WO2012146934A1 - Binding molecules with biased recognition - Google Patents

Abstract

The invention relates to Fc variants with improved binding affinity to the FcγRIIb receptor comprising modified polypeptides in the IgG CH2 region. The invention also relates to methods for increasing the binding affinity to the FcγRIIb receptor relative to other FcγR.

Description

Binding molecules with biased recognition

Introduction

IgG molecules are the most prevalent class of immunoglobulins in humans and the vast majority of therapeutic antibodies are of this class. Four subclasses of IgG proteins are found in humans: lgG1 , lgG2, lgG3 and lgG4. However a fifth, pseudo, immunoglobulin gamma gene (IGHGP) has been described (Takahashi et al., 1982). It has been suggested that IgG molecules cannot use the constant region encoded by the pseudo gamma gene because the gene lacks the class switch region necessary for it to become located 3' proximal to rearranged variable region segments (Bensmana et al., 1988, Kabat et al., 1991 ). The effector function properties of an IgG with a pseudo gamma constant region are unknown.

The Ig monomer is a "Y"-shaped molecule that consists of four polypeptide chains; two identical heavy chains and two identical light chains connected by disulfide bonds. Some parts of an antibody have unique functions. The arms of the Y form the two Fab (fragment, antigen binding) regions and contain the sites that can bind to antigens. Structurally, the fragment of an IgG antibody that consists of four domains from the two heavy chains, two CH2 domains and two CH3 domains is known as the Fc region of the antibody. This region engages with a family of receptors referred to as the Fey receptors (FcyRs). All FcyRs interact with a very similar binding site on Fc. It is known that the constant regions, and in particular the C-regions within the Fc fragment, are responsible for various effector functions of immunoglobulin. Variable regions on the other hand mediate antigen specificity. Effector functions are thus mediated by the Fc fragment and not by the variable regions that provide antigen specificity. Binding to FcyRs elicits a number of responses, for example antibody- dependent cell-mediated cytotoxicity or phagocytosis.

There are three classes of human Fey receptors (FcyR) which allow IgG to interact with cells of the immune system. In addition, the neonatal FcRn plays a role in placental transport of IgG and in preventing IgG degradation therefore prolonging the half life of circulating IgG. The various types of FcyR have different cellular distributions and their properties, including details of functionally significant polymorphisms, are reviewed by van Sorge et al. (2003). FcyRI (high affinity), FcyRllla (intermediate affinity) and FcyRlla and FcyRlllb (low affinity) are all activating receptors and engagement of these receptors can lead to inflammatory reactions including the destruction of target cells. In contrast, the low affinity FcyRI lb is the only inhibitory FcyR and, as such, is of special interest. FcyRI lb is found on B cells, where it acts to prevent activation and proliferation, on mast cells and basophils, where it inhibits FceR I -mediated degranulation, and on macrophages, monocytes and neutrophils, where it can inhibit target cell destruction.

Due to the effector function of the Fc fragment, this region has been of interest to antibody engineering to modify binding to FcyRs. Mutational studies have been carried out in different backgrounds, including lgG1 , lgG2 and lgG4, to generate Fc variants with modified binding to FcyR (WO 99/58572, WO 00/42072, WO 2005/040217, WO 96/16562, Armour et al (1999) and (2003), Clark (1997), all incorporated herein by reference). However, because all of the FcyR interact with a very similar binding site on Fc, and because of the high homology found between the FcyR, identifying mutants that bind selectively or bind with higher affinity to a specific FcyR remains a challenge.

It is generally thought that it is the balance between the interaction through activatory and inhibitory receptors that is important. In particular, increased interaction with FcyRllb, rather than the activating FcyR, would be a useful property of therapeutic IgG and other reagents in the areas of asthma, allergy and auto- and alloimmune diseases. In addition, therapeutic IgG and other reagents, such as Fc-fusion molecules, which are designed to bind to a target cell without causing cell destruction, including blocking antibodies, might benefit from increased affinity for FcyRllb. The present invention is aimed at addressing these needs and provides Fc variants with altered effector function.

Summary of the invention

The inventors have found that modification of an IgG CH2 region to include residues found in IGHGP can alter binding to FcyR, specifically FcyRllb. Specifically, these novel modifications of the CH2 region alter the effector function of the Fc domain and confer increased binding to FcyRllb. Variants of the invention with increased binding to FcyRllb may be used in binding molecules, for example in antibodies for which increased binding to FcyRllb is desirous, and, for example, certain therapeutic applications, such as autoimmune disease and allergies. Alternatively, the Fc variants of the invention may be used in improving cytotoxicity. As described herein, mutants of human lgG1 have been made that incorporate either two (G1 Ap), eight (ψγΔς) or nine (ψγ) residues encoded by an unexpressed fifth human IgG constant region gene (IGHGP) in a range of parent molecules. In experiments that assess binding to the human Fc receptors FcyRI, FcyRlla of allotypes 131 R and 131 H and FcyRllla of allotypes 158F and 158V and in an assay that measures ability to mediate lysis in the presence of human complement, G1Ap and ψγΔς have similar levels of activity. However, the binding of ψγΔς to the inhibitory receptor FcyRllb was 8-fold greater than that of G1Ap. Therefore, lgG1 with ψγΔς displays preferential binding to the inhibitory receptor FcyRllb; a property sought after for some therapeutic antibody molecules. One or more of the six amino acid residues present in ψγΔ , but not G1Ap, must confer FcyRllb specificity and can be inserted into other IgG constant regions to impart the binding preference. As also shown herein, the substitution A327G accounts for the majority of the difference in FcyRllb binding between G1Ap and ψγΔς. The substitution A327G does not appear to improve the FcyRllb binding of a wildtype lgG1 constant region, but it does improve the FcyRllb binding of mutated lgG1 constant regions.

The invention relates to the use of the unexpressed fifth human IgG constant region gene for manipulating FcyR binding of the IgG CH2 region. According to the invention, one or more of the substitutions described herein, when incorporated into an IgG molecule, can cause that IgG molecule to show stronger binding to the inhibitory Fc receptor FcyRllb, whilst, at the same time, not cause the IgG to show stronger binding to the activating Fc receptors FcyRI and FcyRllla. Thus, the invention relates to an Fc variant polypeptide that contains residues of IGHGP and shows increased selective binding to FcyRllb. Binding is increased relative to the parent or starting polypeptide which does not comprise any or all of the modifications of the invention.

The substitutions according to the invention are: residue 292 changed to W; residue 309 changed to V; residue 312 changed to N; residue 317 changed to R; residue 327 changed to G and/or residue 339 changed to T.

The residues are numbered according to the EU system (Kabat et al., 1991 ).

