AU2007332085A1 - Multimeric Fc receptor polypeptides including a modified Fc domain - Google Patents

Description

WO 2008/070927 PCT/AU2007/001934 MULTIMERIC Fc RECEPTOR POLYPEPTIDES INCLUDING A MODIFIED Fc DOMAIN FIELD OF THE INVENTION The present invention relates to a soluble multimeric Fc receptor polypeptide and 5 protein able to inhibit leukocyte Fcy receptors (FcyR) and immunoglobulin G (IgG) interactions. Such a polypeptide and protein is useful in the treatment of inflammatory diseases, particularly immune complex-mediated inflammatory diseases such as rheumatoid arthritis (RA), immune thrombocytopenic purpura (ITP) and systemic lupus erythematosus (SLE). 10 INCORPORATION BY REFERENCE This patent application claims priority from: - PCT/AU2006/001890 entitled "Multimeric Fc receptor polypeptides", filed on 13 December 2006 and - US 11/762,664 entitled "Multimeric Fc receptor polypeptides", filed on 15 13 June 2007. The entire content of these applications are hereby incorporated by reference. BACKGROUND OF THE INVENTION The treatment of autoimmune and other inflammatory diseases such as RA and SLE has entered a new and exciting phase where increased understanding of the 20 molecules involved in the immune system has allowed for the specific inhibition of key inflammatory molecules such as tumour necrosis factor-a (TNFa) and interleukin 1P (IL-13). For example, in recent studies, it has been shown that antibodies can play a powerful role in the pathogenesis of RA, and in human clinical trials, positive responses to the use of anti-CD20 monoclonal antibody 25 (MAb) therapy to eliminate antibody producing B cells have been generating strong evidence of the significant role of antibodies in RA (Emery et al., 2001).

WO 2008/070927 PCT/AU2007/001934 2 Since Fc receptors (FcR) play pivotal roles in immunoglobulin-based effector systems, inhibition of FcR function may provide the basis of effective therapy for a variety of diseases. Moreover, since Fcy receptors (FcyR) are pivotal to effector systems for IgG, targeting the interaction between leukocyte FcyRs and 5 antibodies provides a new opportunity for therapeutic intervention in RA (Nabbe et al., 2003). One approach of achieving such an intervention which is of interest to the present applicant is the use of a soluble form of an FcyR to act as a "decoy" to prevent leukocyte activation by antibodies. Fc receptors (FcR) are leukocyte surface glycoproteins that specifically bind the Fc 10 portion of antibodies. The receptors for IgG, that is FcyR, are the most widespread and diverse, the major types being FcyRI (CD64), FcyRII (CD32) and FcyRIII (CD16). Immune complexes (IC) that are formed in vivo in normal immune responses, and those seen in the pathology of autoimmune diseases such as RA, can simultaneously engage many FcR. For example, in humans, activated 15 macrophages, neutrophils, eosinophils and mast cells can express FcyRI, FcyRIIa, FcyRIIb and FcyRIII (Takai, 2002). However, of these, the FcyRIIa is the major initiator of IC-mediated inflammation and, while all of the FcyR types engage the lower hinge region of the IgG Fc domain and the CH2 domains such that any soluble FcyR decoy polypeptide might inhibit the binding of IgG to all classes of 20 FcyR, the present applicant has realised that since FcyRIIa shows the widest binding specificity and highest selectivity for avid IgG immune complex binding, the development and investigation of a soluble FcyRIIa offers the greatest potential. Indeed, previous studies have shown that a simple recombinant soluble FcyRIIa 25 polypeptide (rsFcyRIla monomer), consisting of FcyRIIa ectodomains (Ierino et al., 1993a), is clearly able to inhibit IC-mediated inflammation. In these studies, the rsFcyRIIa was tested using the Arthus reaction, wherein immune complexes are formed in the dermis by the passive administration of antibody and antigen (Pflum et al., 1979), which is a model of vasculitis (an extra articular complication WO 2008/070927 PCT/AU2007/001934 3 in arthritis) and also occurs in SLE. It was found that while the rsFcyRIla monomer inhibited inflammation and neutrophil infiltration when co administered with the antibody and antigen, large amounts of the rsFcyRIla monomer were required because of a relatively low level of selectivity for the 5 immune complexes. To overcome this problem, the present applicant proposes to use multimeric forms of the rsFcyRIla decoy, and has since found, surprisingly, that not only could such multimeric forms be successfully expressed, they exhibit increased selectivity for immune complexes. Such multimeric rsFcyRIla polypeptides therefore show considerable promise for the treatment of IC 10 mediated inflammatory disease such as RA and SLE. SUMMARY OF THE INVENTION Thus, in a first aspect, the present invention provides a soluble multimeric polypeptide able to inhibit interaction of leukocyte Fcy receptors (FcyR) and immunoglobulin G (IgG), said polypeptide comprising two or more Fc binding 15 regions linked in a head to tail arrangement, at least one of which is derived from an FcyR type receptor, and an Fc domain of an immunoglobulin which has been modified to reduce or prevent binding to said Fc binding regions and/or to alter effector function. Preferably, the polypeptide is a multimer of an Fc binding region derived from an 20 FcyRII type receptor, particularly FcyRIIa. Such a molecule may be considered to be a homomultimer, and one especially preferred molecule of this kind is a homodimer of an Fc binding region derived from an FcyRII type receptor. However, the present invention also contemplates that the molecule may be a multimer of an Fc binding region derived from an FcyR type receptor (e.g. an 25 FcyRII type receptor) and an Fc binding region from another source (e.g. an Fc binding region from another Fc receptor type or a synthetic Fc binding polypeptide). A molecule of this kind may be considered to be a heteromultimer, and one especially preferred molecule of this kind is a heterodimer of an Fc WO 2008/070927 PCT/AU2007/001934 4 binding region derived from an FcyRII type receptor and an Fc binding region derived from an FcyRIII type receptor. The Fc binding regions may be linked through a peptide bond or via a short linker sequence (e.g. a single amino acid or a short peptide of, for example, 2 to 20 5 amino acids in length). In a second aspect, the present invention provides a soluble multimeric protein comprising a polypeptide according to the first aspect. In a third aspect, the present invention provides a polynucleotide molecule comprising a nucleotide sequence encoding a polypeptide according to the first 10 aspect or a protein according to the second aspect. The polynucleotide molecule may consist in an expression cassette or expression vector (e.g. a plasmid for introduction into a bacterial host cell, or a viral vector such as a baculovirus vector for transfection of an insect host cell, or a plasmid or viral vector such as a lentivirus for transfection of a mammalian host cell). 15 Thus, in a fourth aspect, the present invention provides a recombinant host cell comprising a polynucleotide molecule according to the third aspect. In a fifth aspect, the present invention provides a method for producing a polypeptide or protein, the method comprising the steps of; (i) providing a recombinant host cell comprising a polynucleotide 20 molecule according to the third aspect, (ii) culturing said host cell in a suitable culture medium and under conditions suitable for expression of said polypeptide or protein, and (iii) isolating said polypeptide or protein from the culture, and, optionally, from the culture medium.

WO 2008/070927 PCT/AU2007/001934 5 In a sixth aspect, the present invention provides a method of treating a subject for an inflammatory disease, said method comprising administering to said subject a polypeptide according to the first aspect or a protein according to the second aspect, optionally in combination with a pharmaceutically- or veterinary 5 acceptable carrier or excipient. BRIEF DESCRIPTION OF THE FIGURES Figure 1 provides the nucleotide sequence (and translated amino acid sequence) for a head to tail homodimer construct of two FcyRIIa extracellular regions each comprising both FcyRIIa ectodomains, namely ectodomains 1 and 2. The FcyRIIa 10 ectodomains 1 and 2 consist of amino acids 1 to 174 of the FcyRIIa polypeptide sequence with amino acids 1 to 88 comprising domain 1 and amino acids 89 to 174 comprising domain 2 (Hibbs et al., 1988; Homo sapiens Fc fragment of IgG, low affinity IIa receptor (CD32) (FCGR2A), mRNA, ACCESSION NM_021642; and Powell et al., 1999). In the figure, amino acids 1 to 182 are derived from the 15 extracellular region of FcyRIIa, of which amino acids 1 to 174 comprise the FcyRIIa ectodomains 1 and 2 and amino acids 175 to 182 comprise the membrane proximal stalk (which in FcyRIIa links the ectodomains 1 and 2 to the transmembrane sequence). The first of the FcyRIIa extracellular regions comprising the dimer therefore consists of amino acids 1 to 182 and the second of 20 the FcyRIIa extracellular regions consists of amino acids 184 to 362 (corresponding to amino acids 3 to 182 of FcyRIIa). The underlined amino acid represents a non-FcyRIla linker amino acid residue, while the bolded amino acids highlight a C-terminal His6 tag. Figure 2 shows a Western Blot analysis of recombinant soluble (rs) multimeric 25 forms of FcyRIIa expressed from the nucleotide sequence shown in Figure 1. The rsFcyRIIa dimer was substantially stable with only a small amount of rsFcyRIla monomer breakdown product evident. On the other hand, the rsFcyRIla timer and tetramer forms were unstable, being substantially degraded to the rsFcyRIIa WO 2008/070927 PCT/AU2007/001934 6 dimer form. This degradation may be avoided by the use of protease inhibitors during production or by otherwise modifying the sequence of the multimer forms so as to remove cleavage site(s). Figure 3 shows a Coomassie-stained SDS-PAGE (12% acrylamide gel, under non 5 reducing conditions) of fractions collected from the purification of rsFcyRIIa monomer (expressed from mammalian cells) having the expected size of -30 kDa (a), and rsFcyRIla dimer having the expected size of -50 kDa (b). Figure 4 graphically shows the equilibrium binding responses of rsFcyRIla monomer to immobilised (a) IgG monomer (Sandoglobulin) and (b) the model 10 immune complex, heat-aggregated IgG (HAGG). Figure 5 graphically shows the equilibrium binding responses of rsFcyRIla dimer to immobilised (a) IgG monomer (Sandoglobulin) and (b) the model immune complex, HAGG. Figure 6 provides a plot of rsFcyRIla monomer (a) and rsFcyRIla dimer expressed 15 from the nucleotide sequence of Figure 1 (b) binding to immobilised human IgG monomer (Sandoglobulin) following prior reaction in solution with human IgG monomer (Sandoglobulin) and dimer-IgG (Wright et al., 1980), as determined using a standard BIAcore assay protocol. Figure 7 provides plots of (a) the inhibition of dimer-IgG (Wright et al., 1980) 20 binding to human neutrophils (volunteer V5) by purified rsFcyRIIa monomer and rsFcyRIIa dimer calculated as a percentage of the uninhibited dimer-IgG binding activity and (b) the inhibition of dimer-IgG binding to human neutrophils (volunteer V1) by purified rsFcyRIla dimer (expressed from the nucleotide sequence of Figure 1) calculated as a percentage of dimer-IgG binding dimer. 25 Figure 8 provides at (a), a plot of immune-complex (dimer-IgG) stimulated TNF secretion from 24 hour differentiated human MDMs (volunteer V5) in the absence WO 2008/070927 PCT/AU2007/001934 7 and presence of rsFcyRIIa dimer (in supernatant at 2.5 ptg/ml); while at (b), provides a plot of immune-complex (dimer-IgG) stimulated TNF secretion from 24 hour differentiated human MDMs (volunteer V1), in the absence and presence of rsFcyRIIa dimer (2.5 pg/ml). 5 Figure 9 provides a plot of immune-complex (HAGG) stimulated activation of human platelets, as measured by the mean fluorescence intensity (MFI) of P selectin expression in the absence and presence of rsFcyRIIa dimer (30 ptg/ml). Figure 10 provides results from the analysis of the rsFcyRIIa dimer isolated from stably transfected CHO-S cells by SDS-PAGE under (a) non-reducing and (b) 10 reducing conditions, (c) Western blotting using an anti-FcyRIIa antibody, and (d) HPLC. The rsFcyRIIa dimer migrates as a single band at the expected molecular weight (~ 50 kD), reacts with anti-FcyRIIa antibody and was >96% pure as determined by HPLC analysis. Figure 11 provides a plot of immune-complex (HAGG) binding to cell surface 15 expressed human FcyRIIb (on the murine B lymphoma cell line IIA1.6) in the presence of either rsFcyRIla monomer or rsFcyRIIa dimer. Figure 12 provides a plot of activated platelets (positive for both CD41 and CD62P) after treatment with HAGG in the presence of rsFcyRIIa monomer or rsFcyRIla dimer (expressed from the nucleotide sequence of Figure 1), as a 20 percentage of activated platelets following treatment with HAGG alone. - Figure 13 provides a plot of TNF-ca release from MC/9 cells after incubation with OVA immune complexes in the presence of rsFcyRIIa monomer or rsFcyRIIa dimer, as a percentage of TNF-a released in the presence of OVA immune complexes alone. 25 Figure 14 shows Western blot analysis of rsFcyRIla fusion proteins. (1) rsFcyRIIa monomer; (2) rsFcyRIla dimer; (3) rsFcyRIIa monomer fused to IgGi-Fcyl (L234A, WO 2008/070927 PCT/AU2007/001934 8 L235A); (4) the rsFcyRIIa dimer fused to IgGi-Fcyl (L234A, L235A); (5) rsFcyRIla monomer fused to human serum albumin (HSA); (6) the rsFcyRIla dimer fused to HSA; (7) purified rsFcyRIIa monomer standard; and (8) purified rsFcyRIIa dimer standard. 5 Figure 15 provides the results of a HAGG-capture ELISA with rsFcyRIIa monomer and rsFcyRIIa dimer fusions. (a) FcyRIIa monomer standard (Powell et al., 1999) starting at 0.75 pg/ml (monomer std); protein from cells transfected with rsFcyRIla monomer construct (transfection 426 (monomer)); protein from cells transfected with rsFcyRIla monomer fusion to IgG-Fcyl (L234A, L235A) 10 construct (monomer-Fc); and protein from cells transfected with rsFcyRIla monomer fusion to HSA construct (HSA-monomer); (b) rsFcyRIla dimer standard starting at 0.5 ig/ml (dimer std); supernatant from cells transfected with rsFcyRIIa dimer (transfection 427 (dimer)); supernatant from cells transfected with rsFcyRIla dimer fusion to IgG-Fcyl (L234A, L235A) (dimer-Fc); 15 and supernatant from cells transfected with rsFcyRIIa dimer fusion to HSA (HSA-dimer). Figure 16 provides results obtained from a CAPTURE-TAG ELISA on rsFcyRIla monomer and rsFcyRIla dimer fusion proteins to confirm the presence of epitopes that establish that the receptor is properly folded. (A) rsFcyRIIa 20 monomer standard starting at 0.75 pg/ml (monomer std); supernatant from cells transfected with rsFcyRIla monomer (transfection 426 (monomer)); supernatant from cells transfected with rsFcyRIla monomer fusion to IgG-Fcyl (L234A, L235A) (monomer-Fc); and supernatant from cells transfected with rsFcyRIIa monomer fusion to HSA (HSA-monomer); (B) rsFcyRIIa dimer standard ( 25 prepared in-house) starting at 0.5 pg/ml); supernatant from cells transfected with FcyRIIa dimer (transfection 427 (dimer)); supernatant from cells transfected with rsFcyRlIa dimer fusion to IgG-Fcyl (L234A, L235A) (dimer-Fc); and supernatant from cells transfected with rsFcyRIIa dimer fusion to HSA (HSA-dimer).

