunoglobulins) in which a portion of the heavy and/or light chain is identical with or homologous to corresponding sequences in antibodies derived from a particular species or belonging to a particular antibody class or subclass, while the remainder of the chain(s) is identical with or homologous to corresponding sequences in antibodies derived from another species or belonging to another antibody class or subclass, as well as fragments of such 0 Santibodies, so long as they exhibit the desired biological activity Patent No. 4,816,567; and O Morrison et al., Proc. Natl. Acad. Sci. USA 81:6851-6855 (1984)).

"Humanized" forms of non-human murine) antibodies are chimeric antibodies that cN contain minimal sequence derived from non-human immunoglobulin. For the most part, humanized antibodies are human immunoglobulins (recipient antibody) in which residues from a hypervariable region of the recipient are replaced by residues from a hypervariable region of a non-human species (donor antibody) such as mouse, rat, rabbit or nonhuman primate having the tn desired specificity, affinity, and capacity. In some instances, Fv framework region (FR) residues of the human immunoglobulin are replaced by corresponding non-human residues. Furthermore, humanized antibodies may comprise residues that are not found in the recipient antibody or in the donor antibody. These modifications are made to further refine antibody performance. In general, the humanized antibody will comprise substantially all of at least one, and typically two, variable domains, in which all or substantially all of the hypervariable loops correspond to those of a nonhuman immunoglobulin and all or substantially all of the FR regions are those of a human immunoglobulin sequence. The humanized antibody optionally also will comprise at least a portion of an immunoglobulin constant region typically that of a human immunoglobulin. For further details, see Jones et Nature 321:522-525 (1986); Riechmann et Nature 332:323- 329 (1988); and Presta, Curr. Op. Struct. Biol. 2:593-596 (1992).

The term "hypervariable region" when used herein refers to the amino acid residues of an antibody which are responsible for antigen-binding. The hypervariable region comprises amino acid residues from a "complementarity determining region" or "CDR" (Le. residues 24-34 56 (L2) and 89-97 (L3) in the light chain variable domain and 31-35 50-65 (H2) and 95-102 (H3) in the heavy chain variable domain; Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD. (1991)) and/or those residues from a "hypervariable loop" (Le. residues 26-32 50-52 (L2) and 91-96 (L3) in the light chain variable domain and 26-32 53-55 (H2) and 96-101 (H3) in the heavy chain variable domain; Chothia and Lesk J. Mol. Biol. 196:901-917 (1987)).

"Framework" or "FR" residues are those variable domain residues other than the hypervariable region residues as herein defined.

As used herein, the term "immunoadhesin" designates antibody-like molecules which combine the "binding domain" of a heterologous "adhesin" protein a receptor, ligand or enzyme) with an immunoglobulin constant domain. Structurally, the immunoadhesins comprise a fusion of the adhesin amino acid sequence with the desired binding specificity which is other than the antigen recognition and binding site (antigen combining site) of an antibody (Le. is "heterologous") and an immunoglobulin constant domain sequence.

_LI I^ 1 1_ l Y/I

O

O

The term "ligand binding domain" as used herein refers to any native cell-surface receptor O or any region or derivative thereof retaining at least a qualitative ligand binding ability of a z corresponding native receptor. In a specific embodiment, the receptor is from a cell-surface c polypeptide having an extracellular domain that is homologous to a member of the immunoglobulin supergenefamily. Other receptors, which are not members of the immunoglobulin e supergenefamily but are nonetheless specifically covered by this definition, are receptors for cytokines, and in particular receptors with tyrosine kinase activity (receptor tyrosine kinases), t members of the hematopoietin and nerve growth factor receptor superfamilies, and cell adhesion molecules, e.g. L- and selectins.

O

o 10 The term "receptor binding domain" is used to designate any native ligand for a receptor, including cell adhesion molecules, or any region or derivative of such native ligand retaining at least a qualitative receptor binding ability of a corresponding native ligand. This definition, among others, specifically includes binding sequences from ligands for the above-mentioned receptors.

An "antibody-immunoadhesin chimera" comprises a molecule that combines at least one binding domain of an antibody (as herein defined) with at least one immunoadhesin (as defined in this application). Exemplary antibody-immunoadhesin chimeras are the bispecific CD4-IgG chimeras described in Berg et PNAS (USA) 88:4723-4727 (1991) and Chamow et at, J.

Immunol. 153:4268 (1994).

An "isolated" polypeptide is one that has been identified and separated and/or recovered from a component of its natural environment Contaminant components of its natural environment are materials that would interfere with diagnostic or therapeutic uses for the polypeptide, and may include enzymes, hormones, and other proteinaceous or nonproteinaceous solutes. In preferred embodiments, the polypeptide will be purified to greater than 95% by weight of polypeptide as determined by the Lowry method, and most preferably more than 99% by weight, to a degree sufficient to obtain at least 15 residues of N-terminal or internal amino acid sequence by use of a spinning cup sequenator, or to homogeneity by SDS-PAGE under reducing or nonreducing conditions using Coomassie blue or, preferably, silver stain. Isolated polypeptide includes the polypeptide in situ within recombinant cells since at least one component of the polypeptide's natural environment will not be present. Ordinarily, however, isolated polypeptide will be prepared by at least one purification step.

"Treatment" refers to both therapeutic treatment and prophylactic or preventative measures. Those in need of treatment include those already with the disorder as well as those in which the disorder is to be prevented.

A "disorder" is any condition that would benefit from treatment with the polypeptide variant.

This includes chronic and acute disorders or diseases including those pathological conditions 0 0 CA which predispose the mammal to the disorder in question. In one embodiment, the disorder is o cancer.

The terms "cancer" and "cancerous" refer to or describe the physiological condition in Cmammals that is typically characterized by unregulated cell growth. Examples of cancer include but are noi limited to, carcinoma, lymphoma, blastoma, sarcoma, and leukemia. More particular S examples of such cancers include squamous cell cancer, small-cell lung cancer, non-small cell lung cancer, adenocarcinoma of the lung, squamous carcinoma of the lung, cancer of the S peritoneum, hepatocellular cancer, gastrointestinal cancer, pancreatic cancer, glioblastoma, cervical cancer, ovarian cancer, liver cancer, bladdei cancer, hepatoma, breast cancer, colon cancer, colorectal cancer, endometrial or uterine carcinoma, salivary gland carcinoma, kidney cancer, liver cancer, prostate cancer, vulval cancer, thyroid cancer, hepatic carcinoma and various types of head and neck cancer.

A "HER2-expressing cancer" is one comprising cells which have HER2 receptor protein (Semba et al., PNAS (USA) 82:6497-6501 (1985) and Yamamoto et at. Nature 319:230-234 (1986) (Genebank accession number X03363)) present at their cell surface, such that an anti- HER2 antibody is able to bind to the cancer.

The word "label" when used herein refers to a detectable compound or composition which is conjugated directly or indirectly to the polypeptide. The label may be itself be detectable radioisotope labels or fluorescent labels) or, in the case of an enzymatic label, may catalyze chemical alteration of a substrate compound or composition which is detectable.

An "isolated" nucleic acid molecule is a nucleic acid molecule that is identified and separated from at least one contaminant nucleic acid molecule with which it is ordinarily associated in the natural source of the polypeptide nucleic acid. An isolated nucleic acid molecule is other than in the form or setting in which it is found in nature. Isolated nucleic acid molecules therefore are distinguished from the nucleic acid molecule as it exists in natural cells. However, an isolated nucleic acid molecule includes a nucleic acid molecule contained in cells that ordinarily express the polypeptide where, for example, the nucleic acid molecule is in a chromosomal location different from that of natural cells.

The expression "control sequences" refers to DNA sequences necessary for the expression of an operably linked coding sequence in a particular host organism. The control sequences that are suitable for prokaryotes, for example, include a promoter, optionally an operator sequence, and a ribosome binding site. Eukaryotic cells are known to utilize promoters, polyadenylation signals, and enhancers.

Nucleic acid is "operably linked" when it is placed into a functional relationship with another nucleic acid sequence. For example, DNA for a presequence or secretory leader is

O

O

operably linked to DNA for a polypeptide if it is expressed as a preprotein that participates in the o secretion of the polypeptide; a promoter or enhancer is operably linked to a coding sequence if it affects the transcription of the sequence; or a ribosome binding site is operably linked to a Cl coding sequence if it is positioned so as to facilitate translation. Generally, "operably linked" means that the DNA sequences being linked are contiguous, and, in the case of a secretory 0n leader, contiguous and in reading phase. However, enhancers do not have to be contiguous.

Linking is accomplished by ligation at convenient restriction sites. If such sites do not exist, the tl synthetic oligonucleotide adaptors or linkers are used in accordance with conventional practice.

1As used herein, the expressions "cell," "cell line," and "cell culture" are used interchangeably and all such designations include progeny. Thus, the words "transformants" and "transformed cells" include the primary subject cell and cultures derived therefrom without regard for the number of transfers. It is also understood that all progeny may not be precisely identical in DNA content, due to deliberate or inadvertent mutations. Mutant progeny that have the same function or biological activity as screened for in the originally transformed cell are included. Where distinct designations are intended, it will be clear from the context.

The term "molecular complex" when used herein refers to the relatively stable structure which forms when two or more heterologous molecules polypeptides) bind (preferably noncovalently) to one another. The preferred molecular complex herein is an immune complex.

"Immune complex" refers to the relatively stable structure which forms when at least one target molecule and at least one heterologous Fc region-containing polypeptide bind to one another forming a larger molecular weight complex. Examples of immune complexes are antigenantibody aggregates and target molecule-immunoadhesin aggregates. The term "immune complex" as used herein, unless indicated otherwise, refers to an ex vivo complex (Le. other than the form or setting in which it may be found in nature). However, the immune complex may be administered to a mammal, e.g. to evaluate clearance of the immune complex in the mammal.

The term "target molecule" refers to a molecule, usually a polypeptide, which is capable of being bound by a heterologous molecule and has one or more binding sites for the heterologous molecule. The term "binding site" refers to a region of a molecule to which another molecule can bind. The "first target molecule" herein comprises at least two distinct binding sites (for example, two to five separate binding sites) for an analyte an Fc region-containing polypeptide) such that at least two analyte molecules can bind to the first target molecule. In the preferred embodiment of the invention, the two or more binding sites are identical having the same amino acid sequence, where the target molecule is a polypeptide). In Example 1 below, the first target molecule was IgE and had two separate binding sites in the Fc region thereof to which the Fc region-containing polypeptide (an anti-IgE antibody, E27) could bind. Other first

O

O

target molecules include dimers of substantially identical monomors neurotrophins, IL8 and O VEGF) or are polypeptides comprising two or more substantially identical polypeptide chains (e.g.

Z antibodies or immunoadhesins). The "second target molecule" comprises at least two distinct c binding sites (for example, two to five separate binding sites) for the first target molecule such that at least two first target molecules can bind to the second target molecule. Preferably, the two or t more binding sites are identical having the same amino acid sequence, where the target molecule is a polypeptide). In Example 2, the second target molecule was VEGF, which has a n pair of distinct binding sites to which the variable domain of the IgE antibody could bind. Other second target molecules are contemplated, e.g. other dimers of substantially identical monomers 0 10 neurotrophins or IL8) or polypeptides comprising two or more substantially identical domains antibodies or immunoadhesins).

An "analyte" is a substance that is to be analyzed. The preferred analyte is an Fc regioncontaining polypeptide that is to be analyzed for its ability to bind to an Fc receptor.

A "receptor is a polypeptide capable of binding at least one ligand. The preferred receptor is a cell-surface receptor having an extracellular tigand-binding domain and, optionally, other domains transmembrane domain, intracellular domain and/or membrane anchor). The receptor to be evaluated in the assay described herein may be an intact receptor or a fragment or derivative thereof a fusion protein comprising the binding domain of the receptor fused to one or more heterologous polypeptides). Moreover, the receptor to be evaluated for its binding properties may be present in a cell or isolated and optionally coated on an assay plate or some other solid phase.

The phrase "low affinity receptor" denotes a receptor that has a weak binding affinity for a ligand of interest, e.g. having a binding constant of about 50nM or worse affinity. Exemplary low affinity receptors include FcyRII and FcyRIllI.

II. Modes for Carrying Out the Invention The invention herein relates to a method for making a polypeptide variant. The "parent", "starting" or "nonvariant" polypeptide is prepared using techniques available in the art for generating polypeptides comprising an Fc region. In the preferred embodiment of the invention, the parent polypeptide is an antibody and exemplary methods for generating antibodies are described in more detail in the following sections. The parent polypeptide may, however, be any other polypeptide comprising an Fc region, e.g. an immunoadhesin. Methods for making immunoadhesins are elaborated in more detail hereinbelow.

In an alternative embodiment, a variant Fc region may be generated according to the methods herein disclosed and this "variant Fc region" can be fused to a heterologous polypeptide of choice, such as an antibody variable domain or binding domain of a receptor or ligand.

-21-

O

O

The parent polypeptide comprises an Fc region. Generally the Fc region of the parent O polypeptide will comprise a native sequence Fc region, and preferably a human native sequence Fc region. However, the Fc region of the parent polypeptide may have one or more pre-existing l amino acid sequence alterations or modifications from a native sequence Fc region. For example, the Clq binding activity of the Fc region may have been previously altered (other types of Fc region modifications are described in more detail below). In a further embodiment the parent e polypeptide Fc region is "conceptual" and, while it does not physically exist, the antibody engineer en may decide upon a desired variant Fc region amino acid sequence and generate a polypeptide comprising that sequence or a DNA encoding the desired variant Fc region amino acid sequence.

In the preferred embodiment of the invention, however, a nucleic acid encoding an Fc region of a parent polypeptide is available and this nucleic acid sequence is altered to generate a variant nucleic acid sequence encoding the Fe region variant.

DNA encoding an amino acid sequence variant of the starting polypeptide is prepared by a variety of methods known in the art. These methods include, but are not limited to, preparation by site-directed (or oligonucleotide-mediated) mutagenesis, PCR mutagenesis, and cassette mutagenesis of an earlier prepared DNA encoding the polypeptide Site-directed mutagenesis is a preferred method for preparing substitution variants. This technique is well known in the art (see, e.g.,Carter et al. Nucleic Acids Res. 13:4431-4443 (1985) and Kunkel et at, Proc. Natl. Acad.ScL.USA 82:488 (1987)). Briefly, in carrying out sitedirected mutagenesis of DNA, the starting DNA is altered by first hybridizing an oligonucleotide encoding the desired mutation to a single strand of such starting DNA. After hybridization, a DNA polymerase is used to synthesize an entire second strand, using the hybridized oligonucleotide as a primer, and using the single strand of the starting DNA as a template. Thus, the oligonucleotide encoding the desired mutation is incorporated in the resulting double-stranded

DNA.

PCR mutagenesis is also suitable for making amino acid sequence variants of the starting polypeptide. See Higuchi, in PCR Protocols, pp.177-183 (Academic Press, 1990); and Vallette et al., Nuc. Acids Res. 17:723-733 (1989). Briefly, when small amounts of template DNA are used as starting material in a PCR, primers that differ slightly in sequence from the corresponding region in a template DNA can be used to generate relatively large quantities of a specific DNA fragment that differs from the template sequence only at the positions where the primers differ from the template.

Another method for preparing variants, cassette mutagenesis, is based on the technique described by Wells et al., Gene 34:315-323 (1985). The starting material is the plasmid (or other vector) comprising the starting polypeptide DNA to be mutated. The codon(s) in the starting DNA -22i t

O

O

to be mutated are identified. There must be a unique restriction endonuclease site on each side o of the identified mutation site(s). If no such restriction sites exist, they may be generated using Z the above-described oligonucleotide-mediated mutagenesis method to introduce them at C- appropriate locations in the starting polypeptide DNA. The plasmid DNA is cut at these sites to linearize it. A double-stranded oligonucleotide encoding the sequence of the DNA between the restriction sites but containing the desired mutation(s) is synthesized using standard procedures, wherein the two strands of the oligonucleotide are synthesized separately and then hybridized en together using standard techniques. This double-stranded oligonucleotide is referred to as the cassette. This cassette is designed to have 5' and 3' ends that are compatible with the ends of 0 10 the linearized plasmid, such that it can be directly ligated to the plasmid. This plasmid now contains the mutated DNA sequence.

Alternatively, or additionally, the desired amino acid sequence encoding a polypeptide variant can be determined, and a nucleic acid sequence encoding such amino acid sequence variant can be generated synthetically.

The amino acid sequence of the parent polypeptide is modified in order to generate a variant Fc region with altered Fc receptor binding affinity or activity in vitro and/or in vivo and/or altered antibody-dependent cell-mediated cytotoxicity (ADCC) activity in vitro and/or in vivo.

Generally, the modification entails one or more amino acid substitutions. In one embodiment, the replacement residue does not correspond to a residue present in the same position in any of the native sequence Fc regions in Figure 22A. For example, according to this embodiment of the invention, Pro331 of a human IgG3 or IgG1 Fc region is replaced with a residue other than Ser (the corresponding aligned residue found in native sequence human IgG4).

In one embodiment, the residue in the parent polypeptide which is substituted with a replacement residue is not an alanine and/or is not residue Ala339 of an Fc region. In the case of an amino acid substitution, preferably the residue in the parent polypeptide is replaced with an alanine residue. However, the present invention contemplates replacement of the residue of the parent polypeptide with any other amino acid residue. The substitution may, for example, be a "conservative substitution". Such conservative substitutions are shown in Table 1 under the heading of "preferred substitution". More substantial changes may be achieved by making one or more "exemplary substitutions" which are not the preferred substitution in Table 1.

TABLE 1 Original Residue Exemplary Preferred Substitutions Substitution Ala val; leu; ile val Arg lys; gin; asn lys Asn gin; his; lys; arg gin Asp glu glu Cys ser ser Gin asn asn Glu asp asp Gly pro; ala ala His asn; gin; lys; arg arg lie leu; val; met; ala; leu phe; norleucine Leu norleucine; ile; val; ie met; ala; phe Lys arg; gin; asn arg Met leu; phe; ile leu Phe leu; val; ile; ala; tyr leu Pro ala ala Ser(S) thr thr Thr(T) ser ser Trp tyr; phe tyr Tyr trp; phe; thr; ser phe Val ile; leu; met; phe; leu ala; norleucine Substantial modifications in the biological properties of the Fc region may be accomplished by selecting substitutions that differ significantly in their effect on maintaining (a) the structure of the polypeptide backbone in the area of the substitution, for example,.as a sheet or helical conformation, the charge or hydrophobicity of the molecule at the target site, or (c) the bulk of the side chain. Naturally occurring residues are divided into groups based on common side-chain properties: hydrophobic: norleucine, met, ala, val, leu, ile; neutral hydrophilic: cys, ser, thr; acidic: asp, glu; basic: asn, gin, his, lys, arg; -24residues that influence chain orientation: gly, pro; and aromatic: trp, tyr, phe.

Non-conservative substitutions will entail exchanging a member of one of these classes for a member of another class. Conservative and non-conservative amino acid substitutions are exemplified in Table 8 hereinbelow.

As is demonstrated in Example 4 herein, one can engineer an Fc region variant with altered binding affinity for one or more FcRs. As was shown in that Example, different classes of Fc. region variants can be made as summarized in the following table. Where the variant Fc region has more than one amino acid substitution, generally, but not necessarily, amino acid substitutions in the same class are combined to achieve the desired result.

TABLE 2 CLASSES OF Fc REGION VARIANTS Class FcR binding property Position of Fc region substitution(s) 1A reduced binding to all FcyR 238, 265, 269, 270, 297*, 327, 329 1B reduced binding to both FcyRII and 239, 294, 295, 303, 338, 373, 376, 416, FcyRll 435 2 improved binding to both FcyRII and 256, 290, 312, 326, 330, 339", 378, 430 FcyRIII 3 improved binding to FcyRII and no 255, 258, 267, 276, 280, 283, 285, 286, effect on FcyRlll binding 305, 307, 309, 315, 320, 331, 337, 398 4 improved binding to FcyRll and 268, 272, 301, 322, 340 reduced binding to FcyRill reduced binding to FcyRII and no 292, 324, 335, 414, 419, 438, 439 effect on FcyRIII binding 6 reduced binding to FcyRII and 298, 333 improved binding to FcyRIll 7 no effect on FcyRII binding and 248, 249, 252, 254, 278, 289, 293, 296, reduced binding to FcyRlll 338, 382, 388, 389, 434, 437 8 no effect on FcyRII binding and 334, 360 improved binding to FcyRIl deglycosylated version Preferably combined with other Fc region modification(s) as disclosed herein) Aside from amino acid substitutions, the present invention contemplates other modifications of the parent Fc region amino acid sequence in order to generate an Fc region variant with altered effector function.

0 0 COne may, for example, delete one or more amino acid residues of the Fc region in order o to reduce binding to an FcR. Generally, one will delete one or more of the Fc region residues Z identified herein as effecting FcR binding (see Example 4 below) in order to generate such an Fc Cl region variant. Generally, no more than one to about ten Fc region residues will be deleted according to this embodiment of the invention. The Fc region herein comprising one or more tn amino acid deletions will preferably retain at least about 80%, and preferably at least about and most preferably at least about 95%, of the parent Fc region or of a native sequence human Cn Fc region.

Cl One may also make amino acid insertion Fc region variants, which variants have altered o 10 effector function. For example, one may introduce at least one amino acid residue one to two amino acid residues and generally no more than ten residues) adjacent to one or more of the Fc region positions identified herein as impacting FcR binding. By "adjacent" is meant within one to two amino acid residues of a Fc region residue identified herein. Such Fc region variants may display enhanced or diminished FcR binding andlor ADCC activity. In order to generate such insertion variants, one may evaluate a co-crystal structure of a polypeptide comprising a binding region of an FcR the extracellular domain of the FcR of interest) and the Fc region into which the amino acid residue(s) are to be inserted (see, for example, Deisenhofer, Biochemistry 20(9):2361-2370 (1981); and Burmeister et al., Nature 342:379-383, (1994)) in order to rationally design an Fc region variant with, improved FcR binding ability. Such insertion(s) will generally be made in an Fc region loop, but not in the secondary structure (Le. in a P-strand) of the Fc region.

By introducing the appropriate amino acid sequence modifications in a parent Fc region, one can generate a variant Fc region which mediates antibody-dependent cell-mediated cytotoxicity (ADCC) in the presence of human effector cells more effectively and/or binds an Fc gamma receptor (FcyR) with better affinity than the parent polypeptide. Such Fc region variants will generally comprise at least one amino acid modification in the Fc region. Combining amino acid modifications is thought to be particularly desirable. For example, the variant Fc region may include two, three, four, five, etc substitutions therein, e.g. of the specific Fc region positions identified herein.

Preferably, the parent polypeptide Fc region is a human Fc region, e.g. a native sequence human Fc region human IgG1 (A and non-A allotypes), lgG2, IgG3 or lgG4 Fc region. Such sequences are shown in Fig. 23.