The amino acids are changed relative to the native (wild type) sequence as found in nature in the wild type (wt), but may be made in IgG molecules that contain other changes relative to the native sequence. Preferably, the parent molecule into which the substitutions described herein have been introduced is one that comprises a lower hinge region of the CH2 which is modified compared to a native (wt) region and is less than optimal for FcyR binding, such as a variant that comprises further amino acid modifications, as for example G1Ap. Other exemplary parent molecules that comprise different modifications are listed herein and shown in the examples. The modified binding of the Fc variant according to the invention was found when human lgG1 molecules were engineered to contain various numbers of residues that are encoded by the pseudo gamma gene within their CH2 domains. The CH2 domain has been shown to be critical in the interactions with molecules of the immune system which mediate the IgG effector functions; whether this is through the activation of complement or engagement of Fc receptors (FcyR; reviewed by Clark, 1997). We therefore describe herein the production of mutant IgG and comparison to wild type lgG1 , lgG2 and lgG4 molecules for their ability to bind to Fc receptors and to activate complement.

Therefore, in a first aspect, the invention relates to a variant Fc polypeptide comprising a modified human IgG CH2 region compared to a wild type CH2 region comprising at least one amino acid substitution selected from the group consisting of 292W, 309V, 312N, 317R, 327G and 339T where the residues are numbered according to the EU system and wherein the at least one amino acid substitution does not increase binding to human FcyRI and/or FcyRllla and increases binding to FcyRllb. Thus, relative binding to FcyRllb is increased.

In a second aspect, the invention relates to a method for producing a variant Fc polypeptide having increased selective binding affinity for the FcyRllb receptor wherein said method comprises modifying a polypeptide comprising a human IgG CH2 region by substitution of one or more amino acids with the following amino acids at the following positions: 292W, 309V, 312N, 317R, 327G and 339T.

In a further aspect, the invention relates to a method for increasing the relative binding affinity of a Fc region for the FcyRllb receptor wherein said method comprises modifying a polypeptide comprising a human IgG CH2 region by substitution of one or more amino acids with the following amino acids at the following positions: 292W, 309V, 312N, 317R, 327G and 339T.

In another aspect, the invention relates to a method for engaging an FcyRllb receptor comprising contacting a cell comprising an FcyRllb receptor with an Fc variant as described herein.

In another aspect of the invention, the Fc variant, antibody, fusion protein or pharmaceutical formulation described herein may be used in the treatment of a disease. The invention thus also relates to an Fc variant, antibody, fusion protein pharmaceutical formulation described herein in the manufacture of a medicament for the treatment of a disease and to a method of treatment of a disease, such as an auto- or alloimimmune disease, asthma or an allergy, by administration of a Fc variant as described herein to a patient in need thereof. The invention also relates to a nucleic acid sequence encoding a polypeptide variant of the invention, a vector comprising such nucleic acid sequence a host cell transformed with such a vector.

Descriptions of Figures

The invention is further described in the non-limiting figures below. Figure 1

Alignment of the human lgG1 constant region amino acid sequence with those of the mutants ψγ, Θ1 Δρ and HJyAq. The asterisks indicate positions at which all four constant regions have the same amino acid residues.

Figure 2

Comparison of human lgG1 CH2 domain amino acid sequence with those of the mutants ψγ, Θ1 Δρ and i|JYAq, which highlights the differences between the sequences. A dot indicates identity with the lgG1 sequence residue. Residues are numbered according to the EU system (Kabat et ai, 1991 ).

Figure 3

Reducing SDS-PAGE of antibody samples. Samples (1 μg unless stated) were prepared in 5% 2-mercaptoethanol and 2% SDS and heated at 95C to reduced and denature the proteins. Standards were pre-stained broad-range protein markers (New England Biolabs, Hitchin, UK). Gels were stained using Coomassie Brilliant Blue.Part a: 12.5% acrylamide separating gel. Lane 1 , lgG1 ; lane 2, ψγ; lane 3, G1 Ap; lane 4, ijJYAq; lane 5, 1 .6 μg lgG2; lane 6, lgG2; lane 7, 0.86 μg lgG4; lane 8, 1 .3 μg lgG4; lane 9, lgG1 ; lane 10, standards. Part b: 8.5% acrylamide separating gel. Lane 1 , lgG1 ; lane 2, G1 Ap; lane 3, ijJYAq; lane 4, ψγ; lane 5, lgG1 ; lane 6, standards.

Figure 4

Binding of monomeric IgG samples to human FCYRI. Cells of the line B2KA, which expresses FCYRI, were sequentially incubated in dilutions of test antibodies, biotin- conjugated goat anti-human κ light chain antibodies and FITC-conjugated ExtrAvidin. Samples were analysed by flow cytometry. For each sample, geometric mean fluorescence of 20 000 cells was determined and plotted against antibody concentration.

Figure 5 Binding of pre-complexed IgG samples to human FcyRlla, allotype 131 R. CHO cells, which had been transfected to express FcyRlla of the 131 R allotype, were incubated with test antibodies that were cross-linked by goat F(ab')₂ anti-human κ molecules and then with FITC-conjugated F(ab')₂ fragments of rabbit anti-goat IgG, F(ab')₂-specific antibodies. Samples were analysed by flow cytometry. For each sample, geometric mean fluorescence of 20 000 cells was determined and plotted against antibody concentration.

Figure 6

Binding of pre-complexed IgG samples to human FcyRlla, allotype 131 H. As figure 5 but using CHO cells that had been transfected to express FcyRlla of the 131 H allotype.

Figure 7

Binding of pre-complexed IgG samples to human FcyRllb. CHO cells, which had been transfected to express FcyRllb, were incubated with test antibodies that were cross-linked by FITC-conjugated goat F(ab')₂ anti-human κ molecules. Samples were analysed by flow cytometry. For each sample, geometric mean fluorescence of 20 000 cells was determined and plotted against antibody concentration.

Figure 8

Binding of pre-complexed IgG samples to human FcyRllla, allotype 158F. As figure 5 but using CHO cells that had been transfected to express FcyRllla of the 158F allotype.

Figure 9

Binding of pre-complexed IgG samples to human FcyRllla, allotype 158V. As figure 5 but using CHO cells that had been transfected to express FcyRllla of the 158V allotype.

Figure 10

Ability of IgG to activate complement-mediated cell lysis. Antibodies to antigens such as the CD52 antigen as exemplified by the CAMPATH-1 have been used in the art. The CAMPATH-1 antibody was originally developed as an IgM antibody which was very effective in lysing lymphocytes in-vitro using human serum as a complement source (Hale et al 1983). The antigen was identified as CD52 which is a small GPI-anchored glycoprotein expressed by lymphocytes and monocytes but not by haemopioetic stem cells. It represents an exceptionally good target for complement lysis. CAMPATH-1 H IgG (CD52 specificity) were incubated with human PBMC and complement for 1 .5 h at 37C, the cell proliferation reagent, WST-1 , was added and incubated continued for 4 h. 450 nm absorbance readings correlate with numbers of metabolically active cells and are shown as mean (± SD) for IgG tested in triplicate.

Part a: A450 reading for no antibody control was 0.566 (± 0.031 ) and for no cell control was 0.010 (± 0.009).

Part b: A450 reading for no antibody control was 1 .266 (± 0.228) and for no cell control was 0.005 (± 0.007).

