WO2016166139A1

WO2016166139A1 - Bispecific fusion proteins for enhancing immune responses of lymphocytes against tumor cells

Info

Publication number: WO2016166139A1
Application number: PCT/EP2016/058085
Authority: WO
Inventors: Helmut Salih; Ludger Grosse-Hovest; Gundram Jung; Samuel KOERNER
Original assignee: Eberhard Karls Universität Tübingen
Priority date: 2015-04-14
Filing date: 2016-04-13
Publication date: 2016-10-20

Abstract

The present invention relates to recombinant fusion proteins. Said fusion protein comprises a binding protein with a binding site that specifically binds to a receptor specific for T cells or natural killer cells, an extracellular fragment of a transmembrane protein that is an immune receptor and that binds to a target cell or to a ligand or antigen expressed by a target cell, and a linking polypeptide connecting the binding protein and the extracellular fragment of a transmembrane protein, wherein said polypeptide comprises at least a portion of a Fc domain. The invention further relates to uses of said fusion proteins, such as medical uses. Furthermore, the invention relates to nucleic acids encoding for said fusion proteins, host cells comprising said nucleic acids, and methods of production of said fusion proteins.

Description

BISPECIFIC FUSION PROTEINS FOR ENHANCING IMMUNE RESPONSES OF LYMPHOCYTES AGAINST TUMOR CELLS

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] The present application claims the benefit of priority to European patent application 15 163 460.7 "Bispecific Fusion Proteins For Enhancing Immune Responses Of Lymphocytes Against Tumor Cells" filed with the European Patent Office on 14 April 2015, the contents of which is hereby incorporated by reference in its entirety for all purposes.

FIELD OF THE INVENTION

[0002] The present invention is in the field of fusion proteins and relates to fusion proteins comprising molecule binding protein with at least one binding site that specifically binds to a T-cell or natural killer cell, an extracellular fragment of a transmembrane protein that is an immune receptor that binds to a target cell or to a ligand or antigen expressed by a target cell, and a linking polypeptide comprising at least a portion of an Fc domain (for example, a CH2 domain or a CH3 domain) connecting the binding protein and the extracellular fragment of a transmembrane protein. The present application also refers to pharmaceutical compositions and medical uses of these fusion proteins, for example, in the treatment of cancer.

BACKGROUND OF THE INVENTION

[0003] Monoclonal antibodies (mAbs) have by now become an established tool in therapy for many tumors. Some of these monoclonal antibodies, such as Rituxan^® or Herceptin^®, bind to tumor-specific antigens and interact via their Fc-part with Fc-receptor (FcR) positive cells, such as natural killer cells (NK-cells), by which these cells are activated and antibody dependent cellular cytotoxicity (ADCC) against the target tumor cell is mediated (Adams and Weiner, 2005; Bianchini and Gianni, 2014). A different strategy is pursued by bispecific T-cell engagers (BiTEs) (Staerz and Bevan, 1986; Fanger et al, 1991; Kellner et al. 2011), such as the CD19xCD3 bispecific antibody Blinatumomab. This antibody binds to CD 19 of a target B-cell via its first antigen binding site and to CD3 of a cytotoxic T-cell via its second antigen binding site (Bargou et al, 2008; Topp et al, 2011). As a result, these two cell types are linked and the cytotoxic T-cell is activated, by which a cytotoxic activity of the T-cell on the target B-cell is exerted. When employing such bispecific CD3 specific antibodies, it is crucial that these constructs do not bind to FcR. Otherwise, a T-cell activation independent of the binding of the second antigen can be induced anywhere where FcR carrying cells are present, for instance in the entire hematopoietic, lymphatic and reticuloendothelial system, which would lead to a systemic T-cell activation (Tibben et al, 1996).

[0004] Despite of the above-mentioned examples, no immune-stimulatory antibodies exist for various tumor entities, such as acute myeloid leukemia (AML) or many solid tumors. In want of suitable target antigens with tumor-associated expression, ligands of immune receptors NKG2D, RANK and GITR have been identified as potential targets.

[0005] NKG2D is an activating receptor found on NK cells and CD8 T cells (both αβ and γδ) which is encoded by the KLRKl gene. Of its ligands (NKG2DL), 8 have been identified by now, which belong to the group of MHC class I chain-related proteins (MICA and MICB) or HCMV UL16-binding proteins (ULBP1, ULBP2, ULBP3, ULBP4, ULBP and ULBP6). The expression of NKG2L is most widely tumor-restricted, but the expression profile of the individual ligands can vary strongly between different tumor entities. NKG2DL can also be set free in soluble form by tumor cells, whereby the released NKG2DL can systemically inhibit NKG2D-mediated anti-tumor immune-response (Salih et al, 2002; Groh et al, 2002; Spear et al, 2013; Raulet et al, 2013).

[0006] Receptor Activator of NF-κΒ (RANK) is a member of the tumor necrosis factor receptor sub-family. It is the receptor for RANK-Ligand (RANKL) and part of the RANK/RANKL/OPG signaling pathway that regulates osteoclast differentiation and activation. Furthermore, the RANK/RANKL molecule system has immunomodulatory effects. RANKL is expressed by tumor cells in, inter alia, chronic lymphoid leukemia (CLL), multiple myeloma (MM) and acute myeloid leukemia (AML), whereas RANK can be expressed on NK cells. The interaction of RANKL expressed on malignant hematopoietic cells with RANK on NK cells was shown to inhibit anti-tumor immune-responses of NK cells (Leibbrandt and Penninger, 2008; Schmiedel et al. 2011; Schmiedel et al., J Immunol. 2013).

[0007] GITR (glucocorticoid-induced TNF receptor family-related protein) is another receptor expressed among others on NK cells. Its ligand GITRL is expressed and released, among others, by malignant cells in leukemia and solid tumors and has also been shown to impair NK cell reactivity against GITRL-expressing cells (Baltz et al., 2007, 2008; Baessler et al, 2009; Buechele et al, 2012).

[0008] In order to target these tumor-expressed ligands, fusion proteins have been disclosed, in which the extracellular portions of one of the immune receptors NKG2D, RANK or GITR is fused to the Fc-part of human IgG. The human IgG was even modified for enhanced ADCC by introducing S239D/I332E amino acid substitution. These fusion proteins bind to NKG2DL, RANKL or GITRL on target tumor cells with their immune receptor portion and mediate NK ADCC via their enhanced Fc-portion (Schmiedel et al., Cancer Res.. 2013; Schmiedel et al., Mol Ther. 2013, Raab et al, J. Immunol. 2014, Steinbacher et al, Int. J. Cancer 2014). In addition, the International Patent Application WO 2011/085178 as well as Zhang & Sentman Cancer Res 201 \) discloses a fusion protein of a CD3 specific scFv- fragment and NKG2DL linked via polyglycine. No Fc portion was present in this construct, as the Fc region was supposed to interfere with the mode of action of this fusion protein. Moreover, WO 2007/048849 discloses a fusion protein comprising a full length antiCD3 antibody and an ectodomain of NKG2D. Specific mutations that alter the properties of the antibody or the ectodomain of NKG2D are not disclosed.

[0009] The objective of the present invention is to provide a class of fusion protein molecules that can mediate a target-restricted immune response of immune cells against target tumor cells that overcome at least some of the above-discussed difficulties and that can generally be used in therapy, amongst others.

SUMMARY OF THE INVENTION

[0010] The invention provides a recombinant fusion protein. This recombinant fusion protein comprises binding protein with a binding site that specifically binds to a receptor specific for immune cells such as T cells or natural killer cells. The recombinant fusion protein further comprises an extracellular fragment of a transmembrane protein that is an immune receptor and that has the ability to bind to a target cell or to a ligand or antigen expressed by a target cell. Said antigen ligand or antigen expressed by a target cell may be bound to the target cell or may be soluble, i.e. set free by the target cell and thus may be not bound to the target cell. Still further, the recombinant fusion protein comprises a linking polypeptide connecting the binding protein and the extracellular fragment of a transmembrane protein, wherein said polypeptide comprises at least a portion of an Fc domain. In a particular embodiment, the linking polypeptide (that comprises the at least one portion of an Fc domain) consists of at least a portion of a CH2 domain and optionally a hinge region. In addition, the recombinant fusion protein does not comprise a immunoglobulin heavy chain comprising VH- H-CH 1 -CH2-CH3 or VH-H-CH1-CH2-CH3-CH4 (wherein the terms "VH" "H", CHI", "CH2", "CH3" and "CH4" are used in their regular meaning to denote the variable domain, the hinge region, the first constant domain, the second constant domain, the third constant domain and the fourth constant domain of an immunoglobulin heavy chain). Furthermore, the binding protein moiety of the fusion protein does not bind to the extracellular fragment of a transmembrane protein comprised in said fusion protein.

[0011] In specific embodiments, the linking polypeptide consists of a CH2 domain and optionally a hinge region. In preferred embodiments, the linking polypeptide consists of a CH2 domain and a hinge region. In preferred embodiments, the linking polypeptide comprising or consisting of the CH2 domain and the hinge region comprises a mutation or deletion in at least one amino acid residue that is able to mediate binding to Fc receptors. Said amino acid residue is preferably selected from the group consisting of sequence position 233, 234, 235, 236, 265, 297, 327, and 330 (numbering of sequence positions according to the EU-index) and the mutation or deletion is preferably selected from the group consisting of Glu233Pro, Leu234Val, Leu235Ala, deletion of Gly236, Asp265Gly, Asn297Gln, Ala327Gln, and Ala330Ser. In preferred embodiments, the linking polypeptide comprising or consisting of the CH2 domain and optionally the hinge region comprises a mutation or deletion in at least one amino acid residue that is able to mediate dimerization of immunoglobulins. Such an amino acid residue is preferably selected from the group consisting of sequence positions 220, 226, and 229 of the hinge region (numbering of residues according to EU-numbering) and the mutation or deletion is preferably selected from the group consisting of Cys220Ser, Cys226Ser, or Cys229Ser.

[0012] In preferred embodiments, the linking polypeptide comprises at least 50-125 consecutive amino acids corresponding to positions 216-340 (EU-index) of human IgG as set forth in SEQ ID NO: 01 , wherein at least 50-125 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71 , 72, 73, 74, 75, 76, 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 , 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 1 19, 120, 121, 122, 123, 124, or 125. In preferred embodiments, the linking polypeptide comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50-125 consecutive amino acids of SEQ ID NO: 01, wherein at least 50-125 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61 , 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, or 125.

[0013] In preferred embodiments, the linking polypeptide comprises at least 50-125 consecutive amino acids of SEQ ID NO: 02 or of SEQ ID NO: 03, wherein at least 50-125 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71 , 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, or 125. In preferred embodiments, the linking polypeptide comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50-125 consecutive amino acids of SEQ ID NO: 02 or of SEQ ID NO: 03, wherein at least 50-125 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61 , 62, 63, 64, 65, 66, 67, 68, 69, 70, 71 , 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, or 125.

[0014] In specific embodiments, the binding protein is an antibody molecule. In preferred embodiments, the antibody molecule has a single antigen binding site. In specific embodiments, the antibody molecule in the fusion protein of the invention is a Fab fragment or a scFv fragment or a single domain antibody molecule. In preferred embodiments, said binding protein binds to a receptor specific for immune cells such as T cells or natural killer cells, wherein the receptor is preferably CD3, CD16, CD28, CD137/4-1BB, OX40, Nkp44, Nkp30, Nkp40 or Nkp46, preferably CD3 or CD 16 or preferably CD3 on T cells or CD 16 on natural killer (NK) cells or preferably CD3 on cytotoxic T cells.

[0015] In preferred embodiments, the binding protein is a scFv fragment as set forth in SEQ ID NO: 04 or fragments thereof or polypeptides having a sequence identity of at least 80 %, at least 85 %, at least 90 %, at least 95 %, at least 98 %, at least 99 % or 100 % when aligned with the amino acid sequence as set forth in SEQ ID NO: 04 or fragments thereof and which still has single antigen binding site that specifically binds to CD3. In preferred embodiments, the antibody fragment is a scFv fragment as set forth in SEQ ID NO: 05 or fragments thereof or polypeptides having a sequence identity of at least 80 %, at least 85 %, at least 90 %, at least 95 %, at least 98 %, at least 99 % or 100 % when aligned with the amino acid sequence as set forth in SEQ ID NO: 05 or fragments thereof and which still has single antigen binding site that specifically binds to CD 16.

[0016] In preferred embodiments, the antibody fragment is a Fab fragment, with a heavy chain sequence as set forth in SEQ ID NO: 06 and a light chain sequence as set forth in SEQ ID NO: 07 or fragments thereof or heavy or light chain sequences having a sequence identity of at least 80 %, at least 85 %, at least 90 %, at least 95 %, at least 98 %, at least 99 % or 100 % with the amino acid sequence as set forth in SEQ ID NO: 06 or SEQ ID NO: 07 , or fragments thereof which still have an antigen binding site that specifically binds to CD3. In preferred embodiments, the antibody fragment is a Fab fragment, with a heavy chain sequence as set forth in SEQ ID NO: 08 and a light chain sequence as set forth in SEQ ID NO: 09 or fragments thereof or heavy or light chain sequences having a sequence identity of at least 80 %, at least 85 %, at least 90 %, at least 95 %, at least 98 %, at least 99 % or 100 % with the amino acid sequence as set forth in SEQ ID NO: 08 or SEQ ID NO: 09, or fragments thereof which still have an antigen binding site that specifically binds to CD 16. In these embodiments, the heavy chain of the Fab fragment is part of the fusion protein whereas the light chain of the Fab fragment is linked to the heavy chain of the Fab fragment or the light chain of the Fab fragment is part of the fusion protein whereas the heavy chain of the Fab fragment is linked to the light chain of the Fab fragment.

[0017] In specific embodiments, the extracellular fragment of the immune receptor of the fusion protein of the invention binds to a ligand or antigen on a target cell or to a ligand or antigen that is soluble. This ligand or antigen is preferably a tumor-associated ligand or antigen. The ligand or antigen on the target cell may preferably be on the surface of said target cell and it is preferably an antigen or ligand expressed on a tumor cell, more preferably wherein the expression is associated or restricted to a tumor cell. In specific embodiments, the antigen or ligand is NKG2DL or RANKL or GITRL.

[0018] In specific embodiments of the invention, the immune receptor fragment of the fusion protein of the invention is an extracellular fragment of an immune receptor selected from the group of NKG2D (UniProtKB accession number P26718, SEQ ID NO: 10), RANK (UniProtKB accession number Q9Y6Q6, SEQ ID NO: 11), GITR (UniProtKB accession number Q9Y5U5, SEQ ID NO: 12), CD94-NKG2 (UniProtKB accession number Q13241, SEQ ID NO: 13), CTLA-4 (UniProtKB accession number PI 6410, SEQ ID NO: 15), PD-1 (UniProtKB accession number Q15116, SEQ ID NO: 16), BTLA (UniProtKB accession number Q7Z6A9, SEQ ID NO: 17), LAG-3 (UniProtKB accession number PI 8627, SEQ ID NO: 18), TIM-3 (UniProtKB accession number Q8TDQ0, SEQ ID NO: 19), LAIR-1 (UniProtKB accession number Q6GTX8, SEQ ID NO: 20), TIGIT (UniProtKB accession number Q495A1, SEQ ID NO: 21), Siglecl (UniProtKB accession number Q9BZZ2, SEQ ID NO: 22), Siglec2 (UniProtKB accession number P20273, SEQ ID NO: 23), Siglec3 (UniProtKB accession number P20138, SEQ ID NO: 24), Siglec4 (UniProtKB accession number P20916, SEQ ID NO: 25), Siglec5 (UniProtKB accession number 015389, SEQ ID NO: 26), Siglec6 (UniProtKB accession number 043699, SEQ ID NO: 27), Siglec7 (UniProtKB accession number Q9Y286, SEQ ID NO: 28), Siglec8 (UniProtKB accession number Q9NYZ4, SEQ ID NO: 29), Siglec9 (UniProtKB accession number Q9Y336, SEQ ID NO: 30), SigleclO (UniProtKB accession number Q96LC7, SEQ ID NO: 31), Siglecl 1 (UniProtKB accession number Q96RL6, SEQ ID NO: 32), Siglecl2 (UniProtKB accession number Q96PQ1, SEQ ID NO: 33), Siglecl4 (UniProtKB accession number Q08ET2, SEQ ID NO: 34), Siglecl5 (UniProtKB accession number Q6ZMC9, SEQ ID NO: 35), Siglecl6 (UniProtKB accession number A6NMB1, SEQ ID NO: 36) , NKp30 (UniProtKB accession number 014931, SEQ ID NO: 14), NKp40 (UniProtKB accession number Q9NZS2, SEQ ID NO: 85), NKp44 (UniProtKB accession number 095944, SEQ ID NO: 86), NKp46 (UniProtKB accession number 076036, SEQ ID NO: 87), NKp80 (UniProtKB accession number Q9NZS2, SEQ ID NO: 88), OPG (UniProtKB accession number 000300, SEQ ID NO: 89).

[0019] In specific embodiments, the extracellular fragment is of NKG2D and has the ability to bind to at least a portion of at least one NKG2DL, preferably alternatively to more than one NKG2DL, most preferably to any NKG2DL, wherein the group of NKG2DL comprises MICA (UniProtKB accession number Q29983, SEQ ID NO: 37), MICB (UniProtKB accession number Q29980, SEQ ID NO: 38), ULBP1 (UniProtKB accession number Q9BZM6, SEQ ID NO: 39), ULBP2 (UniProtKB accession number Q9BZM5, SEQ ID NO: 40), ULBP3 (UniProtKB accession number Q9BZM4, SEQ ID NO: 41), ULBP4 (UniProtKB accession number Q8TD07, SEQ ID NO: 42), ULBP5 (UniProtKB accession number Q6H3X3, SEQ ID NO: 43) and ULBP6 (UniProtKB accession number Q5VY80, SEQ ID NO: 44). As the person skilled in the art knows, NKG2DL are highly polymorphic in humans, thus specific amino acid sequences as given above for NKG2DL are exemplary but not limiting.

[0020] In specific embodiments, the extracellular fragment of an immune receptor is an extracellular fragment of NKG2D (UniProtKB accession number P26718, SEQ ID NO: 10), wherein the extracellular fragment of NKG2D comprises at least 50-139 consecutive amino acids corresponding to SEQ ID NO: 45, wherein in at least 50-139 means at least 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, 60, 61 , 62, 63, 64, 65, 66, 67, 68, 69, 70, 71 , 72, 73, 74, 75, 76, 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, or 139 positions, or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50-139 consecutive amino acids of SEQ ID NO: 45, wherein in at least 50-139 means at least 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, 60, 61 , 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, or 139 positions.

[0021] In specific embodiments, the extracellular fragment of an immune receptor is an extracellular fragment of RANK and has the ability to bind to at least a portion of RANKL (UniProtKB accession number 014788, SEQ ID NO: 90).

[0022] In specific embodiments, the extracellular fragment is of RANK and comprises at least 90-183 consecutive amino acids corresponding to SEQ ID NO: 46 (= RANK position 25-207), wherein at least 90-183 means 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 11 1, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181 , 182, or 183, or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 90-183 consecutive amino acids of SEQ ID NO: 46, wherein at least 90-183 means 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, or 183.

[0023] In preferred embodiments, the extracellular fragment is of GITR and has the ability to bind to at least a portion of GITRL (UniProtKB accession number Q9UNG2, SEQ ID NO: 91). [0024] In specific embodiments, the extracellular fragment is of GITR comprises at least 50-137 consecutive amino acids corresponding to SEQ ID NO: 47, wherein at least 50- 137 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71 , 72, 73, 74, 75, 76, 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 , 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 1 10, 111 , 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136 or 137, or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50-137 consecutive amino acids of SEQ ID NO: 47, wherein at least 50-137 means 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 , 92, 93, 94, 95, 96, 97, 98, 99, 100, 101 , 102, 103, 104, 105, 106, 107, 108, 109, 110, 11 1, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136 or 137.

[0025] In specific embodiments of the fusion protein of the present invention, the binding protein is located N-terminal of the linker polypeptide and the extracellular fragment of a transmembrane protein is located C-terminal to the linker polypeptide, wherein the extracellular fragment is preferably of a type II transmembrane protein. In specific embodiments of the fusion protein of the present invention the binding protein is located C- terminal to of the linker polypeptide and the extracellular fragment of a transmembrane protein is located N-terminal to the linker polypeptide, wherein the extracellular fragment is preferably of a type I transmembrane protein.

[0026] In specific embodiments of the fusion protein of the present invention, the extracellular fragment of the transmembrane protein is an extracellular fragment of NKG2D and has the ability to bind to at least one NKG2DL, preferably alternatively to more than one NKG2DL, most preferably to any NKG2DL wherein the extracellular fragment of NKG2D is located C-terminal of the linking polypeptide and the binding protein is N-terminal of the linking polypeptide. In specific embodiments, said binding protein binds to CD3 or CD 16, preferably to CD3 on T cells or to CD 16 on natural killer (NK) cells, wherein the binding protein is preferably an antibody molecule, preferably a Fab fragment or a scFv fragment or a single domain antibody, more preferably a Fab fragment. In preferred embodiments, the fragment of NKG2D comprises a sequence as set forth in SEQ ID NO: 45, the linking polypeptide comprises a sequence as set forth in SEQ ID NO: 03, and the antibody fragment is a Fab fragment that binds to CD3 comprising a heavy chain sequence as set forth in SEQ ID NO: 06 and a light chain sequence as set forth in SEQ ID NO: 07 or a Fab fragment that binds to CD 16 comprising a heavy chain sequence as set forth in SEQ ID NO: 08 and a light chain sequence as set forth in SEQ ID NO: 09. In specific embodiments, the extracellular fragment of NKG2D and the linking polypeptide are fused to the heavy chain of a CD3 binding Fab fragment, wherein the fusion protein comprises a sequence as set forth in SEQ ID NO: 48 and wherein the light chain of the Fab fragment comprises a sequence as set forth in SEQ ID NO: 07 and wherein the light chain is linked to the heavy chain in a way that a Fab fragment is formed. In specific embodiments, the extracellular fragment of NKG2D and the linking polypeptide is fused to the heavy chain of a CD 16 binding Fab fragment, wherein the fusion protein comprises a sequence as set forth in SEQ ID NO: 49 and wherein the light chain of the Fab fragment comprises a sequence as set forth in SEQ ID NO: 09 and wherein the light chain is linked to the heavy chain in a way that a Fab fragment is formed. In preferred embodiments, the fusion protein comprises an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 48, and is linked to a Fab fragment light chain comprising an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 07. In preferred embodiments, the fusion protein comprises an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 49, and is linked to a Fab fragment light chain comprising an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 09.

[0027] In specific embodiments of the fusion protein of the present invention, the extracellular fragment of the transmembrane protein is an extracellular fragment of RANK and has the ability to bind RANKL wherein the extracellular fragment of RANK is located N- terminal of the linking polypeptide and the binding protein is C-terminal of the linking polypeptide. In specific embodiments, said binding protein binds to CD3 or CD 16, preferably to CD3 on T cells or to CD 16 on natural killer (NK) cells, wherein said binding protein is preferably an antibody molecule, preferably a scFv fragment or a single domain antibody molecule, more preferably a scFv fragment. In preferred embodiments, the fragment of RANK comprises a sequence as set forth in SEQ ID NO: 46, the linking polypeptide comprises a sequence as set forth in SEQ ID NO: 02, and the binding protein is a scFv fragment that binds to CD3 comprising a sequence as set forth in SEQ ID NO: 04 or a scFv fragment that binds to CD 16 comprising a sequence as set forth in SEQ ID NO: 05. In specific embodiments, the extracellular fragment of RANK and the linking polypeptide are fused to the CD3 binding scFv fragment, wherein the fusion protein comprises a sequence as set forth in SEQ ID NO: 50. In specific embodiments, the extracellular fragment of RANK and the linking polypeptide are fused to the CD 16 binding scFv fragment, wherein the fusion protein comprises a sequence as set forth in SEQ ID NO: 51. In preferred embodiments, the fusion protein comprises an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 50 or SEQ ID NO: 51.

[0028] In specific embodiments of the fusion protein of the present invention, the extracellular fragment of the transmembrane protein is an extracellular fragment of GITR and has the ability to bind GITRL wherein the extracellular fragment of GITR is located N- terminal of the linking polypeptide and the binding protein is C-terminal of the linking polypeptide. In specific embodiments, said binding protein binds to CD3 or CD 16, preferably to CD3 on T cells or to CD 16 on natural killer (NK) cells, wherein said binding protein is preferably an antibody molecule, preferably a scFv fragment or a single domain antibody molecule, more preferably a scFv fragment. In preferred embodiments, the fragment of GITR comprises a sequence as set forth in SEQ ID NO: 47, the linking polypeptide comprises a sequence as set forth in SEQ ID NO: 02, and the binding protein is a scFv fragment that binds to CD3 comprising a sequence as set forth in SEQ ID NO: 04 or a scFv fragment that binds to CD 16 comprising a sequence as set forth in SEQ ID NO: 05. In specific embodiments, the extracellular fragment of GITR and the linking polypeptide are fused to the CD3 binding scFv fragment, wherein the fusion protein comprises a sequence as set forth in SEQ ID NO: 52. In yet another specific embodiment, the extracellular fragment of GITR and the linking polypeptide are fused to the CD 16 binding scFv fragment, wherein the fusion protein comprises a sequence as set forth in SEQ ID NO: 53. In yet another specific embodiment, the fusion protein comprises an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 52 or SEQ ID NO: 53.

[0029] In specific embodiments, the fusion protein of invention binds to a ligand that is soluble and wherein binding of the fusion protein to the soluble ligand or antigen prevents binding of the ligand or antigen to other immune receptors, in particular to receptors from which the immune receptor moiety of the fusion protein is derived from. For instance, in case the immune receptor moiety of the fusion protein comprises an extracellular fragment of NKG2D, binding of the fusion protein to an NKG2DL preferably prevents binding of said NKG2DL to other NKG2D. The same applies mutatis mutandis for fusion proteins in which the immune receptor moiety is derived from other immune receptors, such as, for example, RANK or GITR. In specific embodiments, the binding of the fusion protein to the soluble ligand or antigen neutralizes a physiological or pathophysiological effect of the ligand, preferably an immunomodulatory effect, such as an immune inhibitory effect or an immune activating effect. Preferred binding partners of the fusion protein are soluble NKG2DL or RANKL or GITRL and wherein binding of the fusion protein to NKG2DL or RANKL or GITRL preferably prevents binding of NKG2DL to other NKG2D, or binding of RANKL to other RANK or binding of GITRL to other GITR. In preferred embodiments, the binding of the fusion protein to NKG2DL or RANKL or GITRL neutralizes the immunomodulatory effect of NKG2DL or RANKL or GITRL, wherein the immunomodulatory effect may be either an immune inhibitory effect or immune activating effect. In specific embodiments, the immune receptor moiety of the fusion protein comprises an extracellular fragment of RANK and binds to RANKL, wherein binding preferably neutralizes an effect of RANKL in bone resorption.

[0030] In specific embodiments, the invention provides a pharmaceutical composition comprising a fusion protein of the invention. In specific embodiments, the fusion protein of the invention is used in the treatment of a disease, preferably a proliferative disease, wherein the proliferative disease is preferably cancer, wherein the cancer is preferably selected from the group consisting of adrenal cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, brain and spinal cord tumors, breast cancer, Castleman disease, cervical cancer, colon cancer, endometrial cancer, esophagus cancer, Ewing family of tumors, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumors, gastrointestinal stromal tumor (GIST), gestational trophoblastic disease, Hodgkin disease, Kaposi sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL), chronic myeloid leukemia (CML), chronic myelomonocytic leukemia (CMML), liver cancer, lung cancer, non-small cell lung cancer, small cell lung cancer, lung carcinoid tumor, lymphoma, lymphoma of the skin, malignant mesothelioma, multiple myeloma, myelodysplasia syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroblastoma, non-Hodgkin lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumors, prostate cancer, rectum cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, skin cancer, basal and squamous cell cancer, melanoma, merkel cell cancer, small intestine cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, uterine sarcoma, vaginal cancer, vulvar cancer, Waldenstrom macroglobulinemia, or Wilms tumor. In specific embodiments, the disease is osteoporosis and the fusion protein comprises an extracellular fragment of RANK.

[0031] In specific embodiments, the invention provides a nucleic acid encoding for a fusion protein of the invention. In specific embodiments, said nucleic acid is comprised in a vector. In specific embodiments, the invention provides a host cell comprising said nucleic acid molecule or said vector.

[0032] In specific embodiments, the invention provides a method of producing the fusion protein of the invention, comprising using the nucleic acid encoding the fusion protein for expression of the fusion protein under conditions allowing expression of the fusion protein. In preferred embodiments, the fusion protein is expressed by a host cell or in a cell-free system.

[0033] In specific embodiments, the invention provides a method of treating a disease comprising administering a therapeutically effective amount of the fusion protein of the invention to a subject, wherein the disease is preferably a proliferative disease, and wherein the proliferative disease is preferably cancer, wherein the cancer is preferably selected from the group consisting of adrenal cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, brain and spinal cord tumors, breast cancer, Castleman disease, cervical cancer, colon cancer, endometrial cancer, esophagus cancer, Ewing family of tumors, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumors, gastrointestinal stromal tumor (GIST), gestational trophoblastic disease, Hodgkin disease, Kaposi sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL), chronic myeloid leukemia (CML), chronic myelomonocytic leukemia (CMML), liver cancer, lung cancer, non-small cell lung cancer, small cell lung cancer, lung carcinoid tumor, lymphoma, lymphoma of the skin, malignant mesothelioma, multiple myeloma, myelodysplasia syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroblastoma, non-Hodgkin lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumors, prostate cancer, rectum cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, skin cancer, basal and squamous cell cancer, melanoma, merkel cell cancer, small intestine cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, uterine sarcoma, vaginal cancer, vulvar cancer, Waldenstrom macroglobulinemia, or Wilms tumor. In specific embodiments, the disease is an autoimmune disease, or graft-versus-host-disease or a viral infection. In specific embodiments, the disease is osteoporosis and/or osteopenia and the fusion protein comprises an extracellular fragment of RANK.

[0034] These aspects of the invention will be more fully understood in view of the following description, drawings and non-limiting examples.

BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1: Schematic representation of non-exhaustive illustrative embodiments of fusion proteins of the present invention. In Fig. 1A, extracellular fragment of RANK is fused N-terminal to the linking polypeptide comprising at least a portion of CH2, whereas the binding protein, which is an antibody fragment, is fused C-terminal to the linking polypeptide. The antibody fragment can, for instance, have specificity to CD 16 or CD3 and, as depicted, can be a scFv fragment. In Fig. IB, extracellular portion of GITR is N-terminally fused to the linking polypeptide comprising at least a portion of CH2, whereas the antibody fragment is fused C-terminally to the linking polypeptide. Again, antibody fragment, for instance, can have specificity to CD 16 or CD3. Fig. 1C depicts another specific embodiment, in which a CD 16 or CD3 specific Fab fragment is fused N-terminal to the linking polypeptide comprising at least a portion of CH2 and in which the extracellular fragment of NKG2D is fused C terminal to the linking polypeptide. Although not specifically depicted, the antibody molecule can be any antibody molecule such as scFv, single domain antibody, Fab, or Fv, wherein in the latter cases, either one of heavy chain or light chain can be part of the fusion protein.