To generate an Fc region with improved ADCC activity, the parent polypeptide preferably has pre-existing ADCC activity, it comprises a human IgG1 or human lgG3 Fc region. In one embodiment, the variant with improved ADCC mediates ADCC substantially more effectively than -26- I

C

0 0 an antibody with a native sequence IgG1 or IgG3 Fc region and the antigen-binding region of the o variant. Preferably, the variant comprises, or consists essentially of, substitutions of two or three

F

Z of the residues at positions 298, 333 and 334 of the Fc region. Most preferably, residues at Cpositions 298, 333 and 334 are substituted, with alanine residues). Moreover, in order to generate the Fc region variant with improved ADCC activity, one will generally engineer an Fc region variant with improved binding affinity for FcyRllI, which is thought to be an important FcR for mediating ADCC. For example, one may introduce an amino acid modification a e substitution) into the parent Fc region at any one or more of amino acid positions 256, 290, 298, 312, 326, 330, 333, 334, 360, 378 or 430 to generate such a variant. The variant with improved o 10 binding affinity for FcyRIII may further have reduced binding affinity for FcyRII, especially reduced affinity for the inhibiting FcyRIIB receptor.

The amino acid modification(s) are preferably introduced into the CH2 domain of a FP region, since the experiments herein indicate that the CH2 domain is important for FcR binding activity. Moreover, unlike the teachings of the above-cited art, the instant application contemplates the introduction of a modification into a part of the Fc region other than in the lower hinge region thereof.

Useful amino acid positions for modification in order to generate a variant IgG Fc region with altered Fc gamma receptor (FcyR) binding affinity or activity include any one or more of amino acid positions 238, 239, 248, 249, 252, 254, 255, 256, 258, 265, 267, 268, 269, 270, 272, 276, 278, 280, 283, 285, 286, 289, 290, 292, 293, 294, 295, 296, 298, 301, 303, 305, 307, 309, 312, 315, 320, 322, 324, 326, 327, 329, 330, 331, 333, 334, 335, 337, 338, 340, 360, 373, 376, 378, 382, 388, 389, 398, 414, 416, 419, 430, 434, 435, 437, 438 or 439 of the Fc region.

Preferably, the parent Fc region used as the template to generate such variants comprises a human IgG Fc region. Where residue 331 is substituted, the parent Fc region is preferably not human native sequence lgG3, or the variant Fc region comprising a substitution at position 331 preferably displays increased FcR binding, e.g. to FcyRl.

To generate an Fc region variant with reduced binding to the FcyR one may introduce an amino acid modification at any one or more of amino acid positions 238, 239, 248, 249, 252, 254, 265, 268, 269, 270, 272, 278, 289, 292, 293, 294, 295, 296, 298, 301, 303, 322, 324, 327, 329, 333, 335, 338, 340, 373, 376, 382, 388, 389, 414,416, 419, 434, 435, 437, 438 or 439 of the Fc region.

Variants which display reduced binding to FcyRI, include those comprising an Fc region amino acid modification at any one or more of amino acid positions 238, 265, 269, 270, 327 or 329.

-27- 0 0 CVariants which display reduced binding to FcyRI include those comprising an Fc region o amino acid modification at any one or more of amino acid positions 238, 265, 269, 270, 292, 294, Z 295, 298, 303, 324, 327, 329, 333, 335, 338, 373, 376, 414, 416, 419, 435,438 or 439.

CFc region variants which display reduced binding to FcyRII include those comprising an Fc region amino acid modification at any one or more of amino acid positions 238, 239, 248, 249, ten 252, 254, 265, 268, 269,270, 272, 278, 289, 293, 294, 295, 296,301,303,322, 327, 329,338, 340, 373, 376, 382, 388, 389, 416, 434, 435 or 437.

tCl Variants with improved binding to one or more FcyRs may also be made. Such Fc region variants may comprise an amino acid modification at any one or more of amino acid positions 0 o 10 255, 256, 258, 267, 268,272,276, 280,283, 285, 286, 290, 298, 301, 305, 307, 309,312,315, 320, 322, 326, 330, 331, 333, 334, 337, 340, 360, 378, 398 or 430 of the Fc region.

For example, the variant with improved FcyR binding activity may display increased binding to FcyRIll, and optionally may further display decreased binding to FcyRlI; e.g. the variant may comprise an amino acid modification at position 298 and/or 333 of an Fc region.

Variants with increased binding to FcyRIl include those comprising an amino acid modification at any one or more of amino acid positions 255, 256, 258, 267, 268, 272, 276, 280, 283, 285, 286, 290, 301, 305, 307, 309, 312, 315, 320, 322, 326, 330, 331, 337, 340, 378, 398 or 430 of an Fc region. Such variants may further display decreased binding to FcYRIII. For example, they may include an Fc region amino acid modification at any one or more of amino acid positions 268, 272, 298, 301, 322 or 340.

While it is preferred to alter binding to a FcyR, Fc region variants with altered binding affinity for the neonatal receptor (FcRn) are also contemplated herein. Fc region variants with improved affinity for FcRn are anticipated to have longer serum half-lives, and such molecules will have useful applications in methods of treating mammals where long half-life of the administered polypeptide is desired, to treat a chronic disease or disorder. Fc region variants with decreased FcRn binding'affinity, on the contrary, are expected to have shorter half-lives, and such molecules may, for example, be administered to a mammal where a shortened circulation time may be advantageous, e.g. for In vivo diagnostic imaging or for polypeptides which have toxic side effects when left circulating in the blood stream for extended periods, etc. Fc region variants with decreased FcRn binding affinity are anticipated to be less likely to cross the placenta, and thus may be utilized in the treatment of diseases or disorders in pregnant women.

Fc region variants with altered binding affinity for FcRn include those comprising an Fc region amino acid modification at any one or more of amino acid positions 238, 252, 253, 254, 255, 256, 265, 272, 286, 288, 303, 305, 307, 309, 311, 312, 317, 340, 356, 360, 362, 376, 378, 380, 382, 386, 388, 400, 413, 415, 424, 433, 434, 435, 436, 439 or 447. Those which display S reduced binding to FcRn will generally comprise an Fc region amino acid modification at any one o or more of amino acid positions 252, 253, 254, 255, 288, 309, 386, 388, 400, 415, 433, 435, 436, Z 439 or 447; and those with increased binding to FcRn will usually comprise an Fc region amino l acid modification at any one or more of amino acid positions 238, 256, 265, 272, 286, 303, 305, 307, 311, 312, 317, 340, 356, 360, 362, 376, 378, 380, 382, 413, 424 or 434.

M The polypeptide variant(s) prepared as described above may be subjected to further S modifications, oftentimes depending on the intended use of the polypeptide. Such modifications en may involve further alteration of the amino acid sequence (substitution, insertion and/or deletion of amino acid residues), fusion to heterologous polypeptide(s) and/or covalent modifications.

o 10 Such "further modifications" may be made prior to, simultaneously with, or following, the amino acid modification(s) disclosed above which result in an alteration of Fc receptor binding and/or ADCC activity. In one embodiment, one may combine the Fc region modification herein with Fc region substitutions disclosed in the references cited in the "Related Art" section of this application.

Alternatively or additionally, it may be useful to combine the above amino acid modifications with one or more further amino acid modifications that alter Clq binding and/or complement dependent cytoxicity function of the Fc region.

The starting polypeptide of particular interest herein is usually one that binds to C1q and displays complement dependent cytotoxicity (CDC). The further amino acid substitutions described herein will generally serve to alter the ability of the starting polypeptide to bind to Cl q and/or modify its complement dependent cytotoxicity function, e.g. to reduce and preferably abolish these effector functions. However, polypeptides comprising substitutions at one or more of the described positions with improved Cl q binding andlor complement dependent cytotoxicity (CDC) function are contemplated herein. For example, the starting polypeptide may be unable to bind Clq and/or mediate CDC and may be modified according to the teachings herein such that it acquires these further effector functions. Moreover, polypeptides with pre-existing Clq binding activity, optionally further having the ability to mediate CDC may be modified such that one or both of these activities are enhanced.

To generate an Fc region with altered Clq binding and/or complement dependent cytotoxicity (CDC) function, the amino acid positions to be modified are generally selected from heavy chain positions 270, 322, 326, 327, 329, 331, 333, and 334, where the numbering of the .residues in an IgG heavy chain is that of the EU index as in Kabat et aL, Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD (1991). In one embodiment, only one of the eight above-identified positions is altered in order to generate the polypeptide variant region with altered Clq binding and/or -29-

O

O

c complement dependent cytotoxicity (CDC) function. Preferably only residue 270, 329 or 322 is O altered if this is the case. Alternatively, two or more of the above-identified positions are modified.

Z If substitutions are to be combined, generally substitutions which enhance human Clq binding NC at residue positions 326, 327, 333 and 334) or those which diminish human Clq binding at residue positions 270, 322, 329 and 331) are combined. In the latter embodiment, all four positions 270, 322, 329 and 331) may be substituted. Preferably, further substitutions at two, three or all of positions 326, 327, 333 or 334 are combined, optionally with other Fc region t substitutions, to generate a polypeptide with improved human Clq binding and preferably improved CDC activity in vitro or In viva.

O

o 10 Proline is conserved at position 329 in human IgG's. This residue is preferably replaced with alanine, however substitution with any other amino acid is contemplated, serine, threonine, asparagine, glycine or valine.

Proline is conserved at position 331 in human IgG1, lgG2 and lgG3, but not lgG4 (which has a serine residue at position 331). Residue 331 is preferably replaced by alanine or another amino acid, e.g. serine (for IgG regions other than lgG4), glycine or valine.

Lysine. 322 is conserved in human IgGs, and this residue is preferably replaced by an alanine residue, but substitution with any other amino acid residue is contemplated, e.g. serine, threonine, glycine or valine.

D270 is conserved in human IgGs, and this residue may be replaced by another amino acid residue, e.g. alanine, serine, threonine, glycine, valine, or lysine.

K326 is also conserved in human IgGs. This residue may be substituted with another residue including, but not limited to, valine, glutamic acid, alanine, glycine, aspartic acid, methionine or tryptophan, with tryptophan being preferred.

Likewise, E333 is also conserved in human IgGs. E333 is preferably replaced by an amino acid residue with a smaller side chain volume, such as valine, glycine, alanine or serine, with serine being preferred.

K334 is conserved in human IgGs and may be substituted with another residue such as alanine or other residue.

In human IgG1 and IgG3, residue 327 is an alanine. In order to generate a variant with improved C1q binding, this alanine may be substituted with another residue such as glycine. In lgG2 and lgG4, residue 327 is a glycine and this may be replaced by alanine (or another residue) to diminish C1q binding.

As disclosed above, one can design an Fc region with altered effector function, by modifying Clq binding and/or FcR binding and thereby changing CDC activity and/or ADCC activity. For example, one can generate a variant Fc region with improved Clq binding and

O

O

improved FcyRIll binding; e.g. having both improved ADCC activity and improved CDC activity.

O Alternatively, where one desires that effector function be reduced or ablated, one may engineer Z a variant Fc region with reduced CDC activity and/or reduced ADCC activity. In other C embodiments, one may increase only one of these activities, and optionally also reduce the other activity, e.g. to generate an Fc region variant with improved ADCC activity, but reduced CDC n activity and vice versa.

With respect to further amino acid sequence alterations, any cysteine residue not involved t' in maintaining the proper conformation of the polypeptide variant also may be substituted, generally with serine, to improve the oxidative stability of the molecule and prevent aberrant cross o 10 linking.

Another type of amino acid substitution serves to alter the glycosylation pattern of the polypeptide. This may be achieved by deleting one or more carbohydrate moieties found in the polypeptide, and/or adding one or more glycosylation sites that are not present in the polypeptide.

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

Addition of glycosylation sites to the polypeptide is conveniently accomplished by altering the amino acid sequence such that it contains one or more of the above-described tripeptide sequences (for N-linked glycosylation sites). The alteration may also be made by the addition of, or substitution by, one or more serine or threonine residues to the sequence of the original polypeptide (for O-linked glycosylation sites). An exemplary glycosylation variant has an amino acid substitution of residue Asn 297 of the heavy chain.

Moreover, the class, subclass or allotype of the Fc region may be altered by one or more further amino acid substitutions to generate an Fc region with an amino acid sequence more homologous to a different class, subclass or allotype as desired. For example, a murine Fc region may be altered to generate an amino acid sequence more homologous to a human Fc region; a human non-A allotype IgG1 Fc region may be modified to achieve a human A allotype IgG1 Fc region etc. In one embodiment, the amino modification(s) herein which alter FcR binding and/or ADCC activity are made in the CH2 domain of the Fc region and the CH3 domain is deleted or -31replaced with another dimerization domain. Preferably, however, the CH3 domain is retained o (aside from amino acid modifications therein which alter effector function as herein disclosed).

Z The polypeptide variant may be subjected to one or more assays to evaluate any change C in biological activity compared to the starting polypeptide.

Preferably the polypeptide variant essentially retains the ability to bind antigen compared M to the nonvariant polypeptide, Le. the binding capability is no worse than about 20 fold, e.g. no worse than about 5 fold of that of the nonvariant polypeptide. The binding capability of the en polypeptide variant may be determined using techniques such as fluorescence activated cell sorting (FACS) analysis or radioimmunoprecipitation (RIA), for example.

0 10 The ability of the polypeptide variant to bind an FcR may be evaluated. Where the FcR is a high affinity Fc receptor, such as FcyRI, FcRn or FcyRIIIA-V158, binding can be measured by titrating monomeric polypeptide variant and measuring bound polypeptide variant using an antibody which specifically binds to the polypeptide variant in a standard ELISA format (see Example 2 below). Another FcR binding assay for low affinity FcRs is described in Examples 1 and 4.

To assess ADCC activity of the polypeptide variant, an in vitro ADCC assay, such as that described in Example 4 may be performed using varying effector:target ratios. Useful "effector cells" for such assays include peripheral blood mononuclear cells (PBMC) and Natural Killer (NK) cells. Alternatively, or additionally, ADCC activity of the polypeptide variant may be assessed in vivo, in a animal model such as that disclosed in Clynes et al. PNAS (USA) 95:652-656 (1998).

The ability of the variant to bind Clq and mediate complement dependent cytotoxicity (CDC) may be assessed.

To determine Clq binding, a Clq binding ELISA may be performed. Briefly, assay plates may be coated overnight at 4°C with polypeptide variant or starting polypeptide (control) in coating buffer. The plates may then be washed and blocked. Following washing, an aliquot of human Clq may be added to each well and incubated for 2 hrs at room temperature. Following a further wash, 100 pl of a sheep anti-complement C1q peroxidase conjugated antibody may be added to each well and incubated for 1 hour at room temperature. The plate may again be washed with wash buffer and 100 gil of substrate buffer containing OPD (O-phenylenediamine dihydrochloride (Sigma)) may be added to each well. The oxidation reaction, observed by the appearance of a yellow color, may be allowed to proceed for 30 minutes and stopped by the addition of 100 pl of N H 2

SO

4 The absorbance may then read at (492-405) nm.

An exemplary polypeptide variant is one that displays a "significant reduction in C1q binding" in this assay. This means that about 100.g/ml of the polypeptide variant displays about O C 50 fold or more reduction in Clq binding compared to 100g/ml of a control antibody having a O nonmutated IgG1 Fc region. In the most preferred embodiment, the polypeptide variant "does not bind Clq", Ie. 100pg/ml of the polypeptide variant displays about 100 fold or more reduction Cl in Clq binding compared to 1001 g/ml of the control antibody.

Another exemplary variant is one which "has a better binding affinity for human Clq than the parent polypeptide". Such a molecule may display, for example, about two-fold or more, and en preferably about five-fold or more, improvement in human C1q binding compared to the parent t polypeptide at the ICso values for these two molecules). For example, human C1q binding may be about two-fold to about 500-fold, and preferably from about two-fold or from about five-fold to about 1000-fold improved compared to the parent polypeptide.

To assess complement activation, a complement dependent cytotoxicity (CDC) assay may be performed, e.g. as described in Gazzano-Santoro et al., J. Immunol. Methods 202:163 (1996). Briefly, various concentrations of the polypeptide variant and human complement may be diluted with buffer. Cells which express the antigen to which the polypeptide variant binds may be diluted to a density of -1 x 10' cells /ml. Mixtures of polypeptide variant, diluted human complement and cells expressing the antigen may be added to a flat bottom tissue culture 96 well plate and allowed to incubate for 2 hrs at 37°C and 5% CO 2 to facilitate complement mediated cell lysis. 50 pl of alamar blue (Accumed International) may then be added to each well and incubated overnight at 37 0 C. The absorbance is measured using a 96-well fluorometer with excitation at 530 nm and emission at 590 nm. The results may be expressed in relative fluorescence units (RFU). The sample concentrations may be computed from a standard curve and the percent activity as compared to nonvariant polypeptide is reported for the polypeptide variant of interest.

Yet another exemplary variant "does not activate complement". For example, 0.6 pg/ml of the polypeptide variant displays about 0-10% CDC activity in this assay compared to a 0.6 ig/ml of a control antibody having a nonmutated IgG 1 Fc region. Preferably the variant does not appear to have'any CDC activity in the above CDC assay.

The invention also pertains to a polypeptide variant with enhanced CDC compared to a parent polypeptide, displaying about two-fold to about 100-fold improvement in CDC activity In vitro or in vivo at the ICso values for each molecule being compared).

A. Receptor Binding Assay and Immune Complex A receptor binding assay has been developed herein which is particularly useful for determining binding of an analyte of interest to a receptor where the affinity of the analyte for the receptor is relatively weak, e.g. in the micromolar range as is the case for FcyRIIA, FcyRIIB, FcyRIIIA and FcyRIllB. The method involves the formation of a molecular complex that has an improved avidity for the receptor of interest compared to the noncomplexed analyte. The -33- L 0

O

preferred molecular complex is an immune complex comprising: an Fc region-containing O polypeptide (such as an antibody or an immunoadhesin); a first target molecule which Z comprises at least two binding sites for the Fc region-containing polypeptide; and a second C target molecule which comprises at least two binding sites for the first target molecule.

In Example 1 below, the Fc region-containing polypeptide is an anti-lgE antibody, such t' as the E27 antibody (Figs. 4A-4B). E27, when mixed with human IgE at an 1:1 molar ratio, forms a stable hexamer consisting of three E27 molecules and three IgE molecules. In Example 1 below,-the "first target molecule" is a chimeric form of IgE in which the Fab portion of an anti- VEGF antibody is fused to the human IgE Fc portion and the "second target molecule" is the O 10 antigen to which the Fab binds (Le. VEGF). Each molecule of IgE binds two molecules of VEGF.

VEGF also binds two molecules of IgE per molecule of VEGF. When recombinant human VEGF was added at a 2:1 molar ratio to IgE:E27 hexamers, the hexamers were linked into larger molecular weight complexes via the IgE:VEGF interaction (Fig. The Fc region of the anti-lgE antibody of the resultant immune complex binds to FcR with higher avidity than either uncomplexed anti-lgE or anti-lgE:lgE hexamers.

Other forms of molecular complexes for use in the receptor assay are contemplated.

Examples comprising only an Fc region-containing polypeptide:first target molecule combination include an immunoadhesin:ligand combination such as VEGF receptor (KDR)immunoadhesin:VEGF and a full-length bispecific antibody (bsAb):first target molecule. A further example of an Fc region-containing polypeptide:first target molecule:second target molecule combination include a nonblocking antibody:soluble receptor:ligand combination such as anti-Trk antibody:soluble Trk receptor:neurotrophin (Urfer et al. J. Biol. Chem. 273(10):5829-5840 (1998)).

Aside from use in a receptor binding assay, the immune complexes described above have further uses including evaluation of Fc region-containing polypeptide function and immune complex clearance in vivo. Hence, the immune complex may be administered to a mammal (e.g.

in a pre-clinical animal study) and evaluated for its half-life etc.

To determine receptor binding, a polypeptide comprising at least the binding domain of the receptor of interest the extracellular domain of an a subunit of an FcR) may be coated on solid phase, such as an assay plate. The binding domain of the receptor alone or a receptorfusion protein may be coated on the plate using standard procedures. Examples of receptorfusion proteins include receptor-glutathione S-transferase (GST) fusion protein, receptor-chitin binding domain fusion protein, receptor-hexaHis tag fusion protein (coated on glutathione, chitin, and nickel coated plates, respectively). Alternatively, a capture molecule may be coated on the assay plate and used to bind the receptor-fusion protein via the non-receptor portion of the fusion -34- 0 0 c protein. Examples include anti-hexaHis F(ab') 2 coated on the assay plate used to capture o receptor-hexaHis tail fusion or anti-GST antibody coated on the assay plate used to capture a

F

Z receptor-GST fusion. In other embodiments, binding to cells expressing at least the binding C domain of the receptor may be evaluated. The cells may be naturally occurring hematopoietic cells that express the FcR of interest or may be transformed with nucleic acid encoding the FcR en or a binding domain thereof such that the binding domain is expressed at the surface of the cell to be tested.

en The immune complex described hereinabove is added to the receptor-coated plates and incubated for a sufficient period of time such that the analyte binds to the receptor. Plates may o 10 then be washed to remove unbound complexes, and binding of the analyte may be detected according to known methods. For example, binding may be detected using a reagent an antibody or fragment thereof) which binds specifically to the analyte, and which is optionally conjugated with a detectable label (detectable labels and methods for conjugating them to polypeptides are described below in the section entitled "Non-Therapeutic Uses for the Polypeptide Variant").

As a matter of convenience, the reagents can be provided in an assay kit, a packaged combination of reagents, for combination with the analyte in assaying the ability of the analyte to bind to a receptor of interest. The components of the kit will generally be provided in predetermined ratios. The kit may provide the first target molecule and/or the second target molecule, optionally complexed together. The kit may further include assay plates coated with the receptor or a binding domain thereof the extracellular domain of the oc subunit of an FcR).

Usually, other reagents, such as an antibody that binds specifically to the analyte to be assayed, labeled directly or indirectly with an enzymatic label, will also be provided in the kit. Where the detectable label is an enzyme, the kit will include substrates and cofactors required by the enzyme a substrate precursor which provides the detectable chromophore or fluorophore). In addition, other additives may be included such as stabilizers, buffers assay and/or wash lysis buffer) and the like. The relative amounts of the various reagents may be varied widely to provide for concentrations in solution of the reagents that substantially optimize the sensitivity of the assay. Particularly, the reagents may be provided as dry powders, usually lyophilized, including excipients that on dissolution will provide a reagent solution having the appropriate concentration.

The kit also suitably includes instructions for carrying out the assay.

B. Antibody Preparation In the preferred embodiment of the invention, the Fc region-containing polypeptide which is modified according to the teachings herein is an antibody. Techniques for producing antibodies follow: 0 0 Antigen selection and preparation O Where the polypeptide is an antibody, it is directed against an antigen of interest.

Z Preferably, the antigen is a biologically important polypeptide and administration of the antibody Sto a mammal suffering from a disease or disorder can result in a therapeutic benefit in that mammal. However, antibodies directed against nonpolypeptide antigens (such as tumoren associated glycolipid antigens; see US Patent 5,091,178) are also contemplated.