Figure 1 1

Non-limiting examples of monolclonal antibodies in clinical development (Glennie and Winkel, supplementary material, 2003).

Figure 12

This compares the FcyRllb binding of wildtype lgG1 (G1 ) with G1Ar. The assay was carried out as described for figure 7.

Figure 13

This compares the FcyRllb binding of G1Ae with G1 Aer. The assay was carried out as described for figure 7.

Figure 14

This compares the FcyRllb binding of G1 Ab with G1 Abr. The assay was carried out as described for figure 7.

Figure 15

This compares the FcyRllb binding of G1 Ap with G1 Apr and also ψγΔς. The assay was carried out as described for figure 7.

Detailed disclosure of the invention

The present invention will now be further described. In the following passages, different aspects of the invention are defined in more detail. Each aspect so defined may be combined with any other aspect or aspects unless clearly indicated to the contrary. In particular, any feature indicated as being preferred or advantageous may be combined with any other feature or features indicated as being preferred or advantageous.

The practice of the present invention will employ, unless otherwise indicated, conventional techniques of immunology, molecular biology, chemistry, biochemistry, recombinant DNA technology, which are within the skill of the art. Such techniques are explained fully in the literature.

In a first aspect, the invention relates to an Fc variant polypeptide comprising a modified human IgG CH2 region compared to a wild type CH2 region comprising at least one amino acid substitution selected from the group consisting of 292W, 309V, 312N, 317R, 327G and 339T. The numbering of the residues in the IgG Fc region as used herein is that of the EU index as in Kabat et al (1991 ).

As shown herein, one or more of the substitutions selected from the group consisting of 292W, 309V, 312N, 317R, 327G and 339T mediates specific binding to FcyRllb. Thus, the variant of the invention shows increased relative binding affinity to FcyRllb compared to other FcyR. Specifically, the substitutions as defined above do not substantially increase relative binding to FcyRI and FcyRllla, but increase relative binding to FcyRllb when compared to a parent molecule.

This is shown in figures 4-15 and in the examples. In the case of the FcyRllb receptor, the binding of ψγΔς (which incorporates eight residues encoded by IGHGP) was approximately 8-fold higher than that of G1 Ap (which incorporates two residues encoded by IGHGP). The difference in binding between ψγΔς and G1 Ap must be due to one or more of the six amino acid residues which differ between ψγΔς and G1 Aq. Accordingly, one or more of the substitutions selected from the group consisting of 292W, 309V, 312N, 317R, 327G and 339T mediates specific binding to FcyRllb and can be used in an IgG CH2 domain to confer specific binding of the domain to FcyRllb whilst at the same time not increasing binding to FcyRI and/or FcyRllla. An Fc variant according to the invention comprising one or more or all of substitutions selected from the group consisting of 292W, 309V, 312N, 317R, 327G and 339T thus confers increased relative binding to FcyRllb whilst not increasing binding to FcyRI and/or FcyRllla.

Binding is assessed relative to the parent or starting polypeptide which does not have said substitution(s). The parent or starting polypeptide may be IgG, preferably human IgG, or an engineered variant. In one embodiment, said parent polypeptide comprises further modifications which alter Fc binding to Fc receptors and/or other modifications. Preferred parent polypeptides are set out below.

Binding ability can be measured by methods described in the art, for example in WO 2005/040217 and WO 99/58572). An increase in affinity or binding ability may be at least 2, 3, 4, 5, 6, 7, 8, 9 or 10 fold or more. Preferably, increased binding is defined to have occurred when a lower concentration of variant IgG than original (parent) IgG- from which the variant IgG is derived and which does not comprise substitution(s) of the invention-is needed to achieve the same level of binding signal. The fold difference in binding is the difference between the two concentrations which give the same binding signal.

The Fc variant polypeptide of the invention may comprise all of the following substitutions 292W, 309V, 312N, 317R, 327G and 339T. In other embodiments, the Fc variant polypeptide may comprise only one of 292W, 309V, 312N, 317R, 327G or 339T. In another embodiment, the variant comprises two, three, four or five of the substitutions selected from 292W, 309V, 312N, 317R, 327G or 339T. Thus, any of these substitutions and possible combinations thereof may be included in a variant Fc polypeptide of the invention. In one embodiment, the Fc variant polypeptide comprises the substitutions 327G and/or 339T. In a preferred embodiment, the variant CH2 region comprises the substitution 327G. In one embodiment, the variant CH2 comprises the substitution 327G, but none of the other substitutions of the invention. In another embodiment, the variant CH2 region comprises the substitution 327G combined with 292W, 309V, 312N, 317R and/or 339T.

In one embodiment, the CH2 region of the Fc variant polypeptide does not comprise any additional amino acid modifications in addition to the one or more of the substitutions listed above. In another embodiment, the substitutions the Fc variant polypeptide comprises one or more additional amino acid modifications, for example a substitution, deletion or addition of one or more amino acids. Thus, the variant CH2 region may include in addition to one or more of the substitutions listed above, for example in addition to the substitution 327G, 1 , 2, 3, 4, 5, 6, 7, 8 or nine additional substitutions when compared with the native (wt) CH2 region. These additional modifications may preferably result in optimised variants for therapeutic properties with altered affinity for one or more FcyR, altered affinity for one or more FcyR at different pH values or increased/decreased half life. The additional modifications may preferably reside in the CH2 domain or other parts of the Fc variant as explained below although other modifications, for example modifications that improve IgG stability, solubility or other desired properties, are also within the scope of the invention.

Preferred additional modifications in the variant Fc region present in addition to one or more of the changes listed above include, but are not limited to, the following modifications of human lgG1 :

(i) P232T and L234P;

(ii) H268E; (iii) E233P, L234V, L235A and deletion of G236 or

(iv) E233P, L234V and L235A.

Also included are 'null allotype' variants of human IgG molecules (Greenwood et al 1993).

The variant Fc polypeptide may be derived from any parent human IgG CH2 region. Preferably, the variant is derived from a human lgG1 , lgG2, lgG3 or lgG4 CH2 region. In one embodiment, the variant is derived from lgG1 . The variant Fc polypeptide may be derived from a wild type CH2 region, but may also be derived from a parent polypeptide with a mutant CH2 region resulting in an Fc polypeptide with one or more additional modification(s), for example as defined in (i) to (iv) above.

Thus, in one embodiment, the variant polypeptide comprises the substitution A327G and is derived from a parent polypeptide comprising one of the following modifications:

(i) P232T and L234P;

(ii) H268E;

(iii) E233P, L234V, L235A and deletion of G236 or

(iv) E233P, L234V and L235A.

In a further embodiment, the variant polypeptide comprises the substitution A327G and is derived from G1 Ae (lgG1 containing H268E substitution), giving G1Aer, G1Ab (lgG1 containing substitutions E233P L234V L235A and G236 deleted), giving G1 Abr or G1Ap (lgG1 containing P232T and L234P substitutions), giving G1 Apr.