WO 2008/070927 PCT/AU2007/001934 9 Figure 17 provides a schematic diagram of (a) rsFcyRIIa monomer; (b) rsFcyRIla dimer; (c) a dimer of a rsFcyRIIa monomer fusion to IgG-Fcyl (L234A, L235A), wherein the dimerisation is effected through the Fc domains of the rsFcyRIla monomer fusion polypeptides, providing a molecule having two Fc binding 5 regions (i.e. a protein that is dimeric for the Fc binding region or, otherwise, has a "valency of two"); (d) a dimer of rsFcyRIIa dimer fusion to IgG-Fcyl (L234A, L235A), wherein dimerisation is effected through the two Fc domains of the rsFcyRIIa dimer fusion polypeptides, providing a molecule having four Fc binding regions (i.e. a protein that is tetrameric for the Fc binding region or, 10 otherwise, has a "valency of four"); (e) rsFcyRIIa monomer fusion to HSA; and (f) rsFcyRIIa dimer fusion to HSA. In the figure, D1 and D2 refers to, respectively, ectodomains 1 and 2, the solid bar shown adjacent to D2 represents a linker sequence, the dark loop at the top of the dimerised Fc domains in (c) and (d) represents disulphide linkages, and H 6 refers to a His tag); 15 Figure 18 shows the effect of the rsFcyRIIa dimer (with no fusion partner) in a mouse model of arthritis. Mice treated with arthritogenic anti-collagen antibody in the absence (black square) and presence of the FcyRIIa dimer (white diamond); Figure 19 provides the amino acid sequence of an embodiment of the present invention, namely an rsFcyRIla dimer fusion to an Fc domain derived from IgG2a 20 (this fusion protein is hereinafter referred to as the D2 protein); Figure 20 provides the nucleotide sequence encoding the D2 protein of Figure 19; Figure 21 illustrates the plasmid used to express the nucleotide sequence (encoding the D2 protein) of Figure 20; Figure 22 shows an analysis of the purified D2 protein of Figure 19 by SDS-PAGE 25 (panel A) and by Western blot (panel B); WO 2008/070927 PCT/AU2007/001934 10 Figure 23 shows the effect of the D2 protein (of Figure 19) on TNF-a release in a MC/9 mast cell assay; Figure 24 shows the effect of the D2 protein (of Figure 19) on human neutrophil activation; and 5 Figure 25 shows the effect of the D2 protein (of Figure 19) on human platelet activation. DETAILED DESCRIPTION OF THE INVENTION The present invention provides a soluble multimeric polypeptide and protein able to inhibit interaction of leukocyte Fcy receptors (FcyR) and immunoglobulin 10 G (IgG) which comprises two or more Fc binding regions, one of which is essentially derived from an FcyR type receptor. Such a polypeptide and protein offers an increase in selectivity for immune complexes over that previously observed with soluble monomeric polypeptides such as rsFcyRII monomer, and thereby provides considerable promise as a "decoy" molecule for the treatment of 15 IC-mediated inflammatory disease such as RA and SLE. In a first aspect, the present invention therefore provides a soluble multimeric polypeptide able to inhibit interaction of leukocyte Fcy receptors (FcyR) and immunoglobulin G (IgG), said polypeptide comprising two or more Fc binding regions linked in a head to tail arrangement, at least one of which is derived from 20 an FcyR type receptor, and an Fc domain of an immunoglobulin which has been modified to reduce or prevent binding to the said Fc binding regions and/or to alter effector function. As used herein, the term "soluble" indicates that the polypeptide (or protein) is not bound to a cellular membrane, and is, accordingly, characterised by the 25 absence or functional disruption of all or a substantial part of the transmembrane WO 2008/070927 PCT/AU2007/001934 11 (i.e. lipophilic) domain, so that the polypeptide (or protein) is devoid of any membrane anchoring function. The cytoplasmic domains may also be absent. As used herein, the term "Fc binding region" refers to any part or parts of an Fc receptor that is able to bind with an Fc domain of an immunoglobulin (e.g. an Fc 5 fragment produced by papain hydrolysis of an immunoglobulin) including genetically modified versions thereof, as well as synthetic Fc binding polypeptides. The at least one Fc binding region derived from an FcyR type receptor may be derived, for example, from an FcyR having low affinity for IgG, that is an affinity 10 for IgG of less than 5 x 107 M-1. Such low affinity receptors include FcyRII type receptors (e.g. FcyRIIa including the polymorphic variants, FcyRIla-H131 and FcyRIIa-R131 (Stuart et al., 1987; Brooks et al., 1989; Seki et al., 1989), FcyRIIb and FcyRIIc), FcyRIII type receptors (e.g. FcyRIIIa and FcyRIIIb), truncated forms of FcyRI type receptors (e.g. FcyRla and FcyRlb) such as truncated polypeptides 15 comprising the first and second of the three ectodomains of an FcyRI receptor (Hulett et al., 1991; Hulett et al., 1998), and genetically modified versions of FcyR which normally have high affinity for IgG but by virtue of the modifications (e.g. one or more amino acid substitution(s), deletion(s) and/or addition(s)) show a reduced affinity for IgG of less than 5 x 107 M-1). 20 Preferably, the polypeptide is a homomultimer of an Fc binding region derived from an FcyR receptor such as a low affinity FcyR. A suitable Fc binding region consists of all or an Fc binding part or parts of one or more ectodomains of an FcyR receptor. Persons skilled in the art will be able to readily identify Fc binding ectodomains of FcyR receptors since these domains belong to the IgG 25 domain superfamily (Hulett et al., 1994, Hulett et al., 1995, Hulett et al., 1998, and Tamm et al., 1996) and are typically characterised by "a tryptophan sandwich" (e.g. residues W90 and W113 of FcyRIIa) and other residues (e.g. in FcyRIIa; WO 2008/070927 PCT/AU2007/001934 12 residues of the ectodomain 1 and ectodomain 2 linker, and the BC (W113-V119), C'E (F132-P137) and FG (G159-Y160) loops of ectodomain 2 (Hulett et al., 1994)). More preferably, the polypeptide is a homomultimer of an Fc binding region of FcyRIIa. A suitable Fc binding region from FcyRIIa consists of all or an Fc 5 binding part or parts of the ectodomains 1 and 2 of FcyRIIa. The FcyRIIa ectodomains 1 and 2 are found within amino acids 1 to 172 of the FcyRIIa amino acid sequence (Hibbs et al., 1988, and ACCESSION NM_021642). An example of an Fc binding part of the FcyRIIa ectodomains 1 and 2 is a fragment comprising amino acids 90 to 174 of the FcyRIIa amino acid sequence, which includes 10 residues of the ectodomain 1 and ectodomain 2 linker and BC (W113-V119), C'E (F132-P137) and FG (G159-Y160) loops of ectodomain 2. X-ray crystallography studies has revealed that within this fragment, amino acids 113-116, 129, 131, 133, 134, 155, 156 and 158-160 make important contributions to the fragment surface that is able to bind to the Fc domain of IgG (International patent specification no 15 WO 2005/075512). The polypeptide may also be a heteromultimer of an Fc binding region derived from an FcyRII type receptor and an Fc binding region from another source (e.g. an Fc binding region from another Fc receptor type such as another FcyR type or an Fc binding region from other immunoglobulin receptors such as receptors for 20 IgA and IgE). One especially preferred molecule of this kind is a heterodimer of an Fc binding region derived from an FcyRII type receptor (particularly, FcyRIIa) and an Fc binding region derived from an FcyRIII type receptor. Fc binding regions considered as having been "derived from" a particular Fc receptor include Fc binding regions having an amino acid sequence which is 25 equivalent to that of an Fc receptor as well as Fc binding regions which include one or more amino acid modification(s) of the sequence of the Fc binding region as found in an Fc receptor. Such amino acid modification(s) may include amino acid substitution(s), deletion(s), addition(s) or a combination of any of those WO 2008/070927 PCT/AU2007/001934 13 modifications, and may alter the biological activity of the Fc binding region relative to that of an Fc receptor (e.g. the amino acid modification(s) may enhance selectivity or affinity for immune complexes; such modifications at amino acids 133, 134, 158-161 are described in International patent specification no WO 5 96/08512). On the other hand, Fc binding regions derived from a particular Fc receptor may include one or more amino acid modification(s) which do not substantially alter the biological activity of the Fc binding region relative to that of an Fc receptor. Amino acid modification(s) of this kind will typically comprise conservative amino acid substitution(s). Exemplary conservative amino acid 10 substitutions are provided in Table 1 below. Particular conservative amino acid substitutions envisaged are: G, A, V, I, L, M; D, E, N, Q; S, C, T; K, R, H: and P, Nca-alkylamino acids. In general, conservative amino acid substitutions will be selected on the basis that they do not have any substantial effect on (a) the structure of the polypeptide backbone of the Fc binding region at the site of the 15 substitution, (b) the charge or hydrophobicity of the polypeptide at the site of the substitution, and/or (c) the bulk of the amino acid side chain at the site of the substitution. Where an Fc binding region including one or more conservative amino acid substitutions is prepared by synthesis, the Fc binding region may also include an amino acid or amino acids not encoded by the genetic code, such as y 20 carboxyglutamic acid and hydroxyproline and D-amino acids.

WO 2008/070927 PCT/AU2007/001934 14 Table 1 Exemplary conservative amino acid substitutions Conservative Substitutions Ala Val*, Leu, Ile Arg Lys*, Gln, Asn Asn Gln*, His, Lys, Arg, Asp Asp Glu*, Asn Cys Ser Gin Asn*, His, Lys, Glu Asp*, y-carboxyglutamic acid (Gla) Gly Pro His Asn, Gln, Lys, Arg* Ile Leu*, Val, Met, Ala, Phe, norleucine (Nle) Leu Nle, Ile*, Val, Met, Ala, Phe Lys Arg*, Gln, Asn, ornithine (Orn) Met Leu*, Ile, Phe, Nle Phe Leu*, Val, Ile, Ala Pro Gly*, hydroxyproline (Hyp),Ser, Thr Ser Thr Thr Ser Trp Tyr Tyr Trp, Phe*, Thr, Ser Val Ile, Leu*, Met, Phe, Ala, Nle *indicates preferred conservative substitutions WO 2008/070927 PCT/AU2007/001934 15 The Fc binding regions are preferably linked through a peptide bond or via a short linker sequence (e.g. a single amino acid or a short peptide of, for example, 2 to 20 amino acids in length). However, it may in certain circumstances be preferable or desirable to link the Fc binding regions through other suitable 5 linkage means (e.g. by chemical cross-linking). The Fc binding regions of the polypeptide of the present invention are linked in a "head to tail" arrangement. That is, the C-terminal ("tail") of a first Fc binding region will be linked to the N-terminal ("head") of a second Fc binding region in a tandem manner. There is at least two Fc binding regions, typically 2 to 4 Fc 10 binding regions, linked in this manner, however the polypeptide may have up to 10 or more (e.g. 20) Fc binding regions linked in a head to tail arrangement. The Fc binding regions will typically be linked through a peptide bond or via a short linker sequence (e.g. a single amino acid or a short peptide of, for example, 2 to 20 amino acids in length or, more preferably, 2 to 15 amino acids in length, 2 to 10 15 amino acids in length, 2 to 8 amino acids in length, or, most preferably, 2 to 5 amino acids in length). Suitable short linker sequences may be short random sequences or may comprise short non-Fc binding region fragments of FcyR (e.g. short fragments of 20 or fewer amino acids from the proximal region of the membrane stalk of FcyR). The linker sequence may be a synthetic linker sequence 20 such as, for example, GGGGSGGGGS (SEQ ID NO: 4) which has a low susceptibility to proteolysis. Such a linker sequence may be provided in the form of 2 to 5 tandem "Gly4Ser" units. Linking the Fc binding regions through a peptide bond or a short linker sequence allows for the production of the polypeptide using recombinant expression systems. 25 Thus, in a first particularly preferred embodiment of a polypeptide according to the invention, the polypeptide comprises two to four Fc binding regions derived from FcyRIIa linked in a head to tail arrangement.