Where the antigen is a polypeptide, it may be a transmembrane molecule receptor) n or ligand such as a growth factor. Exemplary antigens include molecules such as renin; a growth hormone, including human growth hormone and bovine growth hormone; growth hormone o 10 releasing factor; parathyroid hormone; thyroid stimulating hormone; lipoproteins; alpha-1antitrypsin; insulin A-chain; insulin B-chain; proinsulin; follicle stimulating hormone; calcitonin; luteinizing hormone; glucagon; clotting factors such as factor VIIIC, factor IX, tissue factor (TF), and von Willebrands factor; anti-clotting factors such as Protein C; atrial natriuretic factor; lung surfactant; a plasminogen activator, such as'urokinase or human urine or tissue-type plasminogen activator bombesin; thrombin; hemopoietic growth factor; tumor necrosis factor-alpha and -beta; enkephalinase; RANTES (regulated on activation normally T-cell expressed and secreted); human macrophage inflammatory protein (MIP-1-alpha); a serum albumin such as human serum albumin; Muellerian-inhibiting substance; relaxin A-chain; relaxin B-chain; prorelaxin; mouse gonadotropin-associated peptide; a microbial protein, such as beta-lactamase; DNase; IgE; a cytotoxic T-lymphocyte associated antigen (CTLA), such as CTLA-4; inhibin; activin; vascular endothelial growth factor (VEGF); receptors for hormones or growth factors; protein A or D; rheumatoid factors; a neurotrophic factor such as bone-derived neurotrophic factor (BDNF), neurotrophin-3, or -6 (NT-3, NT-4, NT-5, or NT-6), or a nerve growth factor such as NGF-p; platelet-derived growth factor (PDGF); fibroblast growth factor such as aFGF and bFGF; epidermal growth factor (EGF); transforming growth factor (TGF) such as TGF-alpha and TGFbeta, including TGF-pl, TGF-p2, TGF-p3, TGF-p4, or TGF-p5; insulin-like growth factor-I and -II (IGF-I and IGF-II); des(1-3)-IGF-l (brain IGF-I), insulin-like growth factor binding proteins; CD proteins such as CD3, CD4, CD8, CD19 and CD20; erythropoietin; osteoinductive factors; immunotoxins; a bone morphogenetic protein (BMP); an interferon such as interferon-alpha, -beta, and -gamma; colony stimulating factors (CSFs), M-CSF, GM-CSF, and G-CSF; interleukins (ILs), IL-1 to IL-10; superoxide dismutase; T-cell receptors; surface membrane proteins; decay accelerating factor; viral antigen such as, for example, a portion of the AIDS envelope; transport proteins; homing receptors; addressins; regulatory proteins; integrins such as CD1 1a, CD lb, CD1 c, CD18, an ICAM, VLA-4 and VCAM; a tumor associated antigen such as HER2, HER3 or HER4 receptor; and fragments of any of the above-listed polypeptides.

-36- 0 Preferred molecular targets for antibodies encompassed by the present invention include O CD proteins such as CD3, CD4, CD8, CD19, CD20 and CD34; members of the ErbB receptor Z family such as the EGF receptor, HER2, HER3 or HER4 receptor; cell adhesion molecules such N as LFA-1, Mac1, p150.95, VLA-4, ICAM-1, VCAM, a4/p7 integrin, and avlp3 integrin including either a or P subunits thereof anti-CD11 a, anti-CD18 or anti-CD11 b antibodies); growth 0 factors such as VEGF; tissue factor alpha interferon (a-IFN); an interleukin, such as IL-8; tn IgE: blood group antigens; flk2/flt3 receptor; obesity (OB) receptor; mpl receptor; CTLA-4; protein C etc.

1 Soluble antigens or fragments thereof, optionally conjugated to other molecules, can be used as immunogens for generating antibodies. For transmembrane molecules, such as receptors, fragments of these the extracellular domain of a receptor) can be used as the immunogen. Alternatively, cells expressing the transmembrane molecule can be used as the immunogen. Such cells can be derived from a natural source cancer cell lines) or may be cells which have been transformed by recombinant techniques to express the transmembrane molecule. Other antigens and forms thereof useful for preparing antibodies will be apparent to those in the art.

(ii) Polyclonal antibodies Polyclonal antibodies are preferably raised in animals by multiple subcutaneous (sc) or intraperitoneal (ip) injections of the relevant antigen and an adjuvant. It may be useful to conjugate the relevant antigen to a protein that is immunogenic in the species to be immunized, keyhole limpet hemocyanin, serum albumin, bovine thyroglobulin, or soybean trypsin inhibitor using a bifunctional or derivatizing agent, for example, maleimidobenzoyl sulfosuccinimide ester (conjugation through cysteine residues), N-hydroxysuccinimide (through lysine residues), glutaraldehyde, succinic anhydride, SOC 2 or R'N=C=NR, where R and R' are different alkyl groups.

Animals are immunized against the antigen, immunogenic conjugates, or derivatives by combining, 100 pg or 5 pg of the protein or conjugate (for rabbits or mice, respectively) with 3 volumes of Freund's complete adjuvant and injecting the solution intradermally at multiple sites.

One month later the animals are boosted with 1/5 to 1/10 the original amount of peptide or conjugate in Freund's complete adjuvant by subcutaneous injection at multiple sites. Seven to 14 days later the animals are bled and the serum is assayed for antibody titer. Animals are boosted until the titer plateaus. Preferably, the animal is boosted with the conjugate of the same antigen, but conjugated to a different protein and/or through a different cross-linking reagent.

Conjugates also can be made in recombinant cell culture as protein fusions. Also, aggregating agents such as alum are suitably used to enhance the immune response.

-37- (Ill) Monoclonal antibodies 0 Monoclonal antibodies may be made using the hybridoma method first described by S Kohler et al., Nature, 256:495 (1975), or may be made by recombinant DNA methods (U.S.

C Patent No. 4,816,567).

In the hybridoma method, a mouse or other appropriate host animal, such as a hamster or macaque monkey, is immunized as hereinabove described to elicit lymphocytes that produce or are capable of producing antibodies that will specifically bind to the protein used for tn immunization. Alternatively, lymphocytes may be immunized in vitro. Lymphocytes then are fused with myeloma cells using a suitable fusing agent, such as polyethylene glycol, to form a 0 10 hybridoma cell (Goding, 'Monoclonal Antibodies: Principles and Practice, pp.59-103 (Academic Press, 1986)).

The hybridoma cells thus prepared are seeded and grown in a suitable culture medium that preferably contains one or more substances that inhibit the growth or survival of the unfused, parental myeloma cells. For example, if the parental myeloma cells lack the enzyme hypoxanthine guanine phosphoribosyl transferase (HGPRT or HPRT), the culture medium for the hybridomas typically will include hypoxanthine, aminopterin, and thymidine (HAT medium), which substances preventthe growth of HGPRT-deficient cells.

Preferred myeloma cells are those that fuse efficiently, support stable high-level production of antibody by the selected antibody-producing cells, and are sensitive to a medium such as HAT medium. Among these, preferred myeloma cell lines are murine myeloma lines, such as those derived from MOPC-21 and MPC-11 mouse tumors available from the Salk Institute Cell Distribution Center, San Diego, California USA, and SP-2 or X63-Ag8-653 cells available from the American Type Culture Collection, Rockville, Maryland USA. Human myeloma and mousehuman heteromyeloma cell lines also have been described for the production of human monoclonal antibodies (Kozbor, J. Immunol., 133:3001 (1984); Brodeur et al., Monoclonal Antibody Production Techniques and Applications, pp. 51-63 (Marcel Dekker, Inc., New York, 1987)).

Culture medium in which hybridoma cells are growing is assayed for production of monoclonal antibodies directed against the antigen. Preferably, the binding specificity of monoclonal antibodies produced by hybridoma cells is determined by immunoprecipitation or by an in vitro binding assay, such as radioimmunoassay (RIA) or enzyme-linked immunoabsorbent assay (ELISA).

After hybridoma cells are identified that produce antibodies of the desired specificity, affinity, and/or activity, the clones may be subcloned by limiting dilution procedures and grown by standard methods (Goding, Monoclonal Antibodies: Principles and Practice, pp.59-103 -38-

O

O

C (Academic Press, 1986)). Suitable culture media for this purpose include, for example, D-MEM O or RPMI-1640 medium. In addition, the hybridoma cells may be grown in vivo as ascites tumors Z in an animal.

The monoclonal antibodies secreted by the subclones are suitably separated from the culture medium, ascites fluid, or serum by conventional immunoglobulin purification procedures n such as, for example, protein A-Sepharose, hydroxylapatite chromatography, gel electrophoresis, dialysis, or affinity chromatography.

en DNA encoding the monoclonal antibodies is readily isolated and sequenced using conventional procedures by using oligonucleotide probes that are capable of binding 0 10 specifically to genes encoding the heavy and light chains of the monoclonal antibodies). The hybridoma cells serve as a preferred source of such DNA. Once isolated, the DNA may be placed into expression vectors, which are then transfected into host cells such as E. coll cells, simian COS cells, Chinese hamster ovary (CHO) cells, or myeloma cells that do not otherwise produce immunoglobulin protein, to obtain the synthesis of monoclonal antibodies in the recombinant host cells. Recombinant production of antibodies will be described in more detail below.

In a further embodiment, antibodies or antibody fragments can be isolated from antibody phage libraries generated using the techniques described in McCafferty et Nature, 348:552- 554 (1990). Clackson et aL, Nature, 352:624-628 (1991) and Marks et al., J. Mo. Biol., 222:581-597 (1991) describe the isolation of murine and human antibodies, respectively, using phage libraries. Subsequent publications describe the production of high affinity (nM range) human antibodies by chain shuffling (Marks et Bio/Technology, 10:779-783 (1992)), as well as combinatorial infection and in vivo recombination as a strategy for constructing very large phage libraries (Waterhouse et al., Nuc. Acids. Res., 21:2265-2266 (1993)). Thus, these techniques are viable alternatives to traditional monoclonal antibody hybridoma techniques for isolation of monoclonal antibodies.

The DNA also may be modified, for example, by substituting the coding sequence for human heavy- and light-chain constant domains in place of the homologous murine sequences Patent No. 4,816,567; Morrison, et Proc. Natl Acad. Sci. USA, 81:6851 (1984)), or by covalently joining to the immunoglobulin coding sequence all or part of the coding sequence for a non-immunoglobulin polypeptide.

Typically such non-immunoglobulin polypeptides are substituted for the constant domains of an antibody, or they are substituted for the variable domains of one antigen-combining site of an antibody to create a chimeric bivalent antibody comprising one antigen-combining site having specificity for an antigen and another antigen-combining site having specificity for a different antigen.

0 c( (iv) Humanized and human antibodies o A humanized antibody has one or more amino acid residues introduced into it from a Z source which is non-human. These non-human amino acid residues are often referred to as "import" residues, which are typically taken from an "import" variable domain. Humanization can be essentially performed following the method of Winter and co-workers (Jones et al., Nature, e321:522-525 (1986); Riechmann etal., Nature, 332:323-327 (1988); Verhoeyen etal, Science, 239:1534-1536 (1988)), by Substituting rodent CDRs or CDR sequences for the corresponding n sequences of a human antibody. Accordingly, such "humanized" antibodies are chimeric antibodies Patent No. 4,816,567) wherein substantially less than an intact human variable domain has been substituted by the corresponding sequence from a non-human species. In practice, humanized antibodies are typically human antibodies in which some CDR residues and possibly some FR residues are substituted by residues from analogous sites in rodent antibodies.

The choice of human variable domains, both light and heavy, to be used In making the humanized antibodies is very important to reduce antigenicity. According to the so-called "best-fit" method, the sequence of the variable domain of a rodent antibody is screened against the entire library of known human variable-domain sequences. The human sequence which is closest to that of the rodent is then accepted as the human framework (FR) for the humanized antibody (Sims et al., J. Immunol., 151:2296 (1993); Chothia et at., J. Mol. BioL, 196:901 (1987)).

Another method uses a particular framework derived from the consensus sequence of all human antibodies of a particular subgroup of light or heavy chains. The same framework may be used for several different humanized antibodies (Carter et atl, Proc. Natl. Acad. Sci. USA, 89:4285 (1992); Presta et aL, J. Immnot, 151:2623 (1993)).

It is further important that antibodies be humanized with retention of high affinity for the antigen and other favorable biological properties. To achieve this goal, according to a preferred method, humanized antibodies are prepared by a process of analysis of the parental sequences and various conceptual humanized products using three-dimensional models of the parental and humanized sequences. Three-dimensional immunoglobulin models are commonly available and are familiar to those skilled in the art. Computer programs are available which illustrate and display probable three-dimensional conformational structures of selected candidate immunoglobulin sequences. Inspection of these displays permits analysis of the likely role of the residues in the functioning of the candidate immunoglobulin sequence, the analysis of residues that influence the ability of the candidate immunoglobulin to bind its antigen. In this way, FR residues can be selected and combined from the recipient and import sequences so that the desired antibody characteristic, such as increased affinity for the target antigen(s), is achieved.

O

O

SIn general, the CDR residues are directly and most substantially involved in influencing antigen O binding.

Z Alternatively, it is now possible to produce transgenic animals mice) that are Cl capable, upon immunization, of producing a full repertoire of human antibodies in the absence of endogenous immunoglobulin production. For example, it has been described that the tn homozygous deletion of the antibody heavy-chain joining region (JH) gene in chimeric and germline mutant mice results in complete inhibition of endogenous antibody production. Transfer of the tn human germ-line immunoglobulin gene array in such germ-line mutant mice will result in the production of human antibodies upon antigen challenge. See, Jakobovits et al., Proc. Natl.

O

O 10 Acad. Scl. USA, 90:2551 (1993); Jakobovits et Nature, 362:255-258 (1993); Bruggermann et al., Year in Immuno., 7:33 (1993); and Duchosal et a. Nature 355:258 (1992). Human antibodies can also be derived from phage-display libraries (Hoogenboom et aL, J. Mol. Blot., 227:381 (1991); Marks et al., J. Mol. Blol., 222:581-597 (1991); Vaughan etal. Nature Blotech 14:309 (1996)).

Multispecific antibodies Multispecific antibodies have binding specificities for at least two different antigens. While such molecules normally will only bind two antigens bispecific antibodies, BsAbs), antibodies with additional specificities such as trispecific antibodies are encompassed by this expression when used herein. Examples of BsAbs include those with one arm directed against a tumor cell antigen and the other arm directed against a cytotoxic trigger molecule such as anti-FcyRI/antianti-p185HE2/FcyRIII (CD16), anti-CD3/anti-malignant B-cell (1D10), anti-CD3/antip1 8 5 HER2, anti-CD3/anti-p97, anti-CD3/anti-renal cell carcinoma, anti-CD3/anti-OVCAR-3, anti- CD3/L-D1 (anti-colon carcinoma), anti-CD3/anti-melanocyte stimulating hormone analog, anti- EGF receptor/anti-CD3, anti-CD3/anti-CAMA1, anti-CD3/anti-CD19, anti-CD3/MoV18, anti-neural cell ahesion molecule (NCAM)/anti-CD3, anti-folate binding protein (FBP)/anti-CD3, anti-pan carcinoma associated antigen (AMOC-31)/anti-CD3; BsAbs with one arm which binds specifically to a tumor antigen and one arm which binds to a toxin such as anti-saporin/anti-ld-1, anti- CD22/anti-saporin, anti-CD7/anti-saporin, anti-CD38/anti-saporin, anti-CEAlanti-ricin A chain, antiinterferon-a (IFN-a)/anti-hybridoma idiotype, anti-CEA/anti-vinca alkaloid; BsAbs for converting enzyme activated prodrugs such as anti-CD30/anti-alkaline phosphatase (which catalyzes conversion of mitomycin phosphate prodrug to mitomycin alcohol); BsAbs which can be used as fibrinolytic agents such as anti-fibrin/anti-tissue plasminogen activator (tPA), anti-fibrin/antiurokinase-type plasminogen activator (uPA); BsAbs for targeting immune complexes to cell surface receptors such as anti-low density lipoprotein (LDL)/anti-Fc receptor FcyRI, FcyRll or FcyRIII); BsAbs for use in therapy of infectious diseases such as anti-CD3/anti-herpes simplex I 1 I_

O

O

virus (HSV), anti-T-cell receptorCD3 complex/anti-influenza, anti-FcyR/anti-HIV; BsAbs for tumor O detection in vitro or In vivo such as anti-CEAlanti-EOTUBE, anti-CEA/anti-DPTA, antip185H/anti-hapten; BsAbs as vaccine adjuvants; and BsAbs as diagnostic tools such as antic rabbit IgG/anti-ferritin, anti-horse radish peroxidase (HRP)/anti-hormone, anti-somatostatin/antisubstance P, anti-HRP/anti-FITC, anti-CEA/anti-p-galactosidase. Examples of trispecific antibodies include anti-CD3/anti-CD4/anti-CD37, anti-CD3/anti-CD5/anti-CD37 and anti-CD3/anti- CD8/anti-CD37. Bispecific antibodies can be prepared as full length antibodies or antibody en fragments F(ab') 2 bispecific antibodies).

Methods for making bispecific antibodies are known in the art. Traditional production of full length bispecific antibodies is based on the coexpression of two immunoglobulin heavy chainlight chain pairs, where the two chains have different specificities (Millstein et at., Nature, 305:537-539 (1983)). Because of the random assortment of immunoglobulin heavy and light chains, these hybridomas (quadromas) produce a potential mixture of 10 different antibody molecules, of which only one has the correct bispecific structure. Purification of the correct molecule, which is usually done by affinity chromatography steps, is rather cumbersome, and the product yields are low. Similar procedures are disclosed in WO 93/08829, and in Traunecker et EMBO 10:3655-3659 (1991).

According to a different approach, antibody variable domains with the desired binding specificities (antibody-antigen combining sites) are fused to immunoglobulin constant domain sequences. The fusion preferably is with an immunoglobulin heavy chain constant domain, comprising at least part of the hinge, CH2, and CH3 regions. It is preferred to have the first heavy-chain constant region (CH1) containing the site necessary for light chain binding, present in at least one of the fusions. DNAs encoding the immunoglobulin heavy chain fusions and, if desired, the immunoglobulin light chain, are inserted into separate expression vectors, and are co-transfected into a suitable host organism. This provides for great flexibility in adjusting the mutual proportions of the three polypeptide fragments in embodiments when unequal ratios of the three polypeptide chains used in the construction provide the optimum yields. It is, however, possible to insert the coding sequences for two or all three polypeptide chains in one expression vector when the expression of at least two polypeptide chains in equal ratios results in high yields or when the ratios are of no particular significance.

In a preferred embodiment of this approach, the bispecific antibodies are composed of a hybrid immunoglobulin heavy chain with a first binding specificity in one arm, and a hybrid immunoglobulin heavy chain-light chain pair (providing a second binding specificity) in the other arm. It was found that this asymmetric structure facilitates the separation of the desired bispecific compound from unwanted immunoglobulin chain combinations, as the presence of an

O

O

c1 immunoglobulin light chain in only one half of the bispecific molecule provides for a facile way of O separation. This approach is disclosed in WO 94/04690. For further details of generating Z bispecific antibodies see, for example, Suresh et al, Methods in Enzymology, 121:210 (1986).

Cl According to another approach described in W096/27011, the interface between a pair of antibody molecules can be engineered to maximize the percentage of heterodimers which are recovered en from recombinant cell culture. The preferred interface comprises at least a part of the C3 domain of an antibody constant domain. In this method, one or more small amino acid side chains from c)n the interface of the first antibody molecule are replaced with larger side chains tyrosine or tryptophan). Compensatory "cavities" of identical orsimilar size to the large side chain(s) are o 10 created on the interface of the second antibody molecule by replacing large amino acid side chains with smaller ones alanine or threonine). This provides a mechanism for increasing the yield of the heterodimer over other unwanted end-products such as homodimers.

Bispecific antibodies include cross-linked or "heteroconjugate" antibodies. For example, one of the antibodies in the heteroconjugate can be coupled to avidin, the other to biotin. Such antibodies have, for example, been proposed to target immune system cells to unwanted cells (US Patent No. 4,676,980), and for treatment of HIV infection (WO 91/00360, WO 92/200373, and EP 03089). Heteroconjugate antibodies may be made using any convenient cross-linking methods. Suitable cross-linking agents are well known in the art, and are disclosed in US Patent No. 4,676,980, along with a number of cross-linking techniques.

Antibodies with more than two valencies are contemplated. For example, trispecific antibodies can be prepared. Tutt et al. J. Immunol. 147: 60 (1991).

While the polypeptide of interest herein is preferably an antibody, other Fc regioncontaining polypeptides which can be modified according to the methods described herein are contemplated. An example of such a molecule is an immunoadhesin.

C. Immunoadhesin Preparation The simplest and most straightforward immunoadhesin design combines the binding domain(s) of the adhesin the extracellular domain (ECD) of a receptor) with the Fc region of an immunoglobulin heavy chain. Ordinarily, when preparing the immunoadhesins of the present invention, nucleic acid encoding the binding domain of the adhesin will be fused Cterminally to nucleic acid encoding the N-terminus of an immunoglobulin constant domain sequence, however N-terminal fusions are also possible.

Typically, in such fusions the encoded chimeric polypeptide will retain at least functionally active hinge, CH 2 and CH3 domains of the constant region of an immunoglobulin heavy chain.

Fusions are also made to the C-terminus of the Fc portion of a constant domain, or immediately N-terminal to the CH1 of the heavy chain or the corresponding region of the light chain. The _CI_ 4_ -I

O

O

precise site at which the fusion is made is not critical; particular sites are well known and may be O selected in order to optimize the biological activity, secretion, or binding characteristics of the Z immunoadhesin.

C In a preferred embodiment, the adhesin sequence is fused to the N-terminus of the Fc region of immunoglobulin G, It is possible to fuse the entire heavy chain constant region t to the adhesin sequence. However, more preferably, a sequence beginning in the hinge region just upstream of the papain cleavage site which defines IgG Fc chemically residue 216, t' taking the first residue of heavy chain constant region to be 114), or analogous sites of other immunoglobulins is used in the fusion. In a particularly preferred embodiment, the adhesin amino acid sequence is fused to the hinge region and CH 2 and C.3 or the CH, hinge, C, 2 and

CH

3 domains, of an IgG heavy chain.

For bispecific immunoadhesins, the immunoadhesins are assembled as multimers, and particularly as heterodimers or heterotetramers. Generally, these assembled immunoglobulins will have known unit structures. A basic four chain structural unit is the form in which IgG, IgD, and IgE exist. A four chain unit is repeated in the higher molecular weight immunoglobulins; Ig M generally exists as a pentamer of four basic units held together by disulfide bonds, IgA globulin, and occasionally IgG globulin, may also exist in multimeric form in serum. In the case of multimer, each of the four units may be the same or different.

Various exemplary assembled immunoadhesins within the scope herein are schematically diagrammed below:

ACL-ACL;

ACH-(ACH, ACL-ACH, ACL-VHCH, or VLCL-ACH); ACL-ACH-(ACL-ACH, ACL-VHCH, VLCL-ACH, or VLCL-VHCH) ACL-VHCH-(ACH, or ACL-VHCH, or VLCL-ACH); VLCL-ACH-(ACL-VHCH, or VLCL-ACH); and (A-Y)-(VLC-VHCH)2, wherein each A represents identical or different adhesin amino acid sequences; VL is an immunoglobulin light chain variable domain; VH is an immunoglobulin heavy chain variable domain; CL is an immunoglobulin light chain constant domain; CH is an immunoglobulin heavy chain constant domain; n is an integer greater than 1; Y designates the residue of a covalent cross-linking agent.

In the interests of brevity, the foregoing structures only show key features; they do not indicate joining or other domains of the immunoglobulins, nor are disulfide bonds shown.