In one embodiment, the invention relates to a polypeptide designated G1Ar, G1Aer, G1 Abr and G1 Apr as shown in figure 1 .

In one embodiment, the variant polypeptide may consist, or consist essentially of, the CH2 sequences discussed above. However, Fc variant polypeptides that comprise additional sequences, domains or regions are also within the scope of the invention. Thus, in one embodiment, the variant polypeptide may comprise the CH2 sequences discussed above and additionally further sequences. In one embodiment, the modifications made in the variant Fc polypeptide consist of the substitution A327G in combination with one of the modifications listed in (i) to (iv), but no other changes compared to the native CH2 sequences. In another embodiment, additional modifications compared to the native sequence are incorporated. According to the different aspects of the invention, the variant polypeptide of the invention may comprise an entire Fc fragment. Also, the variant polypeptide of the invention may comprise the entire constant region of a human IgG heavy chain, comprising the CH2 sequences described herein. Thus, any of the CH2 sequences described herein may be combined with natural or modified CH3, natural or modified hinge region and CH1 sequences. Thus, for example, a variant polypeptide according to the invention may comprise the human IgG CH2, CH1 and CH3 regions.

The invention also relates to the use of a variant Fc region as described herein in conferring increased binding to FcyRllb whilst at the same time not exhibiting increased binding to FcyRI and/or FcyRllla compared to binding of the parent polypeptide. Variants according to the invention may therefore be used in binding molecules where increased binding to FcyRllb (compared to a parent molecule) is desirable.

Furthermore, in another aspect of the invention, the variant Fc polypeptide as described herein may be combined with a binding domain capable of specifically binding to a target molecule. Thus, in aspect, the invention relates to fusion proteins comprising a variant Fc region as described herein. In the fusion protein, one or more polypeptide is operably linked to an Fc region of the invention. The invention also relates to the use of the Fc variant as defined herein in a method for making a fusion protein with increased binding to FcyRllb. For example, the Fc variant of the invention may be linked to a Fab binding domain. The binding domain may comprise more than one polypeptide chain in association e.g. covalent or otherwise (e.g. hydrophobic interaction, ionic interaction, or linked via sulphide bridges). For instance, it may comprise a light chain in conjunction with a heavy chain. Thus, in one embodiment, the fusion protein may be an antibody and the invention may thus encompasses an antibody with a variant Fc polypeptide described herein.

Generally, the one or more polypeptide operably linked to an Fc region of the invention may be any protein or small molecule, for example a binding domain derived from any molecule with specificity for another molecule and capable of binding said molecule. The binding domain will have an ability to interact with a target molecule which will preferably be another polypeptide, but may be any target (e.g. carbohydrate, lipid (such as phospholipid) or nucleic acid). Preferably, the interaction will be specific. Typically, the target will be antigen present on a cell, or a receptor with a soluble ligand. This may be selected as being a therapeutic target, whereby it is desired to bind it with a molecule having the properties discussed above. The target may be present on or in a target cell, for example a target cell which it is desired to lyse, or in which it is desired to induce apoptosis. Protein fusion partners may thus include, but are not limited to, the variable region of any antibody, the target-binding region of a receptor, an adhesion molecule, a ligand, an enzyme, a cytokine, an antigen, a chemokine, or any other protein or protein domain. Small molecule fusion partners may include any therapeutic agent that directs the Fc fusion to a therapeutic target. Such targets may be any molecule, preferably an extracellular receptor, which is implicated in disease. Thus, the invention relates to an immunoadhesin comprising immunoglobulin constant domains, including the Fc variant as described herein, and a binding domain capable of binding a target molecule.

The one or more polypeptide operably linked to an Fc region or protein fusion partners may be based on native sequences or may comprise modifications that are not present in native sequences.

The term "antibody" as used herein covers monoclonal antibodies (including full length monoclonal antibodies), polyclonal antibodies, multispecific antibodies (e.g., bispecific antibodies), and antibody fragments so long as they exhibit the desired biological activity. Thus the term includes molecules having more than one type of binding domain, such as bispecific antibodies.

"Humanized" forms of non-human (e.g., murine) antibodies are chimeric antibodies that contain minimal sequence derived from non- human immunoglobulin. For example, 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 desired specificity, affinity, and capacity.

Methods of producing antibodies (and hence binding domains) are well known to the skilled person. They include immunising a mammal (e.g. human, mouse, rat, rabbit, horse, goat, sheep, camel or monkey) with a suitable target protein or a fragment thereof. Antibodies may be obtained from immunised animals using any of a variety of techniques known in the art, and might be screened, preferably using binding of antibody to antigen of interest. For instance, Western blotting techniques or immunoprecipitation may be used.

The antibody or fusion protein described herein may be conjugated to another molecule. Suitable conjugates include, but are not limited to, labels as described below, drugs and cytotoxic agents including, but not limited to, cytotoxic drugs (e.g., chemotherapeutic agents) or toxins or active fragments of such toxins. Suitable toxins and their corresponding fragments include diphtheria A chain, exotoxin A chain, ricin A chain, abrin A chain, curcin, crotin, phenomycin, enomycin and the like. Cytotoxic agents also include radiochemicals made by conjugating radioisotopes to antibodies, or binding of a radionuclide to a chelating agent that has been covalently attached to the antibody. Additional embodiments utilize calicheamicin, auristatins, geldanamycin, maytansine, and duocarmycins and analogs. Labels may include immune labels, (e.g. an epitope incorporated as a fusion partner that is recognized by an antibody, isotopic labels (e.g. radioactive or heavy isotopes), magnetic labels and small molecule labels (e.g. fluorescent and calorimetric dyes, enzymatic groups or molecules such as biotin that enable other labeling methods). Covalent modifications are also within the scope of the invention. Such modifications may include glycosylation.

The binding molecules of the present invention comprise a binding domain capable of binding to a target molecule, including an antigen. Virtually any antigen may be targeted to the variants of the invention, depending on the desired application. The variant Fc polypeptides of the invention may be incorporated into clinical candidates or products that have been approved for use. Such products may include therapeutic antibodies. These are well known in the art and for example disclosed in a recent review by Reichert, 201 1 (in particular in tables 1 to 8 and most particularly tables 3, 4, 5 and 6) which is incorporated herein by reference in its entirety. Examples of such candidates or products which may be used according to the invention are also shown herein in figure 1 1 (Glennie et al, 2003). In particular, variants of the invention are useful in applications where blocking or inhibition is of more importance than is direct cyctoxity or killing.