WO 2008/070927 PCT/AU2007/001934 16 In a second particularly preferred embodiment of a polypeptide according to the invention, the polypeptide comprises two Fc binding regions from FcyRIIa linked in a head to tail arrangement. And in a third particularly preferred embodiment of a polypeptide according to 5 the invention, the polypeptide comprises two FcyRIIa extracellular regions each comprising ectodomains 1 and 2, wherein said extracellular regions are linked in a head to tail arrangement through a linker comprising 1 to 20 amino acids. In certain embodiments of the invention, the Fc binding regions within the polypeptide are linked through a peptide linker constituting the membrane 10 proximal stalk region of FcyRIIa, which is represented by the sequence PSMGSSSP (SEQ ID NO: 7). Equivalent linkers that adopt a similar secondary structure are also useful, including equivalents that incorporate conservative amino acid substitutions. Further, truncations and extensions of this amino acid sequence, having one or two fewer or additional amino acids, are also useful. 15 Suitable linkers generally are those that permit the multimeric polypeptide to adopt a structure in which each Fc binding region can participate in the binding of an Fc domain-bearing molecule. In this manner, the linker permits, for example, a polypeptide comprising two linked Fc binding regions to bind a greater quantity of Fc domain-bearing molecules than are bound by a 20 corresponding monomer. The selection of linkers suitable to this end can be made based on simple binding experiments, as exemplified herein. The polypeptide of the present invention may further comprise a carrier protein (i.e. such that the polypeptide is a "fusion" of the carrier protein and said two or more linked Fc binding regions and modified Fc domain). The carrier protein 25 may be any suitable carrier protein well known to persons skilled in the art, but preferably, is human serum albumin (HSA) or another carrier protein commonly used to improve bioavailability (i.e. through increasing the serum half life of the polypeptide when administered to a subject). Conveniently, the carrier protein WO 2008/070927 PCT/AU2007/001934 17 can be fused to the polypeptide by expressing the polypeptide as a fusion protein with the said carrier protein in accordance with any of the methods well known to persons skilled in the art. The polypeptide of the present invention may further comprise other useful 5 linked molecules, for example, ethylene glycol (i.e. to produce a PEGylated polypeptide) to improve bioavailability, complement regulating molecules such a CD46, CD55 and CD59, cytokines (e.g. to enable delivery of cytokines to sites of inflammation) and cytokine receptors. The polypeptide of the present invention comprises an Fc domain of an 10 immunoglobulin. This Fc domain is capable of binding to another Fc domain (i.e. to form a dimer) and thereby provides a means of linking two or more polypeptides according to the invention to form a protein. Thus, by using, for example, a polypeptide comprising two or more linked Fc binding regions linked, that is fused to an Fc domain capable of binding to 15 another Fc domain, which may be the same or different and which is itself fused to another polypeptide comprising two or more linked Fc binding regions, a protein can be produced which comprises at least four Fc binding regions (in other words, a soluble multimeric protein comprising four or more Fc binding regions). 20 The Fc domain may be selected from any immunoglobulin (e.g. an IgG such as IgGi, IgG2a or IgG4). The IgG4 Fc domain, in the wild type form, has relatively low affinity for FcyRII receptors, and may therefore be used in the polypeptide of the invention without modification to avoid self-annealing to the linked Fc binding regions (i.e. derived from an FcyRII type receptor) of the polypeptide. 25 More desirably, however, the selected Fc domain is modified (e.g. by amino acid substitution(s) at residues critical for binding with Fc receptors) to prevent self annealing (i.e. "self-binding") of the Fc domain to linked Fc binding regions, as well as, preferably, to prevent binding to Fc receptors in vivo (i.e. the modified Fc WO 2008/070927 PCT/AU2007/001934 18 domain preferably shows a reduced affinity for binding endogenous Fc receptors other than neonatal Fc receptors (FcRn), including, for example, FcyRI, FcyRII and Fc'yRIII). As well, the selected Fc domain is desirably modified to alter effector function, such as to reduce complement binding and/or to reduce or abolish 5 complement dependent cytotoxicity (CDC). Such modifications have been extensively described by, for example, Clark and colleagues, who have designed and described a series of mutant IgG1, IgG2 and IgG4 Fc domains and their FcyR binding properties (Armour et al., 1999; Armour et al., 2002, the content of which is incorporated herein by reference in this application). For example, any one or 10 more of the amino acids at positions 234, 235, 236, 237, 297, 318, 320 and 322 can be modified (e.g. by amino acid substitution) to alter affinity for an effector ligand, such as an Fc receptor or the C1 component of complement (Winter et al. in US Patent Nos 5,624,821 and 5,648,260). Also, one or more of the amino acids at positions 329, 331 and 322 can be modified (e.g. by amino acid substitution) to 15 alter Clq binding and/or reduce or abolish CDC (as described, for instance, by Idusogie et al. in US Patent No 6,194,551), and/or to reduce or abolish antibody dependent cell mediated cytotoxicity (ADCC). In one especially preferred modified Fc domain, the Fc domain is derived from IgG1 (Wines et al., 2000) and comprises amino acid modification at amino acid 234 20 and/or 235, namely Leu 2 34 and/or Leu23 5 . These leucine residues are within the lower hinge region of IgG1 where the Fc receptor engages with the Fc domain. One or both of the leucine residues may be substituted or deleted to prevent Fc receptor engagement (i.e. binding); for example, one or both of Leu234 and Leu 235 may be substituted with alanine (i.e. L234A and/or L235A) or another suitable 25 amino acid(s) (Wines et al., 2000). In another especially preferred modified Fc domain, the Fc domain is derived from IgG2a and comprises amino acid modification at any one or more of amino acids 235, 318, 320 and 322, namely Leu 235 , Glu 318 , Lys 320 and Lys 3 22. Preferably, WO 2008/070927 PCT/AU2007/001934 19 Leu 235 is substituted with glutamate and Glu 318 , Lys 320 and Lys 322 are substituted with alanine. In a further especially preferred modified Fc domain, the Fc domain is derived from IgG4, including human IgG4, and comprises amino acid modification at any 5 one more of amino acids 228, 233, 234, 235 and 236. Preferably, the amino acid modifications in the IgG4 Fc domain introduce Pro22 8 , Pro 233 , Vap34 , Ala 235 , and a deletion of 236 (i.e. Del 236 ). In a second aspect, the present invention provides a soluble multimeric protein comprising a polypeptide according to the first aspect. 10 In a third aspect, the present invention provides a polynucleotide molecule comprising a nucleotide sequence encoding a polypeptide according to the first aspect or a protein according to the second aspect. The polynucleotide molecule may consist in an expression cassette or expression vector (e.g. a plasmid for introduction into a bacterial host cell, or a viral vector 15 such as a baculovirus vector for transfection of an insect host cell, or a plasmid or viral vector such as a lentivirus for transfection of a mammalian host cell). For a soluble multimeric protein comprising a dimer of a polypeptide according to the present invention, persons skilled in the art will appreciated that the encoding polynucleotide molecule will, upon expression in a host cell, yield a 20 single chain of the fusion polypeptide, which then will yield the desired multimeric protein as a product of host cell secretion. In a fourth aspect, the present invention provides a recombinant host cell comprising a polynucleotide molecule according to the third aspect. The recombinant host cell may be selected from bacterial cells such as E. coli, 25 yeast cells such as P. pastoris, insect cells such as Spodoptera Sf9 cells, mammalian WO 2008/070927 PCT/AU2007/001934 20 cells such as Chinese hamster ovary (CHO), monkey kidney (COS) cells and human embryonic kidney 293 (HEK 293) cells, and plant cells. In a fifth aspect, the present invention provides a method for producing a polypeptide or protein, the method comprising the steps of; 5 (i) providing a recombinant host cell comprising a polynucleotide molecule according to the fourth aspect, (ii) culturing said host cell in a suitable culture medium and under conditions suitable for expression of said polypeptide or protein, and (iii) isolating said polypeptide or protein from the culture, and, 10 optionally, from the culture medium. The polypeptide or protein may be isolated using any of the methods well known to persons skilled in the art. For example, the polypeptide or protein may be readily isolated using metal affinity chromatography techniques or using immobilised IgG or Heat-aggregated IgG (HAGG) chromatography techniques. 15 In a sixth aspect, the present invention provides a method of treating a subject for an inflammatory disease, said method comprising administering to said subject a polypeptide according to the first aspect or a protein according to the second aspect, optionally in combination with a pharmaceutically- or veterinary acceptable carrier or excipient. 20 The method is suitable for treatment of inflammatory diseases such as IC mediated inflammatory diseases including RA, ITP, SLE, glomerulonephritis and heparin-induced thrombocytopenia thrombosis syndrome (HITTS). The subject will typically be a human, but the method of the sixth aspect may also be suitable for use with other animal subjects such as livestock (e.g. racing horses) 25 and companion animals.

WO 2008/070927 PCT/AU2007/001934 21 The term "pharmaceutically- or veterinary-acceptable carrier or excipient" is intended to refer to any pharmaceutically- or veterinary-acceptable solvent, suspending agent or vehicle for delivering the polypeptide or protein of the present invention to the subject. 5 The polypeptide or protein may be administered to the subject through any of the routes well known to persons skilled in the art, in particular intravenous (iv) administration, intradermal (id) administration and subcutaneous (sc) administration and oral and nasal administration. For subcutaneous administration, the administration may be achieved through injection or by a 10 catheter inserted below the skin. Alternatively, subcutaneous administration may be achieved through sustained release implant compositions or injectable depot-forming compositions. Typically, the polypeptide or protein will be administered at a dose in the range of 0.5 to 15 mg/kg body weight of the subject per day. Persons skilled in the art 15 will, however, realise that the amount of an "effective dose" (i.e. a dose amount that will be effective in treating an inflammatory disease) will vary according to a number of factors including the age and general health of the subject and the severity of the inflammatory disease to be treated. It is well within the skill of persons skilled in the art to identify or optimise an appropriate effective dose 20 amount for each particular subject. In further aspects of the present invention, there is provided a composition comprising a polypeptide according to the first aspect or a protein according to the second aspect, optionally in combination with a pharmaceutically- or veterinary-acceptable carrier or excipient, and the use of a polypeptide according 25 to the first aspect or a protein according to the second aspect in the manufacture of a medicament for the treatment of an inflammatory disease. Further, the polypeptide and protein of the present invention are also useful in applications other than the treatment of a subject for an inflammatory disease.