~I ~11I1 -Il- C--\rr i C i

O

O

Cl However, where such domains are required for binding activity, they shall be constructed to be O present in the ordinary locations which they occupy in the immunoglobulin molecules.

iZ Alternatively, the adhesin sequences can be inserted between immunoglobulin heavy CNl chain and light chain sequences, such that an immunoglobulin comprising a chimeric heavy chain is obtained. In this embodiment, the adhesin sequences are fused to the 3' end of an en immunoglobulin heavy chain in each arm of an immunoglobulin, either between the hinge and the

COH

2 domain, or between the C 2 and C3 domains. Similar constructs have been reported by n Hoogenboom, et Mol. Immunol. 28:1027-1037 (1991).

S1Although the presence of an immunoglobulin light chain is not required in the S 10 immunoadhesins of the present invention, an immunoglobulin light chain might be present either covalently associated to an adhesin-immunoglobulin heavy chain fusion polypeptide, or directly fused to the adhesin. In the former case, DNA encoding an immunoglobulin light chain is typically coexpressed with the DNA encoding the adhesin-immunoglobulin heavy chain fusion protein.

Upon secretion, the hybrid heavy chain and the light chain will be covalently associated to provide an immunoglobulin-like structure comprising two disulfide-linked immunoglobulin heavy chain-light chain pairs. Methods suitable for the preparation of such structures are, for example, disclosed in U.S. Patent No. 4,816,567, issued 28 March 1989.

Immunoadhesins are most conveniently constructed by fusing the cDNA sequence encoding the adhesin portion in-frame to an immunoglobulin cDNA sequence. However, fusion to genomic immunoglobulin fragments can also be used (see, e.g. Aruffo et Cell61:1303- 1313 (1990); and Stamenkovic et al., Cell 66:1133-1144 (1991)). The latter type of fusion requires the presence of Ig regulatory sequences for expression. cDNAs encoding IgG heavychain constant regions can be isolated based on published sequences from cDNA libraries derived from spleen or peripheral blood lymphocytes, by hybridization or by polymerase chain reaction (PCR) techniques. The cDNAs encoding the "adhesin" and the immunoglobulin parts of the immunoadhesin are inserted in tandem into a plasmid vector that directs efficient expression in the chosen host cells.

D. Vectors, Host Cells and Recombinant Methods The invention also provides isolated nucleic acid encoding a polypeptide variant as disclosed herein, vectors and host cells comprising the nucleic acid, and recombinant techniques for the production of the polypeptide variant.

For recombinant production of the polypeptide variant, the nucleic acid encoding it is isolated and inserted into a replicable vector for further cloning (amplification of the DNA) or for expression. DNA encoding the polypeptide variant is readily isolated and sequenced using conventional procedures by using oligonucleotide probes that are capable of binding C specifically to genes encoding the polypeptide variant). Many vectors are available. The vector O components generally include, but are not limited to, one or more of the following: a signal S sequence, an origin of replication, one or more marker genes, an enhancer element, a promoter, and a transcription termination sequence.

Signal sequence component tE The polypeptide variant of this invention may be produced recombinantly not only directly, but also as a fusion polypeptide with a heterologous polypeptide, which is preferably a signal en sequence or other polypeptide having a specific cleavage site at the N-terminus of the mature protein or polypeptide. The heterologous signal sequence selected preferably is one that is 0 10 recognized and processed cleaved by a signal peptidase) by the host cell. For prokaryotic host cells that do not recognize and process the native polypeptide variant signal sequence, the signal sequence is substituted by a prokaryotic signal sequence selected, for example, from the group of the alkaline phosphatase, penicillinase, Ipp, or heat-stable enterotoxin II leaders. For yeast secretion the native signal sequence may be substituted by, the yeast invertase leader, a factor leader (including Saccharomyces and Kluyveromyces a-factor leaders), or acid phosphatase leader, the C. alblcans glucoamylase leader, or the signal described in WO 90/13646. In mammalian cell expression, mammalian signal sequences as well as viral secretory leaders, for example, the herpes simplex gD signal, are available.

The DNA for such precursor region is ligated in reading frame to DNA encoding the polypeptide variant.

(li) Origin of replication component Both expression and cloning vectors contain a nucleic acid sequence that enables the vector to replicate in one or more selected host cells. Generally, in cloning vectors this sequence is one that enables the vector to replicate independently of the host chromosomal DNA, and includes origins of replication or autonomously replicating sequences. Such sequences are well known for a variety of bacteria, yeast, and viruses. The origin of replication from the plasmid pBR322 is suitable for most Gram-negative bacteria, the 2p plasmid origin is suitable for yeast, and various viral origins (SV40, polyoma, adenovirus, VSV or BPV) are useful for cloning vectors in mammalian cells. Generally, the origin of replication component is not needed for mammalian expression vectors (the SV40 origin may typically be used only because it contains the early promoter).

(Ill) Selection gene component Expression and cloning vectors may contain a selection gene, also termed a selectable marker. Typical selection genes encode proteins that confer resistance to antibiotics or other toxins, ampicillin, neomycin, methotrexate, or tetracycline, complement auxotrophic

O

O

C deficiencies, or supply critical nutrients not available from complex media, the gene O encoding D-alanine racemase for Bacilli.

Z One example of a selection scheme utilizes a drug to arrest growth of a host cell. Those S cells that are successfully transformed with a heterologous gene produce a protein conferring drug resistance and thus survive the selection regimen. Examples of such dominant selection t' use the drugs neomycin, mycophenolic acid and hygromycin.

Another example of suitable selectable markers for mammalian cells are those that enable n the identification of cells competent to take up the polypeptide variant nucleic acid, such as DHFR, thymidine kinase, metallothionein-I and -II, preferably primate metallothionein genes, adenosine o 10 deaminase, ornithine decarboxylase, etc.

For example, cells transformed with the DHFR selection gene are first identified by culturing all of the transformants in a culture medium that contains methotrexate (Mtx), a competitive antagonist of DHFR. An appropriate host cell when wild-type DHFR is employed is the Chinese hamster ovary (CHO) cell line deficient in DHFR activity.

Alternatively, host cells (particularly wild-type hosts that contain endogenous DHFR) transformed or co-transformed with DNA sequences encoding polypeptide variant, wild-type DHFR protein, and another selectable marker such as aminoglycoside 3'-phosphotransferase (APH) can be selected by cell growth in medium containing a selection agent for the selectable marker such as an aminoglycosidic antibiotic, kanamycin, neomycin, or G418. See U.S.

Patent No. 4,965,199.

A suitable selection gene for use in yeast is the trpl gene present in the yeast plasmid YRp7 (Stinchcomb et al., Nature, 282:39 (1979)). The trpl gene provides a selection marker for a mutant strain of yeast lacking the ability to grow in tryptophan, for example, ATCC No. 44076 or PEP4-1. Jones, Genetics, 85:12 (1977). The presence of the trpl lesion in the yeast host cell genome then provides an effective environment for detecting transformation by growth in the absence of tryptophan. Similarly, Leu2-deficient yeast strains (ATCC 20,622 or 38,626) are complemented by known plasmids bearing the Leu2 gene.

In addition, vectors derived from the 1.6 pm circular plasmid pKD1 can be used for transformation of Kluyveromyces yeasts. Alternatively, an expression system for large-scale production of recombinant calf chymosin was reported for K. lactis. Van den Berg, Bio/Technology, 8:135 (1990). Stable multi-copy expression vectors for secretion of mature recombinant human serum albumin by industrial strains of Kluyveromyces have also been disclosed. Fleer et al, Blo/Technology, 9:968-975 (1991).

r ILI_ C I_ t C~I l)l*hl l l i 0 (iv) Promoter component 0 Expression and cloning vectors usually contain a promoter that is recognized by the host organism and is operably linked to the polypeptide variant nucleic acid. Promoters suitable for use C with prokaryotic hosts include the phoA promoter, p-lactamase and lactose promoter systems, alkaline phosphatase, a tryptophan (trp) promoter system, and hybrid promoters such as the tac promoter. However, other known bacterial promoters are suitable. Promoters for use in bacterial t systems also will contain a Shine-Dalgamo sequence operably linked to the DNA encoding t' the polypeptide variant.

1 r Promoter sequences are known for eukaryotes. Virtually all eukaryotic genes have an ATrich region located approximately 25 to 30 bases upstream from the site where transcription is initiated. Another sequence found 70 to 80 bases upstream from the start of transcription of many genes is a CNCAAT region where N may be any nucleotide. At the 3' end of most eukaryotic genes is an AATAAA sequence that may be the signal for addition of the poly A tail to the 3' end of the coding sequence. All of these sequences are suitably inserted into eukaryotic expression vectors.

Examples of suitable promoting sequences for use with yeast hosts include the promoters for 3-phosphoglycerate kinase or other glycolytic enzymes, such as enolase, glyceraldehyde-3phosphate dehydrogenase, hexokinase, pyruvate decarboxylase, phosphofructokinase, glucose- 6-phosphate isomerase, 3-phosphoglycerate mutase, pyruvate kinase, triosephosphate isomerase, phosphoglucose isomerase, and glucokinase.

Other yeast promoters, which are inducible promoters having the additional advantage of transcription controlled by growth conditions, are the promoter regions for alcohol dehydrogenase 2, isocytochrome C, acid phosphatase, degradative enzymes associated with nitrogen metabolism, metallothionein, glyceraldehyde-3-phosphate dehydrogenase, and enzymes -responsible for maltose and galactose utilization. Suitable vectors and promoters for use in yeast expression are further described in EP 73,657. Yeast enhancers also are advantageously used with yeast promoters.

Polypeptide variant transcription from vectors in mammalian host cells is controlled, for example, by promoters obtained from the genomes of viruses such as polyoma virus, fowlpox virus, adenovirus (such as Adenovirus bovine papilloma virus, avian sarcoma virus, cytomegalovirus, a retrovirus, hepatitis-B virus and most preferably Simian Virus 40 (SV40), from heterologous mammalian promoters, the actin promoter or an immunoglobulin promoter, from heat-shock promoters, provided such promoters are compatible with the host cell systems.

The early and late promoters of the SV40 virus are conveniently obtained as an restriction fragment that also contains the SV40 viral origin of replication. The immediate early -48rlc~ r- C U I'

O

O

promoter of the human cytomegalovirus is conveniently obtained as a Hindll E restriction o fragment. A system for expressing DNA in mammalian hosts using the bovine papilloma virus as Z a vector is disclosed in U.S. Patent No. 4,419,446. A modification of this system is described in C U.S. Patent No. 4,601,978. See also Reyes et al., Nature 297:598-601 (1982)on expression of human 1-interferon cDNA in mouse cells under the control of a thymidine kinase promoter from Sherpes simplex virus. Alternatively, the rous sarcoma virus long terminal repeat can be used as en the promoter.

e Enhancer element component 1 Transcription of a DNA encoding the polypeptide variant of this invention by higher eukaryotes is often increased by inserting an enhancer sequence into the vector. Many enhancer sequences are now known from mammalian genes (globin, elastase, albumin, a-fetoprotein, and insulin). Typically, however, one will use an enhancer from a eukaryotic cell virus. Examples include the SV40 enhancer on the late side of the replication origin (bp 100-270), the cytomegalovirus early promoter enhancer, the polyoma enhancer on the late side of the replication origin, and adenovirus enhancers. See also Yaniv, Nature 297:17-18 (1982) on enhancing elements for activation of eukaryotic promoters. The enhancer may be spliced into the vector at a position 5' or 3' to the polypeptide variant-encoding sequence, but is preferably located at a site 5' from the promoter.

(vl) Transcription termination component Expression vectors used in eukaryotic host cells (yeast, fungi, insect, plant, animal, human, or nucleated cells from other multicellular organisms) will also contain sequences necessary for the termination of transcription and for stabilizing the mRNA. Such sequences are commonly available from the 5' and, occasionally untranslated regions of eukaryotic or viral DNAs or cDNAs. These regions contain nucleotide segments transcribed as polyadenylated fragments in the untranslated portion of the mRNA encoding the polypeptide variant. One useful transcription termination component is the bovine growth hormone polyadenylation region. See W094/11026 and the expression vector disclosed therein.

(vii) Selection and transformation of host cells Suitable host cells for cloning or expressing the DNA in the vectors herein are the prokaryote, yeast, or higher eukaryote cells described above. Suitable prokaryotes for this purpose.include eubacteria, such as Gram-negative or Gram-positive organisms, for example, Enterobacteriaceae such as Escherichla, E. coll, Enterobacter, Erwinla, Klebslella, Proteus, Salmonella, Salmonella typhimurlum, Serratia, Serratla marcescans, and Shigella, as well as Bacilli such as B. subtlls and B. Iicheniformis B. lichenlforms 41P disclosed in DD 266,710 published 12 April 1989), Pseudomonas such as P. aeruginosa, and -49-

O

O

Streptomyces. One preferred E. coil cloning host is E. coil 294 (ATCC 31,446), although other O strains such as E. coll B, E. coll X1776 (ATCC 31,537), and E. coll W3110 (ATCC 27,325) are Z suitable. These examples are illustrative rather than limiting.

C-l In addition to prokaryotes, eukaryotic microbes such as filamentous fungi or yeast are suitable cloning or expression hosts for polypeptide variant-encoding vectors. Saccharomyces e cerevislae, or common baker's yeast, is the most commonly used among lower eukaryotic host microorganisms. However, a number of other genera, species, and strains are commonly en available and useful herein, such as Schizosaccharomyces pombe; Kluyveromyces hosts such as, K. lactis, K. fragilis (ATCC 12,424), K. bulgaricus (ATCC 16,045), K. wickeramil o 10 (ATCC 24,178), K. waltii (ATCC 56,500), K. drosophilarum (ATCC 36,906), K thermotolerans, and K. marxlanus; yarrowla (EP 402,226); Pichla pastoris (EP 183,070); Candida; Trichoderma reesia (EP 244,234); Neurospora crassa; Schwanniomyces such as Schwanniomyces occidentalis; and filamentous fungi such as, Neurospora, Penicillium, Tolypocladium, and Aspergillus hosts such as A. nidulans and A. niger.

Suitable host cells for the expression of glycosylated polypeptide variant are derived from multicellular organisms. Examples of invertebrate cells include plant and insect cells. Numerous baculoviral strains and variants and corresponding permissive insect host cells from hosts such as Spodoptera frugiperda (caterpillar), Aedes aegypti (mosquito), Aedes albopictus (mosquito), Drosophila melanogaster (fruiffly), and Bombyx mori have been identified. A variety of viral strains for transfection are publicly available, the L-1 variant of Autographa callfornica NPV and the Bm-5 strain of Bombyx mori NPV, and such viruses may be used as the virus herein according to the present invention, particularly for transfection of Spodoptera fruglperda cells.

Plant cell cultures of cotton, corn, potato, soybean, petunia, tomato, and tobacco can also be utilized as hosts.

However, interest has been greatest in vertebrate cells, and propagation of vertebrate cells in culture (tissue culture) has become a routine procedure. Examples of useful mammalian host cell lines are monkey kidney CV1 line transformed by SV40 (COS-7, ATCC CRL 1651); human embryonic kidney line (293 or 293 cells subcloned for growth in suspension culture, Graham et J. Gen VIrol. 36:59 (1977)); baby hamster kidney cells (BHK, ATCC CCL Chinese hamster ovary cells/-DHFR (CHO, Urlaub et al, Proc. Natl Acad. Sci. USA 77:4216 (1980)); mouse sertoli cells (TM4, Mather, Bol. Reprod. 23:243-251 (1980)); monkey kidney cells (CV1 ATCC CCL 70); African green monkey kidney cells (VERO-76, ATCC CRL-1587); human cervical carcinoma cells (HELA, ATCC CCL canine kidney cells (MDCK, ATCC CCL 34); buffalo rat liver cells (BRL 3A, ATCC CRL 1442); human lung cells (W138, ATCC CCL 75); human

O

liver cells (Hep G2, HB 8065); mouse mammary tumor (MMT 060562, ATCC CCL51); TRI cells O (Mather et aL, Annals N.Y. Acad. Sc. 383:44-68 (1982)); MRC 5 cells; FS4 cells; and a human Z hepatoma line (Hep G2).

cN Host cells are transformed with the above-described expression or cloning vectors for polypeptide variant production and cultured in conventional nutrient media modified as appropriate for inducing promoters, selecting transformants, or amplifying the genes encoding the desired sequences.

S(viii) Culturing the host cells 8 The host cells used to produce the polypeptide variant of this invention may be cultured 0 10 in a variety of media. Commercially available media such as Ham's F10 (Sigma), Minimal Essential Medium (Sigma), RPMI-1640 (Sigma), and Dulbecco's Modified Eagle's Medium ((DMEM), Sigma) are suitable for culturing the host cells. In addition, any of the media described in Ham et aL, Meth. Enz. 58:44 (1979), Bames etal., Anal. Blochem.102:255 (1980), U.S. Pat. Nos. 4,767,704; 4,657,866; 4,927,762; 4,560,655; or 5,122,469; WO 90/03430; WO 87/00195; or U.S. Patent Re. 30,985 may be used as culture media for the host cells. Any of these media may be supplemented as necessary with hormones and/or other growth factors (such as insulin, transferrin, or epidermal growth factor), salts (such as sodium chloride, calcium, magnesium, and phosphate), buffers (such as HEPES), nucleotides (such as adenosine and thymidine), antibiotics (such as GENTAMYCIN T M drug), trace elements (defined as inorganic compounds usually present at final concentrations in the micromolar range), and glucose or an equivalent energy source. Any other necessary supplements may also be included at appropriate concentrations that would be known to those skilled in the art. The culture conditions, such as temperature, pH, and the like, are those pieviously used with the host cell selected for expression, and will be apparent to the ordinarily skilled artisan.

(ix) Polypeptide variant purification When using recombinant techniques, the polypeptide variant can be produced intracellularly, in the periplasmic space, or directly secreted into the medium. If the polypeptide variant is produced intracellularly, as a first step, the particulate debris, either host cells or lysed fragments, is removed, for example, by centrifugation or ultrafiltration. Carter et aL, Blo/Technology 10:163-167 (1992) describe a procedure for isolating antibodies which are secreted to the periplasmic space of E. coli. Briefly, cell paste is thawed in the presence of sodium acetate (pH EDTA, and phenylmethylsulfonylfluoride (PMSF) over about 30 min. Cell debris can be removed by centrifugation. Where the polypeptide variant is secreted into the medium, supernatants from such expression systems are generally first concentrated using a commercially available protein concentration filter, for example, an Amicon or Millipore Pellicon ultrafiltration unit. A protease inhibitor such as PMSF may be included in any of the foregoing 0 steps to inhibit proteolysis and antibiotics may be included to prevent the growth of adventitious contaminants.

The polypeptide variant composition prepared from the cells can be purified using, for example, hydroxylapatite chromatography, gel electrophoresis, dialysis, and affinity chromatography, with affinity chromatography being the preferred purification technique. The n suitability of protein A as an affinity ligand depends on the species and isotype of any C- immunoglobulin Fc region that is present in the polypeptide variant. Protein A can be used to 0 purify polypeptide variants that are based on human yl, y2, or y4 heavy chains (Lindmark et a., C- 10 J. Immunol. Meth. 62:1-13 (1983)). Protein G is recommended for all mouse isotypes and for human y3 (Guss etal., EMBO J. 5:15671575 (1986)). The matrix to which the affinity ligand is attached is most often agarose, but other matrices are available. Mechanically stable matrices such as controlled pore glass or poly(styrenedivinyl)benzene allow for faster flow rates and shorter processing times than can be achieved with agarose. Where the polypeptide variant comprises a CH3 domain, the Bakerbond ABX

T

l resin T. Baker, Phillipsburg, NJ) is useful for purification. Other techniques for protein purification such as fractionation on an ion-exchange column, ethanol precipitation, Reverse Phase HPLC, chromatography on silica, chromatography on heparin SEPHAROSET chromatography on an anion or cation exchange resin (such as a polyaspartic acid column), chromatofocusing, SDS-PAGE, and ammonium sulfate precipitation are also available depending on the polypeptide variant to be recovered.

Following any preliminary purification step(s), the mixture comprising the polypeptide variant of interest and contaminants may be subjected to low pH hydrophobic interaction chromatography using an elution buffer at a pH between about 2.5-4.5, preferably performed at low salt concentrations (e.g.,from about 0-0.25M salt).

E. Pharmaceutical Formulations Therapeutic formulations of the polypeptide variant are prepared for storage by mixing the polypeptide variant having the desired degree of purity with optional physiologically acceptable carriers, excipients or stabilizers (Remington's Pharmaceutical Sciences 16th edition, Osol, A.

Ed. (1980)), in the form of lyophilized formulations or aqueous solutions. Acceptable carriers, excipients, or stabilizers are nontoxic to recipients at the dosages and concentrations employed, and include buffers such as phosphate, citrate, and other organic acids; antioxidants including ascorbic acid and methionine; preservatives (such as octadecyldimethylbenzyl ammonium chloride; hexamethonium chloride; benzalkonium chloride, benzethonium chloride; phenol, butyl or benzyl alcohol; alkyl parabens such as methyl or propyl paraben; catechol; resorcinol; cyclohexanol; 3-pentanol; and m-cresol); low molecular weight (less than about 10 residues) -52-

O

O

polypeptide; proteins, such as serum albumin, gelatin, or immunoglobulins; hydrophilic polymers O such as polyvinylpyrrolidone; amino acids such as glycine, glutamine, asparagine, histidine, Z arginine, or lysine; monosaccharides, disaccharides, and other carbohydrates including glucose, CN mannose, or dextrins; chelating agents such as EDTA; sugars such as sucrose, mannitol, trehalose or sorbitol; salt-forming counter-ions such as sodium; metal complexes Zn-protein c complexes); and/or non-ionic surfactants such as TWEEN T M PLURONICST or polyethylene en glycol (PEG).

c The formulation herein may also contain more than one active compound as necessary for the particular indication being treated, preferably those with complementary activities that do not adversely affect each other. Such molecules are suitably present in combination in amounts that are effective for the purpose intended.

The active ingredients may also be entrapped in microcapsule prepared, for example, by coacervation techniques or by interfacial polymerization, for example, hydroxymethylcellulose or gelatin-microcapsule and poly-(methylmethacylate) microcapsule, respectively, in colloidal drug delivery systems (for example, liposomes, albumin microspheres, microemulsions, nano-particles and nanocapsules) or in macroemulsions. Such techniques are disclosed in Remington's Pharmaceutical Sciences 16th edition, Osol, A. Ed. (1980).

The formulations to be used for in vivo administration must be sterile. This is readily accomplished by filtration through sterile filtration membranes.

Sustained-release preparations may be prepared. Suitable examples of sustainedrelease preparations include semipermeable matrices of solid hydrophobic polymers containing the polypeptide variant, which matrices are in the form of shaped articles, films, or microcapsule. Examples of sustained-release matrices include polyesters, hydrogels (for example, poly(2-hydroxyethyl-methacrylate), or poly(vinylalcohol)), polylactides Pat. No.