For example, the FC polypeptides of the invention may find use in an antibody that is substantially similar to an antibody or Fc fusion protein selected from the following non- limiting list Naptumomab estafenatox, Girentuximab, Zanolimumab, Obinutuzumab, Brentuximab vedotin, Tremelimumab, Zalutumumab, Farletuzumab, Trastuzumab emtansine, Dalotuzumab, Ramucirumab, Ofatumumab, Tositumomab-1131 , Ibritumomab tiuxetan, Rituximab, Gemtuzumab ozogamicin, Alemtuzumab, Ipilimumab, Panitumumab, Cetuximab, Trastuzumab, Bevacizumab, Vedolizumab, Otelixizumab, Teplizumab, Epratuzumab, Reslizumab, REGN88, Briakinumab, AIN-457, Belimumab, Eculiziumab, Muromonab-CD3, Basiliximab, Daclizumab, Efalizumab, Tocilizumab, Ustekinumab, Canakinumab, Omalizumab, Natalizumab,Golimumab, Certolizumab pegol, Adalimumab, Infliximab, Pagibaximab, Solanezumab, Bapineuzumab, Raxibacumab, Abciximab, Denosumab, Motavizumab, Palivizumab, Ranibizumab, Aflibercept, AMG 386, Atacicept, Factor Vlll-Fc, Factor IX-Fc, Belatacept, Abatacept, Rilonacept, Alefacept, Etanercept and Romiplostim. The variants of the invention are prepared by methods known to the skilled person. Sequences for human C regions have been published (Clark 1997) and can be obtained from databases such as Swiss Prot. DNA encoding the variants is prepared by methods known in the art, which include, for example, preparation by site-directed mutagenesis. The invention also relates to a pharmaceutical formulation comprising an Fc variant, antibody or fusion protein described herein and a suitable pharmaceutical carrier. Such carriers are known in the art.

In another aspect of the invention, the Fc variant, antibody, fusion protein pharmaceutical formulation described herein may be used in the treatment of a disease. The invention also relates to an Fc variant, antibody, fusion protein pharmaceutical formulation described herein in the manufacture of a medicament for the treatment of a disease. In one embodiment, the invention relates to an Fc variant, antibody, fusion protein or pharmaceutical formulation described herein for the treatment of a disease. Preferably, the disease is an auto- or alloimimmune disease, asthma or an allergy. The invention also relates to a method of treatment of a disease, such as an auto- or alloimimmune disease, asthma or an allergy, by administration of a Fc variant as described herein to a patient in need thereof.

A non-limiting list of autoimmune diseases includes allogenic islet graft rejection, alopecia, ankylosing spondylitis, antiphospholipid syndrome, autoimmune Addison's disease, antineutrophil cytoplasmic autoantibodies (ANCA), autoimmune diseases of the adrenal gland, autoimmune hemolyticanemia, autoimmune hepatitis, autoimmune myocarditis, autoimmune neutropenia, autoimmune oophoritis and orchitis, autoimmune thrombocytopenia, autoimmune urticaria, Behcet's disease, bullous pemphigoid, cardiomyopathy, Castleman's syndrome, celiac spruce-dermatitis, chronic fatigue immune disfunction syndrome, chronic inflammatory demyelinating polyneuropathy, Churg-Strauss syndrome, cicatricalpemphigoid, CREST syndrome, cold agglutinin disease, Crohn's disease, dermatomyositis, discoid lupus, essential mixed cryoglobulinemia, factor VIII deficiency, fibromyalgia-fibromyositis, glomerulonephritis, Grave's disease, Guillain-Barre, Goodpasture's syndrome, graft-versus-host disease (GVHD), Hashimoto's thyroiditis, hemophilia A, idiopathic pulmonary fibrosis, idiopathic thrombocytopenia purpura (ITP), IgA neuropathy, IgM polyneuropathies, immune mediated thrombocytopenia, juvenile arthritis, Kawasaki's disease, lichen plantus, lupus erthematosis, Meniere's disease, mixed connective tissue disease, multiple sclerosis, type I diabetes mellitus, myasthenia gravis, pemphigus vulgaris, pernicious anemia, polyarteritisnodosa, polychrondritis, polyglandular syndromes, polymyalgia rheumatica, polymyositis and dermatomyositis, primary agammaglobinulinemia, primary biliary cirrhosis, psoriasis, psoriatic arthritis, Reynauld's phenomenon, Reiter's syndrome, rheumatoid arthritis, sarcoidosis, scleroderma, Sjorgen's syndrome, solid organ transplant rejection, stiff-man syndrome, systemic lupus erythematosus, takayasu arteritis, temporal arteristis/giant cell arteritis, thrombotic thrombocytopenia purpura, ulcerative colitis, uveitis, vasculitides such as dermatitis herpetiformisvasculitis, vitiligo, and Wegner's granulomatosis.

Allergies include, but are not limited to, allergic rhinitis, eczema and food allergies.

The Fc variant, antibody, fusion protein or pharmaceutical formulation described herein is administered at a therapeutically effective dosage. The concentration of the therapeutically active IgG variant may vary from about 0.1 to 100mg/kg body weight, depending on the therapy. The dosage depends on the disease to be treated and other factors, such as body weight, diet etc. Administration of the pharmaceutical composition comprising an Fc variant of the present invention, preferably in the form of a sterile aqueous solution, may be done in a variety of ways, including, but not limited to orally, subcutaneously, intravenously, intranasally, intraotically, transdermal^, topically (e.g., gels, salves, lotions, creams, etc.), intraperitoneal^, intramuscularly, intrapulmonary, vaginally, parenterally, rectally, or intraocularly. In some instances, for example for the treatment of wounds, inflammation, etc., the Fc variant may be directly applied as a solution or spray. As is known in the art, the pharmaceutical composition may be formulated accordingly depending upon the manner of introduction.

In another aspect, the invention relates to a method for engaging an FcyRllb receptor comprising contacting a cell comprising an FcyRllb receptor with an Fc variant as described herein.

In another aspect, the invention relates to methods for engineering optimised FcyRllb ligands. Specifically, the invention relates to a method for producing a variant Fc polypeptide having increased selective binding affinity for the FcyRllb receptor. In another aspect, the invention relates to increasing the binding affinity of a Fc region for a Fey receptor. These methods comprises modifying a polypeptide comprising a human IgG CH2 region by substitution of one or more amino acids at the following positions: 292 with W, 309 with V, 312 with N, 317 with R, 327 with G and/or 339 with T. The method comprises providing a nucleic acid comprising a polynucleotide sequence encoding a human IgG CH2 region, modifying the codon corresponding to one or more of positions 292, 309, 312, 317, 327 and/or 339 to encode the following amino acids 292W, 309V, 312N, 317R, 327G and/or 339T and allowing expression of said modified polynucleotide sequence in a host cell to produce the variant of the invention. The methods may comprise introducing one or more of the modifications above. In another embodiment, two, three, four or five or all of the substitutions selected from 292W, 309V, 312N, 317R, 327G or 339T are made. Thus, any of these substitutions and possible combination thereof may be combined in a variant Fc polypeptide of the invention. In one embodiment, the substitutions 327G and/or 339T are made. In a preferred embodiment, the substitution 327G is made. In one embodiment the substitution 327G, but none of the other substitutions are made. In another embodiment, the variant CH2 region comprises the substitution 327G combined with 292W, 309V, 312N, 317R and/or 339T.

In one embodiment, further substitutions may be introduced for example the following modifications of human lgG1 :

(i) P232T and L234P;

(ii) H268E;

(iii) E233P, L234V, L235A and deletion of G236 or

(iv) E233P, L234V and L235A.