WO 2008/070927 PCT/AU2007/001934 22 That is, they can be used in diagnostic assays for detecting circulating immune complexes (IC) associated with the pathology of autoimmune diseases such as RA and SLE, wherein the polypeptide or protein can be used in a step of "capturing" IC (e.g. by binding the polypeptide or protein to a suitable substrate 5 such as an ELISA plate) in place of the typical precipitation step (with polyethylene glycol) employed in such assays. After capturing IC from a sample (e.g. a serum or synovial fluid sample from a subject) to be assayed, the captured IC can be detected by using the polypeptide or protein of the present invention in a form whereby it is linked to a molecule which might serve as a marker or 10 reporter (e.g. radio-labelled molecules, chemiluminescent molecules, bioluminescent molecules, fluorescent molecules or enzymes such as horseradish peroxidase which can generate detectable signals). Alternatively, the captured IC could be detected or "probed" using antibodies specific for certain autoantigens (e.g. citrullene in RA, DNA in SLE, La/SS-B in Sjogren's syndrome, and DNA 15 topoisomerase I in scleroderma) to enable the determination of the level of specific autoantigens in circulating IC, which might allow for the development of assays for autoimmune diseases with improved diagnostic or prognostic results. Moreover, in a similar manner, IC captured by the polypeptide or protein of the present invention bound to a suitable substrate, could be detected or "probed" 20 using antibodies specific for certain antigens of infectious pathogens (e.g. bacteria such as Staphylococcus and Streptococcus, parasites such as P. falciparum (malaria) and viruses such as hepatitis C virus (HCV), Epstein-Barr virus (EBV), human immunodeficiency virus (HIV) and arbovirus causative of Dengue fever), to provide information useful in identifying the causative pathogen of an 25 infection, disease prognosis and/or the management of an infection. Still further, the polypeptide and protein of the present invention are also useful in various bioassays wherein they can usefully inhibit the release of tumour necrosis factor (TNF) from cells including macrophages, dendritic cells (DC) and neutrophils. Moreover, when linked to a molecule which might serve as a WO 2008/070927 PCT/AU2007/001934 23 marker or reporter such as those mentioned above, the polypeptide or protein can be used in in vivo imaging of sites of inflammation. Yet further, the polypeptide and protein of the present invention are useful for the removal of circulating IC associated with IC-mediated inflammatory diseases, 5 wherein the polypeptide or protein is bound to a suitable substrate such as an inert bead, fibre or other surface and exposed to a biological fluid (particularly blood) from a subject containing IC complexes such that IC are captured and subsequently removed from the biological fluid. The treated biological fluid, which is substantially depleted of IC, can then be returned to the subject from 10 which it was obtained. In order that the nature of the present invention may be more clearly understood, preferred forms thereof will now be described with reference to the following non-limiting examples. EXAMPLES 15 Example 1 Production, purification and characterisation of FcR multimer polypeptides Materials and Methods Construction of FcyRIIa multimer expression vectors The Fc binding region comprising the ectodomains 1 and 2 of human FcyRIIa 20 were amplified by using the thermostable polymerase Pwo (Roche), the clone Hu3.0 (Hibbs et al, 1988, ACCESSION NM021642) as cDNA template and the primers oBW10 GTAGCTCCCCCAAAGGCTG (SEQ ID NO: 1) and oBW11 CTACCCGGGTGAAGAGCTGCCCATG (SEQ ID NO: 2). The half SnaBI (all DNA modifying enzymes were from New England Biolabs) and SmaI sites are 25 underlined. The blunt PCR product was ligated using T4 DNA ligase into the vector pPIC9 (Invitrogen, Life Technologies) at the EcoRI site filled in with Klenow WO 2008/070927 PCT/AU2007/001934 24 fragment of DNA polymerase I yielding the vector pBAR14. To produce the vector pBAR28 encoding the tandem ectodomains of FcyRIIa, pBAR14 was digested with SnaBI into which site the SnaBI/SmaI fragment of pBAR14 was ligated. 5 A baculovirus vector for expressing FcyRIIa multimerised ectodomains was constructed as follows: The fragment encoding the FcyRIIa leader sequence and ectodomains 1 and 2 were obtained from pVL-1392 (Powell et al, 1999, and Maxwell et al, 1999) by digest with EcoRI and XbaI, and then ligated into the EcoRI/XbaI sites of modified pBACPAK9 (Invitrogen Life Tech) in which the 10 BamHI site in the multiple cloning site had first been eliminated by digest with BamHI, filling in using Klenow fragment of DNA polymerase and re-ligation. This construct, vector pBAR69, was digested with BamHI to which was ligated the BamHI fragment of pBAR28 yielding vectors pBAR71, pBAR72 and pBAR73 encoding rsFcyRIIa dimer, timer and tetramer respectively. Insert sizes were 15 defined by EcoRI/XbaI digest and the correct orientation of the multimerising BamHI fragment was screened by PvuII digest using standard protocols. The mammalian expression vectors encoding FcyRIIa monomer and dimer were produced as follows: The FcyRIIa cDNA clone Hu3.0 (Hibbs et al, 1988, and DEFINITION: Homo sapiens Fc fragment of IgG, low affinity Ila, receptor 20 (CD32)(FCGR2A), mRNA, ACCESSION NM_021642) was amplified using accuprime Pfx PCR (Invitrogen, Life Technologies) and cloned into the GatewayTM vector pDONRTM221 (Invitrogen, Life Technologies) using the BP clonaseTM reaction according to the manufacturer's instructions (Invitrogen Life Tech) yielding pNB6. PCR using polymerase accuprime Pfx of pNB6 with the 25 primers oBW11 and oBW302 TCTCATCACCACCATCACCACGTCTAGACCCAGCTTTCTTGTACAAAG (SEQ ID NO: 3), digest with SmaI and ligation with T4 ligase yielded pBAR390 encoding the rsFcyRIla with C-terminal hexahistidine tag. Digestion of pBAR390 with BamHI and ligation of the BamHI fragment of pBAR28 yielded vector WO 2008/070927 PCT/AU2007/001934 25 pBAR397, encoding rsFcyRIIa dimer. Pvu II digest was then used to screen for the orientation of the dimerising BamHI fragment and sequencing with ABI BigDye3.1 (Applied Biosytems) confirmed the target sequence. The Gateway LR clonase reaction (Invitrogen, Life Technologies) was then used to transfer the 5 FcyRIIa monomer (pBAR390) or dimer (pBAR397) into Gateway T M reading frame A cassette (Invitrogen, Life Technologies) adapted expression vector pAPEX3P (Evans et al, 1995, and Christiansen et al, 1996) to give the expression vectors pBAR426 and pBAR427. Likewise, the Gateway LR clonase reaction was used to transfer the FcyRIIa monomer (pBAR390) or dimer (pBAR397) into GatewayTM 10 reading frame-A cassette (Invitrogen, Life Technologies) adapted expression vector pIRESneo (Clontech). Figure 1 shows the polynucleotide sequence (and translated amino acid sequence) for the "head to tail" dimer construct of FcyRIIa within pBAR397 used to construct the expression vector pBAR427. The two repeats are shown as amino acids 1 to 174 (i.e. the first Fc binding region) and 184 15 to 362 (i.e. the second Fc binding region) and are linked via a short (8 amino acid sequence; residues 175 to 182) fragment of the FcyRIIa membrane proximal stalk plus an additional valine residue (residue 183 shown underlined in Figure 1). Amino acids -31 to -1 of the sequence shown in Figure 1 represent the natural leader sequence of FcyRIla. 20 Production of rsFcyRIIa monomer and dimer polypeptides Expression of recombinant soluble FcyRIIa (rsFcyRIIa) monomer and dimer polypeptides in HEK 293E cells was performed by transfection with 5 pg of plasmid DNA (pBAR426, pBAR427) in 10cm 2 wells and Lipofectamine 2000 reagent (Invitrogen, Life Technologies) or Transit reagent (BioRad Laboratories) 25 according to the manufacturer's instructions. After 48 hours, the transfected cells were then selected by incubation in 4pg/ml puromycin. Puromycin selected cells were then grown in 1% FCS supplemented CD293 media (Invitrogen, Life Technologies) to stationary phase. The recombinant product was subsequently purified by chromatography over immobilised Nickel (Qiagen) or cobalt WO 2008/070927 PCT/AU2007/001934 26 (Clontech) columns and further purified using Superdex 200 or Superdex G75 (Amersham/Pharmacia) size exclusion chromatography. Comparison of affinity measurements of rsFcyRIIa monomer and dimer polypeptides Using a standard BlAcore assay protocol (Wines et al, 2001; Wines et al, 2003), 5 affinity measurements for purified rsFcyRIIa monomer and dimer were conducted; the rsFcyRIIa monomer or dimer was injected at varying concentrations over immobilised human IgG monomer (Sandoglobulin, Novartis) or heat-aggregated IgG (HAGG, Wines et al, 1988; Wines et al, 2003) for 60 minutes, after which time the surface was regenerated (Wines et al, 2003). The 10 immobilisation of the human IgG monomer on the biosensor surface causes it to be a multivalent array which mimics an immune complex. Comparison of inhibitory activity of rsFcyRIIa monomer and dimer polypeptides Purified rsFcyRIla monomer and dimer were incubated with increasing concentrations of a solution of human IgG monomer (Sandoglobulin) and dimer 15 IgG (Wright et al, 1985). The amount of free receptor polypeptide was then measured by injecting over immobilised human IgG monomer in accordance with a standard BlAcore assay protocol. Inhibition of immune-complex binding to human cells by rsFcyRIla monomer and dimer polypeptides 20 Binding of small immune-complexes (represented by dimer-IgG) to human neutrophils (volunteers V1 and V5) was determined in the absence and presence of purified rsFcyRlla monomer and dimer polypeptides by flow cytometry analysis (Current Protocols in Immunology, Wiley Interscience). Inhibition of TNF secretion from immune-complex stimulated MDMs (monocyte-derived 25 macrophages) by rsFcyRIIa monomer and dimer polypeptides WO 2008/070927 PCT/AU2007/001934 27 In a first experiment, peripheral mononuclear cells were extracted from human blood (volunteer V5), positively sorted for CD14 expression using an automacs sorter (Miltenyi Biotec) and allowed to differentiate for 24 hours in the presence of M-CSF to MDMs (monocyte-derived macrophages) prior to stimulation with 5 varying concentrations of small immune-complexes (represented by dimer-IgG), in the absence and presence of rsFcyRIla dimer (in supernatant at 2.5 pg/ml). TNF secretion from the MDMs was then measured by human TNF ELISA according to manufacturers' protocol (BD Pharmingen). In a second experiment, MDMs were similarly produced ex vivo from human blood (this time from 10 volunteer V1) and allowed to differentiate for 24 hours prior to stimulation with varying concentrations of small immune-complexes (i.e. dimer-IgG), in the absence and presence of rsFcyRIla dimer (in supernatant at 2.5 pg/ml). Inhibition of immune complex activation of platelets by rsFcyRIIa dimer polypeptides Washed platelets were prepared by low speed centrifugation of whole blood 15 (Thai et al, 2003) and stimulated with heat-aggregated IgG (HAGG). Activation of platelets was measured by increased surface expression of P-selectin (CD62P) by flow cytometry (Lau et al, 2004). Results Expression of rsFcyRIIa monomer and multimer polypeptides from insect cells 20 Western blot analysis of infected cell supernatants demonstrated successful production of dimer and timer forms of recombinant soluble FcyRIIa (Figure 2). Although some timer polypeptide was detected this was largely cleaved to the dimeric form and the tetramer was not detected being largely cleaved yielding a dimer form. Since the FcyRIIa dimer was intrinsically the most stable, this was 25 further characterised and developed in a mammalian expression system (i.e. HEK293E cells).

WO 2008/070927 PCT/AU2007/001934 28 However, since native FcyRIIa can be shed from leukocyte cell surfaces by proteolysis (Astier et al, 1994), one strategy for minimising proteolysis of the trimers, tetramers and larger multimers would be to eliminate or, more preferably, replace the membrane proximal stalk linker sequence linking the 5 FcyRIIa extracellular regions. For example, the proteolytic susceptibility of membrane proximal stalk linker sequence could be reduced by one or more amino acid modifications (e.g. one or more amino acid substitution(s), deletion(s) and/or addition(s)) or by otherwise replacing that linker sequence with a synthetic linker sequence such as, for example, GGGGSGGGGS (SEQ ID NO: 4) 10 which has a low susceptibility to proteolysis. Another strategy for successfully producing trimers, tetramers and larger multimers, would be to link an expressed dimer polypeptide to one or more monomer or other dimer polypeptide(s) by chemical cross-linking. Multimers of FcyRIIa dimers may also be produced by expressing the dimer polypeptide as a 15 fusion protein with an Fc domain (e.g. an IgG Fc domain) which is of itself dimeric and will thus dimerise any fusion partner. Expression of rsFcyRIla monomer and dimer polypeptides from mammalian cells Protein yield for purified rsFcyRIIa monomer was 3 mg/l (construct pBAR 426) and for the rsFcyRIIa dimer, to -0.5 mg/l (construct pBAR 427). Figure 3 shows 20 Coomassie-stained SDS-PAGE (12% acrylamide gel, under non-reducing conditions) of fractions collected from the purification of rsFcyRIIa monomer and dimer. The rsFcyRIIa monomer had the expected size of -30 kDa (a), while the rsFcyRIIa dimer had the expected size of -60 kDa (b). Comparison of affinity measurements of rsFcyRIla monomer and dimer polypeptides 25 The results of the assays are shown in Figures 4 and 5. The assays indicated that rsFcyRIIa monomer has a single-binding site with affinity dissociation constant (KD) of 1.7 pM for human IgG monomer and 1.05 pM for HAGG. In the case of WO 2008/070927 PCT/AU2007/001934 29 the rsFcyRIIa dimer, the binding data best fitted a two-binding site model with affinity dissociation constants (KD) of 3.2 nM (KD1) and 100 nM (KD2) for immobilised human IgG monomer and 2.73 nM (KD1; approximated 300-fold lesser than the KD of monomeric rsFcyRIIa) and 99 nM (KD2) for HAGG. 5 Comparison of inhibitory activity of rsFcyRIIa monomer and dimer polypeptides The experiments conducted to compare the inhibitory activity of the rsFcyRIIa monomer and dimer polypeptides showed that, in solution, rsFcyRIIa monomer (Figure 6a) does not distinguish between human IgG monomer and small immune-complexes (i.e. represented by dimer-IgG). In contrast, rsFcyRIIa dimer 10 (Figure 6b) in solution, selectively binds to small immune-complexes (i.e. dimer IgG) over human IgG monomer. Inhibition of immune-complex binding to human cells by rsFcyRIIa monomer and dimer polypeptides 15 The results of the inhibition assays are shown in Figure 7a and 7b, and these indicate that rsFcyRIIa dimer (IC 5 o= 1.1 pg/ml) is -10-fold more active than rsFcyRIIa monomer (IC 5 o= 10.5 ptg/ml) at inhibiting small immune-complexes (i.e. dimer-IgG) from binding to human neutrophils. Further, the results showed that the inhibition of small immune-complexes (dimer-IgG) from binding to human 20 neutrophils by rsFcyRIIa dimer was reproducible with neutrophils from two different individuals, with an IC 5 o of 0.9-1.1 pg/mI. Inhibition of TNF secretion from immune-complex stimulated MDMs (monocyte-derived macrophages) by rsFcyRIla monomer and dimer polypeptides The results are shown in Figures 8a and 8b. The rsFcyRIIa dimer appeared to 25 inhibit immune-complex (i.e. dimer-IgG) stimulated TNF secretion from 24 hour differentiated human MDMs.