3,773,919), copolymers of L-glutamic acid and y ethyl-L-glutamate, non-degradable ethylene-vinyl acetate, degradable lactic acid-glycolic acid copolymers such as the LUPRON DEPOTm (injectable microspheres composed of lactic acid-glycolic acid copolymer and leuprolide acetate), and poly-D-(-)-3-hydroxybutyric acid. While polymers such as ethylene-vinyl acetate and lactic acid-glycolic acid enable release of molecules for over 100 days, certain hydrogels release proteins for shorter time periods. When encapsulated antibodies remain in the body for a long time, they may denature or aggregate as a result of exposure to moisture at 37 0 C, resulting in a loss of biological activity and possible changes in immunogenicity. Rational strategies can be devised for stabilization depending on the mechanism involved. For example, if the aggregation mechanism is discovered to be intermolecular S-S bond formation through thio-disulfide interchange, stabilization may be achieved by modifying sulfhydryl residues; lyophilizing from -53- 0 0 acidic solutions, controlling moisture content, using appropriate additives, and developing specific polymer matrix compositions.

F. Non-Therapeutic Uses for the Polypeptide Variant The polypeptide variant of the invention may be used as an affinity purification agent. In this process, the polypeptide variant is immobilized on a solid phase such a Sephadex resin or ec filter paper, using methods well known in the art. The immobilized polypeptide variant is c contacted with a sample containing the antigen to be purified, and thereafter the support is S washed with a suitable solvent that will remove substantially all the material in the sample except the antigen to be purified, which is bound to the immobilized polypeptide variant. Finally, the support is washed with another suitable solvent, such as glycine buffer, pH 5.0, that will release the antigen from the polypeptide variant.

The polypeptide variant may also be useful in diagnostic assays, for detecting expression of an antigen of interest in specific cells, tissues, or serum.

For diagnostic applications, the polypeptide variant typically will be labeled with a detectable moiety. Numerous labels are available which can be generally grouped into the following categories: Radioisotopes, such as 5S, 14C, 121, 3 H, and 131. The polypeptide variant can be labeled with the radioisotope using the techniques described in Current Protocols in Immunology, Volumes 1 and 2, Coligen et al., Ed. Wiley-lnterscience, New York, New York, Pubs. (1991) for example and radioactivity can be measured using scintillation counting.

Fluorescent labels such as rare earth chelates (europium chelates) or fluorescein and its derivatives, rhodamine and its derivatives, dansyl, Lissamine, phycoerythrin and Texas Red are available. The fluorescent labels can be conjugated to the polypeptide variant using the techniques disclosed in Current Protocols in Immunology, supra, for example. Fluorescence can be quantified using a fluorimeter.

Various enzyme-substrate labels are available and U.S. Patent No. 4,275,149 provides a review of some of these. The enzyme generally catalyzes a chemical alteration of the chromogenic substrate that can be measured using various techniques. For example, the enzyme may catalyze a color change in a substrate, which can be measured spectrophotometrically.

Alternatively, the enzyme may alter the fluorescence or chemiluminescence of the substrate.

Techniques for quantifying a change in fluorescence are described above. The chemiluminescent substrate becomes electronically excited by a chemical reaction and may then emit light which can be measured (using a chemiluminometer, for example) or donates energy to a fluorescent acceptor. Examples of enzymatic labels include luciferases firefly luciferase and bacterial luciferase; U.S. Patent No. 4,737,456), luciferin, 2,3-dihydrophthalazinediones, malate dehydrogenase, urease, peroxidase such as horseradish peroxidase (HRPO), alkaline phosphatase, P-galactosidase, glucoamylase, lysozyme, saccharide oxidases glucose oxidase, galactose oxidase, and glucose-6-phosphate dehydrogenase), heterocyclic oxidases (such as uricase and xanthine oxidase), lactoperoxidase, microperoxidase, and the like.

Techniques for conjugating enzymes to antibodies are described in O'Sullivan et al., Methods for the Preparation of Enzyme-Antibody Conjugates for use in Enzyme Immunoassay, in Methods In Enzym. (ed J. Langone H. Van Vunakis), Academic press, New York, 73:147-166 (1981).

Examples of enzyme-substrate combinations include, for example: Horseradish peroxidase (HRPO) with hydrogen peroxidase as a substrate, wherein the hydrogen peroxidase oxidizes a dye precursor (e.g.,orthophenylene diamine (OPD) or 3,3',5,5'-tetramethyl benzidine hydrochloride (TMB)); (ii) alkaline phosphatase (AP) with para-Nitrophenyl phosphate as chromogenic substrate; and (iii) p-D-galactosidase (p-D-Gal) with a chromogenic substrate p-nitrophenyl-3- D-galactosidase) or fluorogenic substrate 4-methylumbelliferyl-p-D-galactosidase.

Numerous other enzyme-substrate combinations are available to those skilled in the art.

For a general review of these, see U.S. Patent Nos. 4,275,149 and 4,318,980.

Sometimes, the label is indirectly conjugated with the polypeptide variant. The skilled artisan will be aware of various techniques for achieving this. For example, the polypeptide variant can be conjugated with biotin and any of the three broad categories of labels mentioned above can be conjugated with avidin, or vice versa. Biotin binds selectively to avidin and thus, the label can be conjugated with the polypeptide variant in this indirect manner. Alternatively, to achieve indirect conjugation of the label with the polypeptide variant, the polypeptide variant is conjugated with a small hapten digoxin) and one of the different types of labels mentioned above is conjugated with an anti-hapten polypeptide variant anti-digoxin antibody). Thus, indirect conjugation of the label with the polypeptide variant can be achieved.

In another embodiment of the invention, the polypeptide variant need not be labeled, and the presence thereof can be detected using a labeled antibody which binds to the polypeptide variant.

The polypeptide variant of the present invention may be employed in any known assay method, such as competitive binding assays, direct and indirect sandwich assays, and immunoprecipitation assays. Zola, Monoclonal Antibodies: A Manual of Techniques, pp.147- 158 (CRC Press, Inc. 1987).

O

O

The polypeptide variant may also be used forIn vivo diagnostic assays. Generally, the o polypeptide variant is labeled with a radionuclide (such as In, "Tc, 4"C, 1311, 3 H 32 P or 3S) so that the antigen or cells expressing it can be localized using immunoscintiography.

C G. In Vivo Uses for the Polypeptide Variant It is contemplated that the polypeptide variant of the present invention may be used to treat a mammal e.g. a patient suffering from, or predisposed to, a disease or disorder who could benefit from administration of the polypeptide variant. The conditions which can be treated with en the polypeptide variant are many and include cancer where the polypeptide variant binds the HER2 receptor, CD20 or vascular endothelial growth factor (VEGF)); allergic conditions such as asthma (with an anti-IgE antibody); and LFA-1-mediated disorders where the polypeptide variant is an anti-LFA-1 or anti-ICAM-1 antibody) etc.

Where the antibody binds the HER2 receptor, the disorder preferably is HER2-expressing cancer, e.g. a benign or malignant tumor characterized by overexpression of the HER2 receptor.

Such cancers include, but are not limited to, breast cancer, squamous cell cancer, small-cell lung cancer, non-small cell lung cancer, gastrointestinal cancer, pancreatic cancer, glioblastoma, cervical cancer, ovarian cancer, bladder cancer, hepatoma, colon cancer, colorectal cancer, endometrial carcinoma, salivary.gland carcinoma, kidney cancer, liver cancer, prostate cancer, vulval cancer, thyroid cancer, hepatic carcinoma and various types of head and neck cancer.

According to the teachings herein, one may prepare a polypeptide with a variant Fc region which has improved, or diminished, ADCC activity. Such molecules will find applications in the treatment of different disorders.

For example, the polypeptide variant with improved ADCC activity may be employed in the treatment of diseases or disorders where destruction or elimination of tissue or foreign microorganisms is desired. For example, the polypeptide may be used to treat cancer; inflammatory disorders; infections bacterial, viral, fungal or yeast infections); and other conditions (such as goiter) where removal of tissue is desired, etc.

Where the polypeptide variant has diminished ADCC activity, such variants may be used to treat diseases or disorders where a Fc region-containing polypeptide with long half-life is desired, but the polypeptide preferably does not have undesirable effector function(s). For example, the Fc region-containing polypeptide may be an anti-tissue factor (TF) antibody; anti-lgE antibody; and anti-integrin antibody an anti-ca47 antibody). The desired mechanism of action of such Fc region-containing polypeptides may be to block ligand-receptor binding pairs.

Moreover, the Fc-region containing polypeptide with diminished ADCC activity may be an agonist antibody.

-56- The polypeptide variant is administered by any suitable means, including parenteral, 0 subcutaneous, intraperitoneal, intrapulmonary, and intranasal, and, if desired for local immunosuppressive treatment, intralesional administration. Parenteral infusions include Cl intramuscular, intravenous, intraarterial, intraperitoneal, or subcutaneous administration. In addition, the polypeptide variant is suitably administered by pulse infusion, particularly with 0n declining doses of the polypeptide variant. Preferably the dosing is given by injections, most e preferably intravenous or subcutaneous injections, depending in part on whether the en administration is brief or chronic.

0 w For the prevention or treatment of disease, the appropriate dosage of polypeptide variant will depend on the type of disease to be treated, the severity and course of the disease, whether the polypeptide variant is administered for preventive or therapeutic purposes, previous therapy, the patient's clinical history and response to the polypeptide variant, and the discretion of the attending physician. The polypeptide variant is suitably administered to the patient at one time or over a series of treatments.

Depending on the type and severity of the disease, about 1 tpg/kg to 15 mg/kg 0.1of polypeptide variant is an initial candidate dosage for administration to the patient, whether, for example, by one or more separate administrations, or by continuous infusion. A typical daily dosage might range from about 1 p.g/kg to 100 mg/kg or more, depending on the factors mentioned above. For repeated administrations over several days or longer, depending on the condition, the treatment is sustained until a desired suppression of disease symptoms occurs. However, other dosage regimens may be useful. The progress of this therapy is easily monitored by conventional techniques and assays.

The polypeptide variant composition will be formulated, dosed, and administered in a fashion consistent with good medical practice. Factors for consideration in this context include the particular disorder being treated, the particular mammal being treated, the clinical condition of the individual patient, the cause of the disorder, the site of delivery of the agent, the method of administration, the scheduling of administration, and other factors known to medical practitioners.

The "therapeutically effective amount" of the polypeptide variant to be administered will be governed by such considerations, and is the minimum amount necessary to prevent, ameliorate, or treat a disease or disorder. The polypeptide variant need not be, but is optionally formulated with one or more agents currently used to prevent or treat the disorder in question. The effective amount of such other agents depends on the amount of polypeptide variant present in the formulation, the type of disorder or treatment, and other factors discussed above. These are generally used in the same dosages and with administration routes as used hereinbefore or about from 1 to 99% of the heretofore employed dosages.

-57-

O

0 The invention will be more fully understood by reference to the following examples. They o should not, however, be construed as limiting the scope of this invention. All literature and patent citations mentioned herein are expressly incorporated by reference.

C EXAMPLE 1 Low Affinity Receptor Binding Assay This assay determines binding of an IgG Fc region to recombinant FcyRIIA, FcyRIlB and en FcyRIIIA a subunits expressed as His6-glutathione S transferase (GST)-tagged fusion proteins.

Since the affinity of the Fc region of IgG1 for the FcyRI is in the nanomolar range, the binding of SIgG1 Fc variants can be measured by titrating monomeric IgG and measuring bound IgG with a polyclonal anti-lgG in a standard ELISA format (Example 2 below). The affinity of the other members of the FcyR family, Le. FcyRIIA, FcyRIIB and FcyRIIIA for IgG is however in the micromolar range and binding of monomeric IgG1 for these receptors can not be reliably measured in an ELISA format.

The following assay utilizes Fc variants of recombinant anti-lgE E27 (Figures 4A and 4B) which, when mixed with human IgE at a 1:1 molar ratio, forms a stable hexamer consisting of three anti-lgE molecules and three IgE molecules. A recombinant chimeric form of IgE (chimeric IgE) was engineered and consists of a human IgE Fc region and the Fab of an anti-VEGF antibody (Presta et al. Cancer Research 57:4593-4599 (1997)) which binds two VEGF molecules per mole of anti-VEGF. When recombinant human VEGF is added at a 2:1 molar ratio to chimeric IgE:E27 hexamers, the hexamers are linked into larger molecular weight complexes via the chimeric IgE Fab:VEGF interaction. The E27 component of this complex binds to the FcyRIIA, FcyRIIB and FcyRIIIA a subunits with higher avidity to permit detection in an ELISA format.

MATERIALS AND METHODS Receptor Coat Fcy receptor a subunits were expressed as GST fusions of His6 tagged extracellular domains (ECDs) in 293 cells resulting in an ECD-6His-GST fusion protein (Graham et a. J. Gen. Virol. 36:59-74 (1977) and Gorman et al. DNA Prot. Eng. Tech. 2:3-10 (1990)) and purified by Ni-NTA column chromatography (Qiagen, Australia) and buffer exchanged into phosphate buffered saline (PBS). Concentrations were determined by absorption at 280nm using extinction coefficients derived by amino acid composition analysis. Receptors were coated onto Nunc F96 maxisorb plates (cat no. 439454) at 10Ong per well by adding 100 ld of receptor-GST fusion at 1 pg/ml in PBS and incubated for 48 hours at 4°C. Prior to assay, plates are washed 3x with 250 il of wash buffer (PBS pH 7.4 containing 0.5% TWEEN 20

T

M) and blocked with 250 il of assay buffer (50mM Tris buffered saline, 0.05% TWEEN 20 T M 0.5% RIA grade bovine albumin (Sigma A7888), and 2mM EDTA pH 7.4).

-58-

O

O

Immune Complex Formation: Equal molar amounts of E27 and recombinant 0 chimeric IgE which binds two moles recombinant human VEGF per mole of chimeric IgE are added to a 12 x 75mm polypropylene tube in PBS and mixed by rotation for 30 minutes at 25 0

C.

Cl E27 (anti-lgE) Ichimeric IgE (IgE) hexamers are formed during this incubation. Recombinant human VEGF (165 form, MW 44,000) is added at a 2:1 molar ratio to the IgE concentration and 0n mixed by rotation an additional 30 minutes at 25 0 C. VEGF- chimeric IgE binding links cn E27:chimeric IgE hexamers into larger molecular weight complexes which bind FcyR a subunit ECD coated plates via the Fc region of the E27 antibody.

0 E27:chimeric IgE:VEGF: (1:1:2 molar ratio) complexes are added to FcyR a subunit 10 coated plates at E27 concentrations of 5 pg and 1 pg total IgG in quadruplicate in assay buffer and incubated for 120 minutes at 25 0 C on an orbital shaker.

Complex Detection: Plates are washed 5x with wash buffer to remove unbound complexes and IgG binding is detected by adding 100 p horse radish peroxidase (HRP) conjugated goat anti-human IgG heavy chain specific (Boehringer Mannheim 1814249) at 1:10,000 in assay buffer and incubated for 90min at 25*C on an orbital shaker. Plates are washed with wash buffer to remove unbound HRP goat anti-human IgG and bound anti-lgG is detected by adding 100 p1 of substrate solution (0.4mg/ml o-phenylenedaimine dihydrochloride, Sigma P6912, 6 mM H 2 0 2 in PBS) and incubating for 8 min at 25 0 C. Enzymatic reaction is stopped by the addition of 100 pl 4.5N H 2

SO

4 and colorimetric product is measured at 490 nm on a 96 well plate densitometer (Molecular Devices). Binding of E27 variant complexes is expressed as a percent of the wild type E27 containing complex.

EXAMPLE 2 Identification of Unique Clq Binding Sites in a Human IgG Antibody In the present study, mutations were identified in the CH2 domain of a human IgG1 antibody, "C2B8" (Reff et al, Blood 83:435 (1994)), that ablated binding of the antibody to C1q but did not alter the conformation of the antibody nor affect binding to each of the FcyRs. By alanine scanning mutagenesis, five variants in human IgG1 were identified, D270K, D270V, K322A P329A, and P331, that were non-lytic and had decreased binding to Clq. The data suggested that the core Clq binding sites in human IgG1 is different from that of murine IgG2b.

In addition, K322A, P329A and P331A were found to bind normally to the CD20 antigen, and to four Fc receptors, FcyRI, FcyRII, FcyRIll and FcRn.

MATERIALS AND METHODS Construction of C2B8 Variants: The chimeric light and heavy chains of antibody C2B8 (Reff et al., Blood 83:435 (1994)) subcloned separately into previously described PRK vectors (Gorman et al., DNA Protein Eng. Tech. 2:3 (1990)) were used. By site directed -59- I I I II IIIIIII

O

0 mutagenesis (Kunkel et al., Proc. Natl. Acad.ScLUSA 82:488 (1987)), alanine scan variants of o the Fc region in the heavy chain were constructed. The heavy and light chain plasmids were co- Z transfected into an adenovirus transformed human embryonic kidney cell line as previously C described (Werther et al., J. Immunol. 157:4986 (1996)). The media was changed to serum-free 24 hours after transfection and the secreted antibody was harvested after five days. The antibodies were purified using Protein A-SEPHAROSE CL-4BT (Pharmacia), buffer exchanged tn and concentrated to 0.5 ml with PBS using a Centricon-30 (Amicon), and stored at 4 0 C. The ec C concentration of the antibody was determined using total Ig-binding ELISA.

o Clq Binding ELISA: Costar 96 well plates were coated overnight at 4°C with the indicated concentrations of C2B8 in coating buffer (0.05 M sodium carbonate buffer), pH 9. The plates were then washed 3x with PBS/ 0.05% TWEEN 20 T M pH 7.4 and blocked with 200p 1 of ELISA diluent without thimerosal (0.1M NaPO4 0.1 M NaCI 0.1% gelatin 0.05% TWEEN 201/ 0.05% ProClin300) for 1 hr at room temperature. The plate was washed 3x with wash buffer, an aliquot of 1001l of 2 Vig/ml Clq (Quidel, San Diego, CA) was added to each well and incubated for 2 hrs at room temperature. The plate was then washed 6x with wash buffer. 100 ll of a 1:1000 dilution of sheep anti-complement Clq peroxidase conjugated antibody (Biodesign) was added to each well and incubated for 1 hour at room temperature. The plate was again washed 6x with wash buffer and 100 1p of substrate buffer (PBS/0.012% H 2 0 2 containing OPD (O-phenylenediamine dihydrochloride (Sigma)) was added to each well. The oxidation reaction, observed by the appearance of a yellow color, was allowed to proceed for 30 minutes and stopped by the addition of 100 p1 of 4.5 N H 2

SO

4 The absorbance was then read at (492-405) nm using a microplate reader (SPECTRA MAX 250T, Molecular Devices Corp.). The appropriate controls were run in parallel the ELISA was performed without Clq for each concentration of C2B8 used and also the ELISA was performed without C2B8). For each variant, Clq binding was measured by plotting the absorbance (492-405) nm versus concentration of C2B8 in pg/ml using a 4-parameter curve fitting program (KALEIDAGRAPH

T

and comparing EC, values.

Complement Dependent Cytotoxlcity (CDC) Assay. This assay was performed essentially as previously described (Gazzano-Santoro et al., J. Immunol Methods 202:163 (1996)). Various concentrations of C2B8 (0.08-20 jig/ml) were diluted with RHB buffer (RPMI 1640/20mM HEPES (pH 7.2)/2mM Glutamine/0.1% BSA/100 pg/ml Gentamicin). Human complement (Quidel) was diluted 1:3 in RHB buffer and WIL2-S cells (available from the ATCC, Manassas, VA) which express the CD20 antigen were diluted to a density of 1 x 106 cells /ml with RHB buffer. Mixtures of 150 gl containing equal volumes of C2B8, diluted human complement and WIL2-S cells were added to a flat bottom tissue culture 96 well plate and allowed to incubate for 2 hrs at 37°C and 5% CO2 to facilitate complement mediated cell lysis. 50 il of alamar blue

O

O

(Accumed International) was then added to each well and incubated overnight at 37°C. The o absorbance was measured using a 96-well fluorometer with excitation at 530 nm and emission at 590 nm. As described by Gazzano-Santoro et al., the results are expressed in relative C fluorescence units (RFU). The sample concentrations were computed from a C2B8 standard curve and the percent activity as compared to wild type C2B8 is reported for each variant.

Binding Potency of the C2B8 Variants: The binding of C2B8 and variants to the cn CD20 antigen were assessed by a method previously described (Reff et al., (1994), supra; Sreviewed in Gazzano-Santoro et al., (1996), supra). WIL2-S cells were grown for 3-4 days to a cell density of 1 x 106 cells/ml. The cells were washed and spun twice in FACS buffer (PBS/0.1% BSA/0.02% NaN 3 and resuspended to a cell density of 5 x 106 cells/ml. 200 p1 of cells (5 x 106 cells/mi) and 20 pl of diluted C2B8 samples were added to a 5 ml tube and incubated at room temperature for 30 minutes with agitation. The mixture.was then washed with 2 ml of cold FACS buffer, spun down and resuspended in 200 pi of cold FACS buffer. To the suspension, 10 p1 of goat anti-human IgG-FITC (American Qualex Labs.) was added and the mixture was incubated in the dark at room temperature for 30 minutes with agitation. After incubation, the mixture was washed with 2 ml of FACS buffer, spun down and resuspended in 1 ml of cold fixative buffer (1 formaldehyde in PBS). The samples were analyzed by flow cytometry and the results expressed as relative fluorescence units (RFU) were plotted against antibody concentrations using a 4parameter curve fitting program (KALEIDAGRAPHnM). The ECso values are reported as a percentage of that of the C2B8 reference material.

FcTR Binding ELISAs: FcyRI ca subunit-GST fusion was coated onto Nunc F96 maxisorb plates (cat no. 439454) by adding 100 pl of receptor-GST fusion at 1 pg/ml in PBS and incubated for 48 hours at 4 0 C. Prior to assay, plates are washed 3x with 250 il of wash buffer (PBS pH 7.4 containing 0.5% TWEEN 20T) and blocked with 250 1l of assay buffer (50mM Tris buffered saline, 0.05% TWEEN 20T, 0.5% RIA grade bovine albumin (Sigma A7888), and 2mM EDTA pH Samples diluted to 10 g/ml in 1 ml of assay buffer are added to FcyRI a subunit coated plates and incubated for 120 minutes at 25 0 C on an orbital shaker. Plates are washed 5x with wash buffer to remove unbound complexes and IgG binding is detected by adding 100 p1 horse radish peroxidase (HRP) conjugated goat anti-human IgG heavy chain specific (Boehringer Mannheim 1814249) at 1:10,000 in assay buffer and incubated for 90min at 25°C on an orbital shaker. Plates are washed 5x with wash buffer to remove unbound HRP goat anti-human IgG and bound anti-lgG is detected by adding 100 pl of substrate solution (0.4mg/ml ophenylenedaimine dihydrochloride, Sigma P6912, 6 mM H 2 0 2 in PBS) and incubating for 8 min at 25 0 C. Enzymatic reaction is stopped by the addition of 100 pl 4.5N H 2 SO, and colorimetric -61product is measured at 490 nm on a 96 well plate densitometer (Molecular Devices). Binding of o variant is expressed as a percent of the wild type molecule.

FcyRII and II binding ELISAs were performed as described in Example 1 above.