In one embodiment, the substitution A327G is made and one of (i) to (iv). In another aspect, the invention relates to an isolated nucleic acid sequence encoding an Fc variant polypeptide with one or more modifications described herein. This is for example as defined in SEQ ID No. 1 .

Also within the scope of the invention is a vector comprising such nucleic acid sequence and a host cell expressing a polypeptide as disclosed herein.

In another embodiment the invention relates to an Fc variant obtained or obtainable by a method described above.

In another embodiment, the invention relates to the use of a modified CH2 domain in increasing binding of an Fc polypeptide or a protein comprising an Fc region to the FcyRllb receptor. The modified Fc domain comprises one or more residues derived from the IGHGP. Specifically, the modified Fc domain comprises one or more amino acid substitution selected from the group consisting of 292W, 309V, 312N, 317R, 327G and 339T. Preferably, the modified Fc domain comprises 327G. As with other aspects of the invention, additional modifications may be present in the variant.

The disclosure of all references cited herein is specifically incorporated by reference.

Examples The invention is further described in the non-limiting examples below. Preparation and basic characterisation of antibodies

The preparation of antibodies with human lgG1 , lgG2 and lgG4 constant regions and variable regions that are identical to those of the humanised CAMPATH-1 H and specific for the antigen CD52 has already been described (Armour et al., 1999). Three novel antibodies were constructed with the same variable regions and mutated human lgG1 constant regions. These were:

i) ψγ wherein the CH2 domain is that encoded by the human pseudo gamma gene and as such contains nine amino acid residues that differ from the wild type human lgG1 sequence (see figures 1 and 2). The gene encoding this constant region was constructed by site-directed mutagenesis of DNA encoding the human lgG2 CH2 domain and substitution of this DNA into an expression vector for CAMPATH-1 H lgG1 .

ii) G1 Ap wherein only the lower hinge region of the CH2 domain is that encoded by the human pseudo gamma gene and as such contains two amino acid residues that differ from the wild type human lgG1 sequence (see figures 1 and 2). The gene encoding this constant region was constructed by site-directed mutagenesis of the expression vector for CAMPATH-1 H lgG1 .

iii) ijjyAq wherein the CH2 domain is that encoded by the human pseudo gamma gene except at residue position 301 (EU numbering system; Kabat et al., 1991 ) where the human lgG1 residue was retained. This antibody thus contains eight amino acid residues that differ from the wild type human lgG1 sequence (see figures 1 and 2) and the gene encoding this constant region was constructed by site- directed mutagenesis of the expression vector for CAMPATH-1 H ψγ.

We also investigated the effect of the modification A327G.

We have made several lgG1 molecules in which residue 327 has been changed to G. This is one of the substitutions of the invention. We have made the substitution A327G, which we have named Ar, on variants of the human lgG1 constant regions, namely: lgG1 wildtype, giving G1Ar

G1 Ae (lgG1 containing H268E substitution), giving G1 Aer

G1 Ab (lgG1 containing substitutions E233P L234V L235A and G236 deleted), giving G1 Abr G1 Ap (lgG1 containing P232T and L234P substitutions), giving G1 Apr. The expression vectors for the novel heavy chains were each co-transfected with a vector for the expression of the CAMPATH-1 H kappa light chain into the rat myeloma cell line, YB2/0, and antibody produced as previously described (Armour et al., 1999).

A sandwich ELISA, which employed goat anti-human IgG (Fc-specific) antibodies and HRPO-conjugated, goat anti-human κ light chain antibodies (Sigma-Aldrich, Poole, UK), was used to confirm that the concentrations of all six antibodies were correct relative to each other. Reducing SDS-PAGE (Laemmli, 1970) was used to assess the sizes and integrity of the antibody heavy and light chain. Figures 3a and 3b show representative polyacrylamide gel images. Figure 3a illustrates that each antibody sample contained intact heavy and light chains with apparent relative molecular weights of approximately 50-55 kDa and 25kDa respectively. The polyacrylamide gel of figure 3b was subjected to a longer electrophoresis period than that of figure 3a and the light chain molecules have been lost off the bottom of the gel. This enables easier visualisation of the apparent relative molecular weights of the mutant heavy chains. It can be seen that the ψγ heavy chain has the highest molecular weight, the heavy chains of ψγΔς and G1 are of the lowest and similar molecular weights and G1 Ap has an intermediate molecular weight.

Measurement of antibody binding to FcyR

The ability of the antibodies to bind to various human FcyR was measured using cell lines that have been transfected with appropriate cDNA expression vector constructs to express a single type of FcyR on their surface. Antibody binding was detected using fluorescently- labelled reagents and flow cytometry. For each assay, cells were detached from flasks using Cell Dissociation Buffer (Invitrogen, Paisley, UK), collected by centrifugation and washed in PBS containing 0.1 % (w/v) bovine serum albumin, 0.1 % (w/v) sodium azide (wash buffer). Cells were re-suspended in 1 ml wash buffer and 1 ml ACCUMAX™ (PAA, Yeovil, UK) and incubated for 8min at 20C. Cells were diluted to approximately 1 x10^s cells/ml using wash buffer and 100 μΙ samples distributed to wells of a 96-well round-bottomed assay plate. The cells were pelleted by centrifugation before being resuspended in the prepared antibody solutions as detailed below.

For the high affinity receptor, FcyRI, the cell line was B2KA (S. Gorman and G. Hale, unpublished) and the binding of monomeric IgG was measured essentially as previously described (Armour et al., 1999). Briefly, three-fold dilutions of test antibodies were made from 100 μg/ml in wash buffer and 100 μΙ of each sample used to resuspend the cells. Cells were incubated with the antibodies for 30 min on ice and then washed three times with 150 μΙ/well wash buffer. Subsequent incubations were in 5 μ9ΛηΙ biotin-conjugated goat anti- human K light chain antibodies (AbDSerotec, Oxford, UK) and in 17 μg/ml fluorosceinisothiocyanate (FITC)-conjugated ExtrAvidin (Sigma-Aldrich). After the final wash step, the cells were fixed in wash buffer containing 1 % (w/v) formaldehyde and kept at 4C until analysis.

A CyAn ADP flow cytometer and Summit v4.3 software (DakoCytomation, Ely, UK) were used to analyse the samples. For each sample, the geometric mean of fluorescence of 20 000 cells was determined after gating based on forward scatter, side scatter and pulse width parameters. Mean fluorescence was plotted against concentration for each antibody and the graph shown (Figure 4) is typical of the results obtained from replicate experiments.