WO 2008/070927 PCT/AU2007/001934 30 Inhibition of immune complex activation of platelets by rsFcyRIla dimer polypeptides Washed human platelets were incubated with HAGG (10 pg/ml) in the presence and absence of rsFcyRIIa dimer at 30 pig/ml for 30 minutes. As shown in Figure 9, it was found that the activation of platelets was inhibited in the presence of the 5 rsFcyRIIa dimer as evidenced by the lesser expression of P-selectin (CD62P). Discussion Recombinant soluble FcyRIIa in the monomeric (rsFcyRIIa monomer) and dimeric (rsFcyRIla dimer) form was successfully expressed in HEK 293E cells. BIAcore equilibrium binding assays demonstrated that the rsFcyRIIa dimer has 10 an ~ 300 fold greater avidity for immobilised IgG (Sandoglobulin) than the monomeric receptor (i.e. the rsFcyRIIa monomer has a KD ~1PM while the rsFcyRIla dimer has a KD ~ 3nM in the interaction with the immobilised IgG). Competition experiments using BIAcore also demonstrated that the rsFcyRIIa dimer selectively binds small immune complexes, and selective inhibitory activity 15 was confirmed in cell based assays using neutrophils from two donors. The rsFcyRIla dimer also proved to be about 10 times more potent an inhibitor of small IgG immune complex binding than the rsFcyRIla monomer, and in a standard platelet assay, the rsFcyRIIa dimer was observed to completely inhibit immune complex activation of platelets (i.e. rsFcyRIIa dimer is a potent inhibitor 20 of cell activation). It is therefore considered that rsFcyRIla dimer and other soluble multimeric proteins and polypeptides according to the invention show considerable promise for the treatment of IC-mediated inflammatory disease such as RA and SLE.

WO 2008/070927 PCT/AU2007/001934 31 Example 2 Production, purification and characterisation of rsFcyRIIa dimer polypeptide Materials and Methods Production of rsFcyRIIa dimer polypeptides 5 The FcyRIIa dimer construct described in Example 1 was cloned into a mammalian expression vector under the control of a modified CMV promoter. Stable CHO-S transfectants were then established as follows: CHO-S cells at 90% confluency were harvested, washed three times, and 2x10 7 cells in 15 ml medium were dispensed into 10 cm petri dishes. Linearised DNA-lipofectamine 2000 10 complexes (1:2.5 ratio) were then incubated for 5 minutes at room temperature and added dropwise to the cells. Subsequently, cells were incubated at 37*C for 48 hours, and then plated out in limiting dilution in 96-well plates in CD-CHO medium supplemented with 600 ptg/ml hygromycin B, 8 mM L-glutamine, 1x HT supplement and 50 pg/ml dextran sulfate. Cells were screened by standard 15 ELISA to detect soluble FcyRIla protein, and the highest expressing lines were subcloned again by limiting dilution. One clone (#6) secreted FcyRIIa dimer at approximately 40 mg/L and was cultured in shaker flasks at 30*C for optimal protein expression. Supernatant containing rsFcyRIIa dimer was concentrated by tangential flow 20 filtration and exchanged into 20 mM sodium phosphate pH 7.4 buffer. The sample was then diluted four-fold in 20 mM sodium phosphate and 0.5 M sodium chloride and purified over a HisTrap FF 2x5ml column (GE Healthcare), eluting in 20 mM sodium phosphate, 0.5 M sodium chloride pH 7.4 and 100 mM imidazole. The eluted material was 25 dialysed and purified by ion exchange chromatography, using a 25 ml Q FF column (GE Healthcare) and eluting in 150 mM sodium chloride. The purified material was then dialysed into phosphate buffered saline.

WO 2008/070927 PCT/AU2007/001934 32 Blocking of HAGG binding to FcyRIIb with rsFcyRIIa dimer polypeptides The ability of purified rsFcyRIla dimer to block immune complex binding to cell surface FcyRIIb was assessed by a flow cytometric assay. Heat-aggregated IgG (HAGG) was incubated with various concentrations of rsFcyRIIa dimer or 5 rsFcyRIIa monomer (R&D Systems, Cat # 1330-CD/CF) for 1 hour at 4*C. These mixtures were then added to 96-well plates containing 105 IIA1.6 cells transfected with human FcyRIIb (IIA1.6 is a mouse B lymphoma line that lacks endogenous FcyR expression). The plates were incubated for 1 hour at 4"C, washed and then stained with an anti-hIgG-FITC conjugate to detect bound HAGG. After washing, 10 the cells were analysed on a FACS Scan flow cytometer using standard protocols. Blocking of HAGG-induced platelet activation with rsFcyRIIa dimer polypeptides Exposure of platelets to HAGG results in activation via FcyRIIa, leading to upregulation of P-selectin (CD62P). The ability of rsFcyRIla dimer or rsFcyRIIa monomer to block this activation was assessed by a flow cytometric assay. Heat 15 aggregated IgG (HAGG) was incubated with various concentrations of rsFcyRIla dimer or rsFcyRIla monomer (R&D Systems, Cat # 1330-CD/CF) for 1 hour at 4*C. The mixture was then added to 96-well plates containing 3x10 7 human platelets, which had been previously washed and resuspended in Tyrodes/Hepes buffer supplemented with 1 mM EDTA. After a 30 minute 20 incubation at room temperature, the cells were washed, fixed, and stained for CD62P and GPIIb (CD41) expression by standard methods and analysed on a FACS Scan flow cytometer. rsFcyRIla dimer polypeptides block immune complex-induced MC/9 activation 25 MC/9 is an FcyR-positive murine mast cell line that becomes activated and releases TNF-a after exposure to immune complexes. The ability of rsFcyRIIa dimer or rsFcyRIIa monomer to block this activation was assessed using immune complexes consisting of ovalbumin and anti-ovalbumin antibody (OVA ICs) as a WO 2008/070927 PCT/AU2007/001934 33 stimulus. OVA ICs (10 pg) were incubated with various concentrations of rsFcyRIIa dimer or rsFcyRIIa monomer (R&D Systems, Cat # 1330-CD/CF) for 1 hour at room temperature. The mixture was then added to 96-well plates containing 2x10 5 MC/9 cells, and incubated overnight at 37"C. Supernatant was 5 collected and the amount of TNF-a measured using a commercial ELISA kit (BD Biosciences). rsFcyRIIa dimer inhibits induced arthritis in human FcyRIIa transgenic mouse model The activity of the rsFcyRIla dimer was assessed in an arthritis model using the transgenic human FcyRIIa mice described in published PCT application 10 W003/104459, incorporated herein by reference. These mice express a transgene that encodes human FcyRIIa. Clinically apparent arthritic disease (determined using the standard arthritis index) is elicited in these mice at least by Day 4 following administration of a single 2mg dose of monoclonal antibody M2139 in PBS, which is an IgG2a that 15 binds specifically to the J1 epitope of collagen II, amino acids 551-564. The monoclonal antibody is produced by hybridomas proven to be arthritogenic (Amirahmadi et al., Arthritis and Rheumatism, June 2005; Nandakumar et al., Arthritis Research and Therapy, May 2004). The mice were treated using the soluble FcyRIIa dimer shown in Figure 1, as 20 follows: Four FcyRIIa Tg mice were injected with 0.5mg soluble FcyRIIa dimer, and a control group of four mice were given PBS, i.p. Two hours later, both groups were injected with 2mgs of M2139 (ip) and a bolus dose of 1mg of dimer or PBS. Dimer (0.5mg/ dose) was given again at both 24 and 48 hours following injection of M2139. Arthritis was scored as usual, with a maximum score possible 25 of 12. The sum of four paws each scored 0-3 (0 = normal; 1 = one affected joint, erythema, minor swelling; 2 = Two or more affected joints, ankle/wrist swelling; 3 = all joints affected, loss of mobility/ankylosis, profound erythema and oedema).

WO 2008/070927 PCT/AU2007/001934 34 Results Production of rsFcyRIIa dimer polypeptides Figure 10 shows the analysis of purified rsFcyRIla material, including SDS-PAGE (under reducing and non-reducing conditions); Western blotting using an anti 5 FcyRIla antibody (R&D systems, catalogue number AF1875) and rabbit anti-goat IgG-peroxidase as the detector antibody; and HPLC. The polypeptide migrates as a single band at the expected molecular weight (- 50 kD), reacts with anti FcyRIla antibody and is >96% pure as determined by HPLC analysis. rsFcyRIIa dimer polypeptides block HAGG binding to FcyRIIb 10 The results of the HAGG binding assay are shown in Figure 11. Both the rsFcyRIla dimer and rsFcyRIla monomer were able to completely block the binding of HAGG to cell surface FcyRIIb. However, the rsFcyRIla dimer (IC50 = 3.9 ng/ml) was over 500-fold more potent than the monomer protein (IC50 = 2082 ng/ml). 15 rsFcyRIIa dimer polypeptides block HAGG-induced platelet activation The results of the platelet activation assay are shown in Figure 12. The percentage of activated platelets (positive for both CD41 and CD62P) after treatment with HAGG alone was defined as 100%. Both rsFcyRIIa dimer and rsFcyRIla monomer were able to significantly reduce HAGG-induced CD62P 20 upregulation. Titration revealed that the rsFcyRIIa dimer (IC50 = 3.9 pg/ml) was 5-fold more potent than the rsFcyRIla monomer (IC50 = 20.9 pg/ml). rsFcyRIIa dimer polypeptides block immune complex-induced MC/9 activation The results of the MC/9activation assay are shown in Figure 13. The amount of 25 TNF-a released after incubation with OVA ICs alone was defined as 100%. Both WO 2008/070927 PCT/AU2007/001934 35 rsFcyRIIa dimer and rsFcyRIla monomer were able to completely suppress TNF-x release induced by immune complexes. Titration revealed that the rsFcyRIIa dimer (IC50 = 2.1 ptg/ml) was 8-fold more potent than the rsFcyRIla monomer (IC50 = 17.7 ptg/ml). 5 rsFcyRIIa dimer inhibits induced arthritis in human FcyRIIa transgenic mouse model As shown in Figure 18, the administration of the rsFcyRIIa dimer provided a dramatic reduction in arthritis score. A second experiment confirmed the reduction in arthritis score mediated by the dimer, albeit in less dramatic fashion. Discussion 10 The rsFcyRIla dimer was successfully expressed in CHO-S cells. Reducing and non-reducing SDS-PAGE showed that the purified rsFcyRIla dimer was approximately 50 kDa in size, and Western blotting showed that the dimer was specifically bound by anti-FcyRIIa antibodies. The rsFcyRIIa dimer was determined to be 96% pure by HPLC. 15 Both the rsFcyRIIa dimer and rsFcyRIla monomer completely blocked binding of HAGG to cell surface FcyRIIb, with the rsFcyRIIa dimer having an approximately 500-fold increased blocking efficiency than the rsFcyRIIa monomer. Similarly, both the rsFcyRIIa dimer and rsFcyRIla monomer significantly reduced HAGG induced platelet activation, with the dimer showing approximately 5-fold higher 20 efficacy than the rsFcyRIla monomer. Further, both the rsFcyRIIa dimer and rsFcyRIIa monomer suppressed mouse mast cell line (MC/9) activation, as measured by TNF-a release, with the dimer showing 8-fold greater efficacy than the monomer. Importantly, the rsFcyRIla dimer ameliorated arthritis in a mouse model of 25 induced arthritis, demonstrating in vivo effectiveness.