C For measuring FcRn binding activity of IgG variants, ELISA plates were coated with 2 pg/ml streptavidin (Zymed, South San Francisco) in 50 mM carbonate buffer, pH 9.6, at 4°C overnight and blocked with PBS-0.5% BSA, pH 7.2 at room temperature for one hour. Biotinylated tcn FcRn (prepared using biotin-X-NHS from Research Organics, Cleveland, OH and used at 1-2 C pg/ml) in PBS-0.5% BSA, 0.05% polysorbate 20, pH 7.2, was added to the plate and incubated 0 for one hour. Two fold serial dilutions of IgG standard (1.6-100 ng/ml) or variants in C 10 BSA, 0.05% polysorbate 20, pH 6.0, were added to the plate and incubated for two hours. Bound IgG was detected using peroxidase labeled goat F(ab') anti-human IgG F(ab') in the above pH buffer (Jackson ImmunoResearch, West Grove, PA) followed by 3 3 benzidine (Kirgaard Perry Laboratories) as the substrate. Plates were washed between steps with PBS-0.05% polysorbate 20 at either pH 7.2 or 6.0. Absorbance was read at 450 nm on a Vmax plate reader (Molecular Devices, Menlo Park, CA). Titration curves were fit with a fourparameter nonlinear regression curve-fitting program (KaleidaGraph, Synergy software, Reading, PA). Concentrations of IgG variants corresponding to the mid-point absorbance of the titration curve of the standard were calculated and then divided by the concentration of.the standard corresponding to the mid-point absorbance of the standard titration curve.

RESULTS AND DISCUSSION By alanine scanning mutagenesis, several single point mutations were constructed in the CH2 domain of C2B8 beginning with E318A, K320A and K322A. All the variants constructed bound normally to the CD20 antigen (Table 3).

Table 3 wt E318A K320A K322A P329A P331A FcRn FcyRI FcyRII FcyRIII *C1q CDC indicates binding and signifies binding abolished *With respect to Ciq binding, each sign is equivalent to approximately 33% binding.

O

O

Where binding of human complement to an antibody with a human Fc was analyzed, the 0 ability of E318A and K320A to activate complement was essentially identical to that of wild type Z C2B8 (Table When compared to wild type C2B8, there appears to be little difference in the N binding of E318A and K320A to Clq. There is only a 10% decrease in the binding of K320A and about a 30% decrease in the binding of E318A to Clq (Fig. The results indicate that the effect en of the E318A and the K320A substitution on complement activation and Clq binding is minimal.

Also, the human IgG1 of C2B8 was substituted for human lgG2 and used as a negative control tn in the Clq binding studies. The lgG2 variant appears to have a much lower affinity for Clq than the E318A and K320A variants (Fig. Thus, the results demonstrate that E318 and K320 do o 10 not constitute the core Clq binding sites for human IgG1. Conversely, the K322A substitution had a significant effect on both complement activity and Clq binding. The K322A variant did not have CDC activity when tested in the above CDC assay and was more than a 100 fold lower than wild type C2B8 in binding to C1q (Fig. In the human system, K322 is the only residue of the proposed core Clq binding sites that appeared to have a significant effect on complement activation and Clq binding.

Since the Duncan and Winter study was performed using mouse lgG2b and the above results reveal that K320 and E318 in human IgG1 are not involved in Clq binding, and without being bound to any one theory, the above data suggest that the Clq binding region in murine IgGs is different from that of the human. To investigate this further and also to identify additional variants that do not bind to Clq and hence do not activate complement, several more point mutations in the vicinity of K322 were constructed as assessed from the three dimensional structure of the C2B8 Fc. Variants constructed, K274A, N276A, Y278A, S324A, P329A, P331A, K334A, and T335A, were assessed for their ability to bind Clq and also to activate complement.

Many of these substitutions had little or no effect on Clq binding or complement activation. In the above assays, the P329A and the P331A variants did not activate complement and had decreased binding to Clq. The P331A variant did not activate complement and was 60 fold lower in binding to Clq (Fig. 3) when compared to wild type C2B8 (Fig. The concentration range of the antibody variants used in Fig. 3 is expanded to 100pg/ml in order to observe saturation of Clq binding to the P331A variant. The mutation P329A results in an antibody that does not activate complement and is more than a 100 fold lower in binding to Clq (Fig. 3) when compared to wild type C2B8 (Fig. 2).

Variants that did not bind to Clq and hence did not activate complement were examined for their ability to bind to the Fc receptors: FcyRI, FcyRIIA, FcyRIIB, FcyRIIIA and FcRn. This particular study was performed using a humanized anti-lgE antibody, an IgG1 antibody with these mutations (see Example 1 above). The results revealed the variants, K322A and P329A, bind -63-

IC

rr r r_ .xlxl l .I

O

O

to all the Fc receptors to the same extent as the wild type protein (Table However, there was o a slight decrease in the binding of P331A to FcyRIIB.

In conclusion, two amino acid substitutions in the COOH terminal region of the CH2 c domain of human IgG1, K322A and P329A were identified that result in more than 100 fold decrease in Clq binding and do not activate the CDC pathway. These two variants, K322A and tc 0, P329A, bind to all Fc receptors with the same affinity as the wild type antibody. Based on the en results, summarized in Table 4, and without being bound to any one theory, it is proposed that the Cl Clq binding epicenter of human IgG1 is centered around K322, P329 and P331 and is different from the murine IgG2b epicenter which constitutes E318, K320 and K322.

Table 4 wt E318A K320A K322A P329A P331A 100 89 102 86 112 103 "FcyRI 100 93 102 90 104 74 *FcyRIIA 100 113 94 109 111 86 "FcyRIIB 100 106 83 101 96 58 aFcRIllI 100 104 72 90 85 73 CDC 100 108 108 none none none 'For binding to the FcyRs the variants were made in the E27 background (anti-IgE).

The results are presented as a percentage of the wild type.

A further residue involved in binding human Clq was identified using the methods described in the present example. The residue D270 was replaced with lysine and valine to generate variants D270K and D270V, respectively. These variants both showed decreased binding to human C1q (Fig. 6) and were non-lytic (Fig. The two variants bound the antigen normally and recruited ADCC.

EXAMPLE 3 Variants with Improved Clq Binding The following study shows that substitution of residues at positions K326, A327, E333 and K334 resulted in variants with at least about a 30% increase in binding to Clq when compared to the wild type antibody. This indicated K326, A327, E333 and K334 are potential sites for improving the efficacy of antibodies by way of the CDC pathway. The aim of this study was to improve CDC activity of an antibody by increasing binding to Clq. By site directed mutagenesis at K326 and E333, several variants with increased binding to C1q were constructed. The residues in order of increased binding at K326 are K<V<E<A<G<D<M<W, and the residues in order of increased binding at E333 are E<Q<D<V<G<A<S. Four variants, K326M, K326D, K326E -64-

O

O

and E333S were constructed with at least a two-fold increase in binding to Clq when compared o to wild type. Variant K326W displayed about a five-fold increase in binding to Clq.

SVariants of the wild type C2B8 antibody were prepared as described above in Example CN 2. A further control antibody, wild type C2B8 produced in Chinese hamster ovary (CHO) cells essentially as described in US Patent 5,736,137, was included in a Clq binding ELISA to confirm that wt C2B8 produced in the 293 kidney cell line had the same Clq binding activity as the CHOen produced antibody (see "CHO-wt-C2B8" in Fig. The Clq binding ELISA, CDC assay, and c CD20 binding potency assay in this example were performed as described in Example 2 above.

SAs shown in Fig. 8, alanine substitution at K326 and E333 in C2B8 resulted in variants with about a 30% increase in binding to Clq.

Several other single point variants at K326 and E333 were constructed and assessed for their ability to bind Clq and activate complement. All the variants constructed bound normally to the CD20 antigen.

With respect to K326, the other single point variants constructed were K326A, K326D, K326E, K326G, K326V, K326M and K326W. As shown in Fig. 9, these variants all bound to Clq with a better affinity than the wild type antibody. K326W, K326M, K326D and K326E showed at least a two-fold increase in binding to Clq (Table Among the K326 variants, K326W had the best affinity for Clq.

Table Variant ECso value Wild type 1.53.

K326V 1.30 K326A 1.03 K326E 1.08 K326G 0.95 K326D 0.76 K326M 0.67 K326W 0.47 E333S 0.81 E333A 0.98 E333G 1-14 E333V 1.18 E333D 1.22 E333Q 1.52 K334A 1.07

O

ci

O

o Substitutions with hydrophobic as well as charged residues resulted in variants with Z increased binding to Clq. Even substitution with glycine which is known to impart flexibility to a C(l chain and is well conserved in nature, resulted in a variant with higher affinity for Clq when compared to the wild type. It would appear that any amino acid substitution at this site would

C'

Sresult in a variant with higher affinity for Clq. As assessed from the three-dimensional structure, e K326 and E333 are in the vicinity of the core Clq binding sites (Fig. C In addition to alanine, E333 was also substituted with other amino acid residues. These o variants, E333S, E333G, E333V, E333D, and E333Q, all had increased binding to Clq when o 10 compared to the wild type (Fig. 11). As shown in Table 5, the order of binding affinity for Clq was as follows: E333S>E333A>E333G>E333V>E333D>E333Q. Substitutions with amino acid residues with small side chain volumes, i.e. serine, alanine and glycine, resulted in variants with higher affinity for Clq in comparison'to the other variants, E333V, E333D and E333Q, with larger side chain volumes. The variant E333S had the highest affinity for Clq, showing a two-fold increase in binding when compared to the wild type. Without being bound to any one theory, this indicates the effect on Clq binding at 333 may also be due in part to the polarity of the residue.

Double variants were also generated. As shown in Figs. 12 and 13, double variants K326M-E333S and K326A-E333A were at least three-fold better at binding human Clq than wild type C2B8 (Fig. 12) and at least two-fold better at mediating CDC compared to wild type C2B8 (Fig. 13). Additivity indicates these are independently acting variants.

As shown in Fig. 14, a further variant with improved Clq binding (50% increase) was made by changing A327 in a human IgG1 constant region to glycine. Conversely, in a human lgG2 constant region, changing G327 to alanine reduced Clq binding of the lgG2 antibody.

EXAMPLE 4 Identification of FcR Binding Sites in Human IgG Antibodies In the present study, the effect of mutating various Fc region residues of an IgG1 antibody with respect to binding FcyRI, FcyRIIA, FcyRB and FcyRIIIIA as well as FcRn was evaluated.

Antibody variants with improved as well as diminished FcR binding were identified.

MATERIALS AND METHODS Construction of IgG1 Variants: Recombinant anti-lgE E27 having the light chain and heavy chain sequences in Figs. 4A and 4B, respectively, was used as the parent antibody in the following experiments. This antibody binds the antigen IgE and has a non-A allotype IgG1 Fc region. By site directed mutagenesis (Kunkel etal., Proc. Natl. Acad.Sc.USA 82:488 (1987)), variants of the Fc region in the heavy chain of the above parent antibody were constructed. The heavy and light chain plasmids were co-transfected into an adenovirus transformed human S"embryonic kidney cell line as previously described (Werther et J. Immunol. 157:4986 (1996)).

0 The media was changed to serum-free 24 hours after transfection and the secreted antibody was S harvested after five days. The antibodies were purified by Protein G SEPHAROSE® (Pharmacia), l buffer exchanged and concentrated to 0.5 ml with PBS using a Centricon-30 (Amicon), and stored at 4°C. Concentration was determined by adsorption at 280 nm using extinction coefficients Sderived by amino acid composition analysis.

Mn High Affinity FcRIA Binding ELISA: FcyRIA was expressed as a GST fusion of His6 Cl tagged extracellular domain in 293 cells and purified by Ni-NTA column chromatography.

O To purify FcyRIA, supematant from transfected 293 cells was removed after three days.

o 10 Protease inhibitors were added; 50 pL Aprotinin (Sigma)/50 mL supematant, and PMSF (1mM).

Supematants were concentrated to 10 mL in a stirred cell (Amicon), and dialyzed overnight at 4°C against 1 liter column buffer (50 mM Tris pH 8.0, 20mM Imidazole, 300 mM NaCI). Additional dialysis was done the following morning against fresh column buffer for 4 hours at 4 0 C. The solution was loaded on to a lmL Ni" column (NTA super flow resin, Qiagen) previously equilibrated with 10 mL column buffer. Columns were washed with 10 mL column buffer, and protein was eluted with 2.5 mL elution buffer (50mM Tris pH 8.0, 250mM Imidazole, 300mM NaCI). Protein was concentrated to 0.5 mL and buffer exchanged into PBS. Concentrations were determined by adsorption at 280 nm using an extinction coefficient derived by amino acid composition analysis.

Purified receptors were coated onto Nunc F96 maxisorb plates (cat no. 439545) at approximately 150 ng per well by adding 100 pL of receptor at 1.5 ug/mL in PBS and incubated for 24 hours at 40C- Prior to assay, plates were washed 3x with 250 pL of wash buffer (phosphate buffered saline pH 7.4 containing 0.5% TWEEN 20®) and blocked with 250 pL of assay buffer mM tris buffered saline, 0.05% TWEEN 20®, 0.5% RIA grade bovine albumin (Sigma A7888), and 2mM EDTA pH 7.4).

100 iuL of E27 was added to the first four wells of the FcyRIA subunit coated plated at a concentration of 10 pg/mL. 80 pL of assay buffer was added to the next four well followed by uL of the 10pg/mL E27 IgG to give a final concentration of 2 Pg/mL. Plates were incubated at for 2 hours on an orbital shaker.

For detection, plates were washed 5x with wash buffer to remove unbound antibody. IgG binding to GST-FcyRIA was detected by adding 100 UL horse radish peroxidase

(HRP)

conjugated protein G (BIORAD) at 1:5000. HRP conjugates were incubated for 1.5 hours at 250C on an orbital shaker. Plates were washed x5 with wash buffer to remove unbound HRP conjugate. Binding was detected by adding 100 jL of substrate solution (0.4mg/mL o- -67-

O

O

Sphenylenedaimine dihydrochlorde, Sigma P6912, 6 mM H202 in PBS) and incubating for minutes at 25 0 C. Enzymatic reaction was stopped by the addition of 100 ~L of 4.5 N H 2

SO

4 and Scolorimetric product was measured at 490 nm on a 96 well plate densitometer (Molecular Devices).

Binding of E27 variants at IgG concentration of 2 pg/mL was expressed as a ratio of wild type E27.

e Fc yRIA THP-1 Assay. 100 pL of E27 was added to the first three wells of a serocluster en plate (Costar) at a concentration of 20 pg/mL in assay buffer (lx PBS, 0.1% BSA, 0.01% NaN 3 92.5 1L of assay buffer was added to the next three wells followed by 7.5 PL of the 20pg/mL E27 10 IgG to give a final concentration of 1.5 pg/mL. To each well, 100 pL of THP-1 cells were added at a concentration of 5 million cells/mL in FACS assay buffer. The plate is incubated on ice for minutes For detection, cells were washed 2x with assay buffer to remove unbound antibody. IgG binding FcyRIA was detected by adding 100 il FITC conjugated fragment of goat antihuman IgG heavy chain specific. (Jackson Immunoresearch) at 1:200. FITC conjugates were incubated with cells for 30 minutes on ice. Cells were washed x3 with assay buffer to remove unbound FITC conjugate. Cells were stained with P.I. (SIGMA) at 2.5 Pg/mL and analyzed by flow cytometry.

Binding of E27 variants at IgG concentration of 1.5 pg/mL was expressed as a ratio of wild type E27.

Data from the plate assay (FcyRIA ELISA).and cell-based assay (FcyRIA THP-1 assay) was averaged to arrive at an FcyRIA-binding activity.

LowAffinity FcyR Binding ELISAs: FcyRIIA, FcyRIIB and FcyRIIIA binding ELISAs were performed as described in Example 1 above, with detection of the stable hexamer (consisting of three anti-lgE molecules and three IgE molecules).

FcRn Binding ELISA: For measuring FcRn binding activity of IgG variants, ELISA plates were coated with 2 pg/ml streptavidin (Zymed, South San Francisco) in 50 mM carbonate buffer, pH 9.6, at 4 0 C overnight and blocked with PBS-0.5% BSA, pH 7.2 at room temperature for one hour. Biotinylated FcRn (prepared using biotin-X-NHS from Research Organics, Cleveland, OH and used at 1-2 pg/ml) in PBS-0.5% BSA, 0.05% polysorbate 20, pH 7.2, was added to the plate and incubated for one hour. Two fold serial dilutions of IgG standard (1.6-100 ng/ml) or variants in PBS-0.5% BSA, 0.05% polysorbate 20, pH 6.0, were added to the plate and incubated for two hours. Bound IgG was detected using peroxidase labeled goat F(ab') 2 anti-human IgG F(ab') 2 in the above pH 6.0 buffer (Jackson ImmunoResearch, West Grove, PA) followed by tetramethyl benzidine (Kirgaard Perry Laboratories) as the substrate. Plates were washed -68-

O

between steps with PBS-0.05% TWEEN 20® at either pH 7.2 or 6.0. Absorbance was read at O 450 nm on a Vmax plate reader (Molecular Devices, Menlo Park, CA). Titration curves were fit Z with a four-parameter nonlinear regression curve-fitting program (KaleidaGraph, Synergy Cl software, Reading, PA). Concentrations of IgG variants corresponding to the mid-point absorbance of the titration curve of the standard were calculated and then divided by the ge concentration of the standard corresponding to the mid-point absorbance of the standard titration curve.

e In Vitro ADCC Assay. To prepare chromium 51-labeled target cells, tumor cell lines were grown in tissue culture plates and harvested using sterile 10 mM EDTA in PBS..SK-BR-3 cells, a 3+ HER2-overexpressing human breast cancer cell line, were used as targets in all assays. The detached cells were washed twice with cell culture medium. Cells (5x10E) were labeled with 200 pCi of chromium51(New England Nuclear/DuPont) at 37 0 C for one hour with occasional mixing.

Labeled cells were washed three times with cell culture medium, then were resuspended to a concentration of 1x10 5 cells/mL. Cells were used either without opsonization, or were opsonized prior to the assay by incubation with rhuMAb HER2 wildtype (HERCEPTINC) or seven Fc mutants (G14, G18, G17, G36, G30, G31 and G34) at 100 ng/mL and 1.25 ng/mL in PBMC assay or ng/mL and 1 ng/mL in NK assay.

Peripheral blood mononuclear cells were prepared by collecting blood on heparin from normal healthy donors and dilution with an equal volume of phosphate buffered saline (PBS). The blood was then layered over LYMPHOCYTE SEPARATION MEDIUM® (LSM: Organon Teknika) and centrifuged according to the manufacturer's instructions. Mononuclear cells were collected from the LSM-plasma interface and were washed three times with PBS. Effector cells were suspended in cell culture medium to a final concentration of 1x10 7 cells/mL.

After purification through LSM, natural killer (NK) cells were isolated from PBMCs by negative selection using an NK cell isolation kit and a magnetic column (Miltenyi Biotech) according to the manufacturer's instructions. Isolated NK cells were collected, washed and resuspended in cell culture medium to a concentration of 2x10 6 cells/mL. The identity of the NK cells was confirmed by flow cytometric analysis.

Varying effectortarget ratios were prepared by serially diluting the effector (either PBMC or NK) cells two-fold along the rows of a microtiter plate (100pL final volume) in cell culture medium. The concentration of effector cells ranged from 1.0x1 0 7 /mL to 2.0 x 10 4 /mL for PBMC and from 2.0x10 6 /mL to 3.9x10 3 /mL for NK. After titration of effector cells, 100pL of chromium 51labeled target cells (opsonized or nonoponsonized) at lx105 cells/mL were added to each well of the plate. This resulted in an initial effector:target ratio of 100:1 for PBMC and 20:1 for NK cells.

All assays were run in duplicate, and each plate contained controls for both spontaneous lysis (no -69-

O

O

Seffector cells) and total lysis (target cells plus 100 pL) 1% sodium dodecyl sulfate, 1 N sodium O hydroxide). The plates were incubated at 37oC for 18 hours, after which the cell culture Z supematants were harvested using a supematant collection system (Skatron Instrument, Inc.) and C counted in a Minaxi auto-gamma 5000 series gamma counter (Packard) for one minute. Results were then expressed as percent cytotoxicity using the formula: n Cytotoxicity (sample cpm-spontaneous lysis)/(total lysis-spontaneous lysis) x 100 Four-parameter curve-fitting was then used to evaluate the data (KaleidaGraph 3.0.5).

C RESULTS 8 A variety of antibody variants were generated which had FcR binding activity that differed from the parent antibody. The FcR binding data for the variants generated is shown in Tables 6 and 7 below. An additional variant, T307Q, also displayed improved FcRn binding compared to E27 parent antibody.