[0036] Figure 2: Detection of the respective immune receptor parts of the fusion proteins. Mouse antibodies against NKG2D, RANK or GITR were immobilized on 24 well plates, blocked with 7.5 % BSA-PBS and washed. Afterwards, cell culture supernatants containing the indicated fusion proteins or media derived from a clone that did not produce fusion protein as negative control were added. Additionally, an IgG antibody or FC-NKG2D (same as FC-NKG2D-ADCC, SEQ ID NO: 54), GITR-FC (same as GITR-FC-ADCC, SEQ ID NO: 55) or RANK-FC (same as RANK-FC-ADCC, SEQ ID NO: 56) (10 μg/mL) fusion protein as indicated served as negative or positive controls, respectively. Subsequently, plates were again washed and anti-human IgH-HRP was added. Plates were developed using the trimethoxybenzoate hydrochloride peroxidase substrate system (KPL, Gaithersburg, MD), Absorbance was measured at 450 nm. This figure shows that all tested fusion proteins comprise the respective extracellular domain of the respective immune receptor. The FC- NKG2D, RANK-FC, and GITR-FC correspond to FC-NKG2D-ADCC, RANK-FC-ADCC, and GITR-FC-ADCC and are fusion proteins of extracellular fragments of NKG2D, RANK, or GITR with Fc-fragments that are modified to have enhanced ADCC. These constructs are described in Schmiedel et al, Mol Ther, Schmiedel et al, Cancer Res, Steinbacher et al, Int J Cancer and Raab et al., J Immunol.

[0037] Figure 3: Binding of the effector arms of the RANK, GITR and NKG2D fusion proteins. Jurkat T cells (Figure 3A, used with CD3 constructs) or Sp2/0-AG14-CD16 transfectants (Figure 3B, used with CD 16 constructs) were incubated with RANK fusion proteins or GITR fusion proteins or NKG2D fusion proteins and binding of fusion proteins was confirmed by FACS.

[0038] Figure 4: Dose dependent binding of bispecific CD16 constructs (CD16- NKG2D) as compared to Fc fusion proteins with optimized IgGl-Fc-part (FC-NKG2D). NK cells (CD 16+) were incubated with the indicated concentrations of the two constructs (FC-NKG2D, Figure 4A, same as FC-NKG2D-ADCC, SEQ ID NO: 54; antiCD 16-NKG2D, Figure 4B, SEQ ID NO: 49, co-expressed with antiCD16 Fab light chain SEQ ID NO: 09) and binding of the fusion proteins was measured using FACS. This figure shows that the immune receptor-antiCD16 fusion proteins have higher affinity to FcR than the even ADCC-optimized Fc-moiety of the immune receptor-FC-ADCC fusion protein. The FC-NKG2D proteins corresponds to FC-NKG2D-ADCC and are fusion proteins of an extracellular fragment of NKG2D with Fc-fragments that are modified to have enhanced ADCC. These constructs are described in Steinbacher et al., Int J Cancer and Raab et al., J Immunol.

[0039] Figure 5: Comparative analysis of the effects of CD16-fusion proteins versus Fc-optimized constructs. Primary NKG2DL (Figure 5A), RANKL (Figure 5B) and GITRL (Figure 5C) expressing leukemic cells from patients with AML were incubated with peripheral blood mononuclear cells (PBMC) of healthy donors in the presence or absence of the indicated FC-optimized or bispecific CD 16 constructs (Figure 5 A: CD16-NKG2D versus NKG2D-ADCC (same as FC-NKG2D-ADCC, SEQ ID NO: 54); Figure 5B: RANK-CD 16 (SEQ ID NO: 51) versus RANK-ADCC (same as RANK-FC-ADCC, SEQ ID NO: 56); Figure 5C: GITR-CD16 (SEQ ID NO: 53) versus GITR-ADCC (same as GITR-FC-ADCC, SEQ ID NO: 55)). The FC-NKG2D-ADCC, RANK-FC-ADCC, and GITR-FC-ADCC are fusion proteins of extracellular fragments of NKG2D, RANK, or GITR with Fc-fragments that are modified to have enhanced ADCC. These constructs are described in Schmiedel et al., Mol Ther, Schmiedel et al, Cancer Res, Steinbacher et al, Int J Cancer and Raab et al, J Immunol. CD16-NKG2D represents the fusion protein with the sequence set forth in SEQ ID NO: 49 co-expressed with an anti-CD 16-Fab light chain as set forth in SEQ ID NO: 09. RANK-CD 16 represents the fusion protein with the sequence set forth in SEQ ID NO: 51. GITR-CD16 represents the fusion protein with the sequence set forth in SEQ ID NO: 53. Then the effects on cytotoxicity of NK cells were analyzed by 2h BATDA Europium release assays. This Figure shows that the immune receptor-antiCD16 fusion proteins achieve stronger NK cell mediated cell lysis compared to the absence of these fusion proteins and also to employing immune receptor-FC-ADCC fusion proteins.

[0040] Figure 6: Activation of T cells by the bispecific CD3-fusion proteins. PBMCs of healthy donors were cultured in the presence of leukemic cells from AML patients (either expressing NKG2DL (Figure 6A), RANKL (Figure 6B) or GITRL (Figure 6C)) for 48 h with the indicated bispecific CD3 constructs or medium as control. T cells were selected by staining with directly labeled antibodies against CD4 (upper panels) and CD8 (lower panels). Up-regulation of CD69 as a marker for T cell activation was analyzed by FACS using specific fluorescence-conjugates (all antibodies from Becton Dickinson). This Figure shows that the immune receptor-antiCD3 constructs promote T-cell activation in the presence of NKG2DL-, RANKL- or GITRL-positive cells. CD3-NKG2D represents the fusion protein with the sequence as set forth in SEQ ID NO: 48 co-expressed with an anti-CD3-Fab light chain as set forth in SEQ ID NO: 07, RANK-CD3 represents a fusion protein with a sequence set forth in SEQ ID NO: 50, and GITR-CD3 represents a fusion protein with a sequence set forth in SEQ ID NO: 52.

[0041] Figure 7: Induction of leukemia cell lysis by T cells by the bispecific CD3 constructs and target antigen-restriction of the induced effects. PBMC of a patient with AML the leukemic cells of which expressed NKG2DL but were negative for GITRL (not shown) were cultured in triplicates in medium containing 10 % autologous patient serum in the presence or absence of the indicated constructs (1 μg/mL). For assessment of T-cell mediated killing of AML blasts, cells were washed after incubation for 72 hours in FACS- buffer containing 50 μg/mL human IgG (Flebogamma, Grifols, Langen, Germany), stained with a CD34 antibody to select the leukemic cells and finally resuspended in FACS-buffer containing 7-AAD (BioLegend, San Diego, USA) and negative control compensation particles (BD Biosciences). Malignant cells were defined as CD34+CD45dim. The percentage of apoptotic (7-AAD positive) cells is given in the dot plots. This experiment shows that the CD3-NKG2D constructs specifically direct T-cell mediated cytotoxicity to NKG2DL expressing cells. CD3-NKG2D represents the fusion protein with the sequence as set forth in SEQ ID NO: 48 co-expressed with an anti-CD3-Fab light chain as set forth in SEQ ID NO: 07, and GITR-CD3 represents a fusion protein with a sequence set forth in SEQ ID NO: 52.

[0042] Figure 8: Modification on linker elements. This figure shows exemplary modifications of the Fc portion comprised in the linking polypeptide. Modifications are shown for fusion proteins comprising type I transmembrane protein moieties and type II transmembrane protein moieties.

[0043] Figure 9: Amino acid sequence of fusion protein with N-terminal anti-CD3 Fab heavy chain und C-terminal NKG2D fragment (SEQ ID NO: 48). Amino acids in bold from 1-122: humanized variable (VDJ) domain of antibody UCHT1 specific for human CD3. Amino acids underlined from 123-220: CHI domain of human IgGl . Amino acids in italic from 221-235: modified hinge region of IgGl; modification performed to prevent dimerization: substitution of C226 and C229 both replaced by serine (italic and underlined); numbering according Kabat (EU-index). Within this construct, the cysteine at position 220 is necessary to form a disulfide bond to CD3 light chain. Amino acids normal font from 236-344: modified CH2 domain of IgGl; modification performed to prevent complement fixation; to prevent binding to Fc-receptors; to prevent binding to glycan receptors and reduction of immunogenicity. Following amino acids (EU-index) were exchanged or deleted: E233-^P; L234^V; L235^A; G236^deleted; D265^G; N297^Q; A327^Q; A330^S. The modification N297-^Q prevents the addition of a glycan structure. Amino acids bold underlined from 345-349: whereas the amino acids 345-347 are the first three amino acids of human CH3 domain of IgGl, the amino acids serine (348) and glycine (349) were introduced to provide a suitable restriction site (BspEI = tccgga coding for SG) in the genetic construct. Amino acids in italic from 350-488: extracellular domain of NKG2D corresponds to amino acid position of NKG2D 78-216 (SEQ ID NO: 45); GenBank accession: NKG2D [Homo sapiens] CAA04925. For the production of this fusion protein, in the mouse non-Ig-producing myeloma cell line Sp2/0, a co-transfection with an expression vector coding for the humanized CD3 light chain became necessary. [0044] Figure 10: Amino acid sequence of humanized light chain of CD3 specific antibody UCHTl (SEQ ID NO: 07). Amino acids underlined 1-107: humanized VJ domain of human CD3 specific antibody UCHTl . Amino acids normal font: 108-214: human kappa constant light chain, cysteine at position 214 forms disulfide bond to cysteine 220 in main- chain

[0045] Figure 11: Amino acid sequence of fusion protein with N-terminal anti- CD16 Fab heavy chain und C-terminal NKG2D fragment (SEQ ID NO: 49). Amino acids in bold from 1-118: Variable (VDJ) domain of antibody 3G8 specific for human CD 16. Amino acids underlined from 119-216: CHI domain of human IgGl . Amino acids in italic from 217-231 : modified hinge region of IgGl; modification performed to prevent dimerization: substitution of C226 and C229 both replaced by serine (italic and underlined); numbering according Kabat (EU-index). Within this construct, the cysteine at position 220 is necessary to form a disulfide bond to CD 16 light chain. Amino acids normal font from 232- 340: modified CH2 domain of IgGl ; modification performed to prevent complement fixation; to prevent binding to Fc-receptors; to prevent binding to glycan receptors and reduction of immunogenicity. Following amino acids were exchanged or deleted: E233-^P; L234- V; L235^A; G236^deleted; D265^G; N297^Q; A327^Q; A330^S. The modification N297-^Q prevents the addition of a glycan structure. Amino acids bold underlined from 341- 345: whereas the amino acids 341-343 are the first three amino acids of human CH3 domain of IgGl, the amino acids serine (344) and glycine (345) were introduced to provide a suitable restriction site (BspEI = tccgga coding for SG) in the genetic construct. Amino acids in italic from 346-484: extracellular domain of NKG2D corresponds to amino acid position of NKG2D 78-216 (SEQ ID NO: 45); GenBank accession: NKG2D [Homo sapiens] CAA04925. For the production of this fusion protein, in the mouse non-Ig-producing myeloma cell line Sp2/0, a co-transfection with an expression vector coding for CD 16 light chain became necessary.

[0046] Figure 12: Amino acid sequence of chimeric light chain of CD16 specific antibody 3G8 (SEQ ID NO: 09). Amino acids underlined 1-111 : mouse VJ domain of human CD 16 specific antibody 3G8. Amino acids normal font: 112-218: human kappa constant light chain. Cysteine at position 218 forms disulfide bond to cysteine 220 in main- chain.

[0047] Figure 13: Amino acid sequence of fusion protein with N-terminal RANK fragment and C-terminal anti-CD3-scFv (SEQ ID NO: 50). Amino acids in bold from 1- 183: extracellular domain of human RANK (Q25 - P207). Amino acids in italic from 184-198: modified hinge region of human IgGl; modifications performed to prevent dimerization: substitution of C226 and C229 (numbering according Kabat EU-index) both replaced by serine (italic underlined); substitution of C220: no disulfide bond to light chain has to be formed, therefore substitution against serine (italic underlined). Amino acids in normal font from 199-307: modified CH2 domain of IgGl; modification performed to prevent complement fixation; to prevent binding to Fc-receptors; to prevent binding to glycan receptors and reduction of immunogenicity. Following amino acids (EU-index) were exchanged or deleted: E233^P; L234^V; L235^A; G236^deleted; D265^G; N297^Q; A327-^Q; A330-^S. The modification N297-^Q prevents the addition of a glycan structure. Amino acids bold underlined from 308-312: whereas the amino acids 308-310 are the first three amino acids of human CH3 domain of IgGl, the amino acids serine (311) and glycine (312) were introduced to provide a suitable restriction site (BspEI = tccgga coding for SG) in the genetic construct. Amino acids from 313-556: humanized scFv fragment of human CD3 specific antibody UCHT-1 (orientation VL-VH). Amino acids in italic from 313-419: variable VJ domain of humanized antibody UCHT-1 (anti human CD3). Amino acids in bold from 420-434: glycine-serine linker. Amino acids underlined from 435-556: variable VDJ domain of humanized antibody UCHT-1 (anti human CD3).

[0048] Figure 14: Amino acid sequence of fusion protein with N-terminal RANK fragment and C-terminal anti-CD 16-scFv (SEQ ID NO: 51). Amino acids in bold from 1- 183: extracellular domain of human RANK (Q25 - P207). Amino acids in italic from 184-198: modified hinge region of human IgGl; modifications performed to prevent dimerization: substitution of C226 and C229 (numbering according Kabat EU-index) both replaced by serine (italic underlined); substitution of C220 no disulfide bond to light chain has to be formed, therefore substitution against serine (italic underlined). Amino acids in normal font from 199-307: modified CH2 domain of IgGl; modification performed to prevent complement fixation; to prevent binding to Fc-receptors; to prevent binding to glycan receptors and reduction of immunogenicity. Following amino acids (EU-index) were exchanged or deleted: E233^P; L234^V; L235^A; G236^deleted; D265^G; N297^Q; A327-^Q; A330-^S. The modification N297-^Q prevents the addition of a glycan structure. Amino acids bold underlined from 308-312: whereas the amino acids 308-310 are the first three amino acids of human CH3 domain of IgGl, the amino acids serine (311) and glycine (312) were introduced to provide a suitable restriction site (BspEI = tccgga coding for SG) in the genetic construct. Amino acids from 313-556: scFv fragment of human CD 16 specific antibody 3G8 (orientation VL-VH). Amino acids in italic from 313-423: variable VJ domain of mouse antibody 3G8 (anti human CD16). Amino acids in bold from 424-438: glycine- serine linker. Amino acids underlined from 439-556: variable VDJ domain of mouse antibody 3G8 (anti human CD 16)

[0049] Figure 15: Amino acid sequence of fusion protein with N- terminal GITR and C-terminal humanized anti-CD3-scFv (SEQ ID NO: 52). Amino acids in bold from 1- 137: extracellular domain of human GITR (Q26 - P162). Amino acids in italic from 138-152: modified hinge region of human IgGl; modifications performed to prevent dimerization: substitution of C226 and C229 (numbering according Kabat EU-index) both replaced by serine (italic underlined); substitution of C220 no disulfide bond to light chain has to be formed, therefore substitution against serine (italic underlined). Amino acids in normal font from 153-261 : modified CH2 domain of IgGl; modification performed to prevent complement fixation; to prevent binding to Fc-receptors; to prevent binding to glycan receptors and reduction of immunogenicity. Following amino acids (EU-index) were exchanged or deleted: E233^P; L234^V; L235^A; G236^deleted; D265^G; N297^Q; A327-^Q; A330-^S. The modification N297-^Q prevents the addition of a glycan structure. Amino acids bold underlined from 262-266: whereas the amino acids 262-264 are the first three amino acids of human CH3 domain of IgGl, the amino acids serine (265) and glycine (266) were introduced to provide a suitable restriction site (BspEI = tccgga coding for SG) in the genetic construct. Amino acids from 267-510: humanized scFv fragment of human CD3 specific antibody UCHT-1 (orientation VL-VH). Amino acids in italic from 267-377: variable VJ domain of humanized antibody UCHT-1 (anti human CD3). Amino acids in bold from 378-392: glycine-serine linker. Amino acids underlined from 393-510: variable VDJ domain of humanized antibody UCHT-1 (anti human CD3)

[0050] Figure 16: Amino acid sequence of fusion protein with N- terminal GITR fragment and C-terminal anti-CD 16-scFv (SEQ ID NO: 53). Amino acids in bold from 1- 137: extracellular domain of human GITR (Q26 - P162). Amino acids in italic from 138-152: modified hinge region of human IgGl; modifications performed to prevent dimerization: substitution of C226 and C229 (numbering according Kabat EU-index) both replaced by serine (italic underlined); substitution of C220 no disulfide bond to light chain has to be formed, therefore substitution against serine (italic underlined). Amino acids in normal font from 153-261 : modified CH2 domain of IgGl; modification performed to prevent complement fixation; to prevent binding to Fc-receptors; to prevent binding to glycan receptors and reduction of immunogenicity. Following amino acids (EU-index) were exchanged or deleted: E233^P; L234^V; L235^A; G236^deleted; D265^G; N297^Q; A327-^Q; A330-^S. The modification N297-^Q prevents the addition of a glycan structure. Amino acids bold underlined from 262-266: whereas the amino acids 262-264 are the first three amino acids of human CH3 domain of IgGl, the amino acids serine (265) and glycine (266) were introduced to provide a suitable restriction site (BspEI = tccgga coding for SG) in the genetic construct. Amino acids from 267-510: scFv fragment of human CD 16 specific antibody 3G8 (orientation VL-VH). Amino acids in italic from 267-377: variable VJ domain of mouse antibody 3G8 (anti human CD16). Amino acids in bold from 378-392: glycine- serine linker. Amino acids underlined from 393-510: variable VDJ domain of mouse antibody 3G8 (anti human CD 16).

[0051] Figure 17: CD3 or CD16 NKG2D fusion proteins with attenuated FcR- binding. Description for the generation of NKG2D fusion proteins either bi- or tetravalent. As well as the possibility to exchange NKG2D against any other ecto-domain of type II transmembrane proteins. A co-transfection and co-expression of the respective light chain is required, therefore a light chain cloning scheme is included in the Figure 17 D-F. Figs. 17A to 17C depict a schematic representation of the cloning procedure for the generation of a heavy chain (main chain) for the fusion proteins depicted in Fig. 17C, either as bivalent or tetravalent bispecific fusion protein with modified attenuated Fc-parts. i) The original vector, based on the plasmid-backbone of pcDNA3 (Invitrogen; CMV promoter and bovine growth hormone termination signal are deleted), is depicted. This plasmid contains the human γΐ isotype Ig heavy chain with regulatory elements of the immunoglobulin heavy chain locus, ii) The exchange of a VDJ (variable domain of the heavy chain) element via the restriction endonuclease site Aatll and Clal is indicated, iii) the simple exchange (via restriction sites Mlul and Spel) of the complete human γΐ isotype Ig heavy chain against the coding sequence for a CHI and hinge- and CH2-modified and CH3 -deleted NKG2D DNA element resulting in a bivalent bispecific fusion protein heavy chain is shown, iv) Exchanging the modified CH1- H-CH2 fragment (via restriction sites Mlul and BspEI) against a hinge and CH2 modified CH1-H-CH2-CH3 element results in a tetravalent bispecific fusion protein heavy chain, v) If, in addition, or only as such the cysteines at position C226 and C229 are exchanged the resulting molecules are bivalent bispecific fusion proteins, vi) Exchanging the NKG2D fragment (via restriction sites BspEI and Spel) against an ecto-domain of any other type II transmembrane protein. Substitutions iv) and v) can be combined. In Fig. 17B and 17C) the regions adjacent to the inserted VDJ fragment and NKG2D-element, respectively, are shown in detail. Figs. 17D-F depicts a schematic representation of the cloning procedure for the generation of the light chain of human monospecific antibodies. Vii) The parental vector, based on the plasmid backbone of pCR-Script (Stratagene; lacZ promoter and termination signal are deleted) contains the VJ region and the C region of human γ -gene as well as regulatory elements of the immunoglobulin light chain locus, viii) Exchange of a VJ (variable domain of the light chain) element via the restriction endonucleases Xhol and Spel. In Figs. 17E and 17F the regions adjacent to the inserted VJ and CL elements are shown in detail. Legend: Boxes represent exons, circles enhancer elements and thin lines UT regions and intron sequences. LI and L2, leader sequences encoded by two different exons (also shown in Figure 17B and 17E); V, variable regions; D, diversity region; J, joining regions; CHI, CH2, CH3, CL exons of constant heavy and light chains, respectively; H, hinge region; ecto, extracellular domain of type II transmembrane protein; X = amino acid modifications. Notl, Aatll, Clal, Mlul, BspEI, Spel, Xhol, Kpnl, Xhol, Spel, Pmll, BsmBI, Sail, restriction endonucleases used for cloning; AmpR and NeoR represent the coding regions for Ampicillin and Neomycin resistance, respectively. The cleavage sites for secretory signal peptides are indicated by |; and exon- intron boundaries by [, ].

[0052] Figure 18: RANK and GITR CD3 or CD16 fusion proteins with attenuated FcR-binding. Description for the generation of GITR and RANK fusion proteins either bi- or tetravalent. As well as the possibility to exchange the N-terminally located ecto- domain against any other ecto-domain of a type I transmembrane protein or respective domain of a soluble protein. Figs. 18A to 18C depict a schematic representation of the cloning procedure for the generation of a heavy chain (main chain) for the bispecific fusion proteins depicted in Figs. 1A and IB , either as bivalent or tetravalent bispecific fusion protein with modified attenuated Fc-parts. i) The original vector, based on the plasmid-backbone of pcDNA3 (Invitrogen; CMV promoter and bovine growth hormone termination signal are deleted), is depicted. This plasmid contains the human γΐ isotype Ig heavy chain with regulatory elements of the immunoglobulin heavy chain locus, ii) The exchange of a VDJ (variable domain of the heavy chain) element against an ecto-domain of any type I transmembrane protein or against domains of soluble proteins via the restriction endonuclease site Aatll and Clal is indicated, iii) the simple exchange (via restriction sites Mlul and Spel) of the complete human γΐ isotype Ig heavy chain against the coding sequence for a scFv fragment, a CH3-deleted, a hinge and CH2 modified and CHI-deleted DNA element resulting in a bivalent bispecific fusion protein main chain is shown, iv) Exchanging the modified H- CH2 fragment (via restriction sites Mlul and BspEI) against a hinge and CH2 modified H- CH2-CH3 element results in a tetravalent bispecific fusion protein main chain or as shown in v). If, in addition, or only as such the cysteines at position C226 and C229 are exchanged the resulting molecules are bivalent bispecific fusion proteins, v) Exchanging the scFv fragment (via restriction sites BspEI and Spel) against a scFv-fragment of any other antigen specificity or of different VH and VL orientation. Substitutions iv) and v) can be combined. In Fig. 18B and 18C) the regions adjacent to the inserted ecto-domain, hinge and scFv-elements, respectively, are shown in detail. Legend: Boxes represent exons, circles enhancer elements and thin lines UT regions and intron sequences. LI and L2, leader sequences encoded by two different exons (also shown in Figure 18B); V, variable regions; D, diversity region; J, joining regions; CHI, CH2, CH3, CL exons of constant heavy and light chains, respectively; H, hinge region; scFv single-chain Fv-fragment; Ecto, extracellular domain of type I transmembrane proteins or soluble proteins, X = amino acid modifications. Notl, Aatll, Clal, Mlul, BspEI, Spel, Xhol, Kpnl, Xhol, Spel, Pmll, BsmBI, Sail, restriction endonucleases used for cloning; AmpR and NeoR represent the coding regions for Ampicillin and Neomycin resistance, respectively. The cleavage sites for secretory signal peptides are indicated by |; and exon- intron boundaries by [, ].

[0053] Figure 19: cDNA sequences of anti-CD16 Fab-NKG2D fusion protein and anti-CD3 Fab-NKG2D. Figure 19A depicts a main chain of anti-CD 16 Fab heavy chain- NKG2D fusion protein (SEQ ID NO: 63). Figure 19B depicts a light chain of chimeric anti- CD 16 antibody (clone 3G8) (SEQ ID NO: 64). Figure 19C depicts a main chain of anti-CD3 Fab heavy chain-NKG2D fusion protein (SEQ ID NO: 65). Figure 19D depicts a light chain of humanized anti-CD3 antibody (clone UCHT1) (SEQ ID NO: 66).

[0054] Figure 20: cDNA sequence of RANK-antiCD16 scFv fusion protein and RANK-antiCD3 scFv fusion protein. AntiCD16 scFv is based on clone 3G8 and is in VL- VH orientation, antiCD3 scFv is based on humanized UCHT1 and is in VL-VH orientation. Figure 20A depicts a RANK-antiCD 16 scFv fusion protein (SEQ ID NO: 67). Figure 20B depicts a RANK-antiCD3 scFv fusion protein (SEQ ID NO: 68).

[0055] Figure 21: cDNA sequence of GITR-antiCD16 scFv fusion protein and GITR-antiCD3 scFv fusion protein. AntiCD16 scFv is based on clone 3G8 and is in VL- VH orientation, antiCD3 scFv is based on humanized UCHT1 and is in VL-VH orientation. Figure 21A depicts a GITR-antiCD16 scFv fusion protein (SEQ ID NO: 69). Figure 21B depicts a GITR-antiCD3 scFv fusion protein (SEQ ID NO: 70).

[0056] Figure 22: Amino acid sequence of FC-NKG2D-ADCC (SEQ ID NO: 54)

[0057] Figure 23: Amino acid sequence of RANK-FC-ADCC (SEQ ID NO:56)

[0058] Figure 24: Amino acid sequence of GITR-FC-ADCC (SEQ ID NO: 55)

[0059] Figure 25: cDNA sequence of FC-NKG2D-ADCC (SEQ ID NO: 108) [0060] Figure 26: cDNA sequence of RANK-FC-ADCC (SEQ ID NO: 110)

[0061] Figure 27: cDNA sequence of GITR-FC-ADCC (SEQ ID NO: 109)

[0062] Figure 28: Variations of Fab fragments comprised in the fusion proteins.

The figure exemplarily illustrates possible variations in Fab fragments comprised in fusion proteins of the invention. In preferred embodiments, the fusion protein comprises an antibody molecule that is a Fab fragment. Commonly, the Fab heavy chain (VH-CH1, same as VDJ- CH1) is fused to the linking polypeptide and the extracellular portion of a transmembrane protein and a Fab light chain is co-expressed together with said fusion protein (Figure 28A (i) and Figure 28B (i)). Figure 28A depicts preferred embodiments of the fusion protein, in which VL-CL (same as VJ-CL) is fused to the linking polypeptide and the extracellular portion of a transmembrane protein. A VH-CH1 (same as VDJ-CH1) fragment is co- expressed with the fusion protein. In order to enable disulfide bond formation, a cysteine has to be introduced to the VH-CH1 fragment, for instance at its C-terminal end, whereas the corresponding cysteine of the CL region, which is now part of the fusion protein may be removed. Such fusion proteins are illustrated under (ii). Illustrative examples of such fusion proteins are shown in Figures 32 and 36. Illustrative examples for such VH-CH1 fragments are shown in Figures 29 and 33. Figure 28B depicts preferred embodiments of the fusion protein, in which only the variable regions of the antibody molecule are exchanged, i.e. in which VL-CH1 (same as VJ-CH1) is fused to the linking polypeptide and the extracellular portion of a transmembrane protein. A VH-CL (same as VDJ-CL) fragment is co-expressed with the fusion protein. Such fusion proteins are illustrated under (ii). Illustrative examples of such fusion proteins are shown in Figures 31 and 35. Illustrative examples for such VH-CH1 fragments are shown in Figures 30 and 34.

[0063] Figure 29: Amino acid and cDNA sequence of antiCD3-VH-CH.This figure exemplarily illustrates sequences of VH-CH fragments co-expressed with the fusion proteins according to Figure 28A which comprise a modified Fab fragment that specifically binds to CD3. Figure 29A; exemplary amino acid sequence of an antiCD3-VH-CH fragment comprising a humanized variable antiCD3 VDJ (bold) fused to human CHI . In order to enable a disulfide bond formation to the hinge region of the fusion protein, a cysteine has been introduced at the C-terminal end of the fragment (SEQ ID NO: 92). Figure 29B: cDNA sequence of the antiCD3-VH-CH fragment (SEQ ID NO: 93).

[0064] Figure 30: Amino acid and cDNA sequence of antiCD3-VH-CL.This figure exemplarily illustrates sequences of VH-CL fragments co-expressed with the fusion proteins according to Figure 28B which comprise a modified Fab fragment that specifically binds to CD3. Figure 30A; exemplary amino acid sequence of an antiCD3-VH-CL fragment comprising a variable antiCD3 VDJ (bold) fused to human CL (SEQ ID NO: 94). Figure 3 OB: exemplary cDNA sequence of the antiCD3-VH-CH fragment (SEQ ID NO: 95).

[0065] Figure 31: Amino acid and cDNA sequence of antiCD3-VL-CHl-hinge- CH2-NKG2D.This figure exemplarily illustrates sequences of VL-CHl-hinge-CH2-NKG2D fusion proteins according to Figure 28B which comprise a modified Fab fragment that specifically binds to CD3. Figure 31 A; exemplary amino acid sequence of an antiCD3-VL- CHl-hinge-CH2-NKG2D fusion protein (SEQ ID NO: 96) comprising a humanized variable antiCD3 VL (bold) linked to a human CHI as binding molecule, a linking polypeptide comprising hinge region and modified CH2 according to Figure 8, and a extracellular portion of NKG2D. Figure 3 IB: exemplary cDNA sequence of the antiCD3-VL-CHl-hinge-CH2- NKG2D fragment (SEQ ID NO: 97).

[0066] Figure 32: Amino acid and cDNA sequence of antiCD3-VL-CL-hinge- CH2-NKG2D.This figure exemplarily illustrates sequences of VL-CL-hinge-CH2-NKG2D fusion proteins according to Figure 28A which comprise a modified Fab fragment that specifically binds to CD3. Figure 32A; exemplary amino acid sequence of an antiCD3-VL- CL-hinge-CH2-NKG2D fusion protein (SEQ ID NO: 98) comprising a humanized variable antiCD3 VL (bold) linked to a human CHI (italic) as binding molecule, a linking polypeptide comprising hinge (underlined) region and modified CH2 according to Figure 8, and a extracellular portion of NKG2D. The cysteine at the C-terminal end of CL, which usually enables disulfide bond formation between the light chain and the heavy chain of an antibody, has been removed in the fusion protein. Figure 32B: exemplary cDNA sequence of the antiCD3-VL-CL-hinge-CH2-NKG2D fragment (SEQ ID NO: 99).

[0067] Figure 33: Amino acid and cDNA sequence of antiCD16-VH-CH.This figure exemplarily illustrates sequences of VH-CH fragments co-expressed with the fusion proteins according to Figure 28A which comprise a modified Fab fragment that specifically binds to CD 16. Figure 33 A; exemplary amino acid sequence of an antiCD3-VH-CH fragment comprising a variable antiCD16 VDJ (bold) fused to human CHI . In order to enable a disulfide bond formation to the hinge region of the fusion protein, a cysteine has been introduced at the C-terminal end of the fragment (SEQ ID NO: 100). Figure 33B: exemplary cDNA sequence of the antiCD16-VH-CH fragment (SEQ ID NO: 101).

[0068] Figure 34: Amino acid and cDNA sequence of antiCD16-VH-CL.This figure exemplarily illustrates sequences of VH-CL fragments co-expressed with the fusion proteins according to Figure 28B which comprise a modified Fab fragment that specifically binds to CD 16. Figure 34A; exemplary amino acid sequence of an antiCD3-VH-CL fragment comprising a variable antiCD3 VDJ (bold) fused to human CL (SEQ ID NO: 102). Figure 34B: exemplary cDNA sequence of the antiCD3-VH-CH fragment (SEQ ID NO: 103).