This receptor has the greatest affinity for IgG of the FcyR (van Sorge et al., 2003). If the binding is assumed to have reached equilibrium then the midpoint of the binding curve is an approximate measure of the affinity of the interaction. For lgG1 in Figure 4, the midpoint, at about 0.5 μg ml, corresponds to an affinity that is in agreement with the expected K_a of 10⁸- 10⁹ M^" . The binding hierarchy for the wild type antibodies is also as expected with lgG4 showing lower binding than lgG1 whilst the signal for lgG2 binding is no better than that of a control sample without test antibody. Of the mutant IgG, G1 Ap showed the strongest binding, which was slightly less than that of lgG4, but a little higher than ijjyAq binding. The binding of ψγ was a further 5-fold lower than Aq but still gave measurable binding at 1 μg/ml.

FcyRllla of allotypes 158F and 158V were expressed as GPI-anchored receptors in CHO (Armour et al., 2010). FcyRlla of allotypes 131 R and 131 H and FcyRllb were also expressed in CHO cells but as transmembrane proteins with native cytoplasmic domains. Briefly, cDNA was synthesised from human PBMC RNA using specific primers and amplified by nested PCR to yield Hind 111 - Xbal DNAs, which comprised the whole receptor coding region including signal sequence and cytoplasmic domain. These were inserted into pcDNA3.1 /Hygro(+) (Invitrogen). cDNA encoding FcyRlla 131 H was not obtained directly but by mutation of FcyRlla 131 R DNA. In the case of FcyRllb, only clones containing the highly homologous FcyRllc cDNA were obtained but limited mutation provided a cDNA encoding a cytoplasmic domain identical to that of FcyRllbl and FcyRllb2 and the cytoplasmic domain of FcyRllb2. Vector DNAs were transfected into CHO cells and receptor-expressing clones isolated as described for the FcyRI I la-expressing cell lines (Armour et al, 2010).

Binding to FcyRI I and III receptors, which were of low and medium affinity, was measured by pre-complexing the test antibodies with equimolar amounts of F(ab')₂ fragments, which recognised the κ chain, before their addition to the cells (Armour et al., 2003). Human lgA1 , κ purified myeloma protein (The Binding Site, Birmingham, UK) was used as a negative control test antibody. For FcyRllb, antibodies were cross-linked with FITC-conjugated goat F(ab')₂ anti-human κ (AbDSerotec) so that an extra detection step was not needed. For the other receptors, complexes were made with goat F(ab')₂ anti-human κ (Rockland, Gilbertsville, PA, USA) and complex binding was detected using FITC-conjugated F(ab')₂ fragments of rabbit anti-goat IgG, F(ab')₂-specific antibodies (Jackson Immuno Research, Newmarket, UK). Samples were analysed by flow cytometry as described for FcyRI above.

For FcyRlla of allotype 131 R (figure 5), the binding hierarchy is lgG1 >ψγΔς> lgG4 ~ G1 Ap > lgG2 >ψγ ~ IgA (background level). For the 131 H allotype of the same receptor (figure 6), the order of binding is lgG1 > lgG2 >ψγΔς with the remaining test antibodies having only background levels of binding.

The binding hierarchy for the inhibitory receptor, FcyRllb, was lgG1 >ψγΔς = lgG4 > G1 Ap = lgG2≥ ψγ> IgA. Here, ψγΔς binds about 8-fold better than G1 Ap.

Binding to cells expressing FcYRIIIa molecules follows the same pattern in both allotypes of the receptor: 158F and 158V (figures 8 and 9 respectively). The hierarchy is lgG1 > G1 Ap ~ ψγΔς> lgG4 > lgG2 >ψγ. However, relative to lgG1 , the other test antibodies show a greater reduction in the binding with the 158F molecule than with the 158V molecule.

As demonstrated in figures 12-15, the parent constant regions G1 Ae, G1Ab and G1 Ap have different levels of FcYRIIb binding in comparison to wildtype lgG1 . These parent constant regions and their variants with the Ar mutation, were combined with a CAMPATH-1 H heavy chain variable region and kappa chain in CD52-specific IgG molecules. The Ar variants were compared to their parent IgG from which they derived in binding to FCYRI lb.

Measurement of antibody activity in complement-mediated lysis.

The abilities of the mutant antibodies to activate complement were measured in an assay using human serum and peripheral blood mononuclear cells (PBMC). The cells express the surface marker CD52 and are thus bound by antibodies with the CAMPATH-1 H variable regions. A cell proliferation reagent was used as a measure of live cells and thus indicated the relative proportion of cells that had not been lysed in the presence of CAMPATH IgG and serum.

Fresh human blood was collected in EDTA tubes and subjected to density-gradient centrifugation on Histopaque®-1077 (Sigma). The serum was defibrinated on glass beads following addition of Ca and Mg salts to overcome the effects of the EDTA and kept on ice. The PBMC layer was collected and the cells washed extensively in IMDM to remove the Histopaque. The PBMC were finally re-suspended in IMDM (without phenol red) and counted.

All assay samples were diluted in IMDM without phenol red to allow better visualisation of the colour change. Antibody dilutions were made in flat-bottomed assay plates. An additional lgG1 antibody, which does not bind PBMCs, was included as a negative control. 25 μΙ serum and 1 .5x10⁵ PBMCs were added to each well, giving a final volume of 100 μΙ/well. Controls contained serum and PBMC but no antibody (no antibody control) or serum only (no cell control). Plates were incubated for 90 min at 37C and 5% C0₂. To each well, 14 μΙ cell proliferation marker WST-1 (Roche, Mannheim, Germany) and 26 μΙ IMDM were added. Plates were then incubated at 37C and 5% C0₂, and absorbance at 450 nm was read after 4 h.

Results of representative complement lysis assays are shown in figure 10. Part a indicates that G1 Ap and ψγΔς do activate complement lysis as there is a reduction in metabolic activity compared to the negative control lgG1 but they do not activate lysis as efficiently as lgG1 . In part b, it is seen that ψγ has a similar level of activity to G1Ap and ψγΔς.

Conclusions

Human lgG1 molecules, which contained various numbers of residues that are encoded by the pseudo gamma gene within their CH2 domains, were produced. There was a difference in apparent molecular weight of the IgG heavy chains by reducing SDS-PAGE with the ψγ heavy chain being the heaviest. It is thought that changes at some positions within the Fc region can cause differences in glycosylation (Lund et al., 1996). One of these positions is residue 301 which is altered from lgG1 wild type sequence in ψγ but not i|JYAq and, indeed, the heavy chain of i|JYAq looks to be the same size as the lgG1 heavy chain.

When looking at the ability of the IgG to activate complement-mediated lysis, all three mutant antibodies had a similar activity which was lower than that of lgG1 . In binding to all of the FCYR, ψγ showed reduced activity relative to lgG1 . The two mutants that contained fewer residues encoded by the pseudo gamma gene - G1 Ap and i|JYAq - had binding activity intermediate to lgG1 and ψγ. In the case of FCYRI and FcYRIIIa 158F or 158V, the activity of G1Ap was very similar or slightly higher than that of i|JYAq. In the case of the FcYRIIa receptors, the activity of ψγΔ was slightly the higher and this effect was more pronounced for FcyRllb where the binding of ψγΔς was approximately 8-fold higher than that of G1 Ap. This difference in binding must be due to one or more of the six amino acid residues which differ between ψγΔς and G1 Aq. When investigating the effect of A327G, we found that the A327G change accounts for the majority of the difference in FcyRl lb binding between G1 Ap and ψγΔς (see figure 1 5).