WO 2008/070927 PCT/AU2007/001934 36 Example 3 Engineering and expression of rsFcyRIla fusion polypeptides comprising an Fc domain derived from IgGi Materials and Methods Construction of rsFcyRIIa fusion expression vectors 5 Polynucleotides encoding soluble monomer FcyRIIa or soluble dimer FcyRIIa were independently fused to a polynucleotide encoding IgGi-Fcyl (L234A, L235A). The C-terminal of the soluble monomer FcyRIIa polypeptide was operably fused to a human IgGi polypeptide at a position on the N-terminal side of the inter 10 chain disulphide bond in the lower hinge that covalently joins the two Fc portions. Fusion at this position generates a monomeric FcyRIIa-IgGi-Fcyl (L234A, L235A) fusion protein which will dimerise with a second Fc domain due to interactions between the covalently associated Fc domains. The IgG hinge region is known for its flexibility, and fusion of the polypeptide comprising the 15 Fc binding region to the N-terminal side of the inter-chain disulphide bond in the lower hinge allows considerable freedom of movement of the Fc binding region. Similarly, the C-terminal of the soluble dimer FcyRIIa polypeptide was operably fused to a human IgGi polypeptide at a position on the N-terminal side of the inter-chain disulphide bond in the lower hinge that covalently joins the two Fc 20 portions. Polynucleotides encoding soluble monomer FcyRIIa or soluble dimer FcyRIIa were independently fused to a polynucleotide encoding human serum albumin (HSA) in an equivalent manner to that previously described in International patent specification no WO 96/08512. As disclosed in that specification, the HSA 25 was fused to the N-terminal of the rsFcyRIIa monomer. In a similar manner, the HSA was fused to the N-terminal of the rsFcyRIIa dimer.

WO 2008/070927 PCT/AU2007/001934 37 Polynucleotides encoding the various fusion polypeptides or proteins were operably inserted into pAPEX 3P-xDEST using standard cloning techniques. Production of rsFcyRIla monomer and rsFcyRIla dimer fusions The rsFcyRIla monomer and dimer fusion expression vectors were transiently 5 transfected into CHOP cells and stably transfected in 293E cells using standard methods. Transiently transfected CHOP cell supernatants were immunoprecipitated using anti-FcyRIIa antibody 8.2 (Powell et al., 1999) and immunoprecipitates were subjected to non-reducing SDS-PAGE (12%). Western blot analysis was then performed using standard methods and utilising rabbit 10 anti FcyRIIa antibody (Maxwell et al., 1999) as a primary antibody and anti rabbit Ig-HRP as a secondary antibody. HAGG-capture ELISA for detection of rsFcyRIIa fusions in transfected CHOP cell supernatants HAGG-capture ELISAs were performed to measure the Fc binding activity of the 15 rsFcyRIIa fusions. To examine the binding activity of the rsFcyRIIa monomer fusions, a known FcyRIIa monomer standard (Powell et al., 1999) (starting at 0.75 pg) and the protein from an rsFcyRIla monomer transfected cell (transfection 426) titrated and compared with the binding of protein from cells transfected with rsFcyRIla monomer fusion to IgG-Fcyl (L234A, L235A) (monomer-Fc) and 20 protein from cells transfected with rsFcyRIIa monomer fusion to HSA (HSA monomer). To examine the binding activity of rsFcyRIIa dimer fusions, a known FcyRIIa dimer standard (starting at 0.5 pg/ml) and protein from a cell transfected with rsFcyRIIa dimer (transfection 427) were titrated and compared with the binding 25 of protein from cells transfected with rsFcyRIIa dimer fusion to IgG-Fcyl (L234A, L235A) (dimer-Fc) and protein from cells transfected with rsFcyRIIa dimer fusion to HSA (HSA-dimer).

WO 2008/070927 PCT/AU2007/001934 38 Capture-Tag ELISA for detection of rsFcyRIla fusions in transfected CHOP cell supernatants Using a standard ELISA method, plates were coated with anti FcyRIIa antibody 8.2. The rsFcyRIIa fusions were added to the wells and contacted with the 8.2 5 antibody. The secondary antibody was anti FcyRIIa antibody 8.7-HRP (Powell et al., 1999; lerino et al., 1993a), which is specific for a different FcyRIIa epitope than antibody 8.2. The monomeric rsFcyRIla samples tested included a known rsFcyRIIa monomer (monomer standard starting at 0.75 ptg/ml), the supernatant from rsFcyRIla 10 monomer transfected cell (transfection 426), the supernatant from cells transfected with rsFcyRIla monomer fusion to IgG-Fcyl (L234A, L235A) (monomer-Fc) and the supernatant from cells transfected with rsFcyRIla monomer fusion to HSA (HSA-monomer). The dimeric rsFcyRIIa samples tested included a known FcyRIla dimer (dimer 15 standard starting at 0.5 pig/mi), the supernatant from rsFcyRIIa dimer transfected cell (transfection 427), the supernatant from cells transfected with rsFcyRIla dimer fusion to IgG-Fcyl (L234A, L235A) (monomer-Fc) and the supernatant from cells transfected with rsFcyRIIa dimer fusion to HSA (HSA-monomer). Results 20 Expression of rsFcyRIIa monomer and dimer fusions On the basis of the activity of purified rsFcyRIla monomer and rsFcyRIla dimer, the rsFcyRIIa monomer-IgG-Fcyl (L234A, L235A) fusion was secreted at higher levels (approximately 12 ptg/ml in 293E cells) than the rsFcyRIla dimer-IgG-Fcyl (L234A, L235A) fusion (approximately 4 jig/ml in 293E cells). 25 As shown in Figure 14, Western blot analysis indicated that the fusion proteins were present in the supernatant at the expected molecular weight sizes and that WO 2008/070927 PCT/AU2007/001934 39 they could be successfully produced as distinct proteins without evidence of degradation products. HAGG-capture ELISA for detection of rsFcyRIIa fusions in transfected CHOP cell supernatants 5 As shown in Figure 15(a), rsFcyRIIa monomer fusion to IgG-Fcyl (L234A, L235A) (monomer-Fc) was detectably bound in the assay, while rsFcyRIla monomer fusion to HSA (HSA-monomer) was observed to bind poorly. This result may be explained by the fact that the rsFcyRIla monomer fusion to IgG-Fcyl (L234A, L235A) will be a dimer (of the Fc binding region) as a consequence of 10 dimerisation between the heavy chains of the fused Fc domains whereas rsFcyRIla monomer fusion to HSA remains monomeric for the Fc binding region. As shown in Figure 15(b), purified rsFcyRIIa dimer fusion to IgG-Fcyl (L234A, L235A) (dimer-Fc) showed binding activity similar to the dimer standard, and rsFcyRIIa dimer fusion to HSA (HSA-monomer) had detectable, but lower, 15 binding activity. In this case, the rsFcyRIIa dimer fusion to IgG-Fcyl (L234A, L235A) was, due to dimerisation between the heavy chains of the fused Fc domains, tetrameric (or "tetravalent") for the Fc binding region, whereas rsFcyRIIa dimer fusion to HSA remains dimeric for the Fc binding region. Capture-tag ELISA for detection of rsFcyRIIa fusions in transfected CHOP cell 20 supernatants As shown in Figure 16 (a), rsFcyRIIa monomer fusion to IgG-Fcyl (L234A, L235A) and the rsFcyRIIa monomer fusion to HSA are both captured and detectable in this assay. As shown in Figure 16(b), rsFcyRIIa dimer fusion to IgG Fcyl (L234A, L235A) and the rsFcyRIIa dimer fusion to HSA are also both 25 captured and detectable in this assay. Clearly, both the 8.2 epitope used to capture these receptors and the 8.7 epitope used to detect the captured receptors are intact indicating correct folding of the fusions.

WO 2008/070927 PCT/AU2007/001934 40 Discussion The rsFcyRIIa monomer and rsFcyRIIa dimer fusion constructs were expressed from the vector p-APEX 3P-xDST and expressed transiently in CHOP cells and stably in 293E cells. The expressed fusions presented as distinct proteins on 5 Western blot with no evidence of degradation products. The rsFcyRIIa dimer fusion to IgG-Fcyl (L234A, L235A) may show lower expression levels than its monomeric counterpart. However, the expression level of rsFcyRIla dimer fusion to HSA was nearly equivalent to the expression level of rsFcyRIla monomer fusion to HSA, as determined by Western blot (Figure 14), 10 and therefore shows considerable promise as a means for large scale production of rsFcyRIla dimer. Of interest, the rsFcyRIIa dimer fusions showed higher HAGG and anti-FcyRIIa antibody 8.2 binding activity than monomeric counterparts. As mentioned above, this may be explained by the fact that the rsFcyRIIa dimer fusions were dimeric or tetrameric (in the case of the rsFcyRIla 15 dimer fusion to IgG-Fcyl (L234A, L235A)) for the Fc binding region, and as a consequence, possessed a higher apparent binding affinity (avidity) because of this multi-valency. It is anticipated that tetrameric molecules may bind to immune complexes with such affinity that the binding will be substantially irreversible. 20 Example 4 Engineering and expression of rsFcyRIIa fusion polypeptides comprising an Fc domain derived from IgG2a In this example, rsFcyRIIa dimer fusion proteins were prepared using a modified murine IgG2a Fc domain as a fusion partner. The dimer fusion protein was designated D2. The activity of this protein was compared with the rsFcyRIla 25 dimer lacking a fusion partner (as described in Example 1 and 2), and with a rsFcyRIla monomer fusion protein, wherein the fusion partner was the modified murine IgG2a Fc domain.

WO 2008/070927 PCT/AU2007/001934 41 Design of recombinant soluble D2FcyRIIaFc (D2) protein The translated amino acid sequence (SEQ ID NO: 8) and nucleotide sequence (SEQ ID NO: 9) of the D2 protein are shown in Figures 19 and 20, respectively. The D2 protein comprises a dimer of a polypeptide consisting of the native 5 FcyRIIa signal sequence (amino acids 1-31), the extracellular domains of an FcyRIla protein (amino acids 32-205), a short linker corresponding to the FcyRIIa membrane proximal stalk plus an additional valine residue (residues 206-214), a second FcyRIIa protein (residues 215-385), a repeat of the membrane proximal stalk linker (residues 386-393) and a mouse IgG2a Fc domain (hinge-CH2-CH3) 10 region (residues 394-625). The IgG2a Fc domain contains the following four mutations, which were introduced to reduce Fc receptor binding and complement fixation: Leu-413 to Glu (corresponding to position 235 in the EU numbering system), Glu-496 to Ala (corresponding to EU position 318), Lys-498 to Ala (corresponding to EU position 320) and Lys-500 to Ala (corresponding to 15 EU position 322). Construction of the D2 expression vector The cDNA encoding the signal peptide and extracellular domains of human FcyRIIa was amplified by PCR using a previously constructed plasmid (FcyRIIa d/pAPEX-dest) as a template and primers 1 and 4 as shown in Table 1. The 20 mutated mouse IgG2a Fc region was amplified by PCR using a previously constructed plasmid (CD200IgG2aFc-d) as a template and primers 2 and 3 as shown in Table 2.