[0069] Figure 35: Amino acid and cDNA sequence of antiCD16-VL-CHl-hinge- CH2-NKG2D.This figure exemplarily illustrates sequences of VL-CHl-hinge-CH2-NKG2D fusion proteins according to Figure 28B which comprise a modified Fab fragment that specifically binds to CD 16. Figure 35 A; exemplary amino acid sequence of an antiCD16-VL- CH 1 -hinge-CH2-NKG2D fusion protein (SEQ ID NO: 104) comprising a variable antiCD16 VL (bold) linked to a human CHI as binding molecule, a linking polypeptide comprising hinge region and modified CH2 according to Figure 8, and a extracellular portion of NKG2D. Figure 35B: exemplary cDNA sequence of the antiCD16-VL-CHl-hinge-CH2-NKG2D fragment (SEQ ID NO: 105).

[0070] Figure 36: Amino acid and cDNA sequence of antiCD16-VL-CL-hinge- CH2-NKG2D.This figure exemplarily illustrates sequences of VL-CL-hinge-CH2-NKG2D fusion proteins according to Figure 28A which comprise a modified Fab fragment that specifically binds to CD 16. Figure 36A; exemplary amino acid sequence of an antiCD16-VL- CL-hinge-CH2-NKG2D fusion protein (SEQ ID NO: 106) comprising a variable antiCD16 VL (bold) linked to a human CHI (italic) as binding molecule, a linking polypeptide comprising hinge (underlined) region and modified CH2 according to Figure 8, and a extracellular portion of NKG2D. The cysteine at the C-terminal end of CL, which usually enables disulfide bond formation between the light chain and the heavy chain of an antibody, has been removed in the fusion protein. Figure 36B: exemplary cDNA sequence of the antiCD16-VL-CL-hinge-CH2-NKG2D fragment (SEQ ID NO: 107).

[0071] Figure 37: Lysis of leukemia cells by autologous T and NK cells upon exposure to CD3-NKG2D and CD16-NKG2D, respectively. PBMC of a leukemia patient with AML cells expressing NKG2DL were directly cultured after isolation in medium containing 10 % autologous patient serum in the presence or absence of the indicated constructs (1 μg/mL each). For assessment of T cell and NK cell mediated killing of AML blasts, cells were washed after 72 hours in FACS-buffer containing 50 μg/mL human IgG (Flebogamma, Grifols, Langen, Germany), stained with a CD34 antibody to select the leukemic cells and finally resuspended in FACS-buffer containing 7-AAD (BioLegend, San Diego, USA) and quantification beads (BD Biosciences). Malignant cells were defined as CD34+. Results obtained with 3 different AML patients indicated as percent of viable cells as compared to an untreated control (100%) are shown in Figure 37. [0072] Figure 38: Titration of bispecific CD16- and CD3-NKG2D fusion proteins to determine saturating concentrations. T cells (Figure 38 A) and NK cells (Figure 38B) of a healthy donor were incubated with the indicated concentrations of CD3-NKG2D- and CD16-NKG2D-, respectively, followed by anti-human-PE conjugate and then counterstained with CD4/CD8 for T cells or CD56/CD3 for NK cells followed by FACS analysis. This shows that both constructs reach saturating concentrations at 100 pmol/ml.

[0073] Figure 39: T cells (Figure 39A) and NK cells (right panel) of a healthy donor were incubated with CD3-NKG2D and CD16-NKG2D (lOOpmol/ml) followed by an anti- human-PE conjugate and then counterstained for CD4/CD8 or CD56/CD3 to determine specific binding by FACS analysis. This figure shows that the CD3-NKG2D specifically binds to T cells and not NK cells, whereas the CD16-NKG2D construct specifically binds to NK cells and not T cells.

[0074] Figure 40: Proliferation inducing capacity of CD3-NKG2D constructs. PBMC from a healthy donor were incubated with NALM16 (NKG2DL⁺, FLT3⁺) leukemia cells as target cells in the presence of the indicated constructs (^g/ml) for 3d. Subsequently the number of CD8+ (Figure 40A) and CD4+ (Figure 40B) T cells was determined by FACS using quantification beads (BD Pharmingen). This experiment demonstrates that the CD3- NKG2D construct is able to induce proliferation of T cells depending on binding of the construct to target cells. A bispecific FLT3-CD3 antibody served as positive control. As expected, no effect on T cell proliferation was observed with the CD16-NKG2D construct (which does not stimulate T cells) or a bispecific CD19-CD3 antibody serving as negative control due to lack of target antigen expression on the leukemic cells.

Brief Description of the Sequence Listings

SEQ ID NO: 01 hinge and CH2 corresponding to positions 216-340 of human IgG (amino acid)

SEQ ID NO: 02 hinge and CH2 corresponding to positions 216-340 of human IgG, comprising mutations for function depletion, type I (amino acid)

SEQ ID NO: 03 hinge and CH2 corresponding to positions 216-340 of human IgG, comprising mutations for function depletion, type II (amino acid)

SEQ ID NO: 04 antiCD3-scFv (amino acid)

SEQ ID NO : 05 antiCD 16-scFv (amino acid)

SEQ ID NO: 06 antiCD3-Fab heavy chain (amino acid)

SEQ ID NO: 07 antiCD3-Fab light chain (amino acid)

SEQ ID NO: 08 antiCD16-Fab heavy chain (amino acid)

SEQ ID NO : 09 antiCD 16-Fab light chain (amino acid)

SEQ ID NO : 10 NKG2D (amino acid) SEQ ID NO: 1 1 RANK (amino acid)

SEQ ID NO: 12 GITR (amino acid)

SEQ ID NO: 13 CD94-NKG2 (amino acid)

SEQ ID NO: 14 NKp30 (amino acid)

SEQ ID NO: 15 CTLA-4 (amino acid)

SEQ ID NO: 16 PD-1 (amino acid)

SEQ ID NO: 17 BTLA (amino acid)

SEQ ID NO: 18 LAG-3 (amino acid)

SEQ ID NO: 19 TIM-3 (amino acid)

SEQ ID NO: 20 LAIR-1 (amino acid)

SEQ ID NO: 21 TIGIT (amino acid)

SEQ ID NO: 22 Siglecl (amino acid)

SEQ ID NO: 23 Siglec2 (amino acid)

SEQ ID NO: 24 Siglec3 (amino acid)

SEQ ID NO: 25 Siglec4 (amino acid)

SEQ ID NO: 26 Siglec5 (amino acid)

SEQ ID NO: 27 Siglec6 (amino acid)

SEQ ID NO: 28 Siglec7 (amino acid)

SEQ ID NO: 29 Siglec8 (amino acid)

SEQ ID NO: 30 Siglec9 (amino acid)

SEQ ID NO: 31 SigleclO (amino acid)

SEQ ID NO: 32 Siglecl 1 (amino acid)

SEQ ID NO: 33 Siglecl 2 (amino acid)

SEQ ID NO: 34 Siglecl 4 (amino acid)

SEQ ID NO: 35 Siglecl 5 (amino acid)

SEQ ID NO: 36 Siglecl 6 (amino acid)

SEQ ID NO: 37 MICA (amino acid)

SEQ ID NO: 38 MICB (amino acid)

SEQ ID NO: 39 ULBP1 (amino acid)

SEQ ID NO: 40 ULBP2 (amino acid)

SEQ ID NO: 41 ULBP3 (amino acid)

SEQ ID NO: 42 ULBP4 (amino acid)

SEQ ID NO: 43 ULBP5 (amino acid)

SEQ ID NO: 44 ULBP6 (amino acid)

SEQ ID NO: 45 NKG2D position 78-216 (amino acid)

SEQ ID NO: 46 RANK position 25-207 (amino acid)

SEQ ID NO: 47 GITR position 26-162 (amino acid)

SEQ ID NO: 48 N-antiCD3Fab heavy chain-hinge/CH2-NKG2D-C (amino acid)

SEQ ID NO: 49 N-antiCD16Fab heavy chain-hinge/CH2-NKG2D-C (amino acid)

SEQ ID NO: 50 N-RANK-hinge/CH2-antiCD3scFv-C (amino acid)

SEQ ID NO: 51 N-RANK-hinge/CH2-antiCD16scFv-C (amino acid)

SEQ ID NO: 52 N-GITR-hinge/CH2-antiCD3scFv-C (amino acid)

SEQ ID NO: 53 N-GITR-hinge/CH2-antiCD16scFv-C (amino acid)

SEQ ID NO: 54 FC-NKG2D-ADCC (amino acid) SEQ ID NO: 55 GITR-FC-ADCC (amino acid)

SEQ ID NO: 56 RAN -FC-ADCC (amino acid)

SEQ ID NO: 57 Strep-tag (amino acid)

SEQ ID NO: 58 Strep-tag II (amino acid)

SEQ ID NO: 59 myc-tag (amino acid)

SEQ ID NO: 60 FLAG-tag (amino acid)

SEQ ID NO: 61 His6-tag (amino acid)

SEQ ID NO: 62 HA-tag (amino acid)

SEQ ID NO: 63 main chain of anti-CD 16 Fab heavy chain-NKG2D fusion protein (cDNA)

SEQ ID NO: 64 light chain of chimeric anti-CD16 antibody (clone 3G8) (cDNA)

SEQ ID NO: 65 main chain of anti-CD3 Fab heavy chain-NKG2D fusion protein (cDNA)

SEQ ID NO: 66 light chain of humanized anti-CD3 antibody (clone UCHTl) (cDNA)

SEQ ID NO: 67 RANK-ant iCD 16 scFv fusion protein (cDNA)

SEQ ID NO: 68 RANK-antiCD3 scFv fusion protein (cDNA)

SEQ ID NO: 69 GITR-antiCD16 scFv fusion protein (cDNA)

SEQ ID NO: 70 GITR-antiCD3 scFv fusion protein (cDNA)

SEQ ID NO: 71 3G8-H-for primer (nucleotide)

SEQ ID NO: 72 3G8-H-rev primer (nucleotide)

SEQ ID NO: 73 GITR-for primer (nucleotide)

SEQ ID NO: 74 GITR-rev primer (nucleotide)

SEQ ID NO: 75 RANK- for primer (nucleotide)

SEQ ID NO: 76 RANK-rev primer (nucleotide)

SEQ ID NO: 77 universal for (Aatll) primer (nucleotide)

SEQ ID NO: 78 universal rev (Clal) primer (nucleotide)

SEQ ID NO: 79 3G8-L-for (Xhol) primer (nucleotide)

SEQ ID NO: 80 3G8-L-rev (Spel) primer (nucleotide)

SEQ ID NO : 81 UCHT 1 -for (BspEI) primer (nucleotide)

SEQ ID NO: 82 UCHT 1 -rev (Spel) primer (nucleotide)

SEQ ID NO: 83 NKG2D-for (BspEI) primer (nucleotide)

SEQ ID NO: 84 NKG2D-rev (Spel) primer (nucleotide)

SEQ ID NO: 85 Nkp40 (amino acid)

SEQ ID NO: 86 NKp44 (amino acid)

SEQ ID NO: 87 NKp46 (amino acid)

SEQ ID NO: 88 NKp80 (amino acid)

SEQ ID NO: 89 OPG (amino acid)

SEQ ID NO: 90 RANKL (amino acid)

SEQ ID NO : 91 GITRL (amino acid)

SEQ ID NO: 92 antiCD3-VH-CHl (amino acid)

SEQ ID NO: 93 antiCD3-VH-CHl (cDNA)

SEQ ID NO: 94 antiCD3-VH-CL (amino acid)

SEQ ID NO: 95 antiCD3-VH-CL (cDNA)

SEQ ID NO: 96 antiCD3-VL-CHl-hinge-CH2-NKG2D (amino acid)

SEQ ID NO: 97 antiCD3-VL-CHl-hinge-CH2-NKG2D (cDNA)

SEQ ID NO: 98 antiCD3-VL-CL-hinge-CH2-NKG2D (amino acid) SEQ ID NO 99 antiCD3-VL-CL-hinge-CH2-NKG2D (cDNA)

SEQ ID NO 100 antiCD16-VH-CH (amino acid)

SEQ ID NO 101 antiCD16-VH-CH (cDNA)

SEQ ID NO 102 antiCD16-VH-CL (amino acid)

SEQ ID NO 103 antiCD16-VH-CL (cDNA)

SEQ ID NO 104 antiCD16-VL-CHl-hinge-CH2-NKG2D (amino acid)

SEQ ID NO 105 antiCD 16- VL-CH 1 -hinge-CH2-NKG2D (cDNA)

SEQ ID NO 106 antiCD16-VL-CL-hinge-CH2-NKG2D (amino acid)

SEQ ID NO 107 antiCD 16- VL-CL-hinge-CH2-NKG2D (cDNA)

SEQ ID NO 108 FC-NKG2D-ADCC (cDNA)

SEQ ID NO 109 GITR-FC-ADCC (cDNA)

SEQ ID NO 110 RANK-FC-ADCC (cDNA)

Detailed Description of the Invention

Recombinant fusion protein

[0075] The invention provides recombinant fusion proteins consisting of a binding protein with a binding site, an extracellular fragment of a transmembrane protein that is an immune receptor and a linking polypeptide connecting the binding protein and the extracellular fragment of a transmembrane protein, wherein said polypeptide consists of at least a portion of a CH2 domain and optionally at least a portion of a hinge region. The binding protein and the immune receptor are both equipped with specific binding sites, in which the binding protein has the ability to bind to receptors on immune cells, such as T cells or NK cells, whereas the immune receptor fragment has the ability to bind to an antigen or ligand on a target cell. In binding both, the immune cell and the target cell, the fusion protein brings the immune cell and the target cell in close proximity to each other in a way that an immune response of the immune cell directed against the target cell is mediated. It is envisioned by the invention that the binding protein moiety of the fusion protein does not bind to the extracellular fragment of a transmembrane protein comprised in said fusion protein. It is envisioned that the immune receptor fragment may also bind to a ligand or antigen that is expressed by a target cell and that said ligand or antigen may be bound to the target cell or may be soluble, i.e. set free by the target cell and thus may not be bound to the target cell.

[0076] The term "polypeptide" as used herein refers to a linear chain of amino acid residues, wherein the linear chain of amino acid residues comprise at least 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acid residues. The individual amino acid residues adjacent in the linear sequence of the polypeptide are bonded together by peptide bonds. [0077] The term "protein" as used herein refers to biological molecules consisting of one or more polypeptides.

[0078] The term "amino acid" or "amino acid residue" refers to an a- or β-amino carboxylic acid.

[0079] When used in connection with a protein or peptide, the term "amino acid" or "amino acid residue" typically refers to an a-amino carboxylic acid having its art recognized definition such as an amino acid selected from the group consisting of: L-alanine (Ala or A); L-arginine (Arg or R); L-asparagine (Asn or N); L-aspartic acid (Asp or D); L-cysteine (Cys or C); L-glutamine (Gin or Q); L-glutamic acid (GIu or E); glycine (Gly or G); L-histidine (His or H); L-isoleucine (ILE or I): L-leucine (Leu or L); L-lysine (Lys or K); L-methionine (Met or M); L-phenylalanine (Phe or F); L-proline (Pro or P); L-serine (Ser or S); L-threonine (Thr or T); L-tryptophan (Trp or W); L-tyrosine (Tyr or Y); and L- valine (Val or V), although modified, synthetic, or rare amino acids such as e.g. taurine, ornithine, selenocysteine, homocystine, hydroxyproline, thioproline, iodo-tyrosine, 3-nitro-tyrosine, ornithine, citrulline, canavanine, 5 -hydroxy-tryptophane, carnosine, cycloleucine, 3,4-dihydroxy phenylalanine, N-acetylcysteine, pro lino 1, allylglycine or acetidine-2-carboxylic acid may be used as desired. Generally, amino acids can be grouped as having a nonpolar side chain (e.g., Ala, Cys, ILE, Leu, Met, Phe, Pro, Val); a negatively charged side chain (e.g., Asp, GIu); a positively charged side chain (e.g., Arg, His, Lys); or an uncharged polar side chain (e.g., Asn, Cys, Gin, Gly, His, Met, Phe, Ser, Thr, Trp, and Tyr).

[0080] As used herein, "recombinant" refers to the alteration of genetic material by human intervention. Typically, recombinant refers to the manipulation of DNA or RNA in a virus, cell, plasmid or vector by molecular biology (recombinant DNA technology) methods, including cloning and recombination. A recombinant cell, polypeptide, or nucleic acid can be typically described with reference to how it differs from a naturally occurring counterpart (the "wild-type").

[0081] The term "fusion protein" as used herein refers to proteins created through the joining of two or more genes or fragments thereof that originally coded for separate proteins or that originally coded for the same protein or through the joining of fragments of the same gene. The term "recombinant fusion protein" refers to fusion proteins created artificially by recombinant DNA technology comprising joint amino acid sequences from two or more genes or fragments thereof or fragments of the same gene and which is not found in nature.

[0082] A "fragment" as used herein refers to a portion of a parental protein. Such a fragment can comprise consecutive amino acids of the parental protein. A "fragment" can also refer to a protein in which fragments of a parental protein are fused together. Those fragments do not necessarily originate from the same polypeptide chain. Such a fragment may be for example a scFv fragment, which is referred to as an antibody fragment or molecule, in which the variable domains of a heavy chain and a light chain of an immunoglobulin are fused together. A fragment can also comprise modifications such as amino acid substitutions, amino acid deletions or amino acid insertions compared to the parental protein.

Immune receptor

[0083] The term "antigen on a target cell" or "ligand on a target cell" or as used herein refers to an antigen or ligand that is or can be presented on a surface that is located on or within target cells. These antigens or ligands can be presented on the cell surface with an extracellular part, which is often combined with a transmembrane and cytoplasmic part of the molecule. The antigen or ligand can also be a protein, typically a membrane protein or transmembrane protein with an extracellular part which is expressed by the target cell and is located on the cell surface of a target cell or can be expressed in soluble form and be set free, e.g. secreted by the target cell. These antigens or ligands can in some embodiments be presented or expressed only by tumor cells and not by normal, i.e. non-tumor cells. Tumor antigens can be exclusively expressed on tumor cells or may represent a tumor specific mutation compared to non-tumor cells. In such an embodiment a respective antigen may be referred to as a "tumor-specific" antigen or ligand. In this context, "tumor restricted expression" means that an antigen or ligand is expressed only by one or more tumor cells. Other ligands or antigens are presented by both tumor cells and non-tumor cells, which may be referred to as "tumor-associated" ligands or antigens. These tumor-associated antigens can be overexpressed on tumor cells when compared to non-tumor cells or are accessible for antibody binding in tumor cells due to the less compact structure of the tumor tissue compared to non-tumor tissue. In this context, "tumor associated expression" means that an antigen or ligand is expressed by one or more tumor cells and one and more non-tumor cells. In preferred embodiments, the antigen or ligand according to the invention is a tumor-associated antigen or ligand.

[0084] The term "ligand" as used herein is used in a biochemical context. The term relates to a substance that forms a complex with a bio molecule which typically serves a biological purpose. In protein- ligand binding, the ligand is usually a molecule that binds to a site on a target protein and which can trigger a signal upon binding. The binding of the ligand is usually reversible. A ligand can, for instance, be a substrate, inhibitor, activator, or neurotransmitters. It can be a small molecule but also a macromolecule, such as a protein or a nucleic acid.

[0085] The term "antigen" as used herein relates to any substance capable of provoking an adaptive immune response and which is characterized by its ability to be bound by the variable Fab region of an antibody. An antigen can originate from within the body or from the external environment. It can for instance be a peptide, a protein, a polysaccharide or a lipid, including parts of bacteria, viruses, and other microorganisms such as coats, capsules, cell walls, flagella, fimbrae, and toxins. In this context, the term "epitope", also known as the "antigenic determinant", refers to the portion of an antigen to which an antibody or T-cell receptor specifically binds, thereby forming a complex. Thus, the term "epitope" includes any molecule or protein determinant capable of specific binding to an immunoglobulin or T-cell receptor. The binding site(s) (paratope) of an antibody molecule described herein may specifically bind to/interact with conformational or continuous epitopes, which are unique for the target structure. Epitopic determinants usually consist of chemically active surface groupings of molecules such as amino acids or sugar side chains and usually have specific three dimensional structural characteristics, as well as specific charge characteristics. In some embodiments, epitope determinants include chemically active surface groupings of molecules such as amino acids, sugar side chains, phosphoryl, or sulfonyl, and, in certain embodiments, may have specific three dimensional structural characteristics, and/or specific charge characteristics. With regard to polypeptide antigens a conformational or discontinuous epitope is characterized by the presence of two or more discrete amino acid residues, separated in the primary sequence, but assembling to a consistent structure on the surface of the molecule when the polypeptide folds into the native protein/antigen (Sela, M., Science (1969) 166, 1365-1374; Laver, W.G., et al. Cell (1990) 61, 553-556). The two or more discrete amino acid residues contributing to the epitope may be present on separate sections of one or more polypeptide chain(s). These residues come together on the surface of the molecule when the polypeptide chain(s) fold(s) into a three-dimensional structure to constitute the epitope. In contrast, a continuous or linear epitope consists of two or more discrete amino acid residues, which are present in a single linear segment of a polypeptide chain. As an illustrative example, a "context-dependent" CD3 epitope refers to the conformation of said epitope. Such a context-dependent epitope, localized on the epsilon chain of CD3, can only develop its correct conformation if it is embedded within the rest of the epsilon chain and held in the right position by heterodimerization of the epsilon chain with either CD3 gamma or delta chain. In contrast thereto, a context-independent CD3 epitope may be an N-terminal 1-27 amino acid residue polypeptide or a functional fragment thereof of CD3 epsilon. Generally, epitopes can be linear in nature or can be a discontinuous epitope. Thus, as used herein, the term "conformational epitope" refers to a discontinuous epitope formed by a spatial relationship between amino acids of an antigen other than an unbroken series of amino acids. The term "epitope" also includes an antigenic determinant of a hapten, which is known as a small molecule that can serve as an antigen by displaying one or more immunologically recognized epitopes upon binding to larger matter such as a larger molecule e.g. a protein.

[0086] In the fusion protein of the present invention, the ligand or antigen on a target cell is targeted, i.e. specifically bound, by an immune receptor fragment rather than by an antibody fragment. An immune receptor may bind to one specific ligand or may bind to more than one specific ligands meaning that it can bind to different molecules. Sometimes, multiple ligands of the same immune receptor are expressed by target cells. In certain embodiments, the expression level of the multiple ligands of the same immune receptor varies in different target cells. Therefore, being able to bind to different ligands on different target cells, the immune receptor is able to bind to various target cells expressing different ligands, which may make the immune receptor more versatile compared to an antibody fragment that typically binds to one specific antigen. In some embodiments, such a target cell is a tumor cell.

[0087] The term "immune receptor", as used herein, is a receptor, usually on a cell membrane, which binds to a substance, such as a ligand, and causes, activates, stimulates or inhibits a response in the immune system. In more specific embodiments of the present invention, the immune receptor is a receptor that is expressed on NK cells or T-cells, wherein the expression may not be limited to NK cells or T-cells. The immune receptor can be activating or inhibiting, meaning that upon stimulation it can either activate immune response or inhibit immune response by the immune cell. In preferred embodiments, the immune receptor is inhibiting. In other preferred embodiments, the immune receptor is activating. Non-exhaustive examples for receptors on NK-cells or T-cells are NKG2D (UniProtKB accession number P26718, SEQ ID NO: 10), RANK (UniProtKB accession number Q9Y6Q6, SEQ ID NO: 11), GITR (UniProtKB accession number Q9Y5U5, SEQ ID NO: 12), CD94- NKG2 (UniProtKB accession number Q 13241, SEQ ID NO: 13), CTLA-4 (UniProtKB accession number PI 6410, SEQ ID NO: 15), PD-1 (UniProtKB accession number Q15116, SEQ ID NO: 16), BTLA (UniProtKB accession number Q7Z6A9, SEQ ID NO: 17), LAG-3 (UniProtKB accession number PI 8627, SEQ ID NO: 18), TIM-3 (UniProtKB accession number Q8TDQ0, SEQ ID NO: 19), LAIR-1 (UniProtKB accession number Q6GTX8, SEQ ID NO: 20), TIGIT (UniProtKB accession number Q495A1, SEQ ID NO: 21), Siglecl (UniProtKB accession number Q9BZZ2, SEQ ID NO: 22), Siglec2 (UniProtKB accession number P20273, SEQ ID NO: 23), Siglec3 (UniProtKB accession number P20138, SEQ ID NO: 24), Siglec4 (UniProtKB accession number P20916, SEQ ID NO: 25), Siglec5 (UniProtKB accession number 015389, SEQ ID NO: 26), Siglec6 (UniProtKB accession number 043699, SEQ ID NO: 27), Siglec7 (UniProtKB accession number Q9Y286, SEQ ID NO: 28), Siglec8 (UniProtKB accession number Q9NYZ4, SEQ ID NO: 29), Siglec9 (UniProtKB accession number Q9Y336, SEQ ID NO: 30), SigleclO (UniProtKB accession number Q96LC7, SEQ ID NO: 31), Siglecl l (UniProtKB accession number Q96RL6, SEQ ID NO: 32), Siglecl2 (UniProtKB accession number Q96PQ1, SEQ ID NO: 33), Siglecl4 (UniProtKB accession number Q08ET2, SEQ ID NO: 34), Siglecl5 (UniProtKB accession number Q6ZMC9, SEQ ID NO: 35), Siglecl6 (UniProtKB accession number A6NMB1, SEQ ID NO: 36), NKp30 (UniProtKB accession number 014931, SEQ ID NO: 14), NKp40 (UniProtKB accession number Q9NZS2, SEQ ID NO: 85), NKp44 (UniProtKB accession number 095944, SEQ ID NO: 86), NKp46 (UniProtKB accession number 076036, SEQ ID NO: 87), NKp80 (UniProtKB accession number Q9NZS2, SEQ ID NO: 88). A further example of an immune receptor within the meaning of the term as used herein is OPG (UniProtKB accession number 000300, SEQ ID NO: 89). Further examples for immune receptors are described in Pegram et al., (2011), which is incorporated herein by reference. In a preferred embodiment, such a receptor is NKG2D, RANK or GITR. The immune receptor fragments in the embodiments of the present invention also have further the ability to bind to an antigen or ligand on a target cell, preferably on a target tumor cell, preferably an antigen or ligand expressed by a target tumor cell and which is located on the surface of the tumor cell, preferably wherein the expression of the antigen or ligand is restricted or widely restricted or associated to a tumor cell. In preferred embodiments, the ligand or antigen is selected from the group consisting of NKG2DL, RANKL and GITRL.

[0088] The term "extracellular fragment" of a protein, as used herein, refers to a fragment comprising at least a portion of an extracellular part of a protein. For instance, an "extracellular fragment of a transmembrane protein" or "extracellular fragment of an immune receptor" is a fragment comprising at least a portion of the extracellular part of the transmembrane protein or the immune receptor.

[0089] A "transmembrane protein" is a type protein that interacts with biological membranes and typically spans the entirety of the biological membrane to which it is permanently attached, which means that the transmembrane proteins span from one side of a membrane through to the other side of the membrane. Transmembrane proteins may be classified by their structure (alpha-helical or beta-barrels) or by their topology. In the latter case, the classification refers to the position of the N- and C-terminal domains. Types I, II, and III are single-pass molecules, while type IV are multiple-pass molecules. Type I transmembrane proteins are anchored to the lipid membrane with a stop-transfer anchor sequence and have their N-terminal domains targeted to the ER lumen during synthesis, meaning that their C-terminal part is directed to the cytosol whereas their N-terminal portion is on the extracellular side. For instance, RANK and GITR are type I transmembrane proteins. In contrast, type II and III are anchored with a signal-anchor sequence. Herein, the C-terminal domain of type II is targeted to the ER lumen, meaning that the N-terminal part is directed to the cytosol and the C-terminal part is directed to the extracellular space. NGK2D is an example for a type II transmembrane protein. Type III, however, have their N-terminal domains targeted to the ER lumen or extracellular space, respectively and the C-terminal part is directed to the cytosol. Type IV can further be subdivided into IV-A, in which the N- terminal domains are targeted to the cytosol and IV-B, where the N-terminal domain is targeted to the ER lumen or extracellular space. In specific embodiments of the present invention, the immune receptors referred to are single-pass transmembrane proteins. In a more specific embodiment, they are type I or type II transmembrane proteins.

[0090] In one specific embodiment, the immune receptor can be NKG2D (UniProtKB accession number P26718, SEQ ID NO: 10), which is a C-type, lectin-like, type II transmembrane glycoprotein which is an activating receptor found on NK cells and CD 8 T cells (both αβ and γδ) and which is encoded by the KLRKl gene. For signal transduction, NKG2D associates with an adaptor molecule which is DAP 10 in human. Each NKG2D homodimer associates with two homodimers of DAP 10 resulting in a hexameric structure. For NK cells, NKG2D acts as a primary activation signal and can override inhibitory signals received by other NK cell receptors whereas NKG2D functions as a co-stimulatory signal in cytotoxic T-cells. There are 8 individual ligands for human NKG2D that have been identified by now. These ligands belong to two families: MHC class I chain-related proteins (MICA (UniProtKB accession number Q29983, SEQ ID NO: 37) and MICB (UniProtKB accession number Q29980, SEQ ID NO: 38)) and HCMV UL16-binding proteins (ULBP1, UniProtKB accession number Q9BZM6, SEQ ID NO: 39), ULBP2 (UniProtKB accession number Q9BZM5, SEQ ID NO: 40), ULBP3 (UniProtKB accession number Q9BZM4, SEQ ID NO: 41), ULBP4 (UniProtKB accession number Q8TD07, SEQ ID NO: 42), ULBP5 (UniProtKB accession number Q6H3X3, SEQ ID NO: 43) and ULBP6 (UniProtKB accession number Q5VY80, SEQ ID NO: 44)). As the person skilled in the art understands, NKG2DL are highly polymorphic in humans. Thus specific amino acid sequences as given above for NKG2DL are exemplary but not limiting. In addition, any yet to be identified NKG2DL is also encompassed by the present invention, meaning also such novel NKG2DL may be bound by NKG2D or a fusion protein comprising a portion of NKG2D. NKG2DL are widely expressed by many tumor cells from diverse tissue origins. MICA for example is expressed by almost all primary glioma isolates and many primary tumor isolates from carcinoma (lung, breast, kidney, prostate, ovary, and colon), melanoma, and some primary leukemia cells. In addition, MICA is also expressed by 75% of primary cutaneous melanoma isolates and 50% of metastatic melanoma lesions. ULBP1-3 is also expressed by almost all primary glioma isolates. The expression of NKG2DL may be tumor-associated, but the expression profile of the individual ligands can vary strongly between different tumor entities. NKG2DL can also be set free in soluble form by tumor cells, whereby the released NKG2DL can systemically inhibit NKG2D-mediated anti-tumor immune-response. The extracellular part of NKG2D refers to positions 78-216 of the linear amino acid sequence of NKG2D (SEQ ID NO: 10). The amino acid sequence of the extracellular fragment of NKG2D is set forth in SEQ ID NO: 45. In specific embodiments, the extracellular fragment of NKG2D comprised in the fusion protein of the invention comprises at least 50-139 consecutive amino acids corresponding to SEQ ID NO: 45, wherein in at least 50-139 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61 , 62, 63, 64, 65, 66, 67, 68, 69, 70, 71 , 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 11 1, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, or 139 positions, or comprises a sequence having a pairwise sequence identity of at least about 80 %>, at least about 85 %>, at least about 90 %>, at least about 95 %>, at least about 98 %> or at least about 99 %> when aligned to at least 50-139 consecutive amino acids of SEQ ID NO: 45, wherein in at least 50-139 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, or 139 positions.