The Ar mutation (327G) does not appear to improve the FcyRl lb binding of a wildtype lgG1 constant region (Figure 1 2). However, it does improve the FcyRl lb binding of three mutated lgG1 constant regions, namely G1 Ae (Figure 13), G1 Ab (Figure 14) and G1 Ap (Figure 15).

Sequence listing

SEQ ID. No 1

ASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSS LGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGGPSV LFPPKPKDTLMI SRTPEVTCVVVDVS HEDPEVKFNWYVDGVEVHNAKTKPWEEQY STYRVVSVLTVVHQNWLNGREYKCKVSNKGLPAPIEKTI SKTKGQ PREPQVYTLPPSRDELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRW QQGNVFSCSVMHEALHNHYTQKSLSLSPGK

References

Armour, K. L, Clark, M. R., Hadley, A. G. & Williamson, L. M. 1999. Recombinant human IgG molecules lacking Fcgamma receptor I binding and monocyte triggering activities. Eur J Immunol, 29, 2613-24.

Armour KL, van de Winkel JG, Williamson LM, Clark MR. 2003. Differential binding to human FcgammaRl la and FcgammaRl lb receptors by human IgG wild type and mutant antibodies. Mollmmunol, 40, 585-93.

Armour, K. L, Smith, C. S. & Clark, M. R. 2010. Expression of human FcgammaRl lla as a GPI-linked molecule on CHO cells to enable measurement of human IgG binding. J Immunol Methods, 354, 20-33.

Bensmana, M., Huck, S., Lefranc, G. & Lefranc, M. P. 1 988. The human immunoglobulin pseudo-gamma IGHGP gene shows no major structural defect. Nucleic Acids Res, 1 6, 31 08.

Clark, M. R., 1997. IgG effector mechanisms. Chem. Immunol.65, 88-1 10.

Glennie, M.J. and Winkel, J., 2003. Renaissance of cancer therapeutic antibodies. DDT Vol. 8, No.1 1 , 503-51 0. Greenwood J, Clark MR. Effector functions of matched sets of recombinant human IgG subclass antibodies. In: Clark MR, ed. Protein Engineering of Antibody Molecules for Prophylactic and Therapeutic Applications in Man. Nottingham, UK: Academic Titles; 1993:85-100.

Hale G, Bright S, Chumbley G, Hoang T, Metcalf D, Munro AJ, Waldmann H (1983) Removal of T cells from bone marrow for transplantation: amonoclonal antilymphocyte antibody that fixes human complement. Blood, 62: 873-882.

Kabat, E. A. 1991 . Sequences of proteins of immunological interest. U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health.

Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680-5.

Lund, J., Takahashi, N., Pound, J. D., Goodall, M. & Jefferis, R. 1996. Multiple interactions of IgG with its core oligosaccharide can modulate recognition by complement and human Fc gamma receptor I and influence the synthesis of its oligosaccharide chains. J Immunol, 157, 4963-9.

Reichert, J.M., 201 1 . Antobody-based therapeutics to watch in 201 1 . mAbs 3, 1 , 76-99.

Takahashi, N., Ueda, S., Obata, M., Nikaido, T., Nakai, S. & Honjo, T. 1982. Structure of human immunoglobulin gamma genes: implications for evolution of a gene family. Cell, 29, 671 -9.

Van Sorge, N. M., Van Der Pol, W. L. & Van De Winkel, J. G. 2003. FcgammaR polymorphisms: Implications for function, disease susceptibility and immunotherapy. Tissue Antigens, 61 , 189-202.

Claims

Claims

1 . An Fc variant polypeptide comprising a modified human IgG CH2 region compared to a wild type CH2 region comprising at least one amino acid substitution selected from the group consisting of 292W, 309V, 312N, 317R, 327G and 339T where the residues are numbered according to the EU system and wherein the at least one amino acid substitution does not increase binding to human FcyRI and/or FcyRllla and increases binding to FcyRllb.

2. An Fc variant polypeptide according to claim 1 wherein the amino acid substitution is A327G.

3. An Fc variant polypeptide according to claim 1 or 2 wherein the human IgG region is lgG1 , lgG2, lgG3 or lgG4.

4. An Fc variant polypeptide according to a preceding claim wherein the human IgG region is lgG1 .

5. An Fc variant according to a preceding claim comprising further modifications.

6. An Fc variant according to claim 5 wherein the further modifications alter Fc binding.

7. An Fc variant according to claim 6 wherein the further modifications are selected from

(i) P232T and L234P;

(ii) H268E;

(iii) E233P, L234V, L235A and deletion of G236 or

(iv) E233P, L234V and L235A.

8. An antibody comprising an Fc variant polypeptide according to any of claims 1 to 7.

9. An antibody according to claims 8 wherein said antibody is a humanised antibody.

10 A fusion protein comprising an Fc variant polypeptide according to any of claims 1 to 7.

1 1 . A pharmaceutical composition comprising an antibody according to claim 8 or 9 or a fusion protein according to claim 10.

12. A method of treating an autoimmune disease comprising administering an antibody according to claim 8 or 9, fusion protein according to claim 10 or a pharmaceutical composition according to claim 1 1 .

13. The use an antibody according to claim 8 or 9, fusion protein according to claim 10 or a pharmaceutical composition according to claim 1 1 in the manufacture of a medicament for the treatment of an autoimmune disease.

14. A method for activating an FcyRllb receptor comprising contacting a cell comprising an FcyRllb receptor with an Fc variant according to any of claims 1 to 7.

15. A method for producing a variant Fc polypeptide having increased selective binding affinity for the FcyRllb receptor wherein said method comprises modifying a polypeptide comprising a human IgG CH2 region by substitution of one or more amino acids with the following amino acids at the following positions: 292W, 309V, 312N, 317R, 327G and 339T.

16. A method for increasing the relative binding affinity of a Fc region for the FcyRllb receptor wherein said method comprises modifying a polypeptide comprising a human IgG CH2 region by substitution of one or more amino acids with the following amino acids at the following positions: 292W, 309V, 312N, 317R, 327G and 339T.

17. The method according to claim 15 or 16 wherein the substitution is 327G.

18. The method according to any of claims 15 to 17 wherein the method includes introducing one or more additional modification(s).

19. The method according to claim 18 wherein the further modification(s) alter Fc binding.

20. The method according to claim 19 wherein the further modification(s) is selected from

(i) P232T and L234P;

(ii) H268E;

(iii) E233P, L234V, L235A and deletion of G236 or

(iv) E233P, L234V and L235A.

21 . A variant polypeptide obtained or obtainable by the process of any one of claims 15 to 20.

22. A nucleic acid sequence encoding a polypeptide according to any of claims 1 to 7.

23. A vector comprising a nucleic acid sequence according to claim 22.

24. A host cell transformed with a vector according to claim 23.