WO 2008/070927 PCT/AU2007/001934 42 Table 2 Primers used for plasmid construction (restriction enzyme sites are underlined) Primer Sequence 1. Mouse FcyRIIa Forward 5 'GGGATATTGCTAGCGCCACCATGGAGACCCAAATG3' (SEQ ID NO: 10) 2. Mouse IgG2a 5 'TATCTAGACCGGTTATCATTTACCCGGAGTCCGGGAGAAGCTC3' Reverse (SEQ ID NO: 11) 3. FcyRIIa Di mIgG2a 5 'AGCTCTTCACCCCCCAGAGGGCCCACAATCAAGCCCTGTCCTC3' MidFor (SEQ ID NO: 12) 4. FcyRIIa Di mIgG2a 5 'GGCCCTCTGGGGGGTGAAGAGCTGCCCATGCTGG3' MidRev (SEQ ID NO: 13) The FcyRIIa and modified mouse IgG2a Fc PCR products were then amplified by 5 overlapping PCR using primers 1 and 2. Amplification was carried out by using platinum Pfx DNA polymerase (Invitrogen), in 1mM MgSO 4 , 0.4 mM each dNTP, 20 pmol of each primer and 100 ng of template DNA under the following conditions: initial melting at 94 0 C for 5 min, followed by 30 cycles consisting of 94'C for 1.5 min, then 65 0 C for 2 min, then 72"C for 3min. The reactions were then 10 held at 72 0 C for 10 min and cooled to 4"C. The reaction products were electrophoresed through 0.7% agarose gels and visualised with ethidium bromide. The DNA band of interest was excised and purified from agarose gel by using QlAquick Gel Extraction Kit (Qiagen). This purified PCR product was digested with NheI and AgeI restriction enzymes and purified using the Qiaquick 15 PCR Purification Kit (Qiagen). The fragment was then ligated by T4 DNA ligase into the pMPG expression plasmid that had been similarly digested with NheI and AgeL. The ligation reaction (5 ptl) was then transformed into 50 pd of competent Escherichia coli DH5a cells (Invitrogen) according to the manufacturer's WO 2008/070927 PCT/AU2007/001934 43 instructions. Transformants were spread on LB-agar plates containing 100 pg/ml ampicillin, followed by incubation at 37 0 C for 16 hours. Plasmid DNA was purified from small-scale E. coli cultures by mini-prep, and the DNA sequence confirmed. A diagram of the resulting expression plasmid pMPG-D2 FcyRIIA 5 IgG2aFc is shown in Figure 21. Generation of CHO clones expressing D2 pMPG-D2 Fc-yRIla-IgG2aFc plasmid DNA was isolated from a large culture of E.coli using a plasmid Maxi kit (Qiagen), linearised by XbaI, and purified by using QIAGEN tips. CHO-S cells growing in serum-free chemically defined medium 10 were transfected with the linearised plasmid using Lipofectamine 2000 reagent. After 48 hours, the cells were transferred into 96-well plates at different concentrations (10000, 5000, or 2000 cells/well) in medium containing 600 ptg/ml of hygromycin B. Drug-resistant oligoclones were screened by ELISA as follows: 96-well plates were coated with 100 pl of goat anti-mouse IgGFc (Sigma) and 15 incubated overnight at 4 0 C. The wells were washed and blocked with 200 pl of 2% BSA in PBST at room temperature for 1 hour. After washing, 100 pl samples were diluted with 1% BSA in PBST, added to the wells, incubated for 1 hour, washed and then incubated with HRP-conjugated goat anti-mouse IgG (Fc specific) (Sigma) for 1 hour at room temperature. The wells were washed and 20 TMB substrate added and incubated for 3 to 5 min at room temperature. Absorbance was measured at 450 nm, and a standard curve constructed using known amount of purified mouse IgG or D2 FcyRIIa-IgG2aFc. Supernatant samples were also analysed by SDS-PAGE and Western blotting. For SDS-PAGE, samples were resuspended in sample buffer with or without 2-ME and heated at 25 95"C for 10 min and chilled on ice. The samples were then separated on a 8% SDS-PAGE gel. The gel was then stained with Coomassie Blue according to the manufacturer's instructions. For Western blotting, samples were prepared and separated on a SDS-PAGE as described above and then transferred onto ImmunoBlot PVDF Membrane (Bio-Rad) for 1 hr at 100V. The membrane was WO 2008/070927 PCT/AU2007/001934 44 blocked for 1 hr in 5% skim milk in PBS/0.1% Tween-20 and incubated for 1 hr with 0.2 pg/ml goat anti-human FcyRIIa antibody (R&D Systems) and 1 hr with HRP-conjugated rabbit anti-goat IgG (whole molecule from Sigma), then developed using TMB substrate (Vector Laboratories Inc). A second limiting 5 dilution was performed at lower different concentrations (0.25, and 0.5 cell/well) in medium containing 600 pg/ml of hygromycin B. After 2 to 3 weeks, the drug resistant clones were again assessed for recombinant protein production by ELISA and tested by Western blot. Purification of D2 protein 10 CHO transfectants were grown in shaker flasks at 37 0 C. When the cells reached a density of 1.5 to 2x10 6 cells/ml, they were incubated at 30*C for 7 to 10 days with constant agitation. Supernatant was collected, centrifuged at 3000 x g for 30 min at 4*C, and filtered through a series of different autoclaved membrane filter pore sizes (5.0 to 0.2 pm). Tangential flow filtration (Millipore) using a BioMax 10 15 membrane was used to concentrate the supernatant and perform buffer exchange into 20 mM Na-P/148 mM NaCl, pH 7.8. The material was then diluted 9-fold with binding buffer (20 mM Na-P & 3 M NaCl, pH 7.8) and loaded onto a Protein A column (GE Heathcare) at 4 mil/min overnight at 4 0 C. The column was washed with binding buffer (20 volumes at 5 ml/min), and protein eluted with 0.1 M 20 citric acid pH 4.0 at 2 mIl/min. Eluted material was pH adjusted to neutral and dialysed against 4 L of 10 mM Na-P, pH 6.0 at 4*C overnight. It was then loaded onto a macro-prep 40 pm ceramic hydroxyapatite type II (CHT II) column, (Bio Rad). After washing the column with binding buffer, the protein was eluted with 10 mM Na-P, 500 mM NaCl, pH 6.0 (all manipulations at a flow rate of 5 ml/min. 25 The eluted material was then dialysed against 3 X 4 L of PBS, pH 7.4 at 4'C. Protein concentration was determined by absorbance at 280 nM (1.34 extinction coefficient). Figure 22A shows the SDS-PAGE analysis of the final purified material. Western blot analysis is shown in Figure 22B.

WO 2008/070927 PCT/AU2007/001934 45 D2 protein blocks immune complex induced MC/9 mast cell activation The D2 protein was tested for the ability to block immune complex-mediated activation of Fcy receptors in a MC/9 mast cell assay. MC/9 is an FcyR-positive murine mast cell line that becomes activated and releases TNF-a after exposure to 5 immune complexes. 10 pLg of ovalbumin-anti-ovalbumin immune complexes (OVA ICs) were incubated with purified D2 for 1 hour at room temperature. OVA ICs were also incubated with purified BIF (a variant of D2 lacking an Fc tag as described in Examples 1 and 2) and purified M2 protein (a variant of D2 that contains only a single FcyRlIa subunit fused to the modified IgG2a Fc domain). 10 The mixture was then added to 96-well plates containing 2x105 MC/9 cells, and incubated overnight at 37"C. Supernatant was collected and the amount of TNF a measured by commercial ELISA kit. The results are shown in Figure 23, where the amount of TNF-a released in the absence of treatment has been defined as 100%. The D2 and M2 proteins and rsFcyRIIa dimer polypeptide completely 15 suppress OVA IC-mediated activation. The D2 protein, however, is 3-fold more potent than the non-Fc tagged dimer, and 12-fold more potent than Fc-tagged monomer (M2). D2 protein blocks immune complex mediated activation of FcyR in a neutrophil activation assay 20 The D2 protein was also tested for the ability to block immune complex-mediated activation of Fcy receptors in a neutrophil activation assay. Resting human neutrophils express both FcyRIIa and FcyRIIIb and rapidly lose cell surface expression of L-selectin (CD62L) upon activation by immune complexes. OVA ICs (100 pg/ml) were incubated either alone or with titrated amount of purified 25 D2, M2 or BIF for 1 hour on ice. The mixture was then added to 96-well plates containing 2x10 5 /well human neutrophils, which had been purified from peripheral blood by dextran sedimentation and Ficoll density gradient centrifugation. The plates were incubated at 37*C for 15 min, and the reaction WO 2008/070927 PCT/AU2007/001934 46 terminated by addition of an equal volume of ice cold buffer followed by incubation on ice for 5 min. The level of CD62L on the neutrophil cell surface was then determined by flow cytometry. The results are shown in Figure 24, where the percent of cells expressing CD62L in the presence of OVA IC alone is defined 5 as 100% percent activation, and the percent of CD62L-expressing untreated cells as 0% activation. BIF and M2 protein showed similar suppressive activity. The D2 protein, however, is approximately 6-fold more potent. D2 protein blocks immune complex mediated activation of FcyR in a platelet activation assay 10 In addition, the D2 protein was tested for the ability to block immune complex mediated activation of Fcy receptors in a platelet activation assay. Exposure of platelets to heat aggregated IgG (HAGG), a typical immune complex, results in activation via FcyRIIa, leading to upregulation of P-selectin (CD62P). Heat aggregated IgG (HAGG) was incubated with various concentrations of D2 15 protein (or M2 or BIF) for 1 hour at 4*C. The mixture was then added to 96-well plates containing 3x10 7 human platelets, which had been previously washed and resuspended in Tyrodes/Hepes buffer supplemented with 1 mM EDTA. After a 30 minute incubation at room temperature the cells were washed, fixed, and stained for CD62P and GPIIb (CD41) expression by standard techniques and 20 analysed on a FACS Scan flow cytometer. The results of the activation assay are shown in Figure 25. The percentage of activated platelets (positive for both CD41 and CD62P) after treatment with HAGG alone was defined as 100%. BIF and M2 protein showed similar suppressive activity. The D2 protein, however, is approximately 3-fold more potent. 25 Discussion The D2 protein comprises the extracellular domain of two head to tail FcyRIIa proteins fused to the murine IgG2a Fc domain, which is mutated at four amino acids to reduce Fc receptor binding and complement fixation. The D2 protein WO 2008/070927 PCT/AU2007/001934 47 effectively blocked the immune complex mediated activation of MC/9 mast cells, and immune complex mediated activation of FcyR in both a neutrophil activation assay and a platelet activation assay. Such Fc binding dimer fusion proteins may accordingly be effective inhibitors of immune complex-mediated diseases in vivo. 5 Example 5 Engineering and expression of heterodimeric Fc receptor polypeptides Materials and Methods Construction of FcyRIIa-FcyRIII heterodimeric expression vectors The Fc binding region of FcyRIIa and FcyRIII may be independently PCR 10 amplified from cDNA template using appropriate primers as described in Example 1. The regions amplified would encompass the known characteristic residues and motifs of Fc binding regions such as residues of the ectodomain 1 and ectodomain 2 linker (i.e. the D1/D2 junction), and the BC, CE and FG loops. The polynucleotide sequences for these Fc binding regions are well known to 15 persons skilled in the art. Blunt-ended PCR products can be ligated using T4 DNA ligase into the vector pPIC9 (Invitrogen, Life Technologies) at the EcoRI site filled in with Klenow fragment of DNA polymerase I. An operably fused FcyRIIa-FcyRIII heterodimeric polynucleotide may be created from these amplified products 20 using similar PCR and cloning techniques as those described in Example 1. Insert size and orientation may be confirmed by analytical restriction enzyme digestion or DNA sequencing. The operably fused FcyRIIa-FcyRIII heterodimeric polynucleotide can be cloned into various expression vectors. For example, the FcyRIIa-FcyRIII heterodimeric 25 polynucleotide may be ligated into the EcoRI/XbaI sites of modified pBACPAK9 (Invitrogen Life Tech) in which the BamHI site in the multiple cloning site had first been eliminated by digest with BamHI, filling in using Klenow fragment of WO 2008/070927 PCT/AU2007/001934 48 DNA polymerase and re-ligation. Insert sizes may be defined by EcoRI/XbaI digest and the correct orientation of the multimerising BamHI fragment can be screened by PvuII digest using standard protocols. Alternatively, the FcyRIIa-FcyRIII heterodimeric polynucleotide can be cloned 5 into mammalian expression vectors. For example, the Gateway LR clonase reaction (Invitrogen, Life Technologies) may be used to transfer operationally fused multimeric Fc receptor polynucleotide fragments into Gateway T M reading frame-A cassette (Invitrogen, Life Technologies) adapted expression vector pAPEX3P (Evans et al, 1995, and Christiansen et al., 1996) to give mammalian 10 expression vectors expressing the fused Fc receptor multimers. Likewise, the Gateway LR clonase reaction can be used to transfer the operationally fused multimeric Fc receptor polynucleotide fragments into GatewayTM reading frame A cassette (Invitrogen, Life Technologies) adapted expression vector pIRESneo (Clontech). 15 Discussion Multimerisation of Fc binding regions generates molecules having higher avidity interactions with Fc domains. Each monomer in the multimer is able to separately interact with the Fc domain of immunoglobulins to give higher avidities. Multimers containing Fc binding domains derived from different Fc 20 receptors may be generated. For example, multimers could be formed from combinations of the Fc binding regions of FcyRI, FgyRIIa, FcyRIIb, Fc'yRIIIa, FcyRIIIb, FcaRI and FcERI. Heterodimers could also be formed from combinations of these Fc binding regions. For example, FcyRIIa-FcyRIII, FcyRIIa FcyRI, and FcyRI-FcyRIII heterodimers could be formed, as well as heterodimers 25 consisting of other combinations of Fc binding regions. The Fc binding domains of a number of Fc receptors have been defined by mutagenesis or crystallography (IgG and FcyR: Maxwell et al. 1999, Radaev et al., 2001, Sondermann et al., 2000, Hulett et al., 1988, Hulett et al., 1991, Hulett et al., WO 2008/070927 PCT/AU2007/001934 49 1994, Hulett et al., 1995; IgE and FcERI: Garman et al., 2000; IgA and FcaRI interactions: Wines et al., 2001, Herr et al., 2003). Further, comparisons of similar FcR sequences and comparative analysis of Fc receptor structures have been made (Sondermann et al., 2001). These analyses show that related, clearly defined 5 segments of different Fc receptors are capable of interacting with their ligands. Moreover, crystallographic analysis has demonstrated this clearly for the FcyRIIa and IgG interaction in International patent specification no WO 2006/133486 compared to crystallographic analyses of FcyRIII and IgG (Radaev et al. 2001, Sonderman et al., 2001). 10 It is clear that these data together with mutagenesis experiments of other Fc receptors indicate that segments from the connecting region between ectodomain 1 and ectodomain 2 of these related Fc receptors, as well as segments from the BC, C'E and FG loops of the second domains of different receptors, interact with their respective ligands. Incorporation of such Fc binding regions into other 15 polypeptides could confer specificity for that immunoglobulin type on the new polypeptide. To this end, Hulett et al., 1991, Hulett et al., 1995, and Maxwell et al., 1999 have demonstrated that the addition of IgG binding regions into proteins that were otherwise unable to bind IgG acquired specificity for IgG. Similarly, it has been observed that the insertion of a series of IgE binding sequences into 20 proteins unable to bind IgE resulted in protein chimaeras with IgE specificity as previously described in International patent specification no WO 96/08512. It can therefore be predicted that in a similar manner, the inclusion of Fc binding regions that interact with IgG from FcyRI or FcyRIII into a polypeptide or protein could confer IgG binding function to that polypeptide or protein, or similarly, the 25 inclusion of Fc binding regions that interact with IgA from CD89 or FcaRI into a polypeptide or protein could confer IgA binding function to that polypeptide or protein. Such sequences could include the loops of the first extracellular domain of FcaRI of CD89 that are known to interact with IgA, such loops would include the BC, C'E and FG loops of domain 1. Important residues include amino acids WO 2008/070927 PCT/AU2007/001934 50 35, 52 and 81-86 (Wines et al., 2001, Herr et al., 2003). In this way, receptor polypeptides and proteins containing segments capable of interacting with different classes of immunoglobulins are possible. 5 Throughout this specification the word "comprise", or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated element, integer or step, or group of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps. 10 All publications mentioned in this specification are herein incorporated by reference. Any discussion of documents, acts, materials, devices, articles or the like which has been included in the present specification is solely for the purpose of providing a context for the present invention. It is not to be taken as an admission that any or all of these matters form part of the prior art base or were 15 common general knowledge in the field relevant to the present invention as it existed in Australia or elsewhere before the priority date of each claim of this application. It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific 20 embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive.