[0091] Receptor Activator of NF-κΒ (RANK, UniProtKB accession number Q9Y6Q6, SEQ ID NO: 11) is a type I transmembrane protein and a member of the tumor necrosis factor receptor sub-family. It is the receptor for RANK-Ligand (RANKL) and part of the RANK/RANKL/OPG signaling pathway that regulates osteoclast differentiation and activation. Furthermore, the RANK/RANKL molecule system has immune-mo dulatory effects. RANKL (UniProtKB accession number 014788, SEQ ID NO: 11) is expressed by tumor cells in, inter alia, chronic lymphoid leukemia (CLL), multiple myeloma (MM) and acute myeloid leukemia (AML), whereas RANK can be expressed on NK cells. The interaction of RANKL expressed on malignant hematopoietic cells with RANK on NK cells was shown to inhibit anti-tumor immune-response of NK cells. The extracellular part of RANK refers to positions 25-207 of its linear amino acid sequence (SEQ ID NO: 11). The amino acid sequence of the extracellular fragment of RANK is set forth in SEQ ID NO: 46. In specific embodiments, the extracellular fragment of RANK comprised in the fusion protein of the invention comprises at least 90-183 consecutive amino acids corresponding to SEQ ID NO: 46, wherein at least 90-183 means 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181 , 182, or 183, or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 90-183 consecutive amino acids of SEQ ID NO: 46, wherein at least 90-183 means 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, or 183.

[0092] Glucocorticoid-induced tumor necrosis factor receptor (GITR, UniProtKB accession number Q9Y5U5, SEQ ID NO: 12) is another receptor expressed on NK cells. Its ligand GITRL is expressed and released inter alia by malignant cells in leukemia and solid tumors and has also been shown to impair NK cell reactivity against GITRL-expressing cells. The extracellular part of GITR refers to positions 26-162 of its linear amino acid sequence (SEQ ID NO: 12). The amino acid sequence of the extracellular fragment of GITR is set forth in SEQ ID NO: 47. In specific embodiments, the extracellular fragment of GITR comprised in the fusion protein of the invention comprises at least 50-137 consecutive amino acids corresponding to SEQ ID NO: 47, wherein at least 50-137 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136 or 137, or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50-137 consecutive amino acids of SEQ ID NO: 47, wherein at least 50-137 means 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, 60, 61 , 62, 63, 64, 65, 66, 67, 68, 69, 70, 71 , 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136 or 137.

[0093] "Percent (%) sequence identity" with respect to amino acid sequences disclosed herein is defined as the percentage of amino acid residues in a candidate sequence that are pair-wise identical with the amino acid residues in a reference sequence, i.e. a protein molecule or fragment of the present disclosure, after aligning the sequences and introducing gaps, if necessary, to achieve the maximum percent sequence identity, and not considering any conservative substitutions as part of the sequence identity. Alignment for purposes of determining percent amino acid sequence identity can be achieved in various ways that are within the skill in the art, for instance, using publically available computer software such as BLAST, ALIGN, or Megalign (DNASTAR) software. Those skilled in the art can determine appropriate parameters for measuring alignment, including any algorithms needed to achieve maximum alignment over the full length of the sequences being compared. The same is true for nucleotide sequences disclosed herein.

[0094] In some embodiments, the extracellular fragment of the transmembrane protein can be a variant that is modified in a way that it may have one or more amino acid mutations such as substitutions, insertions or deletions with reference to their natural occurring or wild type counterparts. Such modified extracellular fragments may have altered physiochemical properties compared to the corresponding natural occurring or wild type counterparts. For example, such a variant may have a higher binding affinity to at least one of its ligands or have a lower binding affinity to at least one of its ligands. In some embodiments, the variant of the extracellular fragment of the transmembrane protein may have a sequence identity of at least about 40 %, 50 %, 60 %, 70 %, 80 %, 85 %, 90 %, 95 %, 98 %, 99 %, or 99.5 % when aligned to ist natural occurring counterpart. Binding protein

[0095] The term "binding protein" generally refers to a proteinaceous binding molecule that is able to specifically bind to a given target molecule. Any proteinaceous binding molecule that is able for this specific bind can thus be used in the fusion proteins of the present invention.

[0096] The binding protein of the fusion protein of the present invention may, for example, be an immunoglobulin such as an intact (divalent) antibody or a functional "antibody fragment". Such functional antibody fragments comprise at least those parts of an antibody, that form the (antigen) binding site. Illustrative examples of such an antibody fragment are single chain variable fragments (scFv), Fv fragments, single domain antibodies, such as e.g. VHH (camelid) antibodies, di-scFvs, fragment antigen binding regions (Fab), F(ab')₂ fragments, Fab' fragments, diabodies, domain antibodies, (Holt LJ, Herring C, Jespers LS, Woolven BP, Tomlinson IM. Domain antibodies: proteins for therapy. Trends Biotechnol. 2003 Nov; 21(11):484-90), or bispecific "Fabsc"-antibody molecules as described in International patent application WO 2013/092001 comprising a single chain Fv fragment which is connected to an Fab fragment via a CH2 domain to name only a few.

[0097] The binding protein of the fusion protein of the present invention may alternatively be a proteinaceous binding molecule with antibody-like binding properties. Illustrative examples of proteinaceous binding molecules with antibody-like binding properties that can be used as binding proteins include, but are not limited to, an aptamer, a mutein based on a polypeptide of the lipocalin family, a glubody, a protein based on the ankyrin scaffold, a protein based on the crystalline scaffold, an adnectin, an avimer, a EGF- like domain, a Kringle-domain, a fibronectin type I domain, a fibronectin type II domain, a fibronectin type III domain, a PAN domain, a Gla domain, a SRCR domain, a Kunitz/Bovine pancreatic trypsin Inhibitor domain, tendamistat, a Kazal-type serine protease inhibitor domain, a Trefoil (P-type) domain, a von Willebrand factor type C domain, an Anaphylatoxin-like domain, a CUB domain, a thyroglobulin type I repeat, LDL-receptor class A domain, a Sushi domain (complement control protein (CCP) modules), a Link domain, a Thrombospondin type I domain, an immunoglobulin domain or a an immunoglobulin-like domain (for example, domain antibodies or camel heavy chain antibodies), a C-type lectin domain, a MAM domain, a von Willebrand factor type A domain, a Somatomedin B domain, a WAP -type four disulfide core domain, a F5/8 type C domain, a Hemopexin domain, an SH2 domain, an SH3 domain, a Laminin-type EGF-like domain, a C2 domain, "Kappabodies" (111 CR1, Gonzales JN, Houtz EK, Ludwig JR, Melcher ED, Hale JE, Pourmand R, Keivens VM, Myers L, Beidler K, Stuart P, Cheng S, Radhakrishnan R. Design and construction of a hybrid immunoglobulin domain with properties of both heavy and light chain variable regions. Protein Eng. 1997 Aug;10(8):949-57) "Minibodies" ( Martin Fl, Toniatti C, Salvati AL, Venturini S, Ciliberto G, Cortese R, Sollazzo M. The affinity- selection of a minibody polypeptide inhibitor of human interleukin-6. EMBO J. 1994 Nov 15;13(22):5303-9), "Janusins" (Traunecker A, Lanzavecchia A, Karjalainen K. Bispecific single chain molecules (Janusins) target cytotoxic lymphocytes on HIV infected cells. EMBO J. 1991 Dec;10(12):3655-9 and Traunecker A, Lanzavecchia A, Karjalainen K. Janusin: new molecular design for bispecific reagents. Int J Cancer Suppl. 1992;7:51-2), a nanobody, an adnectin, a tetranectin, a microbody, an affilin, an affibody or an ankyrin, a crystallin, a knottin, ubiquitin, a zinc-finger protein, an autofluorescent protein, an ankyrin or ankyrin repeat protein or a leucine-rich repeat protein, an avimer (Silverman Jl, Liu Q, Bakker A, To W, Duguay A, Alba BM, Smith R, Rivas A, Li P, Le H, Whitehorn E, Moore KW, Swimmer C, Perlroth V, Vogt M, Kolkman J, Stemmer WP. Multivalent avimer proteins evolved by exon shuffling of a family of human receptor domains. Nat Biotechnol. 2005 Dec;23(12): 1556-61. Epub 2005 Nov 20); as well as multivalent avimer proteins evolved by exon shuffling of a family of human receptor domains as also described in Silverman et al. (Silverman J, Liu Q, Bakker A, To W, Duguay A, Alba BM, Smith R, Rivas A, Li P, Le H, Whitehorn E, Moore KW, Swimmer C, Perlroth V, Vogt M, Kolkman J, Stemmer WP. Multivalent avimer proteins evolved by exon shuffling of a family of human receptor domains. Nat Biotechnol. 2005 Dec; 23(12): 1556-61. Epub 2005 Nov 20).

[0098] As indicated above, the term "antibody" generally refers to a proteinaceous binding molecule that is based on an immunoglobulin. Typical examples of such an antibody derivatives or functional fragments thereof which retain the binding specificity. Techniques for the production of antibodies and antibody fragments are well known in the art. The term "antibody" also includes immunoglobulins (Ig's) of different classes (i.e. IgA, IgG, IgM, IgD and IgE) and subclasses (such as IgGl, lgG2 etc.). As also mentioned above, illustrative examples of an antibody are Fab fragments, F(ab')₂, Fv fragments, single-chain Fv fragments (scFv), diabodies or domain antibodies (Holt LJ et al, Trends Biotechnol. 21(1 1), 2003, 484- 490). The definition of the term "antibody" thus also includes embodiments such as chimeric, single chain and humanized antibodies.

[0099] An "antibody molecule" as used herein may carry one or more domains that have a sequence with at least about 60 %, at least about 70 %, at least about 75 %, at least about 80 %, at least about 85 %, at least about 90 %, at least about 92 %, at least about 95 %, at least about 96 %, at least about 97 %, at least about 98 % or at least about 99 % sequence identity with a corresponding naturally occurring domain of an immunoglobulin M, an immunoglobulin G, an immunoglobulin A, an immunoglobulin D or an immunoglobulin E. It is noted in this regard, the term "about" or "approximately" as used herein means within a deviation of 20%, such as within a deviation of 10% or within 5% of a given value or range.

[0100] An "immunoglobulin" when used herein, is typically a tetrameric glycosylated protein composed of two light (L) chains of approximately 25 kDa each and two heavy (H) chains of approximately 50 kDa each. Two types of light chain, termed lambda and kappa, may be found in immunoglobulins. Depending on the amino acid sequence of the constant domain of heavy chains, immunoglobulins can be assigned to five major classes: A, D, E, G, and M, and several of these may be further divided into subclasses (isotypes), e.g., IgGl , lgG2, IgG3, IgG4, IgAl, and IgA2. An IgM immunoglobulin consists of 5 of the basic heterotetramer unit along with an additional polypeptide called a J chain, and contains 10 antigen binding sites, while IgA immunoglobulins contain from 2-5 of the basic 4-chain units which can polymerize to form polyvalent assemblages in combination with the J chain. In the case of IgGs, the 4-chain unit is generally about 150,000 Daltons.

[0101] In the IgG class of immunoglobulins, there are several immunoglobulin domains in the heavy chain. By "immunoglobulin (Ig) domain" herein is meant a region of an immunoglobulin having a distinct tertiary structure. In the context of IgG antibodies, the IgG isotypes each have three CH regions: "CHI" refers to positions 118-220, "CH2" refers to positions 237-340, and "CH3" refers to positions 341-447 according to the EU index as in Kabat. By "hinge" or "hinge region" or "antibody hinge region" or "immunoglobulin hinge region" or "H" herein is meant the flexible polypeptide comprising the amino acids between the first and second constant domains of an antibody. Structurally, the IgG CHI domain ends at EU position 220, and the IgG CH2 domain begins at residue EU position 237. Thus for IgG the hinge is herein defined to include positions 221 (D221 in IgGl) to 236 (G236 in IgGl), wherein the numbering is according to the EU index as in Kabat. In some embodiments, the amino acid at position 220 may be assigned to the hinge region instead of the CHI domain. The constant heavy chain, as defined herein, refers to the N-terminus of the CHI domain to the C-terminus of the CH3 domain, thus comprising positions 118-447, wherein numbering is according to the EU index.

[0102] An "antibody molecule with a single binding site" according to the invention is an antibody molecule or antibody fragment that comprises only a single antigen binding site. The antigen binding site is usually formed by a variable domain of an antibody or parts thereof. Examples for such antibody fragments are immunoglobulin fragments such as Fab fragments, Fv fragments, single-chain Fv fragments (scFv), heavy chain antibodies, single domain antibodies, camel antibodies, V_H, or V_L. Such fragments can also be of chimeric antibodies or humanized antibodies, or fragments of proteins that are not derived from immunoglobulins but have antibody-like structural or functional characteristics such as an aptamer or an anticalin. An "antibody molecule" comprised in the invention is preferably an immunoglobulin molecule or fragment with a single binding site, preferably a Fab fragment, a scFv fragment, or a single domain antibody.

[0103] The term "variable" refers to the portions of the immunoglobulin domains that exhibit variability in their sequence and that are involved in determining the specificity and binding affinity of a particular antibody (i.e., the "variable domain(s)"). Variability is not evenly distributed throughout the variable domains of antibodies; it is concentrated in sub- domains of each of the heavy and light chain variable regions. These sub-domains are called "hypervariable regions", "HVR," or "HV," or "complementarity determining regions" (CDRs). The more conserved (i.e., non-hypervariable) portions of the variable domains are called the "framework" regions (FR). The variable domains of naturally occurring heavy and light chains each include four FR regions, largely adopting a β-sheet configuration, connected by three hypervariable regions, which form loops connecting, and in some cases forming part of, the β-sheet structure. The hypervariable regions in each chain are held together in close proximity by the FR and, with the hypervariable regions from the other chain, contribute to the formation of the antigen- binding site (see Kabat et al., see below). Generally, naturally occurring immunoglobulins include six CDRs (see below); three in the VH (HI, H2, H3), and three in the VL (LI, L2, L3). In naturally occurring immunoglobulins, H3 and L3 display the most diversity of the six CDRs, and H3 in particular is believed to play a unique role in conferring fine specificity to immunoglobulins. The constant domains are not directly involved in antigen binding, but exhibit various effector functions, such as, for example, antibody- dependent, cell-mediated cytotoxicity and complement activation.

[0104] The terms "V_H" (also referred to as VH) and "V_L" (also referred to as VL) are used herein to refer to the heavy chain variable domain and light chain variable domain respectively of an immunoglobulin. An immunoglobulin light or heavy chain variable region consists of a "framework" region interrupted by three hypervariable regions. Thus, the term "hypervariable region" refers to the amino acid residues of an antibody which are responsible for antigen binding. The hypervariable region includes amino acid residues from a "Complementarity Determining Region" or "CDR". There are three heavy chains and three light chain CDRs (or CDR regions) in the variable portion of an immunoglobulin. Thus, "CDRs" as used herein refers to all three heavy chain CDRs (CDRH1, CDRH2 and CDRH3), or all three light chain CDRs (CDRL1, CDRL2 and CDRL3) or both all heavy and all light chain CDRs, if appropriate. Three CDRs make up the binding character of a light chain variable region and three make up the binding character of a heavy chain variable region. CDRs determine the antigen specificity of an immunoglobulin molecule and are separated by amino acid sequences that include scaffolding or framework regions. The exact definitional CDR boundaries and lengths are subject to different classification and numbering systems. The structure and protein folding of the antibody may mean that other residues are considered part of the antigen binding region and would be understood to be so by a skilled person. CDRs provide the majority of contact residues for the binding of the immunoglobulin to the antigen or epitope.

[0105] "Framework Region" or "FR" residues are those variable domain residues other than the hypervariable region. The sequences of the framework regions of different light or heavy chains are relatively conserved within a species. Thus, a "human framework region" is a framework region that is substantially identical (about 85% or more, usually 90-95% or more) to the framework region of a naturally occurring human immunoglobulin. The framework region of an antibody, that is the combined framework regions of the constituent light and heavy chains, serves to position and align the CDR's. The CDR's are primarily responsible for binding to an epitope of an antigen.

[0106] The terms "Fab", "Fab region", "Fab portion" or "Fab fragment" are understood to define a polypeptide that includes a V_H, a C_HI , a V_L, and a C_L immunoglobulin domain. Fab may refer to this region in isolation, or this region in the context of an antibody molecule, as well as a full length immunoglobulin or immunoglobulin fragment. Typically a Fab region contains an entire light chain of an antibody. A Fab region can be taken to define "an arm" of an immunoglobulin molecule. It contains the epitope-binding portion of that Ig. The Fab region of a naturally occurring immunoglobulin can be obtained as a proteolytic fragment by a papain-digestion. A "F(ab')₂ portion" is the proteolytic fragment of a pepsin- digested immunoglobulin. A "Fab' portion" is the product resulting from reducing the disulfide bonds of an F(ab')₂ portion. As used herein the terms "Fab", "Fab region", "Fab portion" or "Fab fragment" may further include a hinge region that defines the C-terminal end of the antibody arm (cf. above). This hinge region corresponds to the hinge region found C- terminally of the CHI domain within a full length immunoglobulin at which the arms of the antibody molecule can be taken to define a Y. The term hinge region is used in the art because an immunoglobulin has some flexibility at this region. A "Fab heavy chain" as used herein is understood as that portion or polypeptide of the Fab fragment that comprises a V_H and a C_HI , whereas a "Fab light chain" as used herein is understood as that portion or polypeptide of the Fab fragment that comprises a V_L, and a C_L.

[0107] The term "Fc region" or "Fc fragment" is used herein to define a C-terminal region of an immunoglobulin heavy chain, including native- sequence Fc regions and variant Fc regions. The Fc part mediates the effector function of antibodies, e.g. the activation of the complement system and of Fc-receptor bearing immune effector cells, such as NK cells. In human IgG molecules, the Fc region is generated by papain cleavage N-terminal to Cys226. Although the boundaries of the Fc region of an immunoglobulin heavy chain might vary, the human IgG heavy-chain Fc region is usually defined to stretch from an amino acid residue at position Cys226, or from Pro230, to the carboxyl-terminus thereof. The C-terminal lysine (residue 447 according to the EU numbering system) of the Fc region may be removed, for example, during production or purification of the antibody molecule, or by recombinantly engineering the nucleic acid encoding a heavy chain of the antibody molecule. Native- sequence Fc regions include mammalian, e.g. human or murine, IgGl, IgG2 (IgG2A, IgG2B), IgG3 and IgG4. The Fc region contains two or three constant domains, depending on the class of the antibody. In embodiments where the immunoglobulin is an IgG the Fc region has a CH2 and a CH3 domain.

[0108] The term "single-chain variable fragment" (scFv) is used herein to define an antibody fragment, in which the variable regions of the heavy (VH) and light chains (VL) of a immunoglobulin are fused together, which are connected with a short linker peptide of ten to about 25 amino acids. The linker is usually rich in glycine for flexibility, as well as serine or threonine for solubility, and can either connect the N-terminus of the VH with the C-terminus of the VL, or connect the N-terminus of the VL with the C-terminus of the VH. The scFv fragment retains a specific antigen binding site but lacks constant domains of immunoglobulins.

[0109] An antibody or antibody molecule/fragment or receptor is said to specifically bind or has specific binding affinity to an antigen or ligand when it recognizes its target antigen of ligand in a complex mixture of proteins and/or macromolecules. Antibodies are said to "bind to the same epitope" if the antibodies cross-compete so that only one antibody can bind to the epitope at a given point of time, i.e. one antibody prevents the binding or modulating effect of the other. [0110] The term "specific" in this context, or "specifically binding", also used as "directed to", means in accordance with this invention that the antibody or immune receptor fragment is capable of specifically interacting with and/or binding to a specific antigen or ligand or a set of specific antigens or ligands but does not essentially bind to other antigens or ligands. Such binding may be exemplified by the specificity of a "lock-and-key-principle".

[0111] Typically, binding is considered specific when the binding affinity is higher than 10^~6 M. In particular, binding is considered specific when binding affinity is about 10^~8 to 10^"11 M (KD), or of about 10^"9 to 10^"11 M or even higher. If necessary, nonspecific binding of a binding site can be reduced without substantially affecting specific binding by varying the binding conditions.

[0112] The antibody fragments comprised in the fusion proteins of the present invention specifically binds to receptors on immune cells such as T-cells or NK cells, preferably to receptors capable of activating the immune cell or of stimulating an immune response of the immune cell, in which the immune response is preferably a cytotoxic immune response. In some embodiments, such a receptor may be CD 16, CD2, Ly49m, NCR, CD94:NKG2, TCR, TCRa, TCRp, CD3, CD35/8, CD3y/8, CD247, CD28, CD134, 4-1BB, CD5, or CD95. In more specific embodiments, such a receptor may be CD3 or CD 16. The antibody molecule herein can by any antibody molecule, preferably with a single antigen binding site, as previously defined. In preferred embodiments, such an antibody molecule is a Fab fragment or a scFv fragment or a single domain antibody.

[0113] In some embodiments one of the binding sites of a fusion protein according to the invention is able to bind a T-cell receptor molecule and/or a natural killer cell (NK cell) receptor molecule, preferably a T-cell specific receptor molecule and/or a natural killer cell (NK cell) specific receptor molecule. An example for a T-cell specific receptor is the so called "T-cell receptor" (TCRs), which allows a T cell to bind to and, if additional signals are present, to be activated by and respond to an epitope/antigen presented by another cell called the antigen-presenting cell or APC. The T cell receptor is known to resemble a Fab fragment of a naturally occurring immunoglobulin. It is generally monovalent, encompassing a- and β- chains, in some embodiments it encompasses γ-chains and δ-chains (supra). Accordingly, in some embodiments the TCR is TCR (alpha/beta) and in some embodiments it is TCR (gamma/delta). The T cell receptor forms a complex with the CD3 T-Cell co-receptor. CD3 is a protein complex and is composed of four distinct chains. In mammals, the complex contains a CD3y chain, a CD35 chain, and two CD3s chains. These chains associate with a molecule known as the T cell receptor (TCR) and the ζ-chain to generate an activation signal in T lymphocytes. Hence, in some embodiments a T-cell specific receptor is the CD3 T-Cell co- receptor. In some embodiments a T-cell specific receptor is CD28, a protein that is also expressed on T cells. CD28 can provide co-stimulatory signals, which are required for T cell activation. CD28 plays important roles in T-cell proliferation and survival, cytokine production, and T-helper type-2 development. Yet a further example of a T-cell specific receptor is CD 134, also termed Ox40. CD134/OX40 is being expressed after 24 to 72 hours following activation and can be taken to define a secondary co-stimulatory molecule. Another example of a T-cell receptor is 4-1BB capable of binding to 4-lBB-Ligand on antigen presenting cells (APCs), whereby a co-stimulatory signal for the T cell is generated. Another example of a receptor predominantly found on T-cells is CD5, which is also found on B cells at low levels. A further example of a receptor modifying T cell functions is CD95, also known as the Fas receptor, which mediates apoptotic signaling by Fas-ligand expressed on the surface of other cells. CD95 has been reported to modulate TCR/CD3 -driven signaling pathways in resting T lymphocytes.

[0114] An example of a NK cell specific receptor molecule is CD 16, a low affinity Fc receptor. An example of a receptor molecule that is present on the surface of both T cells and natural killer (NK) cells is CD2 and further members of the CD2-superfamily. CD2 is able to act as a co-stimulatory molecule on T and NK cells.

[0115] A "cytotoxic cell", as used herein refers to an immune cell that is capable of killing other cells, for instance cancer cells, cells that are infected (particularly with viruses) or otherwise damaged cells. Such killing is for instance performed by inducing lysis or apoptosis, for example through release of cytotoxins such as perforin, granzymes or granulysin. Such a cytotoxic cell as used herein may be a cytotoxic T-cell (also known as T_c, cytotoxic T lymphocyte, CTL, T-killer cell, cytolytic T cell, CD8+ T-cells or killer T cell), a natural killer (NK) cell or a natural killer T-cell (NKT cell). In more specific embodiments, such a cytotoxic cell is a cytotoxic T-cell or an NK cell. Another type of T cells are T helper cells, also referred to as CD4+ T cells or CD4 T cells. T helper cells help the activity of other immune cells by releasing T cell cytokines. These cells help suppress or regulate immune responses. They are essential in B cell antibody class switching, in the activation and growth of cytotoxic T cells, and in maximizing bactericidal activity of phagocytes such as macrophages.

[0116] Preferred embodiments of the fusion protein of invention comprise binding protein moieties, preferably antibody molecules, that specifically binds to CD3 CD 16, CD28, CD137/4-1BB, OX40, Nkp44, Nkp30, Nkp40 or Nkp46. Specific embodiments of the fusion protein comprises a binding protein moiety that is an antibody molecule. Said antibody molecule preferably binds to CD3 or CD 16. For instance, such a CD3 binding antibody molecule (also referred to as "anti-CD3 antibody molecule) can be a Fab fragment ("anti-CD3 Fab") or a scFv fragment ("anti-CD3 scFv") or a single domain antibody ("anti-CD3 single domain antibody"). A non-limiting example for such an anti-CD3 Fab comprises a Fab heavy chain sequence set forth in SEQ ID NO: 06 and a Fab light chain sequence set forth in SEQ ID NO: 07. A non-limiting example for an anti-CD3 scFv comprises a sequence set forth in SEQ ID NO: 04. Turning to CD16 binding antibody molecules ("anti-CD16 antibody molecules"), such a molecule can likewise be a Fab fragment ("anti-CD 16 Fab") or a scFv fragment ("anti-CD 16 scFv") or a single domain antibody ("anti-CD 16 single domain antibody"). A non- limiting example for such an anti-CD 16 Fab comprises a Fab heavy chain sequence set forth in SEQ ID NO: 08 and a Fab light chain sequence set forth in SEQ ID NO: 09. A non-limiting example for an anti-CD3 scFv comprises a sequence set forth in SEQ ID NO: 05.

[0117] In preferred embodiments, the binding protein or antibody molecule may bind to a receptor that is mainly expressed on activated immune cells such as activated T cells or NK cells. Targeting an activated immune cell may have the advantage, that immune cells that are already activated will be engaged by the fusion proteins of the invention. Hence, a stronger immune effect such as a stronger cytotoxic effect of the engaged immune cell may be mediated. As an illustrative example, the binding protein or antibody molecule may bind to NKp44, which is mainly expressed on activated immune cells such as activated NK cells. Here, targeting NKp44 will result in a specific targeting of pre-activated NK cells which may therefore mediate a stronger cytotoxic effect than non-preactivated NK cells. Following the same concept, the binding protein or antibody molecule may bind to CD 137, which is mainly expressed on activated immune cells such as activated T cells or NK cells. Targeting CD 137 will also result in mainly engaging pre-activated T cells or NK cells which may therefore mediate a stronger cytotoxic effect than non-preactivated T cells or NK cells. Another advantage of targeting such receptors (e.g. CD 137) is, that immune cells are engaged, which are activated for exerting immune responses directed to malignant cells. Consequently, the immune response mediated by the fusion proteins can be focused on malignant cells and side effects can be minimized. Such a fusion protein may also be employed in sequential therapeutic strategies. As an illustrative example, a tumor-targeting antibody or fusion protein can be combined with a second fusion protein or antibody that activates the host immune system, e.g. by targeting CD 137, similar to the approaches described by Khort et al. 2001, Khort et al. 2012, and Khort et al. 2014. It is further envisioned that the fusion proteins described herein can also be combined with other immunotherapeutic approaches, for example vaccination or cytokine therapy.

[0118] In some embodiments, the binding protein does not itself activate the immune cell, e.g. NK cell or T-cell, upon binding, such as binding to one of its receptors. Instead, only when both the portions of the fusion protein are bound to the receptor on the immune cell and to the antigen or ligand on the target cell, the former will cross-link the activating receptor, triggering the effector cells to kill the specific target cell. In some embodiments, the binding protein activates the receptor on the immune cell upon binding. Standard functional assays to evaluate the target cell -killing capability by lymphocytes in the presence and absence of an binding protein moiety or fusion protein can be set up to assess and/or screen for the ability of the binding protein moiety to activate the receptor to which it binds.

Linking polypeptide

[0119] In the fusion protein of the present invention, the binding protein and the extracellular fragment of the immune receptor are linked via a linking polypeptide. In specific embodiments, this linking polypeptide consists of at least a portion of a CH2 domain and optionally a hinge region. In other specific embodiments, the linking polypeptide comprises at least a portion or a fragment of an Fc domain. In further specific embodiments, the recombinant fusion protein does however not comprise (as a linking polypeptide) an immunoglobulin heavy chain comprising VH-H-CH 1 -CH2-CH3 or VH-H-CH1-CH2 CH3- CH4. In preferred embodiments, the linking peptide, consists of a CH2 domain and optionally a hinge region. In the context of the linking polypeptide, the term "consists of is to be understood such that the functionally essential components of the linking polypeptide are a CH2 domain and a hinge region. Here, the CH2 domain may be a functional fragment of the CH2 domain, meaning that it has at least about 80 %, 85 %, 90 %, 95 %, 98 % or 99 %, of the amino acids as a full length CH2 domain. Also within the meaning of "consists of in this context is that the linking polypeptide can comprise linking amino acids in between the CH2 domain and the hinge region, or N-terminal of the hinge region or C-terminal of the CH2 domain. Typically, these linking amino acids do not exceed a number of about 30, about 25, about 20, about 15, about 10, about 5, about 3, about 2, or about 1 amino acid(s). Also typically, these linking amino acids do not form a structural domain, in the meaning that it does not form an independently stable and folded structure. They can rather be a "spacer" or the "leftovers" of the cloning strategy used for the construction of the nucleic acid molecule(s) that encode a fusion protein of the invention. As an illustrative example, a linking polypeptide having a hinge region, a CH2 domain, and e.g. 15 amino acids derived from a CH3 domain appended C-terminal to the CH2 region is still within the meaning of a polypeptide consisting of a CH2 region and a hinge region. In the context of the fusion protein itself, the term "consists of is understood that the functional entities of the core structure of the fusion protein consists of the functional entities of (a) a binding protein with a binding site that specifically binds to a receptor specific for T cells or natural killer cells, (b) an extracellular fragment of a transmembrane protein that is an immune receptor, and (c) a linking polypeptide. In this context, it is within the meaning of "consist of, that the fusion protein can, apart from the functional entities (a), (b), and (c), further comprise linking amino acids between the functional entities (a), (b), or (c). Typically these linking amino acids do not exceed a number of about 30, about 25, about 20, about 15, about 10, about 5, about 3, about 2, or about 1 amino acid(s). Also typically, these linking amino acids do not form a structural domain, in the meaning that they do not form an independently stable and folded structure. In this context, it is also within the meaning of "consists of that the core structure can be fused to another peptide or polypeptide, for example, an affinity tag for purification, or can be conjugated to another compound, for example a drug or an imaging agent, or a serum half-life extending moiety, such as polyethylene glycol. The linking polypeptide is fused in between the binding protein and the immune receptor fragment, connecting both fragments with each other. Herein, the binding protein can be at the N-terminus of the fusion protein and the immune receptor fragment can be at the C-terminus of the fusion protein or vice versa. In preferred embodiments, the binding protein is an antibody molecule and is at the N-terminus of the fusion protein and an immune receptor fragment derived from a type II transmembrane protein is at the C-terminus of the fusion protein. In such a case, the antibody molecule is preferably a Fab fragment, a scFv fragment or a single domain antibody. In other preferred embodiments, the binding protein is an antibody molecule and is at the C-terminus of the fusion protein and an immune receptor fragment derived from a type I transmembrane protein is at the N-terminus of the fusion protein. In such a case, the antibody molecule is preferably a scFv fragment or a single domain antibody. In some embodiments, the linking polypeptide forms a structural domain of the fusion protein, which means that the linking polypeptide forms a compact three-dimensional structure and can be independently stable and folded, meaning that the linking polypeptide can have an own tertiary structure. Moreover, besides having an own tertiary structure, the linking polypeptide may also have particular functional properties. In other embodiments, the linking polypeptide might merely define a "scaffold" without providing a particular biological function.