2004233493 25 Nov 2004 TABLE 6 CH2 DOMAIN VARIANTS IG2 Res#EU FORn FcyRI FcyRIIA FcyRIIB FCyRIIIA (Kabat) mean sd n mean sd n mean sd mean sd mean ad r= REDUCED BINDING TO ALL FcyR 1 233-236 0.54 (0.20) 3 0:12 (0.06) 6 0.08 (0.01) 0.12 (0.01) 0.04 (0.02) n=2 ELLG PVA- 2 P238A(251) 1.49 (0.17) 3 0.60 (0.05) 5 0.38 0.36 (0.15) 0.07 (0.05) n=4 14 D265A(278) 1.23 (0.14) 4 0.14 (0.04) 6 0.07 (0.01) 0.13 (0.05) 0.09 (0.06) n=4 17 E269A(282) 1.05 0.52 (0.03) 6 0.65 (0.18) 0.75 (0.29) 0.45 (0.13) 18 D270A(283) 1.05 0.76 (0.12) 6 0.06 (0.01) 0.11 (0.05) 0.14 (0.04) 58 N297A(314) 0.80 (0.18) 8 0.15 (0.06) 7 0.05 (0.00) 0.10 (0.02) 0.03 (0.01) n=3 52 A327Q(346) 0.97 0.63 (0.15) 7 0.13 (0.03) 0.14 (0.03) 0.06 (0.01) n=4 64 P329A(348) 0.80 0.48 (0.10) 6 0.08 (0.02) 0.12 (0.08) 0.21 (0.03) n=4 REDUCED BINDING TO FcyRII FcyRIII t= 3 S239A(252) 1.06 0.81 (0.09) 7 0.73 (0.25) 0.76 (0.36) 0.26 (0.08) n=3 33 E294A(311) 0.75 0.90 (0.08) 4 0.87 (0.19) 0.63 (0.17) 0.66 (0.14) 34 0295A(312) 0.79 1.00 (0.11) 4 0.62 (0.20) 0.50 (0.24) 0.25 (0.09) 39 V303A(322) 1.26 (0.21) 3 0.91 (0.11) 5 0.86 (0.10) 0.65 (0.17) 0.33 (0.09) n=8 IMPROVED BINDING TO FcyRII FcyRIII 11 T256A(269) 1.91 (0.43) 6 1.14 (0.14) 4 1.41 (0.27) 2.06 (0.66) 1.32 (0.18) n-9 K290A(307) 0.79 (0.14) 3 1.01 (0.08) 4 129 (0.21) 1.40 0.18) 1.28 (0.21) n--7 K290A(307) 0.79 (0.14) 3 1.01 (0.08) 4 1.29 (0.21) 1.40 (0.19) 1.28 (0.21) n=7 2004233493 25 Nov 2004 44 D312A(331) 1.50 (0.06) 4 1.01 (0.12) 5 1.20 (0.24) 1.19 (0.07) 1.23 (0.14) n-3 51 K326A(345) 1.03 1.04 (0.05) 4 1.26 (0.21) 1.49 (0.27) 1.22 (0.28) 197 A330(349)K 1.28 1.25 1.28 n-1 273 A339(359)T 1.23 1.11 1.23 1.42 n=l EFFECT FcyRII R255A(268) 0.59 (0.19) 4 1.26 (0.26) 8 1.30 (0.20) 1.59 (0.42) 0.98 (0.18) 12 E258A(271) 1.18 1.18 (0.13) 4 1.33 (0.22) 1.65 (0.38) 1.12 (0.12) S267A(280) 1.08 1.20 (0.14) 4 1.64 (0.18) 2.06 (0.35) 1.14 (0.25) n=7 16 H268A(281) 1.02 (0.22) 3 1.05 (0.11) 4 1.22 (0.14) 1.45 (0.23) 0.52 (0.09) n=12 19 E272A(285) 1.34 (0.24) 4 1.04 (0.06) 4 1.24 (0.11) 1.58 (0.19) 0.74 (0.12) n=4 21 N276A(289) 1.15 (0.21) 3 1.05 (0.14) .4 1.29 (0.20) 1.34 (0.40) 0.95 (0.04) n=4 23 D280A(295) 0.82 0.97 (0.06) 4 1.34 (0.14) 1.60 (0.31) 1.09.(0.20) E283A(300) 0.71 0.97 (0.03) 4 1.24 (0.23) 1.20 (0.17) 1.01 (0.14) 26 H285A(302) 0.85 0.96 (0.07) 4 1.26 (0.12) 1.23 (0.15) 0.87 (0.04) n-4 27 N286A(303) 1.24 (0.04) 2 0.94 (0.20) 13 1.28 (0.23) 1.39 (0.14) 1.03 (0.08) 31 R292A(309) 0.81 (0.18) 4 0.93 (0.02) 4 0.27 (0.14) 0.18 (0.07) 0.90 (0.18) n=9 36 S298A(317) 0.80 1.10 (0.04) 3 0.40 (0.08) 0.21 (0.11) 1.30 (0.18) n=12 38 R301A(320) 0.86 1.06 (0.10) 4 1.12 (0.12) 1.26 (0.14) 0.21 (0.08) n-6 38B R301M(320) 0.88 1.06 4 1.29 (0.17) 1.56 (0.12) 0.48 (0.21) n=4 V305A(324) 1.46 (0.48) 6 1.04 (0.19) 10 1.12 (0.12) 1.23 (0.22) 0.84 (0.15) n=4 41 T307A(326) 1.81 (0.32) 6 0.99 (0.14) 4 1.19 (0.37) 1.35 (0.33) 1.12 (0.18) n=12 2004233493 25 Nov 2004 42 L309A(328) 0.63 (0.18) 4 0.93 (0.18) 6 1.13 (0.08) 1.26 (0.12) 1.07 (0.20) n=3 N315A(334) 0.76 (0.14) 3 1.27 (0.36) 6 1.15 (0.06) 1.30 (0.17) 1.07 (0.21) 48 K320A(339) 1.10 0.98 (0.09) 5 1.12 (0.11) 1.22 (0.05) 0.87 (0.17) n=4 49 K322A(341) 0.98 0.94 (0.05) 6 1.15 (0.11) 1.27 (0.24) 0.61 (0.14) S324A(343) 1.08 0.95 (0.05) 4 0.82 (0.22) 0.70 (0.12) 1.12 (0.17) n=4 P331A(350) 0.85 1.30 (0.34) 8 1.29 (0.14) 1.47 (0.28) 1.03 (0.19) n=3 54 E333A(352) 1.03 (0.01) 2 0.98 (0.15) 5 0.92 (0.12) 0.76 (0.11) 1.27 (0.17) 56 T335A(354) 0.98 1.00 (0.05) 4 0.79 (0.22) 0.65 (0.26) 0.92 (0.54) n=3 57 S337A(356) 1.03 1.17 (0.23) 3 1.22 (0.30) 1.26 (0.06) 0.94 (0.18) n=4 EFFECT FcyRIII K248A(261) 0.87 0.95 (0.05) 5 1.06 (0.12) 1.01 (0.12) 0.71 (0.05) n=4 6 D249A(262) 0.93 1.04 (0.10) 4 1.02 (0.12) 0.94 (0.02) 0.66 (0.07) 7 M252A(265) 0.64 (0.13) 4 0.99 (0'10) 5 1.01 (0.18) 1.15 (0.22) 0.65 (0.17) n=6 9 S254A(267) 0.10 0.96 (0.08) 4 0.97 (0.24) 1.15 (0.38) 0,73 (0.14) n-3 16 H268A(281) 1.02 (0.22) 3 1.05 (0.11) 4 1.22 (0.14) 1.45 (0.23) 0.52 (0.09) n-12 19 E272A(285) 1.34 (0.24) 4 1.04 (0.06) 4 1.24 (0.11) 1.58 (0.19) 0.74 (0.12) n=4 22 Y278A(291) 0.90 0.96 (0.02) 4 1.11 (0.08) 1.10 (0.16) 0.67 (0.11) n=4 29 T289A(306) 0.86 0.93 (0.03) 4 0.96 (0.33) 0.83 (0.22) 0.62 (0.19) n=7 32 E293A(310) 0.85 1.11 (0.07) 4 1.08 (0.19) 1.07 (0.20) 0.31 (0.13) n=6 Y296F(313) 0.79 1.07 (0.12) 4 0.97 (0.26) 0.84 (0.18) 0.52 (0.09) 36 S298A(317) 0.80 1.10 (0.04) 3 0.40 (0.08) 0.21 (0.11) 1.30 (0.18) n=12 2004233493 25 Nov 2004 38 R301A(320) 0.86 1.06 (0.10) 4 1.12 (0.12) '1.26 (0.14) 0.21 (0.08) n-6 38B R301M(320) 0.88 1.06 (0.12) 4 1.29 (0.17) 1.56 (0.12) 0.48 (0.21) n=4 49 K322A(341) 0.98 0.94 (0.05) 6 1.15 (0.11) 1.27 (0.24) 0.61 (0.14) 54 E333A(352) 1.03 (0.01) 2 0.98 (0.15) -5 0.92 (0.12) 0.76 (0.11) 1.27 (0.17) K334A(353) 1.05 (0.03) 2 1.10 (0.06) 4 1.01 (0.15) 0.90 (0.12) 1.39 (0.19) n=17 NO EFFECT ON FcyR 4 .K246A(259) 1.03 0.94 (0.06) 4 .1.02 (0.10) 0.92 (0.15) 1.14 (0.38) n=4 4B K246M(259) 0.69 0.83 (0.05) 5 0.83 (0.06) 0.76 (0.05) 0.95 (0.09) n=3 K248M(261) 0.79 0.95 (0.06) 4 0.89 (0.09) 0.83 (0.04) 1.01 (0.23) n=3 8 1253A(266). 0.10 0.96 4 1.14 (0.02) 1.18 (0.06) 1.08 (0.14) n=3 13 T260A(273) 1.09 0.93 (0.09) 4 0.89 (0.14) 0.87 (0.10) 0.89 (0.08) n=4 K274A(287) 1.18 1.02 (0.04) 4 0.86 (0.09) 0.96 (0.10) 1.11 (0.08) n=3 24 V282A(299) 1.13 (0.07) 2 0.96 (0.02) 4 1.15 (0.13) 1.15 (0.20) 1.00 (0.18) n=4 28 K288A(305) 0.38 (0.12) 5 0.88 (0.15) 15 1.15 (0.26) 1.14 (0.20) 1.06 (0.04) n=4 37 Y300F(319) 0.74 (0.10) 2 1.07 (0.15) 4 1.11 (0.04) 1.09 (0.09) 1.01 (0.10) n=3 43 Q311A(330) 1.62 (0.25) 4 0.93 (0.05) 4 1.11 (0.06) 1.19 (0.13) 0.93 (0.17) n=3 46 K317A(336) 1.4-4 (0.18) 4 0.92 (0.17) 6 1.13 (0.05) 1.18 (0.27) 1.10 (0.23) n=4 47 E318A(337) 0.85 0.92 (0.07) 4 1.04 (0.10) 1.17 (0.23) 1.01 (0.05) n=3 •53 A330Q(349) 0.76 0.96 (0.10) 4 1.01 (0.12) 1.02 (0.02) 0.75 (0.18) n=3 i lllli=II=i=i===I 2004233493 25 Nov 2004 TABLE 7 CH3 DOMAIN VARIANTS IG2 Res#EU FcRn FcyRI FcyRIIA FcyRIIB FcyRITIA (Kabat) mean sd n mean ad n mean sd mean sd mean ad E1 K338(358)A 1.14 0.90 (0.05) 3 0.78 (0.09) 0.63 (0.08) 0.15 (0.01) n-2 BlA K338(358)M 0.78 0.99 (0.08) 3 0.99 (0.13) 0.93 (0.15) 0.49 (0.04) n-2 S2 K340(360)A 1.02 1.04 (0.07) 3 1.05 (0.18) 0.96 (0.20) 0.84 (0.11) n=2 B2A K340(360)M 1.20 1.17 (0.11) 3 1.10 (0.12) 1.20 (0.19) 0.75 (0.12) n=2 83 Q342(363)A 1.09 1.13 (0.11) 3 1.01 1.09 (0.23) 0.98 (0.10) n=2 84 R344(365)A 0.77 1.04 (0.08) 3 0.89 (0.14) 0.91 (0.04) 0.97 (0.07) n=4 E345(366)A 1.18 1.06 (0.05) 3 1.03 (0.10) 0.98 (0.10) 0.97 (0.13) n=4 B6 Q347(368)A 0.95 1.04 (0.06) 3 1.00 (0.03) 0.92 (0.02) 1.04 (0.12) n=4 B7 R355(376)A 1.06 1.09 (0.07) 3 0.84 (0.09) 0.87 (0.11) 0.98 (0.09) n=4 B8 E356(377)A 1.21 (0.11) 2 1.05 (0.04) 3 0.90 (0.02) 0.99 (0.13) 0.92 (0.03) n=3 B9 M358(381)A 0.96 1.06 (0.07) 3 1.11 (0.06) 1.16 (0.25) 0.91 (0.09) n=3 810 T359(382) 1.04 1.04 (0.05) 3 1.13 (0.10) 1.15 (0.04) 1.23 (0.26) n=3 311 K360(383)A 1.30 (0.08) 4 1.02 (0.04) 3 1.12 (0.10) 1.12 (0.08) 1.23 (0.16) n=6 812 N361(384)A 1.16 1.00 (0.03) 3 0.82 (0.07) 0.82 (0.12) 1.08 (0.06) n=3 B13 Q362(385)A 1.25 (0.24) 3 1.00 (0.04) 3 1.03 (0.10) 1.02 (0.03) 1.03 (0.16) n=4 B14 Y373(396)A 0.86 0.98 (0.07) 3 0.84 (0.11) 0.75 (0.08) 0.67 (0.04) S375(398)A 1.17 (0.19) 5 0.95 (0.02) 3 1.08 (0.06) 1.14 (0.11) 1.04 (0.05) n=6 516 D376(399)A 1.45 (0.36) 4 1.00 3 0.80 (0.16) 0.68 (0.14) 0.55 (0.10) 2004233493 25 Nov 2004 B17 A378(401)Q 1.32 (0.13) 3 1.06 (0.05) 3 1.40 (0.17) 1.45 (0.17) 1.19 (0.17) B18 E380(405)A 2.19 (0.29) 6 1.04 (0.06) 3 1.18 (0.01) 1.07 (0.05) 0.92 (0.12) n=2 B19 E382(407)A 1.51 (0.18) 4 1.06 (0.03) 3 0.95 (0.11) 0.84 (0.04) 0.76 (0.17) n=3 820 S383(408)A 0.74 1.03 (0.03) 3 0.92 (0.04) 0.94 (0.05) 0.88 (0.07) n=3 B21 N384(410)A 0.88 1.00 (0.01) 3 1.05 (0.19) 1.10 (0.18) 0.96 (0.18) n=8 B22 Q386(414)A 0.70 (0.10) 2 1.14 (0.08) 3 1.08 (0.13) 1.19 (0.25) 0.98 (0.14) n=9 B23 E388(416)A 0.64 (0.12) 2 1.15 (0.09) 3 0.87 (0.03) 0.94 (0.09) 0.62 (0.04) n=3 824 N389(417)A 0.73 1.00 (0.02) 3 0.98 (0.15) 0.81 (0.04) 0.75 (0.02) n=3 N390(418)A 0.87 1.06 (0.04) 3 0.99 (0.10) 0.94 (0.02) 0.87 (0.09) n=3 B26A Y391(419)A 1.14 1.00 (0.08) 3 0.97 (0.10) 0.94 (0.02) 0.86 (0.05) n=3 sB26B Y391(419)F 0.81 (0.10) 2 1.00 (0.01) 3 1.05 (0.12) 1.11 (0.08) 1.01 (0.15) 827 K392(420)A 0.97 1.01 (0.08) 3 0.92 (0.20) 0.94 (0.01) 0.79 (0.22) n=3 B28 L398(426)A 0.94 (0.04) 2 1.13 (0.15) 6 1.17 (0.11) 1.20 (0.08) 0.94 (0.04) n=3 B29 5400(428)A 0.64 (0.07) 3 1.10 (0,09) 3 0.95 (0.04) 0.99 (0.08) 0.83 (0.07) n=2 D401(430)A 1.10 (0.09) 3 1.13 (0.16) 6 1.11 (0.12) 1.19 (0.11) 0.97 (0.10) 831 D413(444)A 1.21 (0.07) 2 1.00 (0.01) 3 0.83 (0.08) 0.84 (0.06) 0.90 (0.16) n=2 832 K414(445)A 1.02 1.00 (0.04) 3 0.64 (0.15) 0.58 (0.18) 0.82 (0.27) n=3 B33 S415(446)A 0.44 1.04 (0.03) 3 0.90 (0.11) 0.88 (0.05) 0.86 (0.18) n=2 B34 R416(447)A 1.08 0.96 (0.04) 3 0.68 (0.05) 0.80 (0.05) 0.71 (0.08) n=2 Q418(449)A 0.77 (0.03) 2 0.98 (0.01) 3 1.00 (0.01) 0.96 (0.02) 0.96 (0.05) n=2 836 Q419(450-A 0.76 (0.01) 2 (0.02) 3 0.68 (0.09) 0.63 0.86 n=3- 836 Q419(450)A 0.76 (0.01) 2 0.97 (0.02) 3 0.68 (0,09) 0.63 (0.07) 0.86 (0.08) n=3 2004233493 25 Nov 2004 B37 N421(452)A 0.98 0.99 (0.01) 3 0.90 (0.03) 0.81.(0.0) 0.87 (0.12) n=2 B38 V422(453)A 1.01 0.98 (0.02) 3 0.89 0.83 (0.05) 0.83 (0.12) n=2 B39 S424(455)A 1.41 (0.14) 3 0.98 (0.03) 3 1.04 (0.06) 1.02 (0.02) 0.88 (0.09) n=2 E430(461).A 0.93 (0.03) 2 1.05 (0.02) 3 1.24 (0.11) 1.28 (0.10) 1.20 (0.18) B41 H433(464)A 0.41 (0.14) 2 0.98 (0.03) 3 0.92 (0.18) 0.79 (0.18) 1.02 (0.15) n=3 B42 N434(465)A 3.46 (0.37) 7 1.00 (0.04) 3 0.97 (0.07) 0.98 (0.13) 0.74 (0.12) B43 H435(466)A 0.10 4 1.25 (0.09) 3 0.77 (0.05) 0.72 (0.05) 0.78 (0.03) n=3 B44 Y436(467)A 0.10 2 0.99 (0.02) 2 0.93 (0.05) 0.91 (0.06) 0.91 (0.15) n=3 T437(468)A 0.99 (0.07) 1.00 (0.02) 3 1.12 (0.18) 1.00 (0.22) 0.77 (0.19) B46 Q438(469)A 0.79 (0.05) 2 1.02 (0.05) 3 0.80 (0.10) 0.72 (0.16) 1.01 (0.17) 1 B47 K439(470)A 0.70 (0.04) 2 0.98 (0.04) 3 0.78 (0.16) 0.68 (0.22) 0.86 n=4 "-4 SB48 S440(471)A 0.99 1.01 (0.02) 3 1.10 (0.15) 1.11 (0.26) 0.93 (0.01) n=3 B49 S442(473)A 0.86 1.02 (0.02) 3 0.98 (0.08) 0.91 (0.11) 0.95 (0.10) S444(475)A 0.80 1.01 (0.02) 3 1.07 (0.03) 1.03 (0.03) 0.88 (0.12) n=2 851 K447(478)A 0.62 (0.12) 3 1.02 (0.03) 3 0.95 (0.05) 0.91 (0.05) 0.84 (0.09) n=2 Variants with increased binding to a FcyR generally had binding values z 1.20 as determined in this Example and those with reduced binding to a FcyR generally had binding values s 0.80 as determined in this Example. Variants with increased binding to FcRn generally had binding values a 1.30 as determined in this Example and those with reduced binding to FcRn generally had binding values 0.70 as determined in this Example.

Aside from alanine variants, various non-alanine substitution variants were made, and the FcR binding activity of those variants is summarized in the following table.

miniIll II IIII1 2004233493 25 Nov 2004 TABLE 8 NON-ALANINE VARIANTS 102 ResiEJ FcRn FcyRI FcyRIIA FcyRIIB FcyRIIIA (Kabat) mean sd n mean sd n mean sd mean sd mean sd 222 D249(262)E 0.97 0.99 0.84 n=1 176 T256(269)G 1.10 (0.03) 1.06 (0.07) 0.96 (0.27) n=2 254 T256(269)N 1.03 0.89 1.13 .n-i 157 D 2 6 5 2 7 8 N 0,02 (0.01) .0.03 (0.01) 0.02 (0.01) 158 D265(278)E 0.11 (0.04) 0.03 (0.01) 0.02 (0.01) n=3 189 S267(280)G R131 1.21 (0.05) 0.97 (0.16) 0.09 (0.02) n-3 H131 0.59 (0.09) n=3 84 H268(281)N 1.33 1.41 0.56 n-1 H268(281)S 1.35 1.38 0.81 n-i 87 H268(281)Y 1.19 1.29 0.76 n-i 168 E269(282)D 0.89 (0.10) 0.73 (0.07) 1.13 (0.21) n=2 169 E269(282)Q 0.08 (0.01) 0.16 (0.00) 0.29 (0.03) n=2 92 D270(283)N 0.06 (0.01) 0.10 (0.02) 0.04'(0.00) n-2 93 D270(283)9 0.55 (0.05) 0.38 (0.05) 1.17 (0.01) n=2 223 E272(285)Q 1.93 1.81' 0.82 n-i 224 E272(285)N 0.43 0.23 0.50 n-i 167 1274(287)Q 0.66 0.94 0.62 n-1 165 N276(289y1< 0.81 0.77 0.61 n-i 2004233493 25 Nov 2004 233 N276(289)Q 1.09 0.79 0.91 n=1 79 D280(295)N 1.26 (0.07) 1.38 (0.04) 1.13 (0.13) n=2 149 0280(295)S 1.07 (0.06) 1.04 (0.08) 1.09 (0.06) n-2 226 E283(300)Q 1.12 1.24 1.19 n=1 227 E283(300)S 1.03 1.07 0.85 n=1 228 E283(300)N 1.18 1.28 0.94 229 E283(300)D 1.14 1.23 0.95 n=1 23 N286(303)Q 1.52 1.13 0.96 n=1.

.237 N286(303)S 1.72 1.38 1.32 n=1 238 N286(303)D 1.41 1.23 0.98 n=1 73 K290(307)Q 1.17 1.26 1.40 n=1 K290(307)S 1.27 1.34 1.26 n=1 77 K290(307)E 1.14 1.10 1.20 1.30 n=1 78 K290(307)R 1.25 1.05 1.15 1.08 n-1l 177 K290(307}G 1.07 1.21 1.23 n=1 R292(309)K 0,71 (0.17) 0.75 (0.10) 1.15 (0.18) n=3 81 R292(309)H 0.21 (0.09) 0.12 (0.01) 0.92 (0.08) n=2 82 R292(309)Q 0.47 (0.12) 0.25 (0.06) 0.45 (0.09) n=3 83 R292(309)N 0.54 (0.16) 0.29 (0.07) 0.88 (0.02) n-3 144 E293(310)Q 0.85 (0.03) 0.77 (0.13) 0.99 (0.04) 145 E293(310)D 0.90 (0.02) 0.88 (0.07) 0.37 (0.07) n=2 147 E293(310)T< 1.13 (0.04) 1.31 (0.17) 0.72 (0.09) n=4 2004233493 25 Nov 2004 173 E294(311)Q 1.01 0.95 0.84 n-1 174 E294(311)D 0.37 0.26 0.14 n-1 185 Y296(313)H 0.90 0.81 0.92 n=1 186 Y296(313)W 0.96 0.93 1.38 n=1 3298(317)G 0.87 (0.17) 0.63 (0.33) 0.46 (0.09) n=4 71 S298(317)T 0.41 (0.21) 0.40 (0.19) 0.89 (0.20) n=3 72- S298(317)N 0.08 (0.01) 0.16 (0.03) 0.06 (0.01) n=2 218 S298(317)V 0.11 (0.06) 0.17 (0.01) 0.33 (0.19) n=3 219 S298(317)L 1.14 (0.12) 1.42 (0.31) 0.34 (0.04) n-3 150 V303(322)L 0.89 (0.05) 0.73 (0.10) 0.76 (0.09) n=4 o 151 V303(322)T 0.64 (0.11) 0.34 (0.05) 0.20 (0.05) n=4 217 E318(337)K 1.03 1.08 0.72 n=1 172 K320(339)R 0.71 0.66 0.68 n=1 202 K320(339)N 1.34 1.40 1.27 n=1 204 K320(339)Q 1.23 1.12 1.17 n=l 205 K320(339)E 1.29 1.34 '1.12 n=1 235 K320(339)R 1.24 0.95 0.86 n-1 155 E322(341)R 0.87 (0.07) 0.87 (0.21) 0.92 (0.15) n=3 156 K322t341)Q 0.87 (0.02) 0.92 (0.23) 0.78 (0.18) n=3 206 K322(341)E 1.38 1.34 0.81 n=i 207 K322(341)N 0.57 .0.36 0.04 n~1 2004233493 25 Nov 2004 213 S324(343)N t.15 1.09 0.97 n-i 214 S324(343)Q 0.82 0.83 0.78 n=1 215 S324(343)K 0.66 0.37 0.77 n-1 216 S324(343)E 0.82 0.73 0.81 n-i 208 K326(345)5 1.44 1.62 1.37 209 K326(345)N 1.04 1.00 1.27 n=1 210 K326(345)Q 1.36 1.41 1.15 n=1 211 K326(345)D 1.68 2.01 1.36 n=1 212 K326(345)E 1.34 (0.27) 1.47 (0.33) 1.26 (0.04) n=2 131 A327(346)S 0.23 (0.06Y- 0.22 (0.05) 0.06 (0.01) n=4 159 A327(346)G 0.92 (0.09) 0.83 (0.10) 0.36 (0.05) n=3 196 A330 (349)D 0.18 0.08 0.07 n-i 197 A330(349)K 1.28 1.25 1.28 n-1 198 ?331C350)S 1.00 0.86 0.86 n-1 199 P331(350)N 0.86 0.65 0.23 200 P331(350)E 1.06 0.91 0.42 n=1 203 P331(350)K 0.94 0.71 0.33 n=1 141 E333(352)Q 0.70 (0.05) 0.64 (0.09) 1.10 (0.03) n=2 142 E333(352)N 0.59 (0.04) 0.52 (0.07) 0.56 (0.10) n=2 143 E333(352)S 0.94 n-i 152 E333(352)K 0.85 (0.14) n-3 153 E333(352)R 0.75 (0.04) 0.66 (0.03) 0.84 (0.05) n=2 2004233493 25 Nov 2004 154 E333(352)D 1.26 (0.04) n=3 178 E333(352)G 0.87 0.76 1.05 n=1 179 K334(353)G 0.76 (0.08) 0.60 (0.13) 0.88 (0.22) 135 K334(353)R 1 1.15 (0.09) 1.33 (0.18) 0.68 (0.07) 136 K334(353)Q 1.08 (0.11) 1.10 (0.21) 1.31 (0.26) n=7 137 K334(353)N 1.16 (0.11) 1.29 (0.30) 1.15 (0.16) n-7 138 K334(353)S 1.01 (0.11) 1.03 1.19 (0.08) n=3 139 K334(353)E 0.74 (0.15) 0.72 (0.12) 1.30 (0.09) n=4 140 K334(353)D 0.51 (0.09) 0.40 (0.03) 1.13 (0.09) n=4 190 K334(353)M 1.18 1.06 1.01 1.35 n=1 S191 K334(353)Y 1.15 1.08 1.05 1.31 n=1 192 K334(353)W 1.16 0.94 0.91 1.07 n-1 193 K334(353)H 1.11 1.09 1.07 1.26 n=1 220 K334(353)V 1.13 (0.11) 1.09 (0.15) 1.34 (0.18) n=3 221 K334(353)L 1.05 1.09 1.38 n=1 171 T335(354)Q 0.86 0.79 0.84 n=1 194 T335(354)E 1.24 1.30 1.19 n=1 195 T335(354)K 1.19 1.14 1.30 n=1 273 A339(359)T 1.23 1.11 1.23 1.42 n=1 The following table summarizes the FcR binding activity of various combination variants.