WO 2008/070927 PCT/AU2007/001934 51 REFERENCES 1. Amirahmadi SF et al. & Rowley MJ. 2005. Arthritogenic anti-type II collagen antibodies are pathogenic for cartilage-derived chondrocytes independent of inflammatory cells. Arthritis Rheum. 52(6):1897-906. 2. Armour KL et al. & Clark MR. 2003. Differential binding to human FcgammaRlla and FcgammaRlIb receptors by human IgG wildtype and mutant antibodies. Mol Immunol. 40(9):585-93. 3. Armour KL et al. & Clark MR. 2002. The contrasting IgG-binding interactions of human and herpes simplex virus Fc receptors. Biochem Soc Trans. 30(4):495-500. 4. Astier A et al. & D Hanau. 1994. Human epidermal Langerhans cells secrete a soluble receptor for IgG (Fc gamma RII/CD32) that inhibits the binding of immune complexes to Fc gamma R+ cells. j Immunol. 152(1):201. 5. Brooks D et al. & Ravetch J. 1989. Structure and expression of human IgG FcRII (CD32). Functional heterogeneity is encoded by the alternatively spliced products of multiple genes. I Exp Med. 170:1369-1385. 6. Christiansen D et al. & BE Loveland. 1996. Engineering of recombinant soluble CD46: an inhibitor of complement activation. Immunology. 87:348. 7. Emery P et al. & G Seydoux. 2001. Rheumatology (Oxford) 40:699. 8. Evans MJ et al. & SP Squinto. 1995. Rapid expression of an anti-human C5 chimeric Fab utilizing a vector that replicates in COS and 293 cells. j Immunol Methods. 1995 184:123. 9. Garman SC et al., 2000. Structure of the Fc fragment of human IgE bound to its high-affinity receptor Fc epsilonRI alpha. Nature 406(6793):259-66.

WO 2008/070927 PCT/AU2007/001934 52 10. Herr AB et al., 2003. Insights into IgA-mediated immune responses from the crystal structures of human FcalphaRI and its complex with IgAl-Fc. Nature 423(6940):614-20. 11. Hibbs ML et al. & PM Hogarth. 1988. Molecular cloning of a human immunoglobulin G Fc receptor. Proc Natl Acad Sci USA 85 (7), 2240. 12. Hulett MD et al. & PM Hogarth. 1991. Chimeric Fc receptors identify functional domains of the murine high affinity receptor for IgG. J Immunol. 47(6):1863-8. 13. Hulett MD et al. & PM Hogarth. 1994. Identification of the IgG binding site of the human low affinity receptor for IgG Fc gamma RII. Enhancement and ablation of binding by site-directed mutagenesis. J Biol Chem. 269(21):15287. 14. Hulett MD et al. & PM Hogarth. 1995. Multiple regions of human Fc gamma RII (CD32) contribute to the binding of IgG. J Biol Chem. 270(36):21188. 15. Hulett MD & PM Hogarth. 1998. The second and third extracellular domains of FcgammaRI (CD64) confer the unique high affinity binding of IgG2a. Mol Immunol. 35(14-15):989. 16. Ierino FL et al. & PM Hogarth. 1993a. Rec. soluble human FcyRII: prodn, characterization, and inhibition of the Arthus reaction. J Exp Med 178:1617. 17. lerino FL et al. & PM Hogarth. 1993b. Mapping epitopes of human Fc gamma RII (CDw32) with monoclonal antibodies and recombinant receptors. J. Immunol. 150:1794-803. 18. Lau LM et al. & DE Jackson. 2004. The tetraspanin superfamily member CD151 regulates outside-in integrin alphallbbeta 3 signaling and platelet function. Blood. 104(8):2368.

WO 2008/070927 PCT/AU2007/001934 53 19. Maxwell KF et al. & PM Hogarth. 1999. Crystal structure of the human leukocyte Fc receptor, Fc gammaRIla. Nat Struct Biol 6:437. 20. Nabbe KC et al. & WB van den Berg. 2003. Coordinate expression of activating Fc gamma receptors I and III and inhibiting Fc gamma receptor type II in the determination of joint inflammation and cartilage destruction during immune complex-mediated arthritis. Arthritis Rheum 48:255. 21. Nandakumar KS et al. & Holmdahl R. 2004. Collagen type II (CII)-specific antibodies induce arthritis in the absence of T or B cells but the arthritis progression is enhanced by CII-reactive T cells. Arthritis Res Ther. 6(6):R544-50. 22. Pflum LR & ML Graeme. 1979. The Arthus reaction in rats, a possible test for anti-inflammatory and anti-rheumatic drugs. Agents Actions 9:184. 23. Powell MS et al. & PM Hogarth. 1999. Biochemical analysis and crystallisation of Fc gamma RIIa, the low affinity receptor for IgG. Immunol Lett. 68(1):17. 24. Radaev S et al., 2001. The structure of a human type III Fcgamma receptor in complex with Fc. J. Biol. Chem. 276(19):16469-77. 25. Sondermann P et al., 2000. The 3.2-A crystal structure of the human IgG1 Fc fragment-Fc gammaRIII complex. Nature 406(6793):267-73. 26. Sondermann P et al., 2001. Molecular basis for immune complex recognition: a comparison of Fc-receptor structures. J. Mol. Biol. 309(3):737-49. 27. Seki T. 1989. Identification of multiple isoforms of the low-affinity human IgG Fc receptor. Immunogenetics. 30:5-12.

WO 2008/070927 PCT/AU2007/001934 54 28. Stuart S et al. & Vaux D. 1987. Isolation and expression of cDNA clones encoding a human receptor for IgG (Fc-RII). J Exp Med. 166:1668-1684. 29. Takai T. 2002. Roles of Fc receptors in autoimmunity. Nat Rev Immunol. 2(8):580. 30. Tamm A et al. & RE Schmidt RE. 1996. The IgG binding site of human FcgammaRIIIB receptor involves CC' and FG loops of the membrane proximal domain. J Biol Chem. 271(7):3659. 31. Thai LeM et al. & DE Jackson. 2003. Physical proximity and functional interplay of PECAM-1 with the Fc receptor Fc gamma RIla on the platelet plasma membrane. Blood. 102(10):3637. 32. Wines BD & SB Easterbrook-Smith. 1988. Enhancement of the binding of Clq to immune complexes by polyethylene glycol. Mol Immunol. 25(3):263. 33. Wines BD et al. & PM Hogarth. 2000. The IgG Fc contains distinct Fc receptor (FcR) binding sites: the leukocyte receptors Fc gamma RI and Fc gamma RIIa bind to a region in the Fc distinct from that recognized by neonatal FcR and protein A. J. Immunol. 164:5313-8 34. Wines BD et al. & PM Hogarth. 2001. The interaction of Fc alpha RI with IgA and its implications for ligand binding by immunoreceptors of the leukocyte receptor cluster. j Immunol. 166(3):1781. 35. Wines BD et al. & PM Hogarth. 2003. Soluble FcgammaRlIa inhibits rheumatoid factor binding to immune complexes. Immunology. 109(2):246. 36. Wright JK et al. & JC Jaton. 1980. Preparation and characterization of chemically defined oligomers of rabbit immunoglobulin G molecules for the complement binding studies. Biochem J. 187(3):767.

Claims (23)

1. A soluble multimeric polypeptide able to inhibit interaction of leukocyte Fcy receptors (FcyR) and immunoglobulin G (IgG), said polypeptide comprising two or more Fc binding regions linked in a head to tail arrangement, at least one of which is derived from an FcyR type receptor, and a modified Fc domain that has substantially no ability to bind said Fc binding regions and permits dimerisation of the said polypeptide.

2. The polypeptide of claim 1, wherein said polypeptide comprises just two linked Fc binding regions, at least one of which is derived from an FcyR type receptor.

3. The polypeptide of claim 1 or 2, wherein said at least one Fc binding region derived from an FcyR type receptor is derived from an FcyRII type receptor.

4. The polypeptide of claim 3, wherein said at least one Fc binding region is derived from FcyRIIa.

5. The polypeptide of any one of claims 1 to 4, wherein each of said linked Fc binding regions is derived from an FcyR type receptor.

6. The polypeptide of claim 5, wherein each of said linked Fc binding regions is derived from the same FcyRII type receptor.

7. The polypeptide of any one of claims 1 to 6, wherein said Fc binding regions are linked through a linker comprising 1 to 20 amino acids.

8. The polypeptide of any one of claims 1 to 7, wherein said modified Fc domain shows altered effector function. WO 2008/070927 PCT/AU2007/001934 56

9. The polypeptide of any one of claims 1 to 8, wherein said Fc domain is derived from IgGi and has been modified by substitution of Leu 234 and/or Leu 2 3 5 .

10. The polypeptide of any one of claims 1 to 8, wherein said Fc domain is derived from IgG2a and has been modified by substitution of any one or more of Leu23 5 , Glu 318 , Lys 32 0 and Lys 3 2 2 .

11. The polypeptide of any one of claims 1 to 8, wherein said Fc domain is derived from IgG4 and has been modified by amino acid modification at any one or more of the amino acids at positions 228, 233, 234, 235 and 236.

12. The polypeptide of any one of claims 1 to 11, further comprising a carrier protein.

13. The polypeptide of claim 12, wherein said carrier protein is human serum albumin (HSA).

14. A soluble multimeric protein comprising a dimer of a polypeptide according to any one of claims 1 to 13.

15. The protein of claim 14 in the form of an Fc fusion dimer protein, wherein the Fc fusion dimer protein comprises a first polypeptide chain and a second polypeptide chain, each polypeptide chain comprising (i) two Fc binding regions derived from FcyRIIa linked in a head to tail arrangement and (ii) a modified Fc domain that has substantially no FcyRIIa binding ability and permits dimerisation of the first and second polypeptide chains.

16. A polynucleotide molecule comprising a nucleotide sequence encoding the polypeptide of any one of claims 1 to 13 or the protein of claim 14 or 15.

17. The polynucleotide molecule of claim 16, wherein said polynucleotide molecule consists in an expression cassette or expression vector. WO 2008/070927 PCT/AU2007/001934 57

18. A recombinant host cell comprising the polynucleotide molecule of claim 16 or 17.

19. A method of producing a polypeptide or protein, said method comprising the following steps: (i) providing a recombinant host cell comprising said polynucleotide molecule of claim 16 or 17, (ii) culturing said host cell in a suitable culture medium and under conditions suitable for expression of said polypeptide or protein, and (iii) isolating said polypeptide or protein from the culture, and optionally, from the culture medium.

20. A method of treating a subject for an inflammatory disease, said method comprising administering to said subject the polypeptide of any one of claims 1 to 12 or the protein of claim 14 or 15, optionally in combination with a pharmaceutically- or veterinary-acceptable carrier or excipient.

21. The method of claim 20, wherein said inflammatory disease is an immune complex (IC)-mediated inflammatory disease.

22. The method of claim 21, wherein said IC-mediated inflammatory disease is selected from the group consisting of rheumatoid arthritis (RA), immune thrombocytopenic purpura (ITP), systemic lupus erythematosus (SLE), glomerulonephritis and heparin-induced thrombocytopenia thrombosis syndrome (HITTS).

23. A method of removing circulating immune complexes (IC) from a subject suffering an immune-complex-mediated inflammatory disease, said method comprising the following steps: WO 2008/070927 PCT/AU2007/001934 58 (i) providing a polypeptide according to any one of claims 1 to 13 or a protein according to claim 14 or 15 bound to a suitable substrate, (ii) treating blood removed from said subject by contacting the blood ex vivo with said substrate-bound polypeptide or protein such that IC present in said blood is bound to the substrate via said polypeptide or protein, (iii) separating the treated blood from the substrate, and (iv) thereafter returning the treated blood to the subject.