[0120] Compared to a polyglycine linker as disclosed by WO 2011/085178, a linking polypeptide of the invention comprising a portion of a Fc fragment, such as a CH2 and/or an CH3 domain, or consisting of an at least a portion of a CH2 domain and optionally a hinge region, has a series of advantages. The fusion proteins of the invention comprising such a linking polypeptide have improved stability and extended serum half lives due to higher molecular weights, have superior expression and production rates, a lower tendency to for aggregations. In addition, purification is facilitated as established methods for purification of polypeptides comprising Fc portions exist, such as affinity chromatography methods targeting CH2 and/or CH3. Furthermore, in case a fusion protein comprises a CH3 domain, dimerization of this fusion protein by spontaneous dimerization and formation of disulfide bonds can be considered. This may provide the possibility to further improve stability and serum half life and reduce tendency for aggregation.

[0121] In specific embodiments, the linking polypeptide consists of at least a portion of a CH2 domain, preferably a CH2 domain and a hinge region. Said linking polypeptide preferably comprises at least a portion of a CH2 domain, which comprises at least a portion of positions 216-340 of the linear sequence of human IgG according to Kabat numbering (EU- index). In another preferred embodiment, the linking polypeptide comprises at least 50-125 consecutive amino acids corresponding to positions 216-340 (EU-index) of human IgG as set forth in SEQ ID NO: 01 , wherein at least 50-125 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 , 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, or 125. In another preferred embodiment, the linking polypeptide comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50-125 consecutive amino acids of SEQ ID NO: 01 , wherein at least 50-125 means at least 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, or 125.

[0122] In more specific embodiments, the CH2 domain comprised in the linking polypeptide comprises at least one mutation or deletion in at least one amino acid residue of the CH2 that mediates binding of FcR, preferably wherein said mutation or deletion reduces binding affinity of CH2 with FcR. Such "function-depleted" Fc portions, either through having mutated CH2 moieties or through not having CH2 moieties have the advantage, that the risk of a systemic T-cell activation due to binding of Fc portions to Fc receptors is lowered.

[0123] The terms "mutated", "mutant" and "mutation" in reference to a nucleic acid or a polypeptide refers to the exchange, deletion, or insertion of one or more nucleotides or amino acids, respectively, compared to the naturally occurring nucleic acid or polypeptide, i.e. to a reference sequence that can be taken to define the wild-type.

[0124] It is understood in this regard that the term "position", when used in accordance with the present invention, means the position of an amino acid within an amino acid sequence depicted herein. This position may be indicated relative to a resembling native sequence, e.g. a sequence of a naturally occurring IgG domain or chain. The term "corresponding" as used herein also includes that a position is not necessarily, or not only, determined by the number of the preceding nucleotides/amino acids. Thus, the position of a given amino acid in accordance with the present invention which may be substituted may vary due to deletion or addition of amino acids elsewhere in the antibody chain.

[0125] Thus, under a "corresponding position" in accordance with the present invention it is to be understood that amino acids may differ in the indicated number but may still have similar neighbouring amino acids. Said amino acids which may be exchanged, deleted or added are also encompassed by the term "corresponding position". In order to determine whether an amino acid residue in a given amino acid sequence corresponds to a certain position in the amino acid sequence of a naturally occurring immunoglobuline domain or chain, the skilled person can use means and methods well-known in the art, e.g., alignments, either manually or by using computer programs such as BLAST2.0, which stands for Basic Local Alignment Search Tool or ClustalW or any other suitable program which is suitable to generate sequence alignments.

[0126] In some embodiments a substitution (or replacement) is a conservative substitution. Conservative substitutions are generally the following substitutions, listed according to the amino acid to be mutated, each followed by one or more replacement(s) that can be taken to be conservative: Ala— Gly, Ser, Val; Arg— Lys; Asn— Gin, His; Asp— Glu; Cys→ Ser; Gin→ Asn; Glu→ Asp; Gly→ Ala; His→ Arg, Asn, Gin; He→ Leu, Val; Leu→ He, Val; Lys→ Arg, Gin, Glu; Met→ Leu, Tyr, He; Phe→ Met, Leu, Tyr; Ser→ Thr; Thr— Ser; Trp— Tyr; Tyr— Trp, Phe; Val— He, Leu. Other substitutions are also permissible and can be determined empirically or in accord with other known conservative or non-conservative substitutions. As a further orientation, the following eight groups each contain amino acids that can typically be taken to define conservative substitutions for one another:

i. Alanine (Ala), Glycine (Gly);

ii. Aspartic acid (Asp), Glutamic acid (Glu);

iii. Asparagine (Asn), Glutamine (Gin);

iv. Arginine (Arg), Lysine (Lys);

v. Isoleucine (He), Leucine (Leu), Methionine (Met), Valine (Val);

vi. Phenylalanine (Phe), Tyrosine (Tyr), Tryptophan (Trp);

vii. Serine (Ser), Threonine (Thr); and

viii. Cysteine (Cys), Methionine (Met)

[0127] If such substitutions result in a change in biological activity, then more substantial changes, such as the following, or as further described below in reference to amino acid classes, may be introduced and the products screened for a desired characteristic. Examples of such more substantial changes are: Ala— Leu, He; Arg— Gin; Asn— Asp, Lys, Arg, His; Asp→ Asn; Cys→ Ala; Gin→ Glu; Glu→ Gin; His→ Lys; He→ Met, Ala, Phe; Leu→ Ala, Met, Norleucine; Lys→ Asn; Met→ Phe; Phe→ Val, He, Ala; Trp→ Phe; Tyr → Thr, Ser; Val→ Met, Phe, Ala.

[0128] In some embodiments, the portion of a CH2 domain according to the invention includes one or more amino acid residues, including two, three, four, five, six, seven, eight, nine, ten, eleven, twelve, thirteen, fourteen, fifteen, sixteen, seventeen or eighteen amino acid residues that are mutated or lacking to modulate Fc-function. In some of these embodiments one or more amino acid residue(s) of the CH2 domain is able to mediate binding to Fc receptors are mutated or lacking. If present, the one or more amino acid residue(s) able to mediate binding to Fc receptors may be an amino acid residue that is able to activate antibody dependent cellular cytotoxicity (ADCC). In some embodiments a respective amino acid residue capable of mediating binding to Fc receptors is substituted by another amino acid, generally when comparing the sequence to the sequence of a corresponding naturally occurring domain in an immunoglobulin, such as an IgG. In some embodiments such an amino acid residue capable of mediating binding to Fc receptors is deleted, generally relative to the sequence of a corresponding naturally occurring domain in an immunoglobulin, such as an IgG.

[0129] In some embodiments, the linking polypeptide consists of a hinge region and a CH2 domain. In such an embodiment, one or more amino acids in the hinge region and/or CH2 domain that enable(s) dimerization or multimerization of the antibody molecule via disulphide bond formation can be mutated or deleted.

[0130] In some embodiments, at least one amino acid residue of the hinge region or the CH2 domain that is able to mediate binding to Fc receptors or mediate disulphide bond formation is lacking or mutated e.g. substituted or deleted. Such amino acid residue(s) can be an amino acid located at one of the positions 226, 228, 229, 230, 231, 232, 233, 234, 235, 236, 237, 238, 265, 297, 327, and 330. In other embodiments, the at least one amino acid residue of the hinge region and/or CH2 domain that is able to mediate binding to Fc receptors or mediate disulphide bond formation and that is lacking or mutated is/are the amino acid residues selected from the group consisting of sequence position 230, 231, 232, 233, 234, 235, 236, 237, 238, 265, 297, 327, and 330 (numbering of sequence positions according to the EU- index). Further one or more amino acid residues of sequence positions 226, 228 and 229, can also be lacking or mutated.

[0131] Again, the numbering of amino acids used corresponds to the sequence positions according to the Kabat numbering [EU-Index]. A corresponding deletion of an amino acid may for example be a deletion of amino acid 228, generally a proline in IgG, a deletion of amino acid 229, generally a cysteine in IgG, a deletion of amino acid 230, generally a proline in IgG, a deletion of amino acid 231, generally an alanine in IgG, a deletion of amino acid 232, generally a proline in IgG, a deletion of amino acid 233, generally a glutamic acid in IgG, a deletion of amino acid 234, generally a leucine in IgG, a deletion of amino acid 235, generally a leucine in IgG, a deletion of amino acid 236, generally a glycine in IgG, a deletion of amino acid 237, generally a glycine in IgG, a deletion of amino acid 238, generally a proline in IgG and a deletion of amino acid 265, generally an aspartic acid in IgG. A corresponding substitution of an amino acid may for example be a substitution of amino acid 220, generally a cysteine in IgG, a substitution of amino acid 226, generally a cysteine in IgG, a substitution of amino acid 228, generally a proline in IgG, a substitution of amino acid 229, generally a cysteine in IgG, a substitution of amino acid 230, generally a proline in IgG, a substitution of amino acid 231, generally an alanine in IgG, a substitution of amino acid 232, generally a proline in IgG, a substitution of amino acid 233, generally a glutamic acid in IgG, a substitution of amino acid 234, generally a leucine in IgG, a substitution of amino acid 235, generally a leucine in IgG, a substitution of amino acid 265, generally an aspartic acid in IgG, a substitution of amino acid 297, generally an asparagine in IgG, a substitution of amino acid 327, generally an alanine in IgG, and a substitution of amino acid 330, generally an alanine in IgG. A respective substitution may be one of substitution Cys220— >Ser, of substitution Cys226— >Ser, substitution Cys229— >Ser, substitution Glu233— >Pro, substitution Leu234— >Val, substitution Leu235— >Ala, substitution Asp265— >Gly, substitution Asn297— >Gln, substitution Ala327— >Gln, substitution Ala327— >Gly, and substitution Ala330— »Ser. As can be taken from the above, in some embodiments one or two of the cysteine residues at positions 226 and 229 in the hinge region are being substituted for another amino acid, for instance substituted for a serine residue. Thereby the formation of a disulphide bond with another main chain can be prevented. Further, and as also explained below, deleting and/ or substituting (mutating) selected amino acid residues in the hinge region and/or the CH2 domain that is able to mediate binding to Fc-receptors can cause a fusion protein of the invention to have less or no activity in terms of antibody-dependent cell- mediated cytotoxicity and fixation of complement.

[0132] In specific embodiments of the present invention, the linking polypeptide comprises at least 50-125 consecutive amino acids of SEQ ID NO: 02 or of SEQ ID NO: 03, wherein at least 50-125 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, or 125. In another preferred embodiment, the linking polypeptide comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50-125 consecutive amino acids of o SEQ ID NO: 02 or of SEQ ID NO: 03, wherein at least 50-125 means at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71 , 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, or 125.

[0133] Another type of amino acid variant of an antibody alters the original glycosylation pattern (if any) of the antibody molecule. By altering is meant deleting one or more carbohydrate moieties found in the antibody, and/or adding one or more glycosylation sites that are not present in the antibody. Glycosylation of antibodies is typically either N- linked or O-linked. N-linked refers to the attachment of the carbohydrate moiety to the side chain of an asparagine residue. The tripeptide sequences asparagine-X-serine and asparagine- X-threonine, where X is any amino acid except proline, are the recognition sequences for enzymatic attachment of the carbohydrate moiety to the asparagine side chain. Thus, the presence of either of these tripeptide sequences in a polypeptide creates a potential glycosylation site. O-linked glycosylation refers to the attachment of one of the sugars N- aceylgalactosamine, galactose, or xylose to a hydroxyamino acid, most commonly serine or threonine, although 5-hydroxyproline or 5 -hydroxy lysine may also be used. Addition of glycosylation sites to the antibody is conveniently accomplished by altering the amino acid sequence such that it contains one or more of the above-described tripeptide sequences (for N- linked glycosylation sites). The alteration may also be made by the addition of, or substitution by, one or more serine or threonine residues to the sequence of the original antibody (for O- linked glycosylation sites).

[0134] In the context of the present invention in some embodiments the linking polypeptide is typically inert, or at least essentially of low influence, with regard to binding to Fc receptors. As said, this is achieved by deleting and/or substituting (mutating) at least one of selected amino acid residues in the hinge region and/or CH2 domain that are able to mediate binding to an Fc-receptor. Such molecules are also referred to herein as "Fc-attenuated" or "function-depleted" CH2 domains or molecules. "Function-depleted" CH2 variants are desirable because binding of the fusion proteins to FcR expressing cells and formation of a functional CD3 or CD 16 binding site on the surface of these cells should be prevented.

[0135] In some embodiments the recognition, and accordingly binding, of this Fc- corresponding portion to a given Fc receptor is of about 2-fold, about 5-fold, about 8-fold, about 10-fold, about 12-fold, about 15-fold, about 20-fold or lower than the Fc domain region of a naturally occurring immunoglobulin. In some embodiments this Fc-corresponding portion is entirely void of its ability of binding to Fc receptors. The binding of a fusion protein of the invention to Fc receptors, including determining a dissociation constant, can easily be determined by the skilled artisan using standard techniques such as surface plasmon resonance, e.g. using a Biacore™ measurement. Any other method of measuring binding may likewise be used, which may for instance rely on spectroscopic, photochemical, photometric or radiological means. Examples for the corresponding detection methods are fluorescence correlation spectroscopy, photochemical cross-linking and the use of photoactive or radioactive labels respectively. Some of these methods may include additional separation techniques such as electrophoresis or HPLC.

[0136] Where required, a substitution or deletion of amino acid residues, as explained above, may be carried out to this effect. Suitable mutations can be taken from Armour et al. (Eur. J. Immunol. [1999] 29, 2613-2624), for example. Further suitable positions for mutations to a sequence of an antibody chain can be taken from the crystal structure data published on the complex between FcyRIII and the human IgGl Fc fragment (Sondermann et al, Nature [2000] 406, 267-273). In addition to measuring the binding affinity as described above in order to assess the level of "Fc attenuation" or loss of binding affinity, it is also possible to functionally assess the (lack of the) ability to mediate binding to an Fc-receptor. In the case of antibody molecules which bind CD3 as one target, it is for example possible to assess the binding through the mitogenity of such CD3 binding antibody molecules on cells. The mitogenity is mediated by binding of CD3 antibodies to the Fc-receptors on accessory cells, such as monocytes. If a Fc-modified antibody molecule that has one binding site for CD3 does not show any mitogenic effect whereas the parent monoclonal anti-CD3 antibody that has a functional Fc part induces strong mitosis in T cells, it is clear that, due to the lack of mitosis, the Fc-modified antibody molecule lacks the ability for Fc binding and can thus be considered as a "Fc knock-out" molecule. A fusion protein of the invention, having corresponding amino acid modifications in its linking polypeptide as the above-discussed Fc- modified antibody molecule, will hence have the same "Fc knock-out". Illustrative examples of a method of assessing anti-CD3 mediated mitogenity have been described by Davis, Vida & Lipsky (J.Immunol (1986) 137, 3758), and by Ceuppens, JL, & van Vaeck, F, (see J.Immunol. (1987) 139, 4067, or Cell. Immunol. (1989) 118, 136). Further illustrative suitable examples of an assay for assessing mitogenity of an antibody have been described by Rosenthal-Allieri et al. (Rosenthal-Allieri MA, Ticcioni M, Deckert M, Breittmeyer JP, Rochet N, Rouleaux M, and Senik A, Bernerd A, Cell Immunol. 1995 163(l):88-95) and Grosse-Hovest et al. (Grosse-Hovest L, Hartlapp I, Marwan W, Brem G, Rammensee H-G, and Jung G, Eur J Immunol. [2003] May; 33(5): 1334-1340). In addition, the lack of Fc binding can be assessed by the ability of an fusion protein of the invention to mediate one or more of the well-known effector functions of the Fc part.

[0137] As noted above, substitutions or deletions of cysteine residues may be carried out in order to introduce or to remove one or more disulphide bonds, including removing a potential or a previously existing disulphide bond. Thereby linkage between a main chain and a chain of lower weight/shorter length of an antibody molecule according to the invention may be controlled including established, strengthened or abolished. By removing one or more cysteine residues a disulphide bridge may be removed. One such disulphide bond is typically defined by a cysteine in the main chain of a first antibody molecule and a cysteine in the hinge region of a second antibody molecule. In this regard, in some embodiments an antibody according to the invention may include an amino acid substitution of a native cysteine residue at positions 220, 226 and/or 229, relative to the sequence of a human IgG immunoglobulin according to the Kabat numbering [EU-Index], by another amino acid residue. [0138] Substitutions or deletions of amino acid residues such as arginine, asparagine, serine, threonine or tyrosine residues may also be carried out to modify the glycosylation pattern of an antibody. As an illustrative example, an IgG molecule has a single N-linked biantennary carbohydrate at Asn297 of the CH2 domain. For IgG from either serum or produced ex vivo in hybridomas or engineered cells, the IgG are heterogeneous with respect to the Asn297 linked carbohydrate. For human IgG, the core oligosaccharide typically consists of GlcNAc₂Man₃GlcNAc, with differing numbers of outer residues.

[0139] As indicated, besides binding of antigens/epitopes, an immunoglobulin is known to have further "effector functions", biological activities attributable to the Fc region (a native sequence Fc region or amino acid sequence variant Fc region) of an immunoglobulin, and vary with the immunoglobulin isotype. Examples of antibody effector functions include: Clq binding and complement dependent cytotoxicity (CDC); Fc receptor binding; antibody- dependent cell-mediated cytotoxicity (ADCC); phagocytosis; down regulation of cell surface receptors (e.g., B cell receptors); and B cell activation. Exerting effector functions of an antibody generally involves recruiting effector cells. Several immunoglobulin effector functions are mediated by Fc receptors (FcRs), which bind the Fc region of an antibody. FcRs are defined by their specificity for immunoglobulin isotypes; Fc receptors for IgG antibodies are referred to as FcyR, for IgE as FcsR, for IgA as FcaR and so on. Any of these effector functions (or the loss of such effector functions) such as ADCC can be used in order to evaluate whether an fusion protein of the invention lacks the ability of Fc binding.

[0140] In this context, it is noted that the term "Fc receptor" or "FcR" defines a receptor, generally a protein that is capable of binding to the Fc region of an antibody. Fc receptors are found on the surface of certain cells of the immune system of an organism, for example natural killer cells, macrophages, neutrophils, and mast cells. In vivo Fc receptors bind to immunoglobulins that are immobilized on infected cells or present on invading pathogens. Their activity stimulates phagocytic or cytotoxic cells to destroy microbes, or infected cells by antibody-mediated phagocytosis or antibody-dependent cell-mediated cytotoxicity. Some viruses such as flaviviruses use Fc receptors to help them infect cells, by a mechanism known as antibody-dependent enhancement of infection. FcRs have been reviewed in Ravetch and Kinet, Annu. Rev. Immunol. 9: 457-92 (1991); Capel et al., Immunomethods 4: 25-34 (1994); and de Haas et al, J. Lab. Clin. Med. 126: 330-41 (1995).

[0141] "Antibody-dependent cell-mediated cytotoxicity" or ADCC refers to a form of cytotoxicity in which secreted Ig bound onto Fc receptors (FcRs) present on certain cytotoxic cells - such as natural killer (NK) cells, neutrophils and macrophages - enable these cytotoxic effector cells to bind specifically to an antigen-bearing target cell and subsequently kill the target cell with cytotoxins. The antibodies "arm" the cytotoxic cells and are required for killing of the target cell by this mechanism. The primary cells for mediating ADCC in humans, NK cells, express FcyRIII only, whereas monocytes express FcyRI, FcyRII and FcyRIII. FcR expression on hematopoietic cells is summarized in Table 3 on page 464 of Ravetch and Kinet, Annu. Rev. Immunol. 9: 457-92 (1991). To assess ADCC activity of a molecule of interest, an in vitro ADCC assay, such as described in US Patent Nos. 5,500,362 or 5,821,337 may be carried out. Useful effector cells for such assays include, but are not limited to, peripheral blood mononuclear cells (PBMC) and natural killer (NK) cells. In some embodiments ADCC activity of the molecule of interest may be assessed in vivo, e.g., in an animal model such as disclosed in Clynes et al, PNAS USA 95: 652-656 (1998).

[0142] Several antibody effector functions are mediated by Fc receptors (FcRs), which bind the Fc region of an antibody. FcRs are defined by their specificity for immunoglobulin isotypes; Fc receptors for IgG antibodies are referred to as FcyR, for IgE as FcsR, for IgA as FcaR and so on. Three subclasses of FcyR have been identified: FcyRI (CD64), FcyRII (CD32) and FcyRIII (CD 16).

[0143] A "peripheral blood mononuclear cell" (PBMC) as used herein is any blood cell having a round nucleus (as opposed to a lobed or no nucleus). A PMBC can be, for instance, a lymphocyte or a monocyte. Method of obtaining PMBC are well known in the art. For example, PMBC can be extracted from whole blood using ficoll, a hydrophilic polysaccharide that separates layers of blood, and gradient centrifugation, which will separate the blood into a top layer of plasma, followed by a layer of PBMCs and a bottom fraction of polymorphonuclear cells (such as neutrophils and eosinophils) and erythrocytes.

[0144] In certain preferred embodiments of the present invention, the extracellular fragment of the immune receptor is a extracellular fragment of a type II transmembrane protein, preferably a extracellular fragment of NKG2D and is located C-terminal of the linking polypeptide comprising at least a portion of a CH2 domain whereas the binding protein, which is preferably and antibody molecule, is located N-terminal of the linking polypeptide. Said extracellular fragment of NKG2D has preferably the ability to bind to at least one NKG2DL, preferably alternatively to more than one NGK2DL, most preferably to an NKG2DL. In preferred embodiments, the extracellular fragment of NKG2D comprises an amino acid sequence as set forth in SEQ ID NO: 45 and/or the fusion protein comprises a linking polypeptide sequence as set forth in SEQ ID NO: 03. In preferred embodiments, the antibody molecule has the ability to bind to CD3 or CD 16. In other preferred embodiments, the antibody molecule is a Fab fragment or a scFv fragment or a single domain antibody. In more specific embodiments, the antibody molecule is a Fab fragment specific for CD3, wherein the Fab fragment preferably has a heavy chain sequence as set forth in SEQ ID NO: 06 and a light chain sequence as set forth in SEQ ID NO: 07. In certain specific embodiments, the fusion protein comprises the heavy chain of the Fab fragment fused to the immune receptor fragment and the linking polypeptide, said fusion protein preferably comprise the amino acid sequence as set forth in SEQ ID NO: 48 and a light chain sequence as set forth in SEQ ID NO: 07. In other specific embodiments, the antibody fragment is a Fab fragment specific for CD 16, wherein the Fab fragment preferably has a heavy chain sequence as set forth in SEQ ID NO: 08 and a light chain sequence as set forth in SEQ ID NO: 09. In certain specific embodiments, the fusion protein comprises the heavy chain of the Fab fragment fused to the immune receptor fragment and the linking polypeptide, said fusion protein preferably comprise the amino acid sequence as set forth in SEQ ID NO: 49 and a light chain sequence as set forth in SEQ ID NO: 09.

[0145] The invention also envisions that if a Fab fragment is comprised in the fusion protein, the linking polypeptide and the extracellular fragment of the transmembrane protein is not necessarily fused to the Fab heavy chain (VH-CH1). Instead the linking polypeptide and the extracellular fragment of the transmembrane protein may be fused to VL-CL. A VH-CH1 (same as VDJ-CH1) fragment is then co-expressed with the fusion protein. In order to enable disulfide bond formation, a cysteine may be introduced to the VH-CH1 fragment, for instance at its C-terminal end, whereas the corresponding cysteine of the CL region, which is now part of the [VL-CL] -[linking polypeptide]-[transmembrane protein] fusion polypeptide may be removed. Such a fusion protein is illustrated under Figure 28A (ii). The invention also envisions that the extracellular fragment of the transmembrane protein may be fused to VL- CH1. A VH-CL fragment (same as VDJ-CL) is then co-expressed with the fusion protein. Such a fusion protein is illustrated under Figure 28B (ii).

[0146] In specific embodiments, the fusion protein comprises VL-CL fused to the linking polypeptide and the transmembrane protein fragment, said fusion protein preferably comprises the amino acid sequence as set forth in SEQ ID NO: 98 or 106 and is preferably co-expressed with a sequence set forth in SEQ ID NO: 92 or 100. In other specific embodiments, the fusion protein comprises VL-CH1 fused to the linking polypeptide and the transmembrane protein fragment, said fusion protein preferably comprises the amino acid sequence as set forth in SEQ ID NO: 96 or 104 and is preferably co-expressed with a sequence set forth in SEQ ID NO: 94 or 102. [0147] In other preferred embodiments of the present invention, the extracellular fragment of the immune receptor is a extracellular fragment of RANK and is located N- terminal of the linking polypeptide comprising at least a portion of a CH2 domain whereas the antibody fragment is located C-terminal of the linking polypeptide. Said extracellular fragment of RANK has preferably the ability to bind to at least a portion of RANKL. In preferred embodiments, the extracellular fragment of RANK comprises an amino acid sequence as set forth in SEQ ID NO: 47 and/or comprises a linking polypeptide sequence as set forth in SEQ ID NO: 02. In preferred embodiments, the antibody fragment has the ability to bind to CD3 or CD 16. In other preferred embodiments, the antibody fragment is a scFv fragment or a single domain antibody. In more specific embodiments, the antibody fragment is a scFv fragment specific for CD3, wherein the scFv fragment preferably has a sequence as set forth in SEQ ID NO: 04. In certain specific embodiments, the fusion protein comprises the amino acid sequence as set forth in SEQ ID NO: 50. In other specific embodiments, the antibody fragment is a scFv fragment specific for CD 16, wherein the scFv fragment preferably has a sequence as set forth in SEQ ID NO: 05. In certain specific embodiments, the fusion protein comprises the amino acid sequence as set forth in SEQ ID NO: 51.

[0148] Osteoprotegerin (OPG, UniProt Accession number 000300, SEQ ID NO: 89) is a cytokine receptor and a member of the tumor necrosis factor (TNF) receptor superfamily. It is another receptor that binds RANKL By binding RANKL, OPG prevents RANK-mediated nuclear kappa B activation which is a central and rapid acting transcription factor for immune-related genes, and a key regulator of inflammation, innate immunity, and cell survival and differentiation. As OPG also binds to RANKL, it is envisioned by the present invention that an extracellular domain of a fusion protein of the invention can be replaced by an extracellular domain of OPG that is similarly able to bind RANKL.

[0149] In preferred embodiments of the present invention, the extracellular fragment of the immune receptor is a extracellular fragment of GITR and is located N-terminal of the linking polypeptide comprising at least a portion of a CH2 domain whereas the antibody fragment is located C-terminal of the linking polypeptide. Said extracellular fragment of GITR has preferably the ability to bind to at least a portion of GITRL. In preferred embodiments, the extracellular fragment of GITR comprises an amino acid sequence as set forth in SEQ ID NO: 47 and/or comprises a linking polypeptide sequence as set forth in SEQ ID NO: 02. In preferred embodiments, the antibody fragment has the ability to bind to CD3 or CD 16. In other preferred embodiments, the antibody fragment is a Fab fragment or a scFv fragment. In more specific embodiments, the antibody fragment is a scFv fragment specific for CD3, wherein the scFv fragment preferably has a sequence as set forth in SEQ ID NO: 04. In certain specific embodiments, the fusion protein comprises the amino acid sequence as set forth in SEQ ID NO: 52. In other specific embodiments, the antibody fragment is a scFv fragment specific for CD 16, wherein the scFv fragment preferably has a sequence as set forth in SEQ ID NO: 05. In certain specific embodiments, the fusion protein comprises the amino acid sequence as set forth in SEQ ID NO: 53.

[0150] In specific embodiments, the fusion proteins the invention can bind to ligands or antigens that are soluble and thus not located on the surface of a target cell. Such ligands or antigens are preferably bound by the immune receptor fragment moiety of the fusion protein. Herein, binding of the ligand by the fusion protein preferably prevents binding of the same ligands to other proteins, such as other receptors, such as immune receptors. Therefore, as binding of the ligand to other receptors are hindered, a potential physiological or pathophysiological effect of said ligand is preferably weakened or neutralized. In certain embodiments, such ligands have immunomodulatory effect, such as an immune activating effect or an immune inhibitory effects, such as effects inhibiting an anti-tumor immune response, for instance by NK cells or T cells or preferably cytotoxic T-cells. In binding such ligands, embodiments of the present invention may neutralize the immunomodulatory effect such as an immune activating effect or an immune inhibitory effect of such ligands. In a preferred embodiment, such a soluble ligand or antigens is a NKG2DL. NKG2DL can be expressed by tumor cells and be set free in soluble form. These soluble NKG2DL are able to systemically inhibit anti-tumor immune responses. In a preferred embodiment, the fusion protein of the present invention is able to bind and neutralize soluble NGK2DL and thus neutralizing or reducing its/their immune inhibitory effect. In another preferred embodiment, such a soluble ligand or antigen is RANKL. RANKL can be expressed by tumor cells and be set free in soluble form. As RANKL is inhibitory to NK cells, NK activity against tumor cells is impaired by soluble RANKL. In a preferred embodiment, the fusion protein of the present invention is able to bind and neutralize soluble RANKL and thus neutralizing its immune inhibitory effect. Apart from its immune inhibitory effect, RANKL is also involved in bone resorption. Therefore, binding of RANKL by the fusion protein can also preferably reduce or neutralize the effect of RANKL in bone resorption. In yet another preferred embodiment, such a soluble ligand or antigen is GITRL. GITRL can be expressed by tumor cells and be set free in soluble form. As GITRL is inhibitory to NK cells, NK activity against tumor cells is impaired by soluble GITRL. In a preferred embodiment, the fusion protein of the present invention is able to bind and neutralize or reduce soluble GITRL and thus neutralizing its immune inhibitory effect.

[0151] It is understood that functional variants of sequences discussed herein can also be used as components of the inventive fusion protein. A "functional variant" of an binding protein or an extracellular fragment of an immune receptor refers to a protein, sequence, or portion that differs from a reference protein, sequence, or portion by one or more amino acid residue substitutions, additions, insertions, and/or deletions, but which at least substantially retains some (and desirably most or even all) of the functional attributes of the protein (in the case of antibody sequences the relevant functional attribute typically is binding to the same target with an affinity that is sufficient for the desired purpose) . A variant is significantly similar in terms of sequence identity with (e.g., exhibits at least about 40%, typically at least about 50%, more typically at least about 60%>, even more typically at least about 70%>, commonly at least about 80%>, frequently as at least about 85%, such as at least about 90%, 95%), or more identity) and usually in possession of other similar physiochemical properties to at least one (referenced) protein or amino acid sequence (which may be referred to as the "parent," which typically is a naturally occurring ("wild-type") molecule or molecule component)

[0152] Advantageous sequence changes with respect to a parent sequence that frequently are sought in the production of variants are those that (1) reduce susceptibility to proteolysis, (2) reduce susceptibility to oxidation, (3) alter binding affinity of the variant sequence (typically desirably increasing affinity), and/or (4) confer or modify other physicochemical or functional properties on the associated variant/analog peptide. The skilled artisan will be aware of these and other factors in the design, production, and selection of variants In the context of antibody CDR variants, for example, it is typically desired that residues required to support and/or orientate the CDR structural loop structure (s) are retained; that residues which fall within about 10 angstroms of a CDR structural loop are unmodified or modified only by conservative amino acid residue substitutions; and/or that the sequence is subject to only a limited number of insertions and/or deletions (if any), such that CDR structural loop-like structures are retained in the variant (a description of related techniques and relevant principles is provided in, e.g., Schiweck, et al . (1997) J. Mol. Biol. 268 (5) : 934-51 ; Morea (1997) Biophys . Chem. 68 (1-3) : 9-16; Shirai, et al . (1996) FEBS Lett. 399(1- 2) :l-8; Shirai, et al . (1999) FEBS Lett. 455 ( 1 - 2 ) : 188 - 97 ; Reckzo, et al . (1995) Protein Eng. 8 (4) : 389-95 ; and Eigenbrot, et al . (1993) J. Mol. Biol. 229 (4) : 969-95) . Medical use/ method of treatments

[0153] As explained above, a fusion protein according to the invention may be directed against a desired target ligand or antigen on the surface of a target cell or a ligand or antigen expressed by a target cell. Depending on the selected ligand or antigen, the fusion protein may be suitable in the treatment or prevention of a disease. Accordingly, in some embodiments, a fusion protein according to the invention may be used in a method of treating and/or preventing a medical condition such as a disorder or disease. Similarly, the fusion protein of the present invention can be used in the treatment of a disease. In some embodiments a disease to be treated or prevented may be a proliferative disease. In some embodiments, the target cell is a tumor cell. In a more specific embodiment, such a proliferative disease is a tumor or cancer. Examples of a proliferative disease include, but are not limited to, adrenal cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, brain and spinal cord tumors, breast cancer, Castleman disease, cervical cancer, colon cancer, endometrial cancer, esophagus cancer, Ewing family of tumors, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumors, gastrointestinal stromal tumor (GIST), gestational trophoblastic disease, Hodgkin disease, Kaposi sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL), chronic myeloid leukemia (CML), chronic myelomonocytic leukemia (CMML), liver cancer, lung cancer, non-small cell lung cancer, small cell lung cancer, lung carcinoid tumor, lymphoma, lymphoma of the skin, malignant mesothelioma, multiple myeloma, myelodysplasia syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroblastoma, non-Hodgkin lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumors, prostate cancer, rectum cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, skin cancer, basal and squamous cell cancer, melanoma, merkel cell cancer, small intestine cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, uterine sarcoma, vaginal cancer, vulvar cancer, Waldenstrom macroglobulinemia, or Wilms tumor. In preferred embodiments, the disease is an autoimmune disease, or graft-versus-host-disease or a viral infection. In preferred embodiment, the disease is osteoporosis and/or osteopenia and the fusion protein comprises an extracellular fragment of RANK.