2004233493 25 Nov 2004 TADLE 9 COMb2NkTION VARIANTS r~lll ~IE= 1G2 ResiW FcRn FcyRI FoyRTIA FeyRlIB FyRIIIA (Kabat) mean sd n mean ad n mean sd mean ad mean ad 96 S267 280)A 1.41 1.72- 0.84 n=1 1268(281)A 134 E333(352)A 0.72 (0.08) 0.62 (0.13) 1.30 (0.12) K334(353)A 1059 T256(269)A 0.44 (0.03) 0.22 (0.04) 1.41 (0.06) n=2 S298(317)A 1051 T256(269)A 0.47 (0.01) 0.30 (0.03) 1.21 (0.26) n=2 D280(295)A 5298(317)A -T307(326)A 106 T256(269)A 0.11 0.08 0.90 n-l D2BO (295)A R292(309)A 5298(317) T307(326)1 107 S298 (317)A 0.34 (0.05) 0.16 (0.08) 1.53 (0.24) £333(352)1 109 5298(317)1k 0.41 (0.07) 0.19 (0108) 1.62 (0.34) -n=6 1334(353)1 110 S298(317)A 0.35 (0.13) 0.18 (0.08) 1.66 (0.42) n=l1 £333(352)A K334(353)1 246 S267(280)A 1.62 (0.15) 2.01 (0.45) 1.04 (0.12) n=2 £258(271)1 247 5267(280)A 1.60 1.72 (0.13) 0.88 (0.07) n=3 R255(268)1 2004233493 25 Nov 2004 248 S267(280)A 1.54 (0.00) 1.96 (0.37) 1.13 (0.07) n=2 0280 (295)A 250 S267(280)A 1.51 (0.13) 1.82 (0.32) 0.95 (0.05) n=3 £272 (285)A 251 S267(280)A 1.67 (0.11) 1.85 (0.10) 0.92 (0.09) n=3 E293(310)A 264 S267(280)A 1.48 (0.12) 2.03 (0.30) 0.89 (0.04) n=2 £258 (271)A 0280 (295)A R255 (268)A 269 E380(405)A 8.55 (0.94) 3 1.02 (0.07) 1.05 (0.11) 1.02 n=2 14434 (465)A 270 .E380(405)A 12.6 0.99 (0.06) 0.99 (0.11) 0.96 n=2 14434 (465)A T307 (326)A p 271 E380(405)A 1.01 (0.01) 2 0.98 1.04 0.92 n=1 L309(328)A 272 N434(465)A 3.15 (0.42) 2 0.94 (0.11) 0.96 (0.17) 0.88 n=2 1(288 (305)A'

DISCUSSION

S This study includes a complete mapping of human IgG1 for human FcyRI, FcyRIIA, N, FcyRIIB, FcyRIIIA, and FcRn. An alanine-scan of all amino acids in human IgG1 Fc (CH2 and CH3 domains) exposed to solvent, based on the crystal structure of human Fc (Deisenhofer, Biochemistry20:2361-2370 (1981)), was performed. Each exposed amino acid in CH2 and CH3 was individually changed to alanine and the variant igG assayed against all five human receptors; all variants were evaluated using humanized anti-lgE E27 IgG1 as the parent polypeptide. FcyRI cn and FcRn are high affinity receptors and monomeric IgG could be evaluated in the assays for these two receptors. FcyRIIA, FcyRIIB and FcyRIIIA are low affinity receptors and required use cn 10 of an immune complex. Hence, an ELISA-type assay was used for FcyRIIA, FcyRIIB, and FcyRIIIA, in which pre-formed hexamers, consisting of three anti-lgE E27 and three IgE molecules S were bound to the FcyR and either anti-human IgG Fc-HRP or protein G-HRP used as detection reagent. In order to increase binding, these hexamers could be linked into multimers by addition of human VEGF (using anti-VEGF IgE). The hexamers bound to the low affinity FcyR significantly better than the IgG monomers; the multimers bound better than the hexamers (Figs. 15A and The hexameric complexes were used since these provided sufficient binding and required less IgG. Complexes formed using other antibody:antigen combinations are also possible reagents, as long as the antigen contains at least two identical binding sites per molecule for the antibody. As an example, VEGF contains two binding sites per VEGF dimer for anti-VEGF A.4.6.1 (Kim et al., Growth Factors 7:53 (1992) and Kim et al. Nature 362:841 (1993)).

VEGF:anti-VEGF multimers also bound to the low affinity FcyRIIA and FcyRIIIA (Figs. 16A and 16B).

Once the complete alanine-scan was performed, several classes of alanine variants were found. Some variants exhibited reduced binding to all FcyR (G14, Fig. 17), while other variants showed reduced binding only to one FcyR (G36, Fig. 17), improved binding only to one FcyR G54,- G55, Fig. 17), or simultaneous reduction to one FcyR with improvement to another (G16, Fig. 17).

Individual alanine variants were also combined in a single variant Fc region; e.g.

combining S298(317)A with K334(353)A improved binding to FcyRIlIA more than either S298(317)A or K334(353)A alone (Figs. 18A and B; and compare variants 36, 55, and 109 in Tables 6 and 9) (residue numbers in parentheses are those of the EU index as in Kabat).

Similarly, combining S298(317)A with E333(352)A improved binding to FcyRIIIA more than either S298(317)A or E333(352)A alone (compare variants 36, 54, and 107 in Tables 6 and 9).

Selected IgG variants were also tested for their binding to FcyR transfected into mammalian cells. The a-chain extracellular portion of human FcyRIIIA was transfected into CHO cells using a GPI-link, whereas for human FcyRIIB the full-length receptor was transfected into 1 CHO cells. For the variants tested, the pattern of binding to the cells was the same as the pattern 0 of binding in the protein:protein (ELISA) assay (Figs. 18A-B and 19A-B).

One application of these variants is to improve the ADCC effector function of an antibody.

This can be achieved by modifying Fc region amino acids at one or more residues which would S 5 lead to improved binding to FcyRIIIA. Improved FcyRIIIA binding would lead to improved binding by NK cells, which carry only FcyRIIIA and can mediate ADCC. Selected alanine variants which were either reduced in binding to FcyRIIIA (variants 17, 18, 34; Table had no effect on n FcyRIIIA binding (variant 31; Table or had improved binding to FcyRIIIA (variants 30, 36; Table 6) were tested in an in vitro ADCC assay using human PBMCs as effector cells. Since the target cells were HER2-overerexpressing SKBR3 cells, the IgG Fc variants used in this assay were generated by substituting the V/VL domains of anti-lgE E27 with those from anti-HER2 antibody; o HERCEPTIN® (humAb4D5-8 in Table 1 of Carter et al. PNAS (USA) 89:4285-4289 (1992)). The pattern of ADCC exhibited by the variants correlated well with the pattern of binding to FcyRIIIA (Figs. 20 and 21). Notably the variant which showed the best improvement in binding to FcyRIl IA in protein:protein assays, variant 36 S298(317)A, also showed improvement in ADCC compared to wildtype HERCEPTIN® at 1.25 ng/ml (Fig. 21).

EXAMPLE Bind of Fc Variants to Polymorphic Fc Receptors Allelic variants of several of the human FcyR have been found in the human population.

These allelic variant forms have been shown to exhibit differences in binding of human and murine IgG and a number of association studies have correlated clinical outcomes with the presence of specific allelic forms (reviewed in Lehmbecher et al. Blood 94(12):4220-4232 (1999)). Several studies have investigated two forms of FcyRIIA, R131 and H131, and their association with clinical outcomes (Hatta et al. Genes and Immunity 1:53-60 (1999); Yap et at Lupus 8:305-310 (1999); and Lorenz et al. European J. Immunogenetics 22:397-401 (1995)).

Two allelic forms of FcyRllA, F158 and V158, are only now being investigated (Lehmbecher et al, supra; and Wu et al. J. Clin. Invest 100(5):1059-1070 (1997)). In this example, selected IgG variants were tested against both allelic forms of FcyRIIA or FcyRIIIA. Fc receptor binding assays were performed essentially as described in the above examples. However, for FcyRllIA-V158, both the low affinity receptor binding assay of Example 1 (which analyzes binding of the IgG complex to FcyRIIIA-VI58); and the high affinity FcyR binding assay of Example 4 (which analyzes binding of IgG monomer to FcyRIIIA-V158) were carried out. The results of these studies are summarized in Table 10 below.

2004233493 25 Nov 2004 TABLE Binding of Variants to FcyRIIA and FeyRIIXA Polymorphic Receptors IgG Complex IgG Complex IgG Complex IgG Complex IgG Monomer IG2 Res#EU FcyRIIA-R131 FcyRIIA-H131 FcyRIIIA-F158 FcyRIIIA-V158 FcyRIIIA-V158 (Kabat) mean sd n mean sd n mean sd n mean sd n mean sd n 11 T256(269)A 1.41 (0.27) 9 1.32 (0.18) 9 0.97 (0.03) 2 1.20 1 254 T256(269)N 1.03 1 1.13 1 0.95 1 0.88 1 14 D265(278)A 0.07 (0.01) 4 0.09 (0.06) 4 0.01 1 S267(280)A 1.64 (0.18) .7 1.05 (0.03) 2 1.14 (0.25) 7 189 S267(280)G 1.21 (0.05) 3 0.59 (0.09) 3 0.09 (0;02) 3 16 H268(281)A 1.22 (0.14)12 1.09 (0.01) 2 0.52 (0.09)12 E283(300)A 1.24 (0.23) 5 1.01 (0.14) 5 0.78 1 226 E283(300)Q 1.12 1 1.19 1 0.89 1 227 E283(300) 1.03 1 0.5 1 0.83 228 E283(300)N 1.0318 1 0.8594 1 0.63 1 228 E283(300)N 1.18 1 0.94 1 0.63 1 229 E283(300)D 1.14 1 0.95 1 0.67 1 K290(307)A 1.29 (0.21) 7 1.28 (0.21) 7 1.12 (0.05) 2 1.13 1 73 K290(307)Q 1.17 1 1.40 1 1.02 1 1.30 1 K290(307)S 1.27 1 1.26 1 1.05 1 1.62 1 77 K290(307)E 1.10 1 1.30 1 0.98 1 1.50 1 78 K290(307)R 1.05 1 1.08 1 1.07 1 1.24 1 177 K290(307)G 1.07 1 1.23 1. 1.11 1 2.29 1 2004233493 25 Nov 2004 31 R292(309)A 0.27 (0.14) 9 0.90 (0.19) 9 0.94 1 R292(309)K 0.71 (0.17) 3 1.15 (0.18) 3 1.64 1 81 R292(309)H 0.21 (0.09) 2 0.92 (0.08) 2 1.21 1 82 R292(309)Q 0.47 (0.12) 3 0.45 (0.09) 3 0.56 1 83 R292(309)N 0.54 (0.16) 3 0.88 (0.02) 3 0.91 1 144 E293(310)Q 0.65 (0.03) 2 0.99 (0.04) 2 1.00 1 0.97 1 33 E294(311)A 0.87 (0.19) 5 0.66 (0.14) 5 0.68 1 173 E294(311)Q 1.01 1 0.84 1 0.79 1 174 E294(311)D 0.37 1 0.14 1 0.26 1 36 S298(317)A 0.40 (0.08)12 1.30 (0.18)12 1.02 (0.04) 2 1.96 1 C 70 S298(317)G 0.87 (0.17) 4 0.46 (0.09) 4 0.88 1 1.88 1 71 S298(317)T 0.41 (0.21) 3 0.89 (0.20) 3 0.96 1 0.75 1 72 S298(317)N 0.08 (0.01) 2 0.06 (0.01) 2 0.66 1 0.17 1 218 S298(317)V 0.11 (0.06) 3 0.33 (0.19) 3 0.88 1 0.39 1 219 S298(317)L 1.14 (0.12) 3 0.34 (0.04) 3 0.83 1 0.67 1 V305(324)A 1.12 (0.12) 4 1.04 1 0.84 (0.15) 4 41 T307(326)A 1.19 (0.37)12 1.37 (0.13) 2 1.12 (0.18)12 N315(334)A 1.15 (0.06) 5 1.11 (0.06) 2 1.07 (0.21) 46 K317(336)A 1.13 (0.05) 4 1.04 1 1.10 (0.23) 4 48 K320(339)A 1.12 (0.11) 4 1.16 1 0.87 (0.17) 4 54 E333(352)A 0.92 (0.12)10 1.27 (0.17)10 1.10 (0.10) 2 1.29 1 2004233493 25 Nov 2004 141 E333(352)Q 0.70 (0.05) 2 1.10 (0.03) 2 1.05 1 1.00 1 142 E333(352)N 0.59 (0.04) 2 0.56 (0.10) 2 0.64 1 0.56 1 143 E333(352)S 0.94 1 0.99 1 1.07 1 152 E333(352)K 0.85 (0.14) 3 0.88. 1 0.81 1 153 E333(352)R 0.75 (0.04) 2. 0.84 (0.05) 2 0,93 1 0.83 1 154 E333(352)D 1.26 (0.04) 3 1.00 1 1.70 1 178 E333(352)G 0.87 1 1.05 1 1.23 1 K334(353)A 1.01 (0.15)17 1.39 (0.19)17 1.07 (0.09) 3 1.60 (0.01) 2 135 K334(353)R 1.15 (0.09) 5 0.68 (0.07) 5 0.88 1 136 K334(353)Q 1.08 (0.11) 7 1.31 (0.26) 7 1.27 (0.01) 2 1.92 1 137 K334(353)N 1.16 (0.11) 7 1.15 (0.16) 7 1.19 (0.06) 2 1.70 1 138 K334(353)S 1.01 (0.11) 3 1.19 (0.08) 3 1.25 1 1.82 1 139 K334(353)E 0.74 (0.15) 4 1.30 (0.09) 4 1.17 1 2.75 1 140 K334(353)D 0.51 (0.09) 4 1.13 (0.09) 4 1.07 1 179 K334(353)G 0.76 (0.08) 5 0.88 (0.22) 5 0.94 1 1.28 1 190 K334(353)M 1.06 1 1.35 1 0.99 1 2.08 1 191 K334(353)Y 1.08 1 1.31 1 0.98 1 1.72 1 192 K334(353)W 0.94 1 1.07 1 0.96 1 1.53 1 193 K334(353)H 1.09 1 1.26 1 0.97 1 2.06 1 220 K334(353)V 1.13 (0.11) 3 1.34 (0.18) 3 1.00 1 2.89 1 221 K334(353)L 1.05 1 1.38 1 0.96 1 3.59 1 2004233493 25 Nov 2004 P331 (350) A 1.29 (0.14) 3 1.03 (0.19) 3 0.96 1 0.78 1 198 P331(350)S 1.00 1 0.06 1 0.54 1 199 P331(350)N 0.86 1 0.23 1 0.24 1 200 P331(350)E 1.06 1 0.42 1 0.36 1 203 2331(350)1< 0.94 1 0.33 i 0.26 1 96 S267(280)A 1.54 (0.12) 3 1.07 (0.06) 2 0.84 1 1-26'8(281)A 110 S298(317}A 0.35 (0.13)11 1.66 (0.42)11 1.19 (0.18) 3 E333 (352)A K334 (353)A 271 E380(405)A 0.98 1 0.92 1 1.10 1 L309(328)A For FcyRIIIA, the pattern of binding of the selected IgG1 variants to the relatively higher affinity FcyRIIIA-V158 was the same as for the relatively lower affinity FcyRIIIA-F158 (the F158 N form was used in assaying all variants). IgG1 variants which showed improved binding to the SFcyRIIIA-F158 form also showed improved binding to the FcyRIIA-V158 form though the improvement was not as pronounced. For FcyRIIA-R131 (used in assaying all variants) and FcyRilA-H131, the binding pattern of the selected IgG1 variants did show some distinct differences. S267(280)A, H268(281)A, and S267(280)A/H268(281)A exhibited improved binding e to FcyRIIA-R131, compared to native IgG1, but not to FcyRIIA-H131. In contrast, S267(280)G showed improved binding to FcyRIlA-R131 but reduced binding to FcyRIIA-H131 (Table e 10 Other variants bound similarly to both allelic FcyRIIA forms: V305(324)A, T307(326)A, N315(324)A, K317(336)A, and K320(339)A.

O

The entire disclosure in the complete specification of our Australian Patent Application No. 28505/00 is by this cross-reference incorporated into the present specification.

-It.-

Claims (22)

1. A variant of a parent polypeptide comprising a human IgG 1 Fc region, which variant mediates antibody-dependent cell-mediated cytotoxicity (ADCC) in the presence of human effector cells more effectively, or binds an Fc gamma receptor III (FcyRIII) with better affinity, than the parent polypeptide and comprises two or more amino acid substitutions in the Fc region.

2. The variant of claim 1 which mediates ADCC about 1.5 fold to about 100 fold more effectively than the parent polypeptide.

3. The variant of claim 1 which binds an FcyRIII with better affinity than the parent polypeptide.

4. The variant of claim 3 which further binds an FcyRII with worse affinity than the parent polypeptide. The variant of claim 1 which comprises at least one amino acid modification in a CH2 domain of the Fc region.

6. The variant of claim I which comprises at least one amino acid modification in the Fc region, other than at amino acid positions 233 to 239.

7. The variant of claim I which comprises an amino acid substitution at any one or more of amino acid positions 256, 290, 298, 312, 326, 330, 333, 334, 360, 378 or 430 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

8. A composition comprising the polypeptide variant of claim 1 and a pharmaceutically acceptable carrier.

9. The composition of claim 8 which is sterile. 00 O 10. A polypeptide comprising a variant Fc region which is not a native sequence Fc region and has increased binding to an Fc gamma receptor (FcyR) which polypeptide Scomprises an amino acid modification at any one or more of amino acid positions 238, 239, 248,249, 252,254, 255,256, 258,265, 267, 268, 269, 270, 272,278, 280, 283, 285,286, 289, 290, 292, 293,294, 295,296, 298, 301,303, 305, 307, 312, 315,324, 327,330, 335,337,338,340, 360, 373,376, 378, 382, 388,389, 398,414, 416, 419, 430, 434, 435, 437, 438 or 439 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

11. The polypeptide of claim 10 wherein the variant Fc region comprises a variant human IgG Fc region.

12. The polypeptide of claim 10 which displays increased binding to an FcyR and comprises an amino acid modification at any one or more of amino acid positions 255, 256, 258,267,268,272, 280, 283,285,286, 290, 298, 301,305,307,312,315, 330, 337, 340, 360, 378, 398 or 430 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

13. The polypeptide of claim 12 which displays increased binding to an FcyRIII.

14. The polypeptide of claim 13 which further displays decreased binding to an FcyRII. The polypeptide of claim 14 which displays increased binding to the FcyRIII and further displays decreased binding to the FcyRII, wherein the polypeptide comprises an amino acid modification at position 298 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

16. The polypeptide of claim 12 which displays increased binding to an FcyRII.

17. The polypeptide of claim 16 which displays increased binding to the FcyRII and comprises an amino acid modification at any one or more of amino acid 255, 256, 258,267,268,272,280, 283,285,286, 290, 301, 305, 307, 312,315,330, 337, 340, 378, 398 or 430 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat. 00 0 O 18. The polypeptide of claim 16 which further displays decreased binding to an N FcyRIII.

19. The polypeptide of claim 18, which displays increased binding to the FcyRII r C 5 and further displays decreased binding to the FcyRIII, wherein the polypeptide comprises an amino acid modification at any one or more of amino acid positions 268, 272, 298, 301, 322 or 340 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat. 10 20. A polypeptide comprising a variant Fc region which is not a native sequence Fc region and has increased binding to an Fc gamma receptor (FcyR), which polypeptide comprises an amino acid modification at any one of more of amino acid positions 256, 290, 298, 312, 330, 360, 378 or 430 of the Fc region, wherein the numbering of the residues in the Fc region is that of the EU index as in Kabat.

21. The variant of claim 7 or claim 20 which comprises two or more amino acid substitutions at the amino acid positions listed therein.

22. The variant of claim 7 or claim 20 which comprises three or more amino acid substitutions at the amino acid positions listed therein.

23. The variant of claim I or claim 20 which comprises an antibody.

24. The polypeptide of claim 20 which comprises an amino acid substitution at position 298. The polypeptide of claim 20 which comprises amino acid substitutions at two or three positions 298, 333 and 334.

26. The polypeptide of claim 25 which comprises amino acid substitutions at positions 298, 333 and 334.

27. The polypeptide of claim 26 which consists of amino acid substitutions at positions 298, 333 and 334. 00 0 O 28. The polypeptide of claim 20 which comprises a variant human IgG IgG2, IgG3 or IgG4 Fc region.

29. The polypeptide of claim 29 which comprises a variant human IgGI Fc region. C A variant of a parent polypeptide comprising a human IgG Fc region, which variant binds an Fc gamma receptor III (FcyRIII) with better affinity than the parent polypeptide and comprises two or more amino acid substitutions in the Fc region, C wherein the Fc region of the variant is not a native sequence Fc region.