[0154] The subject to be treated with the fusion protein can be a human or non-human animal. Such an animal is preferably a mammal, for instance a human, pig, cattle, rabbit, mouse, rat, primate, goat, sheep, chicken, or horse, most preferably a human. [0155] The invention also provides a pharmaceutical composition that includes a fusion protein of the invention and, optionally a pharmaceutically acceptable excipient.

[0156] The fusion protein according to the invention can be administered via any parenteral or non-parenteral (enteral) route that is therapeutically effective for proteinaceous drugs. Parenteral application methods include, for example, intracutaneous, subcutaneous, intramuscular, intratracheal, intranasal, intra vitreal or intravenous injection and infusion techniques, e.g. in the form of injection solutions, infusion solutions or tinctures, as well as aerosol installation and inhalation, e.g. in the form of aerosol mixtures, sprays or powders. An overview about pulmonary drug delivery, i.e. either via inhalation of aerosols (which can also be used in intranasal administration) or intracheal instillation is given by J.S. Patton et al. The lungs as a portal of entry for systemic drug delivery. Proc. Amer. Thoracic Soc. 2004 Vol. 1 pages 338-344, for example). Non-parenteral delivery modes are, for instance, orally, e.g. in the form of pills, tablets, capsules, solutions or suspensions, or rectally, e.g. in the form of suppositories. Fusion protein of the invention can be administered systemically or topically in formulations containing conventional non-toxic pharmaceutically acceptable excipients or carriers, additives and vehicles as desired.

[0157] In one embodiment of the present invention the pharmaceutical is administered parenterally to a mammal, and in particular to humans. Corresponding administration methods include, but are not limited to, for example, intracutaneous, subcutaneous, intramuscular, intratracheal or intravenous injection and infusion techniques, e.g. in the form of injection solutions, infusion solutions or tinctures as well as aerosol installation and inhalation, e.g. in the form of aerosol mixtures, sprays or powders. A combination of intravenous and subcutaneous infusion and /or injection might be most convenient in case of compounds with a relatively short serum half-life. The pharmaceutical composition may be an aqueous solution, an oil-in water emulsion or a water-in-oil emulsion.

[0158] In this regard it is noted that transdermal delivery technologies, e.g. iontophoresis, sonophoresis or microneedle-enhanced delivery, as described in Meidan VM and Michniak BB 2004 Am. J. Ther. 11(4): 312-316, can also be used for transdermal delivery of a fusion protein described herein. Non-parenteral delivery modes are, for instance, oral, e.g. in the form of pills, tablets, capsules, solutions or suspensions, or rectal administration, e.g. in the form of suppositories. The fusion protein of the invention can be administered systemically or topically in formulations containing a variety of conventional non-toxic pharmaceutically acceptable excipients or carriers, additives, and vehicles. [0159] The dosage of the fusion protein applied may vary within wide limits to achieve the desired preventive effect or therapeutic response. It will, for instance, depend on the affinity of the fusion protein for a chosen target as well as on the half-life of the complex between the fusion protein and the ligand in vivo. Further, the optimal dosage will depend on the biodistribution of the fusion protein or a conjugate thereof, the mode of administration, the severity of the disease/disorder being treated as well as the medical condition of the patient. For example, when used in an ointment for topical applications, a high concentration of the fusion protein can be used. However, if wanted, the fusion protein may also be given in a sustained release formulation, for example liposomal dispersions or hydrogel-based polymer microspheres, like PolyActiveTM or OctoDEXTM (cf. Bos et al., Business Briefing: Pharmatech 2003: 1-6). Other sustained release formulations available are for example PLGA based polymers (PR pharmaceuticals), PLA-PEG based hydrogels (Medincell) and PEA based polymers (Medivas).

[0160] Accordingly, the fusion proteins of the present invention can be formulated into compositions using pharmaceutically acceptable ingredients as well as established methods of preparation (Gennaro, A.L. and Gennaro, A.R. (2000) Remington: The Science and Practice of Pharmacy, 20^th Ed., Lippincott Williams & Wilkins, Philadelphia, PA). To prepare the pharmaceutical compositions, pharmaceutically inert inorganic or organic excipients can be used. To prepare e.g. pills, powders, gelatin capsules or suppositories, for example, lactose, talc, stearic acid and its salts, fats, waxes, solid or liquid polyols, natural and hardened oils can be used. Suitable excipients for the production of solutions, suspensions, emulsions, aerosol mixtures or powders for reconstitution into solutions or aerosol mixtures prior to use include water, alcohols, glycerol, polyols, and suitable mixtures thereof as well as vegetable oils.

[0161] The pharmaceutical composition may also contain additives, such as, for example, fillers, binders, wetting agents, glidants, stabilizers, preservatives, emulsifiers, and furthermore solvents or solubilizers or agents for achieving a depot effect. The latter is that fusion proteins may be incorporated into slow or sustained release or targeted delivery systems, such as liposomes and microcapsules.

[0162] The formulations can be sterilized by numerous means, including filtration through a bacteria-retaining filter, or by incorporating sterilizing agents in the form of sterile solid compositions which can be dissolved or dispersed in sterile water or other sterile medium just prior to use. [0163] In some embodiments, the fusion protein of the present invention can be used in methods involving adoptive cell transfer. For example, the administration of the fusion proteins of the invention can be combined with the transfer of immune cells. Said immune cells may be resting or pretreated, such as IL- 12/15/18-pre-activated NK cells as described in Ni et al, 2012 or pre-activated T cells. It is understood that in such embodiments, the fusion protein has to be able to bind to the pre-activated immune cells. As an illustrative example, preparations of T or NK cells may be generated in vitro with allogeneic or autologous T and NK cells to achieve higher numbers of immune effector cells, an approach that is already clinically applied. These cell preparations may be infused to patients together with the fusion protein to enhance the antitumor reactivity of the transfused immune cells. It is also envisioned that fusion proteins of the present invention may be administered in combination with donor lymphocyte infusion such as after hematopoietic stem cell transplantation, with administration of autologous or allogenic polyclonal NK cells, with administration of antigen specific T cells or with administration of chimeric antigen receptor T cells or NK cells.

[0164] Numerous possible applications for the inventive fusion protein exist in medicine. In addition to their use in in vitro diagnostics or drug delivery, a fusion protein of the invention, which binds, for example, tissue- or tumor-specific cellular surface molecules can be generated.

Nucleic acid

[0165] A nucleic acid molecule encoding one or more chains of a fusion protein to the invention may be any nucleic acid in any possible configuration, such as single stranded, double stranded or a combination thereof. Nucleic acids include for instance DNA molecules, RNA molecules, analogues of the DNA or RNA generated using nucleotide analogues or using nucleic acid chemistry, locked nucleic acid molecules (LNA), and protein nucleic acids molecules (PNA). DNA or RNA may be of genomic or synthetic origin and may be single or double stranded. Such nucleic acid can be e.g. mRNA, cRNA, synthetic RNA, genomic DNA, cDNA synthetic DNA, a copolymer of DNA and RNA, oligonucleotides, etc. A respective nucleic acid may furthermore contain non-natural nucleotide analogues and/or be linked to an affinity tag or a label. Exemplary embodiments of the nucleic acids of the invention are set forth in SEQ ID NO: 63, SEQ ID NO: 64, SEQ ID NO: 65, SEQ ID NO: 66, SEQ ID NO: 67, SEQ ID NO: 68, SEQ ID NO: 69 and SEQ ID NO: 70.

[0166] In some embodiments a nucleic acid sequence encoding a chain, such as a main chain and/or a smaller chain of a fusion protein according to the invention is included in a vector such as a plasmid. Where a substitution or deletion is to be included in a fusion protein chain, when compared to a naturally occurring domain or region of a parental protein, the coding sequence of the respective native domain/region, e.g. included in the sequence of an immunoglobulin or an immune receptor, can be used as a starting point for the mutagenesis. For the mutagenesis of selected amino acid positions, the person skilled in the art has at his disposal the various established standard methods for site-directed mutagenesis.

[0167] A nucleic acid molecule encoding a chain, such as a main chain and/or a smaller chain of an antibody according to the invention can be expressed using any suitable expression system, for example in a suitable host cell or in a cell-free system. The obtained fusion protein is enriched by means of selection and/ or isolation. Thus, in one embodiment, the nucleic acid molecule of the present invention can be comprised in a vector. Similarly, the nucleic acid molecule of the present invention may be comprised in a host cell or the vector comprising the nucleic acid molecule of the present invention may be comprised in a host cell.

Production

[0168] An fusion protein of the invention may be produced using any known and well- established expression system and recombinant cell culturing technology, for example, by expression in bacterial hosts (prokaryotic systems), or eukaryotic systems such as yeasts, fungi, insect cells or mammalian cells. A fusion protein of the present invention may be produced in transgenic organisms such as a goat, a plant or a XENOMOUSE transgenic mouse, an engineered mouse strain that has large fragments of the human immunoglobulin loci and is deficient in mouse antibody production. A fusion protein may also be produced by chemical synthesis.

[0169] For recombinant production of a fusion protein of the invention typically a polynucleotide encoding the fusion protein is isolated and inserted into a replicable vector such as a plasmid for further cloning (amplification) or expression. An illustrative example of a suitable expression system is a glutamate synthetase system (such as sold by Lonza Biologies), with the host cell being for instance CHO or NSO. A polynucleotide encoding the fusion protein is readily isolated and sequenced using conventional procedures. Vectors that may be used include plasmid, virus, phage, transposons, minichromosomes of which plasmids are a typical embodiment. Generally such vectors further include a signal sequence, origin of replication, one or more marker genes, an enhancer element, a promoter and transcription termination sequences operably linked to the polynucleotide so as to facilitate expression. Polynucleotides encoding separate chains of the fusion protein may be inserted into separate vectors and transfected into the same host cell or, if desired both polynucleotides can be inserted into the same vector for transfection into the host cell. Both chains can, for example, be arranged, under the control of a dicistronic operon and expressed to result in the functional and correctly folded fusion protein comprising multiple chains as described for antibodies in Skerra, A. (1994) Use of the tetracycline promoter for the tightly regulated production of a murine antibody fragment in Escherichia coli, Gene 151, 131-135, or Skerra, A. (1994) A general vector, pASK84, for cloning, bacterial production, and single-step purification of antibody Fab fragments, Gene 141, 79-8. Thus according to one aspect of the present invention there is provided a process of constructing a vector encoding the polynucleotide chain(s) of a fusion protein of the invention, which method includes inserting into a vector, a polynucleotide encoding either the or a chain of fusion protein of the invention.

[0170] When using recombinant techniques, the fusion protein can be produced intracellularly, in the periplasmic space, or directly secreted into the medium (cf. also Skerra 1994, supra). If the fusion protein is produced intracellularly, as a first step, the particulate debris, either host cells or lysed fragments, are removed, for example, by centrifugation or ultrafiltration. Carter et al., Bio/Technology 10: 163-167 (1992) describe a procedure for isolating antibodies which are secreted to the periplasmic space of E coli. The fusion protein can also be produced in any oxidizing environment. Such an oxidizing environment may be provided by the periplasm of Gram-negative bacteria such as E. coli, in the extracellular milieu of Gram-positive bacteria or in the lumen of the endoplasmatic reticulum of eukaryotic cells (including animal cells such as insect or mammalian cells) and usually favors the formation of structural disulfide bonds. It is, however, also possible to produce a fusion protein of the invention in the cytosol of a host cell such as E. coli. In this case, the polypeptide can either be directly obtained in a soluble and folded state or recovered in form of inclusion bodies, followed by renaturation in vitro. A further option is the use of specific host strains having an oxidizing intracellular milieu, which may thus allow the formation of disulfide bonds in the cytosol (Venturi M, Seifert C, Hunte C. (2002) "High level production of functional antibody Fab fragments in an oxidizing bacterial cytoplasm." J. Mol. Biol. 315, 1-8).

[0171] The fusion protein produced by the cells can be purified using any conventional purification technology, for example, hydroxylapatite chromatography, gel electrophoresis, dialysis, and affinity chromatography, with affinity chromatography being one preferred purification technique. Fusion proteins comprising constant domains of immunogloblulines may be purified via affinity purification with proteins/ligands that specifically and reversibly bind constant domains such as the CHI or the CL domains. Examples of such proteins are immunoglobulin-binding bacterial proteins such as Protein G, Protein A/G or Protein L, wherein Protein L binding is restricted to fusion proteins that contain kappa light chains of an immunoglobulin. An alternative method for purification of fusion protein with κ- light chains of immunoglobulins is the use of bead coupled anti kappa antibodies (KappaSelect). The choice of the purification method that is used for a particular fusion protein of the invention is within the knowledge of the person of average skill in the art.

[0172] It is also possible to equip one of the chains of the fusion protein of the invention with an affinity tag. Affinity tags such as the Strep-tag (SEQ ID NO: 57) or Strep- tag II (SEQ ID NO: 58) (Schmidt, T.G.M. et al. (1996) J. Mol. Biol. 255, 753-766), the myc- tag (SEQ ID NO: 59), the FLAG-tag (SEQ ID NO: 60), the His6-tag (SEQ ID NO: 61) or the HA-tag (SEQ ID NO: 62) allow easy detection and also simple purification of the recombinant fusion protein.

[0173] Thus, a method of producing an fusion protein of the present invention comprises expressing a nucleic acid encoding the fusion protein under conditions allowing expression of the nucleic acid, preferably the fusion protein is expressed in a host cell or a cell-free system.

[0174] The present invention is further characterized by the following items

[0175] Item 1 : A recombinant fusion protein consisting of:

a. A binding protein with a binding site that specifically binds to a receptor specific for T cells or natural killer cells;

b. An extracellular fragment of a transmembrane protein that is an immune receptor and that binds to a target cell or to a ligand or antigen expressed by a target cell; c. A linking polypeptide connecting the binding protein and the extracellular fragment of a transmembrane protein,

wherein said linking polypeptide consists of at least a portion of a CH2 domain and optionally a hinge region.

[0176] Alternative Item 1. A recombinant fusion protein comprising:

b. an extracellular fragment of a transmembrane protein that is an immune receptor and that binds to a target cell or to a ligand or antigen expressed by a target cell;

c. a linking polypeptide connecting the binding protein and the extracellular fragment of a transmembrane protein, wherein said linking polypeptide comprises at least a portion of a Fc domain;

[0177] wherein the recombinant fusion protein does not comprise immunoglobulin heavy chain comprising VH-H-CH 1 -CH2-CH3 or VH-H-CH1-CH2 CH3-CH4.

Item 2. The fusion protein of item 1, wherein the at least a portion of an Fc domain comprises at least a portion of a CH2 domain and/or at least a portion of a CH3 domain.

[0178] Item 3. The fusion protein of item 2, wherein the at least a portion of an Fc domain comprises at least a portion of a CH2 domain.

[0179] Item 4. The fusion protein of any one of items 1-3, wherein the at least a portion of an Fc domain comprises at least a portion of a hinge region.

[0180] Item 5. The fusion protein of any one of items 1-4, wherein the at least a portion of an Fc domain comprises at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, or 124 consecutive amino acids corresponding to SEQ ID NO: 01 or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61 , 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81 , 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101 , 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, or 124 consecutive amino acids of SEQ ID NO: 01.

[0181] Item 6. The fusion protein of any one of items 1-4, wherein the at least a portion of an Fc domain comprises at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, or 124 consecutive amino acids corresponding to SEQ ID NO: 02 or to SEQ ID NO: 03 or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61 , 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, or 124 consecutive amino acids of SEQ ID NO: 02 or to SEQ ID NO: 03. [0182] Item 7. The fusion protein of any one of items 1-6, wherein the at least a portion of an Fc domain comprises a mutation or deletion in at least one amino acid residue of the Fc domain that is able to mediate binding to Fc receptors.

[0183] Item 8. The fusion protein of item 7, wherein the at least one amino acid residue of the Fc domain is selected from the group consisting of sequence position 233, 234, 235, 236, 265, 297, 327, and 330 (numbering of sequence positions according to the EU- index).

[0184] Item 9. The fusion protein of item 8, wherein the at least one amino acid mutation or deletion is selected from the group consisting of Glu233Pro, Leu234Val, Leu235Ala, deletion of Gly236, Asp265Gly, Asn297Gln, Ala327Gln, and Ala330Ser.

[0185] Item 10. The fusion protein of any one of items 1-9, wherein the at least a portion of an Fc domain comprises a mutation or deletion in at least one amino acid residue of the Fc domain that is able to mediate dimerization of immunoglobulins.

[0186] Item 11. The fusion protein of item 10, wherein the at least one amino acid residue of the Fc domain is selected from the group consisting of sequence position 220, 226, and 229 (numbering of sequence positions according to the EU-index).

[0187] Item 12. The fusion protein of item 11, wherein the at least one amino acid mutation or deletion is selected from the group consisting of Cys220Ser, Cys226Ser, or Cys229Ser.

[0188] Item 13. The fusion protein of any one of items 1-12, wherein the binding protein is an antibody molecule.

[0189] Item 14. The fusion protein of item 13, wherein the antibody molecule has a single antigen binding site.

[0190] Item 15. The fusion protein of item 13 or 14, wherein the antibody molecule is a Fab fragment or a scFv fragment or a single domain antibody molecule.

[0191] Item 16. The fusion protein of any one of items 1-15, wherein the binding protein binds to CD3, CD16, CD28, CD137/4-1BB, OX40, Nkp44, Nkp30, Nkp40 or Nkp46.

[0192] Item 17. The fusion protein of any one of items 1-16, wherein the binding protein binds to CD3 or CD 16.

[0193] Item 18. The fusion protein of item 17, wherein the binding protein binds to CD3 on T cells.

[0194] Item 19. The fusion protein of item 18, wherein the binding protein binds to CD3 on cytotoxic T cells. [0195] Item 20. The fusion protein of item 17, wherein the binding protein binds to CD 16 on natural killer (NK) cells.

[0196] Item 21. The fusion protein of any one of the preceding items, wherein the ligand or antigen that binds to the extracellular fragment of the immune receptor is a tumor- associated ligand or antigen.

[0197] Item 22. The fusion protein of any one of the preceding items, wherein the extracellular fragment of the immune receptor binds to a ligand or antigen on the target cell or to a ligand or antigen that is soluble.

[0198] Item 23. The fusion protein of item 22, wherein the extracellular fragment of the immune receptor binds to a ligand or antigen on the surface of a target cell.

[0199] Item 24. The fusion protein of item 23, wherein the ligand or antigen is a NKG2DL or RANKL or GITRL.

[0200] Item 25. The fusion protein of any of the preceding items, wherein the extracellular fragment of an immune receptor is an extracellular fragment of an immune receptor selected from the group of NKG2D (UniProtKB accession number P26718, SEQ ID NO: 10), RANK (UniProtKB accession number Q9Y6Q6, SEQ ID NO: 11) or GITR (UniProtKB accession number Q9Y5U5, SEQ ID NO: 12), CD94-NKG2 (UniProtKB accession number Q 13241, SEQ ID NO: 13), CTLA-4 (UniProtKB accession number PI 6410, SEQ ID NO: 15), PD-1 (UniProtKB accession number Q15116, SEQ ID NO: 16), BTLA (UniProtKB accession number Q7Z6A9, SEQ ID NO: 17), LAG-3 (UniProtKB accession number PI 8627, SEQ ID NO: 18), TIM-3 (UniProtKB accession number Q8TDQ0, SEQ ID NO: 19), LAIR-1 (UniProtKB accession number Q6GTX8, SEQ ID NO: 20), TIGIT (UniProtKB accession number Q495A1, SEQ ID NO: 21), Siglecl (UniProtKB accession number Q9BZZ2, SEQ ID NO: 22), Siglec2 (UniProtKB accession number P20273, SEQ ID NO: 23), Siglec3 (UniProtKB accession number P20138, SEQ ID NO: 24), Siglec4 (UniProtKB accession number P20916, SEQ ID NO: 25), Siglec5 (UniProtKB accession number 015389, SEQ ID NO: 26), Siglec6 (UniProtKB accession number 043699, SEQ ID NO: 27), Siglec7 (UniProtKB accession number Q9Y286, SEQ ID NO: 28), Siglec8 (UniProtKB accession number Q9NYZ4, SEQ ID NO: 29), Siglec9 (UniProtKB accession number Q9Y336, SEQ ID NO: 30), SigleclO (UniProtKB accession number Q96LC7, SEQ ID NO: 31), Siglecl 1 (UniProtKB accession number Q96RL6, SEQ ID NO: 32), Siglecl2 (UniProtKB accession number Q96PQ1, SEQ ID NO: 33), Siglecl4 (UniProtKB accession number Q08ET2, SEQ ID NO: 34), Siglecl 5 (UniProtKB accession number Q6ZMC9, SEQ ID NO: 35), Siglecl6 (UniProtKB accession number A6NMB1, SEQ ID NO: 36), NKp30 (UniProtKB accession number 014931, SEQ ID NO: 14), NKp40 (UniProtKB accession number Q9NZS2, SEQ ID NO: 85), NKp44 (UniProtKB accession number 095944, SEQ ID NO: 86), NKp46 (UniProtKB accession number 076036, SEQ ID NO: 87), NKp80 (UniProtKB accession number Q9NZS2, SEQ ID NO: 88), OPG (UniProtKB accession number 000300, SEQ ID NO: 89).

[0201] Item 26. The fusion protein of item 25, wherein the extracellular fragment of an immune receptor is from an extracellular fragment of NKG2D (UniProtKB accession number P26718), RANK (UniProtKB accession number Q9Y6Q6) or GITR (UniProtKB accession number Q9Y5U5).

[0202] Item 27. The fusion protein of item 26, wherein the extracellular fragment is of NKG2D and has the ability to bind to at least a portion of at least one NKG2DL, preferably alternatively to more than one NKG2DL, most preferably to any NKG2DL, wherein the group of NKG2DL comprises MICA (UniProtKB accession number Q29983), MICB (UniProtKB accession number Q29980), ULBP1 (UniProtKB accession number Q9BZM6), ULBP2 (UniProtKB accession number Q9BZM5), ULBP3 (UniProtKB accession number Q9BZM4), ULBP4 (UniProtKB accession number Q8TD07), ULBP5 (UniProtKB accession number Q6H3X3) and ULBP6 (UniProtKB accession number Q5VY80).

[0203] Item 28. The fusion protein of item 26 or 27, wherein the extracellular fragment is of NKG2D comprises at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, or 139 consecutive amino acids corresponding to SEQ ID NO: 45 or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50, 51 , 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91 , 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, or 139 consecutive amino acids of SEQ ID NO: 45.

[0204] Item 29. The fusion protein of item 26, wherein the extracellular fragment is of RANK and has the ability to bind to at least a portion of RANKL (UniProtKB accession number 014788). [0205] Item 30. The fusion protein of item 26 or 29, wherein the extracellular fragment is of RANK comprises at least 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, or 183 consecutive amino acids corresponding to SEQ ID NO: 46 or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171 , 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, or 183 consecutive amino acids of SEQ ID NO: 46.

[0206] Item 31. The fusion protein of item 26, wherein the extracellular fragment is of GITR and has the ability to bind to at least a portion of GITRL (UniProtKB accession number Q9UNG2).

[0207] Item 32. The fusion protein of item 26 or 31, wherein the extracellular fragment is of GITR comprises at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63,

64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88,

89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109,

110, 111, 112, 1 13, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128,

129, 130, 131, 132, 133, 134, 135, 136 or 137 consecutive amino acids corresponding to SEQ ID NO: 47 or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64,

65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89,

90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110,

111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129,

130, 131, 132, 133, 134, 135, 136 or 137 consecutive amino acids of SEQ ID NO: 47.

[0208] Item 33. The fusion protein of any one of the preceding items, wherein the binding protein is located N-terminal of the linker polypeptide and the extracellular fragment of a transmembrane protein is located C-terminal to the linker polypeptide. [0209] Item 34. The fusion protein of any one of the preceding items, wherein the binding protein is located N-terminal of the linker polypeptide and the extracellular fragment of a transmembrane protein is located C-terminal to the linker polypeptide, wherein said transmembrane protein is a type II transmembrane protein.

[0210] Item 35. The fusion protein of item 33 or 34, wherein the extracellular fragment of the transmembrane protein is an extracellular fragment of NKG2D and has the ability to bind to at least one NKG2DL, preferably alternatively to more than one NKG2DL, most preferably to any NKG2DL.

[0211] Item 36. The fusion protein of any one of items 33-35, wherein the binding protein is an antibody molecule.

[0212] Item 37. The fusion protein of item 36, wherein the antibody molecule is a Fab fragment or a scFv fragment or a single domain antibody molecule that binds to CD3.

[0213] Item 38. The fusion protein of item 37, wherein the antibody molecule binds to CD3 on T cells.

[0214] Item 39. The fusion protein of item 37, wherein the antibody molecule is a Fab fragment that binds to CD3.

[0215] Item 40. The fusion protein of item 39, wherein the Fab fragment binds to CD3 on T cells.

[0216] Item 41. The fusion protein of item 39, wherein the Fab fragment heavy chain comprises a sequence set forth in SEQ ID NO: 06.

[0217] Item 42. The fusion protein of item 39, wherein the Fab fragment light chain comprises a sequence set forth in SEQ ID NO: 07.

[0218] Item 43. The fusion protein of any one of items 39-42, wherein the immune receptor fragment and the at least a portion of an Fc fragment are fused with a heavy chain or a light chain of a Fab fragment.

[0219] Item 44. The fusion protein of item 43, wherein the immune receptor fragment and the at least a portion of an Fc fragment are fused with a heavy chain of a Fab fragment.

[0220] Item 45. The fusion protein of item 44, wherein the fusion protein comprises an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 48.

[0221] Item 46. The fusion protein of item 45, wherein the fusion protein comprises a Fab fragment light chain comprising an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 07.

[0222] Item 47. The fusion protein of item 36, wherein the antibody molecule is a Fab fragment or a scFv fragment or a single domain antibody molecule that binds to CD 16.

[0223] Item 48. The fusion protein of item 47, wherein the antibody molecule binds to CD 16 on natural killer (NK) cells.

[0224] Item 49. The fusion protein of item 47, wherein the antibody molecule is a Fab fragment that binds to CD 16.

[0225] Item 50. The fusion protein of item 49, wherein the Fab fragment binds to CD 16 on natural killer (NK) cells.

[0226] Item 51. The fusion protein of item 49, wherein the Fab fragment heavy chain comprises a sequence set forth in SEQ ID NO: 08.

[0227] Item 52. The fusion protein of item 49, wherein the Fab fragment light chain comprises a sequence set forth in SEQ ID NO: 09.

[0228] Item 53. The fusion protein of item 49, wherein the immune receptor fragment and the at least a portion of an Fc fragment are fused with a heavy chain or a light chain of a Fab fragment.

[0229] Item 54. The fusion protein of item 53, wherein the immune receptor fragment and the at least a portion of an Fc fragment are fused with a heavy chain of a Fab fragment.

[0230] Item 55. The fusion protein of item 54, wherein the fusion protein comprises a polypeptide having an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 49.

[0231] Item 56. The fusion protein of item 55, wherein the fusion protein comprises a Fab fragment light chain comprising an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 09.

[0232] Item 57. The fusion protein of any one of items 1-32, wherein the binding protein is located C-terminal to of the linker polypeptide and the extracellular fragment of a transmembrane protein is located N-terminal to the linker polypeptide.

[0233] Item 58. The fusion protein of any one of items 1-32, wherein the binding protein is located C-terminal to of the linker polypeptide and the extracellular fragment of a transmembrane protein is located N-terminal to the linker polypeptide, wherein said transmembrane protein is a type I transmembrane protein.

[0234] Item 59. The fusion protein of item 58 or 59, wherein the extracellular fragment of the transmembrane protein is selected from either an extracellular fragment of RANK that has the ability to bind at least a portion of RANKL or an extracellular fragment of GITR that has the ability to bind to at least a portion of GITRL.

[0235] Item 60. The fusion protein of any one of items 57-59, wherein the binding protein is an antibody molecule.

[0236] Item 61. The fusion protein of item 60, wherein the antibody molecule is a scFv fragment or a single domain antibody molecule that binds to CD3.

[0237] Item 62. The fusion protein of item 61, wherein the antibody molecule binds to CD3 on T cells.

[0238] Item 63. The fusion protein of item 61, wherein the antibody molecule is a scFv fragment that binds to CD3.

[0239] Item 64. The fusion protein of item 63, wherein the scFv fragment binds to CD3 on T cells.

[0240] Item 65. The fusion protein of item 63, wherein the scFv fragment comprises a sequence as set forth in SEQ ID NO: 04 .

[0241] Item 66. The fusion protein of any one of items 63-65, wherein the fusion protein comprises an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 50 or SEQ ID NO: 52.

[0242] Item 67. The fusion protein of item 60, wherein the antibody molecule is a scFv fragment or a single domain antibody molecule that binds to CD 16.

[0243] Item 68. The fusion protein of item 67, wherein the antibody molecule binds to CD 16 on natural killer (NK) cells.

[0244] Item 69. The fusion protein of item 67, wherein the antibody molecule is a scFv fragment that binds to CD 16.

[0245] Item 70. The fusion protein of item 69, wherein the scFv fragment binds to CD 16 on natural killer (NK) cells.

[0246] Item 71. The fusion protein of item 69, wherein the scFv fragment comprises a sequence as set forth in SEQ ID NO: 05. [0247] Item 72. The fusion protein of item 69, wherein the fusion protein comprises a polypeptide having an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 51 or SEQ ID NO: 53.

[0248] Item 73. The fusion protein of any one of the preceding items, wherein the target cell is a tumor cell.

[0249] Item 74. The fusion protein of any one of the preceding items, wherein the fusion protein binds to a ligand that is soluble.

[0250] Item 75. The fusion protein of item 74, wherein binding prevents binding of the ligand to other immune receptors.

[0251] Item 76. The fusion protein of item 75, wherein binding neutralizes a physiological or pathophysiological effect of the ligand.

[0252] Item 77. The fusion protein of item 76, wherein binding neutralizes an immunomodulatory effect of the ligand.

[0253] Item 78. The fusion protein of item 76, wherein binding neutralizes an immune inhibitory effect of the ligand.

[0254] Item 79. The fusion protein of any of items 74-78, wherein the fusion protein binds to soluble NKG2DL or RANKL or GITRL.

[0255] Item 80. The fusion protein of item 79, wherein binding prevents binding of NKG2DL to other NKG2D, or prevents binding of RANKL to other RANK, or prevents binding of GITRL to other GITR.

[0256] Item 81. The fusion protein of item 80, wherein binding neutralizes a physiological or pathophysiological effect of NKG2DL or RANKL or GITRL.

[0257] Item 82. The fusion protein of 81, wherein binding neutralizes an immunomodulatory effect of NKG2DL or an immunomodulatory effect of RANKL or an immunomodulatory effect of GITRL.

[0258] Item 83. The fusion protein of 82, wherein the immunomodulatory effect is an immune inhibitory effect.

[0259] Item 84. The fusion protein of 81, wherein binding of RANKL neutralizes an effect of RANKL in bone resorption.

[0260] Item 85. A pharmaceutical composition comprising a fusion protein as defined in items 1-84. [0261] Item 86. A fusion protein as defined in any one of items 1-84 for use in the treatment of a disease.

[0262] Item 87. The fusion protein for use of item 86, wherein the disease is a proliferative disease.

[0263] Item 88. The fusion protein for use of item 87, wherein the proliferative disease is cancer.

[0264] Item 89. The fusion protein for use of item 88, wherein the cancer is selected from the group consisting of adrenal cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, brain and spinal cord tumors, breast cancer, Castleman disease, cervical cancer, colon cancer, endometrial cancer, esophagus cancer, Ewing family of tumors, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumors, gastrointestinal stromal tumor (GIST), gestational trophoblastic disease, Hodgkin disease, Kaposi sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL), chronic myeloid leukemia (CML), chronic myelomonocytic leukemia (CMML), liver cancer, lung cancer, non-small cell lung cancer, small cell lung cancer, lung carcinoid tumor, lymphoma, lymphoma of the skin, malignant mesothelioma, multiple myeloma, myelodysplasia syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroblastoma, non-Hodgkin lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumors, prostate cancer, rectum cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, skin cancer, basal and squamous cell cancer, melanoma, merkel cell cancer, small intestine cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, uterine sarcoma, vaginal cancer, vulvar cancer, Waldenstrom macroglobulinemia, or Wilms tumor.

[0265] Item 90. The fusion protein for use of item 86, wherein the disease is osteoporosis.

[0266] Item 91. The fusion protein for use of item 90, wherein the fusion protein comprises an extracellular fragment of RANK.

[0267] Item 92. The fusion protein for use of item 86, wherein the disease is an autoimmune disease.

[0268] Item 93. The fusion protein for use of item 86, wherein the disease is graft- versus-host disease.

[0269] Item 94. The fusion protein for use of item 86, wherein the disease is a viral infection. [0270] Item 95. A nucleic acid encoding for the fusion protein of any of items 1-84.

[0271] Item 96. A nucleic acid of item 95 comprised in a vector.

[0272] Item 97. A host cell comprising the nucleic acid molecule of item 95 or a vector of item 96.

[0273] Item 98. A method of producing the fusion protein of any of items 1-84, comprising using the nucleic acid encoding the fusion protein for expression of the fusion protein under conditions allowing expression of the fusion protein.

[0274] Item 99. The method of item 98, wherein the fusion protein is expressed by a host cell or in a cell- free system.

[0275] Item 100. A method of treating a disease comprising administering a therapeutically effective amount of the fusion protein as defined in any of items 1-84 to a subject.

[0276] Item 101. The method of item 100, wherein the disease is a proliferative disease.

[0277] Item 102. The method of item 101, wherein the proliferative disease is cancer.

[0278] Item 103. The method of item 102, wherein the cancer is selected from the group consisting adrenal cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, brain and spinal cord tumors, breast cancer, Castleman disease, cervical cancer, colon cancer, endometrial cancer, esophagus cancer, Ewing family of tumors, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumors, gastrointestinal stromal tumor (GIST), gestational trophoblastic disease, Hodgkin disease, Kaposi sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL), chronic myeloid leukemia (CML), chronic myelomonocytic leukemia (CMML), liver cancer, lung cancer, non-small cell lung cancer, small cell lung cancer, lung carcinoid tumor, lymphoma, lymphoma of the skin, malignant mesothelioma, multiple myeloma, myelodysplasia syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroblastoma, non-Hodgkin lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumors, prostate cancer, rectum cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, skin cancer, basal and squamous cell cancer, melanoma, merkel cell cancer, small intestine cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, uterine sarcoma, vaginal cancer, vulvar cancer, Waldenstrom macroglobulinemia, or Wilms tumor..

[0279] Item 104. The method of item 100, wherein the disease is osteoporosis. [0280] Item 105. The method of item 104, wherein the fusion protein comprises an extracellular fragment of RANK.

[0281] Item 106. The method of item 100, wherein the disease is an autoimmune disease.

[0282] Item 107. The method of item 100, wherein the disease is Graft-versus-host disease.

[0283] Item 108. The method of item 100, wherein the disease is a viral infection.

Examples

[0284] The invention is further illustrated by the following non-limiting Examples.

[0285] Example 1: Escherichia coli. Cloning and amplification of plasmids was carried out using Escherichia coli DH5a (Invitrogen, Karlsruhe, Germany). The build-up of the respective vectors is depicted in Figs. 17 and 18.

[0286] Example 2: Transfection purification. Transfection (electroporation 1230V; 975 μΕ) of expression vectors encoding main and smaller chains, which can also be referred to as heavy and light chains, of indicated specificities was done in Sp2/0 plasmacytoma cells, obtained from the American Type Culture Collection (ATCC, Manassas, VA). For the buildup of the respective vectors reference is made to Figs. 17 and 18. Cells were cultured in IMDM media, supplemented with 10% fetal calf serum (PAN-Biotech, Aidenbach, Germany), 1 % penicillin and streptomycin (Lonza, Basel, Switzerland). Stable transfectants were selected by adding 1 mg/ml G418 (Invitrogen, Karlsruhe, Germany).

[0287] Bispecific fusion proteins were purified from supernatants of cultures of stably transfected cells via affinity chromatography using KappaSelect (GE Healthcare, Munich, Germany) for (CD3 -Fab-Fcko -NKG2D ; CD 16-Fab-Fcko-NKG2D) and Protein L for (GITR- CD3; GITRxCD16; RANKxCD3; RANKxCD16) (Perbio Science; Bonn, Germany).

[0288] Example 3: Generation of bispecific CD3- or CD 16-Fab-Fc^k0-NKG2D fusion proteins. Immunoglobulin V regions were combined with the desired constant C regions in an expression vector. The cloning procedure indicated here allows the introduction of complete Ig V regions and their expression in lymphoid cells without any alterations of their amino acid sequence. To this end, the nucleotide sequence of a VDJ and VJ fragment of the monospecific CD 16 antibody (clone 3G8) was used to design primer pairs (A A'; D D'; Table 1). The reamplified DNA fragments of the V segments were digested (VJ directly and VDJ after reamplification with primer pair E E' (Table 1) with appropriate restriction nucleases (summarized in Table 1) and then ligated into the expression vectors. Alternatively, the V domains were synthesized as DNA fragments at GeneART, Regensburg, Germany. This method was used for genes coding for the V regions of the humanized antibody directed to CD3 (clone humanized UCHT1). The vectors (Figure 17) contain human heavy and human light constant region genes. Thus, insertion of the amplified and digested V segments reconstitutes the original genomic organization of the Ig genes in the vectors without altering any amino acid of the V regions.

[0289] The original vector for the heavy chain contains the human γΐ isotype Ig heavy chain (Fig. 17A). Restriction sites were introduced at the required positions in introns in order to exchange the Aatll-Clal fragment with the VDJ fragment of the heavy chain of monoclonal antibodies UCHT-1 (anti-CD3) and 3G8 (anti-CD 16) or any other monoclonal antibody. The region relevant for cloning the VDJ fragment is shown enlarged in Figure 17B. The fragment to be exchanged contains parts of the first intron with an Aatll restriction site, the second exon of the leader sequence, the VDJ region and part of the heavy chain intron with the restriction site Clal. For the substitution of all exons of the constant region of the human γΐ heavy chain restriction sites were introduced at the required position in the heavy chain intron (Mlul) and in the 5'-UTR heavy chain polyA-region (pA-region; Spel), as shown in Figure 17A and 17C.

[0290] Furthermore, with the expression vectors constructed, it is possible to exchange the entire constant region of the human Igyl isotype (Mlul-Spel fragment; see Figure 17A) either against constant regions of all other antibody isotypes or against Fc parts with optimized or reduced effector function. In order to generate bispecific fusion proteins as depicted in Fig. 1C, Mlul and Spel flanked DNA fragments containing exons coding for modified constant domains of the Ig heavy chain can be inserted. The Mlul-Spel fragment to be exchanged is shown enlarged in Figure 17C. Adding the second specificity of a bispecific fusion protein, the ecto-domain of NKG2D or ecto-domains of any other type II transmembrane protein can be included via the restriction enzyme sites BspEI and Spel, as also shown in Figure 17A. The region relevant for cloning of the ecto-domain of NKG2D is shown enlarged in Figure 17C. An ecto-domain fragment of NKG2D (F78-V216) was generated by PCR with oligonucleotides G and G' listed in Table 2. The DNA fragment of the NKG2D domain, was digested with the appropriate restriction nucleases (summarized in Table 2) and was then ligated into the expression vector.

[0291] The original vector for the light chain contains the VJ region of the light chain and the C region of human κ gene (Figure 17D). Restriction sites were introduced at the required locations (Xhol and Spel) in order to substitute the light chain Xhol-Spel fragment with the appropriate VJ fragment of the light chain of monoclonal antibodies UCHT-1 (anti- CD3), or 3G8 (anti-CD 16) or any other monoclonal antibody. The region adjacent to the fragment to be exchanged is shown in Figure 17E. This region contains parts of the second exon of the leader sequence, a suitable restriction site (Xhol) for in frame fusion, the VJ region and parts of the kappa chain intron with restriction site Spel. In order to replace the constant domain of the light chain (CL) restriction sites were introduced at the required locations (Pmll and BsmBI). The region adjacent to the fragment to be exchanged is shown enlarged in Figure 17F. This region contains parts of the kappa chain intron, a suitable restriction site (Pmll), the CL region and parts of the 3'-UTR region kappa chain polyA- region (pA-region) with restriction site (BsmBI).

[0292] Thus, bispecific fusion proteins with CD3xNKG2D (UCHTlxNKG2D(F78- V216)) and CD16xNKG2D (UCHTlxNKG2D(F78-V216)) were obtained. Sequences of the corresponding chains are depicted as SEQ ID NO: 48 to SEQ ID NO: 49.

[0293] Co-transfection of the expression vectors encoding the fusion protein main chain and light chain into the non-Ig-producing myeloma cell line Sp2/0 yielded stable transfectomas secreting bispecific fusion proteins which are able to bind specifically to the desired antigen. The functional characterisation of these bispecific molecules is illustrated in the following experiments using CD3xNKG2D and CD16xNKG2D bispecific fusion proteins. Fusion proteins as described herein were used in examples 6, 7, 8, 9, 10, and 11.

Table 1 : Oligonucleotides used for amplification of VDJ and VJ segments and ecto-domains for the insertion into expression vectors

Restriction sites are shown in bold and indicated by letters in parentheses.

Table 2: Oligonucleotides used for amplification of ecto-domains and scFv segments for the insertion into expression vectors

Restriction sites are shown in bold and indicated by letters in parentheses.

[0294] Example 4: Generation of bispecific RANK- or GITR-Fc^k0-CD3 or CD16 fusion proteins.

[0295] The cloning procedure indicated here allows the introduction of ecto-domains of type I transmembrane proteins or domains of soluble proteins and their expression in lymphoid cells. For this the respective domains were combined with the desired constant C regions in an expression vector. To this end, the nucleotide sequence of an ecto-domain fragment was used to design primer pairs (B B' and C C; Table 1). The amplified DNA fragments of the ecto-domain were digested after reamplification with primer pair E E' (Table 1) with appropriate restriction nucleases (summarized in Table 1) and then ligated into the expression vectors. The vectors (Figure 18) contain human heavy and human light constant region genes.

[0296] The original vector for the heavy chain contains the human γΐ isotype Ig heavy chain (Fig. 18A). Restriction sites were introduced at the required positions in introns in order to exchange the Aatll-Clal fragment with the ecto-domain fragment of RANK (Q25- P207) and GITR (Q26-P162) or any other ecto-domain of a type I transmembrane protein or domain of any soluble protein. The region relevant for cloning the domain is shown enlarged in Figure 18B. The fragment to be exchanged contains parts of the first intron with an Aatll restriction site, the second exon of the leader sequence, the ecto-domain and part of the heavy chain intron with the restriction site Clal. For the substitution of all exons of the constant region of the human γΐ heavy chain restriction sites were introduced at the required position in the heavy chain intron (Mlul) and in the 5'-UTR heavy chain polyA-region (pA- region; Spel), as shown in Figure 18A and 18C.

[0297] Furthermore, with the expression vectors constructed, it is possible to exchange the entire constant region of the human Igyl isotype (Mlul-Spel fragment; see Figure 18A) either against constant regions of all other antibody isotypes or against Fc parts with optimized or reduced effector function. In order to generate bispecific fusion proteins as depicted in Figs. 1A and IB, Mlul and Spel flanked DNA fragments containing exons coding for modified constant domains of the Ig heavy chain can be inserted. The Mlul-Spel fragment to be exchanged is shown enlarged in Figure 18C. Adding the second specificity of a bispecific fusion protein, scFv-fragments either in VH-VL or VL-VH orientation can be included via the restriction enzyme sites BspEI and Spel, as also shown in Figure 18A. The region relevant for cloning of a scFv fragment in VL-VH orientation is shown enlarged in Figure 18C. ScFv fragments with the specificity for CD3 (clone humanized UCHT1; VL- VH orientation), were generated by PCR with oligonucleotides F and F' listed in Table 2. Alternatively, they were synthesized as DNA- fragments at GeneArt, Regensburg, Germany. This method was used for genes coding for the antibodies directed to CD 16 (clone 3G8; VL- VH orientation). The DNA fragment of the scFv segments was digested with the appropriate restriction nucleases (summarized in Table 2) and was then ligated into the expression vector.

[0298] Thus, bispecific fusion proteins with GITR(Q26-P162)xCD3, GITR(Q26- P162)xCD16, RANK(Q25-P207)xCD3 and RANK(Q25-P207)xCD16 were obtained. Sequences of the corresponding chains are depicted as SEQ ID NO: 50 to SEQ ID NO 51.

[0299] Transfection of the expression vectors encoding the fusion protein chain into the non-Ig-producing myeloma cell line Sp2/0 yielded stable transfectomas secreting bispecific fusion proteins which are able to bind specifically to the desired antigen. The functional characterization of these molecules is illustrated in the following experiments using GITR(Q26-P162)xCD3, GITR(Q26-P162)xCD16, RANK(Q25-P207)xCD3 and RANK(Q25-P207)xCD16 bispecific fusion proteins. Fusion proteins as described herein were used in examples 6, 7, 9, 10, and 11.

[0300] Example 5: BATDA Europium release assay.

[0301] Cytotoxicity of NK cells was analyzed by a 2-h BATDA Europium release assay. Leukemia cells from patients with >80% blast counts were labeled with a membrane permeable ester of the fluorescence enhancing ligand BATDA (Wallac Oy) which is hydrolyzed when entering the cytoplasm. The hydrophilic ligand TDA results which is no longer capable to penetrate the plasma membrane. Therefore, only TDA of lysed cells is released to the culture supernatant which forms a highly fluorescent chelate complex (EuTDA) with the added Europium solution. Fluorescence intensity correlates with the number of lysed target cells.

[0302] Target cells were labeled with BATDA washed, and placed in 96-well round- bottomed plates at 5,000 per well before addition of NK cells at the indicated E:T ratio. After incubation, 20 of supernatant per well were removed and mixed with 200 DELFIA Europium Solution (Wallac Oy). Cytotoxicity was quantified by measuring the fluorescence of the Europium TDA chelates using a time-resolved fluorometer (VICTOR, Wallac Oy). Maximum release was determined from target cells lysed in 1% Triton X-100. Percentage of lysis was calculated as follows: 100 x (experimental release - spontaneous release) / (maximum release - spontaneous release).

[0303] Example 6: Detection of the respective immune receptor parts of the fusion proteins (Figure 2). 24 well plates were coated either with NKG2D, RANK or GITR mAb (2 μg/mL), blocked with 7.5 % BSA-PBS and washed. Afterwards, cell culture supernatants containing the indicated fusion proteins or media derived from a clone that did not produce fusion protein as negative control were added. Additionally, an IgG antibody or FC-NKG2D (same as FC-NKG2D-ADCC, SEQ ID NO: 54), GITR-FC (same as GITR-FC- ADCC, SEQ ID NO: 55) or RANK-FC (same as RANK-FC-ADCC, SEQ ID NO: 56) (10 μg/mL) fusion protein as indicated served as negative or positive controls, respectively. The fusion proteins FC-NKG2D, GITR-FC or RANK-FC consist of extracellular fragments of NKG2D, GITR, or RANK fused to the Fc-portion of an IgG. Subsequently, plates were again washed and anti-human IgH-HRP (1 : 10,000 in 3.75 % BSA-PBS) was added. Plates were developed using the trimethoxybenzoate hydrochloride peroxidase substrate system (KPL, Gaithersburg, MD), a two component system which develops a deep blue color when reacted with horseradish peroxidase conjugates in ELISA. Absorbance was measured at 450 nm. This figure shows that all tested fusion proteins comprise the respective extracellular domain of the respective immune receptor. Results are depicted in Figure 2. The FC-NKG2D, RANK-FC, and GITR-FC correspond to FC-NKG2D-ADCC, RANK-FC- ADCC, and GITR-FC-ADCC are fusion proteins of extracellular fragments of NKG2D, RANK, or GITR with Fc-fragments that are modified to have enhanced ADCC. These constructs are described in Schmiedel et al, Mol Ther, Schmiedel et al, Cancer Res, Steinbacher et al., Int J Cancer and Raab et al., J Immunol. [0304] Example 7: Binding of the effector arms of the RANK, GITR and NKG2D fusion proteins (Figure 3). Jurkat T cells (Figure 3 A, used with CD3 constructs) or Sp2/0- AG14-CD16 transfectants (Figure 3B, used with CD16 constructs) were incubated for 20 min with 50 cell culture supernatants derived from transfectants producing the indicated fusion proteins with fresh culture medium serving as negative control. After washing, a second staining step with a donkey anti human antibody (1 : 100 for 15 min) was performed. Afterwards cells were washed twice and analyzed by flow cytometry using a FC500 (Beckman Coulter, Krefeld, Germany). Jurkat T cells or Sp2/0-AG14-CD16 transfectants do not express RANKL nor GITRL nor NKG2L: This figure shows that all fusion proteins bind to CD3 or CD 16 via their antibody fragment moiety (anti-CD3 or anti- CD 16, respectively). Results are depicted in Figure 3.

[0305] Example 8: Dose dependent binding of bispecific CD16 constructs (CD16- NKG2D) as compared to Fc fusion proteins with optimized IgGl-Fc-part (FC-NKG2D) (Figure 4). NK cells (CD 16+) were incubated with the indicated concentrations of the two constructs (FC-NKG2D, Figure 4A, same as FC-NKG2D-ADCC, SEQ ID NO: 54; CD16- NKG2D, Figure 4B, SEQ ID NO: 49, co-expressed with antiCD16 Fab light chain SEQ ID NO: 09) for 30 min in 50 FACS buffer. After washing, a second staining step with a donkey anti human antibody (1 : 100 for 15 min) was performed. Afterwards cells were washed twice and analyzed by flow cytometry using a FC500 (Beckman Coulter, Krefeld, Germany). This figure shows that the immune receptor-antiCD16 fusion proteins have higher affinity to FcR than the even ADCC-optimized Fc-moiety of the immune receptor- FC-ADCC fusion protein. The FC-NKG2D proteins corresponds to FC-NKG2D-ADCC and are fusion proteins of an extracellular fragment of NKG2D with Fc-fragments that are modified to have enhanced ADCC. These constructs are described in Steinbacher et al, Int J Cancer and Raab et al., J Immunol.

[0306] Example 9: Comparative analysis of the effects of CD16-fusion proteins versus Fc-optimized constructs (Figure 5). Primary NKG2DL (Figure 5A), RANKL (Figure 5B) and GITRL (Figure 5C) expressing leukemic cells from patients with AML were incubated with peripheral blood mononuclear cells (PBMC) of healthy donors in the presence or absence of the indicated FC-optimized or bispecific CD 16 constructs (Figure 5 A: CD16-NKG2D versus NKG2D-ADCC (same as FC-NKG2D-ADCC, SEQ ID NO: 54); Figure 5B: RANK-CD 16 (SEQ ID NO: 51) versus RANK-FC-ADCC (same as RANK-FC- ADCC, SEQ ID NO: 56); Figure 5C: GITR-CD16 (SEQ ID NO: 53) versus GITR-FC- ADCC (same as GITR-FC-ADCC, SEQ ID NO: 55)). The FC-NKG2D-ADCC, RANK-FC- ADCC, and GITR-FC-ADCC are fusion proteins of extracellular fragments of NKG2D, RANK, or GITR with Fc-fragments that are modified to have enhanced ADCC. These constructs are described in Schmiedel et al, Mol Ther, Schmiedel et al, Cancer Res, Steinbacher et al., Int J Cancer and Raab et al., J Immunol. CD16-NKG2D represents the fusion protein with the sequence as set forth in SEQ ID NO: 49 co-expressed with an anti- CD16-Fab light chain as set forth in SEQ ID NO: 09. Then the effects on cytotoxicity of NK cells were analyzed by 2h BATDA Europium release assays. To this end, leukemia cells from patients with >80 % blast counts were labeled with BATDA (Wallac Oy), washed, and placed in 96-well round-bottom plates at 5,000 per well prior to addition of PBMC. After incubation for 2 h, 20 of supernatant per well were removed and mixed with 200 DELFIA Europium Solution (Wallac Oy). Cytotoxicity was quantified by measuring the fluorescence of the Europium TDA chelates using a time-resolved fluorometer (VICTOR, Wallac Oy). Maximum release was determined from target cells lysed in 1 % Triton X-100. Percentage of lysis was calculated as follows: 100 x (experimental release - spontaneous release) / (maximum release - spontaneous release). This Figure shows that the immune receptor-antiCD16 fusion proteins achieve stronger NK cell mediated cell lysis compared to the absence of these fusion proteins and also to employing immune receptor-FC-ADCC fusion proteins.

[0307] Example 10: Activation of T cells by the bispecific CD3-fusion protein (Figure 6). 0.5xl0⁶ PBMCs of healthy donors were cultured in the presence of 0.5xl0⁶ leukemic cells from AML patients (either expressing NKG2DL (Figure 6A), RANKL (Figure 6 A) or GITRL (Figure 6 A)) for 48 h with the indicated bispecific CD3 constructs or medium as control. T cells were selected by staining with directly labeled antibodies against CD4 and CD8. Upregulation of CD69 as a marker for T cell activation was analyzed by FACS using specific fluorescence-conjugates (all antibodies from Becton Dickinson). This Figure shows that the immune receptor-antiCD3 constructs promote cytotoxic T-cell activation in the presence of NKG2DL-, RANKL- or GITRL-positive cells. CD3-NKG2D represents the fusion protein with the sequence as set forth in SEQ ID NO: 48 co-expressed with an anti-CD3-Fab light chain as set forth in SEQ ID NO: 07, RANK-CD3 represents a fusion protein with a sequence set forth in SEQ ID NO: 50, and GITR-CD3 represents a fusion protein with a sequence set forth in SEQ ID NO: 52.

[0308] Example 11: Induction of leukemia cell lysis by T cells by the bispecific CD3 constructs and target antigen-restriction of the induced effects (Figure 7). 5x10⁶ PBMC of a patient with AML (leukemia cell content 80 %) in which the leukemic cells expressed NKG2DL but were negative for GITRL (not shown) were cultured in triplicates in medium containing 10 % autologous patient serum in the presence or absence of the indicated constructs (1 μg/mL). For assessment of T-cell mediated killing of AML blasts, cells were washed after incubation for 72 hours in FACS-buffer containing 50 μg/mL human IgG (Flebogamma, Grifols, Langen, Germany), stained with a CD34 antibody to select the leukemic cells and finally resuspended in FACS-buffer containing 7-AAD (BioLegend, San Diego, USA) and negative control compensation particles (BD Biosciences). Malignant cells were defined as CD34+CD45dim. The percentage of apoptotic (7-AAD positive) cells is given in the dot plots. This experiment shows that the CD3- NKG2D constructs specifically directs T-cell mediated cytotoxicity to NKG2DL expressing cells. CD3-NKG2D represents the fusion protein with the sequence as set forth in SEQ ID NO: 48 co-expressed with an anti-CD3-Fab light chain as set forth in SEQ ID NO: 07, and GITR-CD3 represents a fusion protein with a sequence set forth in SEQ ID NO: 52.

[0309] Example 12: Lysis of leukemia cells by autologous T and NK cells upon exposure to CD3-NKG2D and CD16-NKG2D, respectively. PBMC of a leukemia patient with AML cells expressing NKG2DL were directly cultured after isolation in medium containing 10 % autologous patient serum in the presence or absence of the indicated constructs (1 μg/mL each). For assessment of T-cell and NK cell mediated killing of AML blasts, cells were washed after 72 hours in FACS-buffer containing 50 μg/mL human IgG (Flebogamma, Grifols, Langen, Germany), stained with a CD34 antibody to select the leukemic cells and finally resuspended in FACS-buffer containing 7-AAD (BioLegend, San Diego, USA) and quantification beads (BD Biosciences). Malignant cells were defined as CD34+. Results obtained with 3 different AML patients indicated as percent of viable cells as compared to an untreated control (100%) are shown in Figure 37. This experiment shows that CD16-NKG2D potently induces AML cell lysis by stimulating NK cells, and this was, in line with their higher effector and proliferative potential, by far exceeded upon stimulation of T cells with CD3-NKG2D. CD3-NKG2D represents the fusion protein with the sequence as set forth in SEQ ID NO: 48 co-expressed with an anti-CD3-Fab light chain as set forth in SEQ ID NO: 07, and CD16-NKG2D represents a fusion protein with a sequence set forth in SEQ ID NO: 49 co-expressed with an anti-CD 16-Fab light chain as set forth in SEQ ID NO: 09. These results are remarkable as it demonstrates that the fusion constructs mediate killing of AML blasts by autologous T and NK cells. This example thus shows the therapeutic efficacy of the fusion constructs in an ex vivo setting that essentially and directly mimics the in vivo conditions in a leukemic patient. [0310] Example 13: Titration of bispecific CD16- and CD3-NKG2D fusion proteins to determine saturating concentrations. T cells (Figure 38 A) and NK cells (Figure 38B) of a healthy donor were incubated with the indicated concentrations of CD3- NKG2D and CD16-NKG2D, respectively, followed by anti-human-PE conjugate and then counterstained with CD4/CD8 for T cells or CD56/CD3 for NK cells followed by FACS analysis. This shows that both constructs reach saturating concentrations at 100 pmol/ml.

[0311] Example 14: Specificity of CD16-NKG2D & CD3-NKG2D with regard to their effector parts. T cells (Figure 39 A) and NK cells (right panel) of a healthy donor were incubated with CD3-NKG2D and CD16-NKG2D (100 pmol/ml) followed by an anti- human-PE conjugate and then counterstained for CD4/CD8 or CD56/CD3 to determine specific binding by FACS analysis. This figure shows that the CD3-NKG2D specifically binds to T cells and not NK cells, whereas the CD16-NKG2D construct specifically binds to NK cells and not T cells.

[0312] Example 15: Proliferation inducing capacity of CD3-NKG2D constructs. PBMC from a healthy donor were incubated with NALM16 (NKG2DL⁺, FLT3⁺) leukemia cells as target cells in the presence of the indicated constructs (^g/ml) for 3d. Subsequently the number of CD8+ (Figure 40A) and CD4+ (Figure 40B) T cells was determined by FACS using quantification beads (BD Pharmingen). This experiment demonstrates that the CD3- NKG2D construct is able to induce proliferation of T cells depending on binding of the construct to target cells, a particularly important feature when aiming to treat patients with substantial tumor mass. A bispecific FLT3-CD3 antibody served as positive control. As expected, no effect on T cell proliferation was observed with the CD16-NKG2D construct (which does not stimulate T cells) or a bispecific CD19-CD3 antibody serving as negative control due to lack of target antigen expression on the leukemic cells.

[0313] The invention illustratively described herein may suitably be practiced in the absence of any element or elements, limitation or limitations, not specifically disclosed herein. Thus, for example, the terms "comprising", "including," containing", etc. shall be read expansively and without limitation. Additionally, the terms and expressions employed herein have been used as terms of description and not of limitation, and there is no intention in the use of such terms and expressions of excluding any equivalents of the features shown and described or portions thereof, but it is recognized that various modifications are possible within the scope of the invention claimed. Thus, it should be understood that although the present invention has been specifically disclosed by exemplary embodiments and optional features, modification and variation of the inventions embodied therein herein disclosed may be resorted to by those skilled in the art, and that such modifications and variations are considered to be within the scope of this invention.

[0314] The invention has been described broadly and generically herein. Each of the narrower species and subgeneric groupings falling within the generic disclosure also form part of the invention. This includes the generic description of the invention with a proviso or negative limitation removing any subject matter from the genus, regardless of whether or not the excised material is specifically recited herein.

[0315] Other embodiments are within the following claims. In addition, where features or aspects of the invention are described in terms of Markush groups, those skilled in the art will recognize that the invention is also thereby described in terms of any individual member or subgroup of members of the Markush group.

[0316] The content of all documents and patent documents cited herein is incorporated by reference in their entirety.

Scientific References cited in the present application

Adams GP & Weiner LM. Monoclonal antibody therapy of cancer, Nat. Biotechnol.

2005;23(9)1147-57.

Baessler T, et al. Glucocorticoid-lnduced TNFR-Related Protein (GITR) ligand subverts immunosurveillance of Acute Myeloid Leukemia in humans. Cancer Res. 2009;69(3): 1037-45.

Baltz KM, et al. Cancer Immunoediting by GITR (Glucocorticoid-lnduced TNF Related Protein) Ligand in Humans: NK Cell / Tumor Cell Interactions. FASEB J. 2007;21(10):2442-54.

Baltz KM, et al. Neutralization of tumor-derived soluble Glucocorticoid-lnduced TNFR- Related (GITR) ligand increases NK cell anti-tumor reactivity. Blood 2008;112(9):3735-43.

Bargou R, et al. Tumor Regression in Cancer Patients by Very Low Doses of a T Cell- Engaging Antibody. Science 2008;321(5891):974-7.

Bianchini G & Gianni L. The immune system and response to HER2 -targeted treatment in breast cancer. Lancet, 2014;15(2):e58-68.

Buechele C et al. Glucocorticoid-induced TNFR-related protein (GITR) ligand modulates cytokine release and NK cell reactivity in chronic lymphocytic leukemia (CLL). Leukemia, 2012;26(5):991-1000.

Eigenbrot C et al . X-ray structures of the antigen-binding domains from three variants of humanized anti-pl85HER2 antibody 4D5 and comparison with molecular modeling. J. Mol. Biol. 1993;229(4):969-95.

Fanger MW et al. Bispecific antibodies and targeted cellular cytotoxicity. Immunol Today 1991;12(2):51-4.

Groh V, et al. Tumour-derived soluble MIC ligands impair expression of NKG2D and T-cell activation. Nature, 2002;419(6908):734-8.

Holt LJ, et al. Domain antibodies: proteins for therapy. Trends Biotechnol. 2003;21(11):484- 90.

Ill CR, et al. Design and construction of a hybrid immunoglobulin domain with properties of both heavy and light chain variable regions. Protein Eng. 1997;10(8):949-57.

Kellner C et al. Heterodimeric bispecific antibody-derivatives against CD 19 and CD 16 induce effective antibody-dependent cellular cytotoxicity against B-lymphoid tumor cells Cancer Lett. 2011 ;303(2): 128-39.

Khort HE et al. CD 137 stimulation enhances the antilymphoma activity of anti-CD20 antibodies. Blood. 2011;117(8):2423-32.

Khort HE et al. Stimulation of natural killer cells with a CD137-specific antibody enhances trastuzumab efficacy in xenotransplant models of breast cancer. J Clin Invest. 2012;122(3): 1066-75. Khort HE et al. Targeting CD 137 enhances the efficacy of cetuximab. J Clin Invest. 2014;124(6):2668-82.

Leibbrandt A & Penninger JM. RANK/RANKL: regulators of immune responses and bone physiology. Ann N Y Acad Sci. 2008;1143: 123-50.

Martin Fl, et al. The affinity- selection of a minibody polypeptide inhibitor of human interleukin-6. EMBO J. 1994;13(22):5303-9).

Morea V et al. Antibody structure, prediction and redesign. Biophys . Chem. 1997;68 (l-3):9- 16.

Ni J, et al. Sustained effector function of IL-12/15/18-preactivated NK cells against established tumors. J. Exp. Med. 2012;209(13):2351-2365.

Pegram HJ et al. Activating and inhibitory receptors of natural killer cells Immunology and Cell Biology. 2011 ;89:216-224.

Raab S et al. Fc-optimized NKG2D-Fc constructs induce NK cell antibody-dependent cellular cytotoxicity against breast cancer cells independently of HER2/neu expression status. J. Immunol. 2014;193(8):4261-72.

Raulet DH, et al. Regulation of ligands for the NKG2D activating receptor. Annu Rev Immunol 2013;31 :413-41.

Reckzo M et al. Prediction of hypervariable CDR-H3 loop structures in antibodies. Protein Eng. 1995;8(4):389-95.

Salih HR, et al. Cutting Edge: Down-regulation of MICA on human tumors by proteolytic shedding. J Immunol. 2002;169(8):4098-102.

Schiweck W & Skerra A. The rational construction of an antibody against cystatin: lessons from the crystal structure of an artificial Fab fragment. J. Mol. Biol. 1997;268(5):934- 51.

Schmiedel BJ et al. A "vicious cycle" of NK-cell immune evasion in acute myeloid leukemia mediated by RANKL? Onco immunology 2013; 2(5):e23850.

Schmiedel BJ et al. RANKL expression, function, and therapeutic targeting in multiple myeloma and chronic lymphocytic leukemia. Cancer Res. 2013;73(2):683-94.

Schmiedel BJ, et al. Receptor Activator for NF-κΒ Ligand in Acute Myeloid Leukemia:

Expression, Funkction and Modulation of NK cell Immunosurveillance. J Immunol. 2013;190(2):821-31.

Schmiedel BJ et al. Generation and preclinical characterization of a Fc-optimized GITR-Ig fusion protein for induction of NK cell reactivity against leukemia. Mol Ther. 2013;21(4):877-86.

Shirai H et al . Structural classification of CDR-H3 in antibodies. FEBS Lett. 1996;399(1- 2): l-8.

Shirai H et al. H3-rules: identification of CDR-H3 structures in antibodies. FEBS Lett;1999.

455(l-2): 188-97. Silverman Jl, et al. Multivalent avimer proteins evolved by exon shuffling of a family of human receptor domains. Nat Biotechnol. 2005 ;23( 12): 1556-61.

Spear P, et al. NKG2D ligands as therapeutic targets. Cancer Immun. 2013; 13:8.

Staerz UD and Bevan MJ. Hybrid hybridoma producing a bispecific monoclonal antibody that can focus effector T-cell activity. Proc Natl Acad Sci U S A 1986;83(5): 1453-7.

Steinbacher J et al. An Fc-optimized NKG2D-immunoglobulin G fusion protein for induction of natural killer cell reactivity against leukemia. Int. J. Cancer. 2014;136(5): 1073-84.

Tibben JG et al. Pharmacokinetics, biodistribution and biological effects of intravenously administered bispecific monoclonal antibody OC/TR F(ab')2 in ovarian carcinoma patients. Int. J. Cancer 1996;66(4):477-83.

Topp MS, et al. Targeted therapy with the T-cell-engaging antibody blinatumomab of chemotherapy-refractory minimal residual disease in B-lineage acute lymphoblastic leukemia patients results in high response rate and prolonged leukemia-free survival. J Clin Oncol. 2011;29(18):2493-8.

Traunecker A, et al. Bispecific single chain molecules (Janusins) target cytotoxic lymphocytes on HIV infected cells. EMBO J. 1991;10(12):3655-9.

Traunecker A, et al. new molecular design for bispecific reagents. Int J Cancer Suppl.

1992;7:51-2.

Zhang T & Sentmann CL. Cancer Immunotherapy Using a Bispecific NK Receptor Fusion Protein that Engages both T Cells and Tumor Cells. Cancer Res. 2011;71(6):2066-76.

List of Figures

Figure 1 : Schematic representation of non-exhaustive illustrative embodiments of fusion proteins of the present invention.

Figure 2: Detection of the respective immune receptor parts of the fusion proteins.

Figure 3 : Binding of the effector arms of the RANK, GITR and NKG2D fusion proteins.

Figure 4: Dose dependent binding of bispecific CD16 constructs (CD16-NKG2D) as compared to Fc fusion proteins with optimized IgGl-Fc-part (FC-NKG2D).

Figure 5: Comparative analysis of the effects of CD16-fusion proteins versus Fc-optimized constructs.

Figure 6: Activation of T cells by the bispecific CD3-fusion proteins.

Figure 7: Induction of leukemia cell lysis by T cells by the bispecific CD3 constructs and target antigen-restriction of the induced effects.

Figure 8: Modification on linker elements.

Figure 9: Amino acid sequence of fusion protein with N-terminal anti-CD3 Fab heavy chain und C-terminal NKG2D fragment (SEQ ID NO: 48).

Figure 10: Amino acid sequence of humanized light chain of CD3 specific antibody UCHTl

(SEQ ID NO: 07).

Figure 11 : Amino acid sequence of fusion protein with N-terminal anti-CD 16 Fab heavy chain und C-terminal NKG2D fragment (SEQ ID NO: 49).

Figure 12: Amino acid sequence of chimeric light chain of CD16 specific antibody 3G8

(SEQ ID NO: 09).

Figure 13: Amino acid sequence of fusion protein with N-terminal RANK fragment and C- terminal anti-CD3-scFv (SEQ ID NO: 50

Figure 14: Amino acid sequence of fusion protein with N-terminal RANK fragment and C- terminal anti-CD 16-scFv (SEQ ID NO: 51).

Figure 15: Amino acid sequence of fusion protein with N-terminal GITR and C-terminal humanized anti-CD3-scFv (SEQ ID NO: 52).

Figure 16: Amino acid sequence of fusion protein with N-terminal GITR fragment and C- terminal anti-CD 16-scFv (SEQ ID NO: 53).

Figure 17: CD3 or CD 16 NKG2D fusion proteins with attenuated FcR-binding.

Figure 18: RANK and GITR CD3 or CD 16 fusion proteins with attenuated FcR-binding.

Figure 19: cDNA sequences of anti-CD 16 Fab-NKG2D fusion protein and anti-CD3 Fab- NKG2D.

Figure 20: cDNA sequence of RANK-antiCD16 scFv fusion protein and RANK-antiCD3 scFv fusion protein. Figure 21 : cDNA sequence of GITR-antiCD16 scFv fusion protein and GITR-antiCD3 scFv fusion protein.

Figure 22: Amino acid sequence of FC-NKG2D- ADCC (SEQ ID NO: 54)

Figure 23: Amino acid sequence of RANK-FC- ADCC (SEQ ID NO:56)

Figure 24: Amino acid sequence of GITR-FC-ADCC (SEQ ID NO: 55)

Figure 25: cDNA sequence of FC-NKG2D- ADCC (SEQ ID NO: 108)

Figure 26 : cDNA sequence of RANK-FC- ADCC (SEQ ID NO : 110)

Figure 27: cDNA sequence of GITR-FC-ADCC (SEQ ID NO: 109)

Figure 28: Variations of Fab fragments comprised in the fusion proteins.

Figure 29: Amino acid and cDNA sequence of antiCD3-VH-CH.

Figure 30: Amino acid and cDNA sequence of antiCD3-VH-CL.

Figure 31 : Amino acid and cDNA sequence of antiCD3-VL-CHl-hinge-CH2-NKG2D.

Figure 32: Amino acid and cDNA sequence of antiCD3-VL-CL-hinge-CH2-NKG2D.

Figure 33: Amino acid and cDNA sequence of antiCDl 6- VH-CH. Figure 34: Amino acid and cDNA sequence of antiCD16-VH-CL.

Figure 35: Amino acid and cDNA sequence of antiCD16-VL-CHl-hinge-CH2-NKG2D.

Figure 36: Amino acid and cDNA sequence of antiCD16-VL-CL-hinge-CH2-NKG2D.

Figure 37: Lysis of leukemia cells by autologous T and NK cells upon exposure to CD3- NKG2D and CD16-NKG2D, respectively.

Figure 38: Titration of bispecific CD16- and CD3-NKG2D fusion proteins to determine saturating concentrations.

Figure 39: Specificity of CD16-NKG2D & CD3-NKG2D with regard to their effector parts.

Figure 40: Proliferation inducing capacity of CD3-NKG2D constructs.

List of Examples

Example 1 : Escherichia coli.

Example 2: Transfection purification.

Example 3: Generation of bispecific CD3- or CD16-Fab-Fcko-NKG2D fusion proteins. Example 4: Generation of bispecific RANK- or GITR-Fcko-CD3 or CD 16 fusion proteins. Example 5 : BATDA Europium release assay.

Example 6: Detection of the respective immune receptor parts of the fusion proteins

(Figure 2).

Example 7: Binding of the effector arms of the RANK, GITR and NKG2D fusion proteins

(Figure 3).

Example 8: Dose dependent binding of bispecific CD 16 constructs (CD16-NKG2D) as compared to Fc fusion proteins with optimized IgGl-Fc-part (FC-NKG2D) (Figure 4).

Example 9: Comparative analysis of the effects of CD16-fusion proteins versus Fc- optimized constructs (Figure 5).

Example 10: Activation of T cells by the bispecific CD3-fusion protein (Figure 6).

Example 11 : Induction of leukemia cell lysis by T cells by the bispecific CD3 constructs and target antigen-restriction of the induced effects (Figure 7).

Example 12: Lysis of leukemia cells by autologous T and NK cells upon exposure to CD3- NKG2D and CD16-NKG2D, respectively (Figure 37).

Example 13: Titration of bispecific CD 16- and CD3-NKG2D fusion proteins to determine saturating concentrations (Figure 38).

Example 14: Specificity of CD16-NKG2D & CD3-NKG2D with regard to their effector parts (Figure 39).

Example 15: Proliferation inducing capacity of CD3-NKG2D constructs (Figure 40).

Claims

What is claimed is:

1. A recombinant fusion protein consisting of:

2. The fusion protein of claim 1, wherein the recombinant fusion protein does not comprise immunoglobulin heavy chain comprising VH-H-CH1-CH2-CH3 or VH-H- CH1-CH2 CH3-CH4.

3. The fusion protein of claim 1 or 2, wherein the linking polypeptide consists of a CH2 domain and optionally a hinge region.

4. The fusion protein of any one of claims 1-3, wherein the linking polypeptide consists of a CH2 domain and a hinge region.

5. The fusion protein of any one of claims 1-4, wherein the linking polypeptide comprises at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, or 124 consecutive amino acids corresponding to SEQ ID NO: 01 or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, or 124 consecutive amino acids of SEQ ID NO: 01.

6. The fusion protein of any one of claims 1-4, wherein the linking polypeptide comprises at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70,

71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93,

94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112,

113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, or 124 consecutive amino acids corresponding to SEQ ID NO: 02 or to SEQ ID NO: 03 or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71,

72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94,

95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113,

114, 115, 116, 117, 118, 119, 120, 121, 122, 123, or 124 consecutive amino acids of SEQ ID NO: 02 or to SEQ ID NO: 03.

7. The fusion protein of any one of claims 1-6, wherein the at least a portion of a CH2 domain and optionally hinge region acting as linking polypeptide comprises a mutation or deletion in at least one amino acid residue that is able to mediate binding to Fc receptors.

8. The fusion protein of claim 7, wherein the at least one amino acid residue of the CH2 domain and the hinge region is selected from the group consisting of sequence position 233, 234, 235, 236, 265, 297, 327, and 330 (numbering of sequence positions according to the EU-index).

9. The fusion protein of claim 8, wherein the at least one amino acid mutation or deletion is selected from the group consisting of Glu233Pro, Leu234Val, Leu235Ala, deletion of Gly236, Asp265Gly, Asn297Gln, Ala327Gln, and Ala330Ser.

10. The fusion protein of any one of claims 1-9, wherein the at least a portion of a CH2 domain and optionally the hinge region acting as linking polypeptide comprises a mutation or deletion in at least one amino acid residue of the the CH2 domain and the hinge region that is able to mediate dimerization of immunoglobulins.

11. The fusion protein of claim 10, wherein the mutated or deleted at least one amino acid residue of the hinge region is selected from the group consisting of sequence position 220, 226, and 229 (numbering of sequence positions according to the EU-index).

12. The fusion protein of claim 11 , wherein the at least one amino acid mutation or deletion is selected from the group consisting of Cys220Ser, Cys226Ser, or Cys229Ser.

13. The fusion protein of any one of claims 1-12, wherein the binding protein is an antibody molecule.

14. The fusion protein of claim 13, wherein the antibody molecule has a single antigen binding site.

15. The fusion protein of claim 13 or 14, wherein the antibody molecule is a Fab fragment or a scFv fragment or a single domain antibody molecule.

16. The fusion protein of any one of claims 1-15, wherein the binding protein binds to CD3, CD16, CD28, CD137/4-1BB, OX40, Nkp44, Nkp30, Nkp40 or Nkp46.

17. The fusion protein of any one of claims 1-16, wherein the binding protein binds to CD3 or CD 16.

18. The fusion protein of claim 17, wherein the binding protein binds to CD3 on T cells.

19. The fusion protein of claim 18, wherein the binding protein binds to CD3 on cytotoxic T cells.

20. The fusion protein of claim 17, wherein the binding protein binds to CD 16 on natural killer (NK) cells.

21. The fusion protein of any one of the preceding claims, wherein the ligand or antigen that binds to the extracellular fragment of the immune receptor is a tumor-associated ligand or antigen.

22. The fusion protein of any one of the preceding claims, wherein the extracellular fragment of the immune receptor binds to a ligand or antigen on the target cell or to a ligand or antigen that is soluble.

23. The fusion protein of claim 22, wherein the extracellular fragment of the immune receptor binds to a ligand or antigen on the surface of a target cell.

24. The fusion protein of claim 23, wherein the ligand or antigen is a NKG2DL or RANKL or GITRL.

25. The fusion protein of any of the preceding claims, wherein the extracellular fragment of an immune receptor is an extracellular fragment of an immune receptor selected from the group of NKG2D (UniProtKB accession number P26718, SEQ ID NO: 10), RANK (UniProtKB accession number Q9Y6Q6, SEQ ID NO: 11) or GITR (UniProtKB accession number Q9Y5U5, SEQ ID NO: 12), CD94-NKG2 (UniProtKB accession number Q 13241, SEQ ID NO: 13), CTLA-4 (UniProtKB accession number PI 6410, SEQ ID NO: 15), PD-1 (UniProtKB accession number Q15116, SEQ ID NO: 16), BTLA (UniProtKB accession number Q7Z6A9, SEQ ID NO: 17), LAG-3 (UniProtKB accession number PI 8627, SEQ ID NO: 18), TIM-3 (UniProtKB accession number Q8TDQ0, SEQ ID NO: 19), LAIR-1 (UniProtKB accession number Q6GTX8, SEQ ID NO: 20), TIGIT (UniProtKB accession number Q495A1, SEQ ID NO: 21), Siglecl (UniProtKB accession number Q9BZZ2, SEQ ID NO: 22), Siglec2 (UniProtKB accession number P20273, SEQ ID NO: 23), Siglec3 (UniProtKB accession number P20138, SEQ ID NO: 24), Siglec4 (UniProtKB accession number P20916, SEQ ID NO: 25), Siglec5 (UniProtKB accession number 015389, SEQ ID NO: 26), Siglec6 (UniProtKB accession number 043699, SEQ ID NO: 27), Siglec7 (UniProtKB accession number Q9Y286, SEQ ID NO: 28), Siglec8 (UniProtKB accession number Q9NYZ4, SEQ ID NO: 29), Siglec9 (UniProtKB accession number Q9Y336, SEQ ID NO: 30), SigleclO (UniProtKB accession number Q96LC7, SEQ ID NO: 31), Siglecl 1 (UniProtKB accession number Q96RL6, SEQ ID NO: 32), Siglecl2 (UniProtKB accession number Q96PQ1, SEQ ID NO: 33), Siglecl4 (UniProtKB accession number Q08ET2, SEQ ID NO: 34), Siglecl5 (UniProtKB accession number Q6ZMC9, SEQ ID NO: 35), Siglecl6 (UniProtKB accession number A6NMB1, SEQ ID NO: 36), NKp30 (UniProtKB accession number 014931, SEQ ID NO: 14), NKp40 (UniProtKB accession number Q9NZS2, SEQ ID NO: 85), NKp44 (UniProtKB accession number 095944, SEQ ID NO: 86), NKp46 (UniProtKB accession number 076036, SEQ ID NO: 87), NKp80 (UniProtKB accession number Q9NZS2, SEQ ID NO: 88), OPG (UniProtKB accession number 000300, SEQ ID NO: 89).

26. The fusion protein of claim 25, wherein the extracellular fragment of an immune receptor is from an extracellular fragment of NKG2D (UniProtKB accession number P26718), RANK (UniProtKB accession number Q9Y6Q6) or GITR (UniProtKB accession number Q9Y5U5).

27. The fusion protein of claim 26, wherein the extracellular fragment is of NKG2D and has the ability to bind to at least a portion of at least one NKG2DL, preferably alternatively to more than one NKG2DL, most preferably to any NKG2DL, wherein the group of NKG2DL comprises MICA (UniProtKB accession number Q29983), MICB (UniProtKB accession number Q29980), ULBP1 (UniProtKB accession number Q9BZM6), ULBP2 (UniProtKB accession number Q9BZM5), ULBP3 (UniProtKB accession number Q9BZM4), ULBP4 (UniProtKB accession number Q8TD07), ULBP5 (UniProtKB accession number Q6H3X3) and ULBP6 (UniProtKB accession number Q5VY80).

28. The fusion protein of claim 26 or 27, wherein the extracellular fragment is of NKG2D comprises at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, or 139 consecutive amino acids corresponding to SEQ ID NO: 45 or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, or 139 consecutive amino acids of SEQ ID NO: 45.

29. The fusion protein of claim 26, wherein the extracellular fragment is of RANK and has the ability to bind to at least a portion of RANKL (UniProtKB accession number 014788).

30. The fusion protein of claim 26 or 29, wherein the extracellular fragment is of RANK comprises at least 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, or 183 consecutive amino acids corresponding to SEQ ID NO: 46 or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, or 183 consecutive amino acids of SEQ ID NO: 46.

31. The fusion protein of claim 26, wherein the extracellular fragment is of GITR and has the ability to bind to at least a portion of GITRL (UniProtKB accession number Q9UNG2).

32. The fusion protein of claim 26 or 31, wherein the extracellular fragment is of GITR comprises at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136 or 137 consecutive amino acids corresponding to SEQ ID NO: 47 or comprises a sequence having a pairwise sequence identity of at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 % or at least about 99 % when aligned to at least 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 134, 135, 136 or 137 consecutive amino acids of SEQ ID NO: 47.

33. The fusion protein of any one of the preceding claims, wherein the binding protein is located N-terminal of the linker polypeptide and the extracellular fragment of a transmembrane protein is located C-terminal to the linker polypeptide.

34. The fusion protein of any one of the preceding claims, wherein the binding protein is located N-terminal of the linker polypeptide and the extracellular fragment of a transmembrane protein is located C-terminal to the linker polypeptide, wherein said transmembrane protein is a type II transmembrane protein.

35. The fusion protein of claim 33 or 34, wherein the extracellular fragment of the transmembrane protein is an extracellular fragment of NKG2D and has the ability to bind to at least one NKG2DL, preferably alternatively to more than one NKG2DL, most preferably to any NKG2DL.

36. The fusion protein of any one of claims 33-35, wherein the binding protein is an antibody molecule.

37. The fusion protein of claim 36, wherein the antibody molecule is a Fab fragment or a scFv fragment or a single domain antibody molecule that binds to CD3.

38. The fusion protein of claim 37, wherein the antibody molecule binds to CD3 on T cells.

39. The fusion protein of claim 37, wherein the antibody molecule is a Fab fragment that binds to CD3.

The fusion protein of claim 39, wherein the Fab fragment binds to CD3 on T cells.

41. The fusion protein of claim 39, wherein the Fab fragment heavy chain comprises a sequence set forth in SEQ ID NO: 06.

42. The fusion protein of claim 39, wherein the Fab fragment light chain comprises a sequence set forth in SEQ ID NO: 07.

43. The fusion protein of any one of claims 39-42, wherein the immune receptor fragment and the at least a portion of the CH2 domain and optionally the hinge region are fused with a heavy chain or a light chain of a Fab fragment.

44. The fusion protein of claim 43, wherein the immune receptor fragment and the at least a portion of the CH2 domain and optionally the hinge region are fused with a heavy chain of a Fab fragment.

45. The fusion protein of claim 44, wherein the fusion protein comprises an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 48.

46. The fusion protein of claim 45, wherein the fusion protein comprises a Fab fragment light chain comprising an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 07.

47. The fusion protein of claim 36, wherein the antibody molecule is a Fab fragment or a scFv fragment or a single domain antibody molecule that binds to CD 16.

48. The fusion protein of claim 47, wherein the antibody molecule binds to CD16 on natural killer (NK) cells.

49. The fusion protein of claim 47, wherein the antibody molecule is a Fab fragment that binds to CD 16.

50. The fusion protein of claim 49, wherein the Fab fragment binds to CD 16 on natural killer (NK) cells.

51. The fusion protein of claim 49, wherein the Fab fragment heavy chain comprises a sequence set forth in SEQ ID NO: 08.

52. The fusion protein of claim 49, wherein the Fab fragment light chain comprises a sequence set forth in SEQ ID NO: 09.

53. The fusion protein of claim 49, wherein the immune receptor fragment and the at least a portion of the CH2 domain and optionally the hinge region are fused with a heavy chain or a light chain of a Fab fragment.

54. The fusion protein of claim 53, wherein the immune receptor fragment and the at least a portion of the CH2 domain and optionally the hinge region are fused with a heavy chain of a Fab fragment.

55. The fusion protein of claim 54, wherein the fusion protein comprises a polypeptide having an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 49.

56. The fusion protein of claim 55, wherein the fusion protein comprises a Fab fragment light chain comprising an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 09.

57. The fusion protein of any one of claims 1-32, wherein the binding protein is located C- terminal to of the linker polypeptide and the extracellular fragment of a transmembrane protein is located N-terminal to the linker polypeptide.

58. The fusion protein of any one of claims 1-32, wherein the binding protein is located C- terminal to of the linker polypeptide and the extracellular fragment of a transmembrane protein is located N-terminal to the linker polypeptide, wherein said transmembrane protein is a type I transmembrane protein.

59. The fusion protein of claim 58 or 59, wherein the extracellular fragment of the transmembrane protein is selected from either an extracellular fragment of RANK that has the ability to bind at least a portion of RANKL or an extracellular fragment of GITR that has the ability to bind to at least a portion of GITRL.

60. The fusion protein of any one of claims 57-59, wherein the binding protein is an antibody molecule.

61. The fusion protein of claim 60, wherein the antibody molecule is a scFv fragment or a single domain antibody molecule that binds to CD3.

62. The fusion protein of claim 61, wherein the antibody molecule binds to CD3 on T cells.

63. The fusion protein of claim 61, wherein the antibody molecule is a scFv fragment that binds to CD3.

64. The fusion protein of claim 63, wherein the scFv fragment binds to CD3 on T cells.

65. The fusion protein of claim 63, wherein the scFv fragment comprises a sequence as set forth in SEQ ID NO: 04.

66. The fusion protein of any one of claims 63-65, wherein the fusion protein comprises an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 50 or SEQ ID NO: 52.

67. The fusion protein of claim 60, wherein the antibody molecule is a scFv fragment or a single domain antibody molecule that binds to CD 16.

68. The fusion protein of claim 67, wherein the antibody molecule binds to CD16 on natural killer (NK) cells.

69. The fusion protein of claim 67, wherein the antibody molecule is a scFv fragment that binds to CD 16.

70. The fusion protein of claim 69, wherein the scFv fragment binds to CD 16 on natural killer (NK) cells.

71. The fusion protein of claim 69, wherein the scFv fragment comprises a sequence as set forth in SEQ ID NO: 05.

72. The fusion protein of claim 69, wherein the fusion protein comprises a polypeptide having an amino acid sequence having at least about 80 %, at least about 85 %, at least about 90 %, at least about 95 %, at least about 98 %, at least about 99 % or about 100 % sequence identity in a pairwise sequence alignment with the sequence as set forth in SEQ ID NO: 51 or SEQ ID NO: 53.

73. The fusion protein of any one of the preceding claims, wherein the target cell is a tumor cell.

74. The fusion protein of any one of the preceding claims, wherein the fusion protein binds to a ligand that is soluble.

75. The fusion protein of claim 74, wherein binding prevents binding of the ligand to other immune receptors.

76. The fusion protein of claim 75, wherein binding neutralizes a physiological or pathophysiological effect of the ligand.

77. The fusion protein of claim 76, wherein binding neutralizes an immunomodulatory effect of the ligand.

78. The fusion protein of claim 76, wherein binding neutralizes an immune inhibitory effect of the ligand.

79. The fusion protein of any of claims 74-78, wherein the fusion protein binds to soluble NKG2DL or RANKL or GITRL.

80. The fusion protein of claim 79, wherein binding prevents binding of NKG2DL to other NKG2D, or prevents binding of RANKL to other RANK, or prevents binding of GITRL to other GITR.

81. The fusion protein of claim 80, wherein binding neutralizes a physiological or pathophysiological effect of NKG2DL or RANKL or GITRL.

82. The fusion protein of 81, wherein binding neutralizes an immunomodulatory effect of NKG2DL or an immunomodulatory effect of RANKL or an immunomodulatory effect of GITRL.

83. The fusion protein of 82, wherein the immunomodulatory effect is an immune inhibitory effect.

84. The fusion protein of 81, wherein binding of RANKL neutralizes an effect of RANKL in bone resorption.

85. A pharmaceutical composition comprising a fusion protein as defined in claims 1-84.

86. A fusion protein as defined in any one of claims 1-84 for use in the treatment of a disease.

87. The fusion protein for use of claim 86, wherein the disease is a proliferative disease.

88. The fusion protein for use of claim 87, wherein the proliferative disease is cancer.

89. The fusion protein for use of claim 88, wherein the cancer is selected from the group consisting of adrenal cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, brain and spinal cord tumors, breast cancer, Castleman disease, cervical cancer, colon cancer, endometrial cancer, esophagus cancer, Ewing family of tumors, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumors, gastrointestinal stromal tumor (GIST), gestational trophoblastic disease, Hodgkin disease, Kaposi sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL), chronic myeloid leukemia (CML), chronic myelomonocytic leukemia (CMML), liver cancer, lung cancer, non-small cell lung cancer, small cell lung cancer, lung carcinoid tumor, lymphoma, lymphoma of the skin, malignant mesothelioma, multiple myeloma, myelodysplasia syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroblastoma, non-Hodgkin lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumors, prostate cancer, rectum cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, skin cancer, basal and squamous cell cancer, melanoma, merkel cell cancer, small intestine cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, uterine sarcoma, vaginal cancer, vulvar cancer, Waldenstrom macroglobulinemia, or Wilms tumor.

90. The fusion protein for use of claim 86, wherein the disease is osteoporosis.

91. The fusion protein for use of claim 90, wherein the fusion protein comprises an extracellular fragment of RANK.

92. The fusion protein for use of claim 86, wherein the disease is an autoimmune disease.

93. The fusion protein for use of claim 86, wherein the disease is graft-versus-host disease.

94. The fusion protein for use of claim 86, wherein the disease is a viral infection.

95. A nucleic acid encoding for the fusion protein of any of claims 1-84.

A nucleic acid of claim 95 comprised in a vector.

97. A host cell comprising the nucleic acid molecule of claim 95 or a vector of claim 96.

I l l

98. A method of producing the fusion protein of any of claims 1-84, comprising using the nucleic acid encoding the fusion protein for expression of the fusion protein under conditions allowing expression of the fusion protein.

99. The method of claim 98, wherein the fusion protein is expressed by a host cell or in a cell- free system.

100. A method of treating a disease comprising administering a therapeutically effective amount of the fusion protein as defined in any of claims 1-84 to a subject.

101. The method of claim 100, wherein the disease is a proliferative disease.

102. The method of claim 101, wherein the proliferative disease is cancer.

103. The method of claim 102, wherein the cancer is selected from the group consisting adrenal cancer, anal cancer, bile duct cancer, bladder cancer, bone cancer, brain and spinal cord tumors, breast cancer, Castleman disease, cervical cancer, colon cancer, endometrial cancer, esophagus cancer, Ewing family of tumors, eye cancer, gallbladder cancer, gastrointestinal carcinoid tumors, gastrointestinal stromal tumor (GIST), gestational trophoblastic disease, Hodgkin disease, Kaposi sarcoma, kidney cancer, laryngeal and hypopharyngeal cancer, leukemia, acute lymphocytic leukemia (ALL), acute myeloid leukemia (AML), chronic lymphocytic leukemia (CLL), chronic myeloid leukemia (CML), chronic myelomonocytic leukemia (CMML), liver cancer, lung cancer, non-small cell lung cancer, small cell lung cancer, lung carcinoid tumor, lymphoma, lymphoma of the skin, malignant mesothelioma, multiple myeloma, myelodysplasia syndrome, nasal cavity and paranasal sinus cancer, nasopharyngeal cancer, neuroblastoma, non-Hodgkin lymphoma, oral cavity and oropharyngeal cancer, osteosarcoma, ovarian cancer, pancreatic cancer, penile cancer, pituitary tumors, prostate cancer, rectum cancer, retinoblastoma, rhabdomyosarcoma, salivary gland cancer, sarcoma, skin cancer, basal and squamous cell cancer, melanoma, merkel cell cancer, small intestine cancer, stomach cancer, testicular cancer, thymus cancer, thyroid cancer, uterine sarcoma, vaginal cancer, vulvar cancer, Waldenstrom macroglobulinemia, or Wilms tumor.

104. The method of claim 100, wherein the disease is osteoporosis.

105. The method of claim 104, wherein the fusion protein comprises an extracellular fragment of RANK.

106. The method of claim 100, wherein the disease is an autoimmune disease.

107. The method of claim 100, wherein the disease is Graft-versus-host disease.

108. The method of claim 100, wherein the disease is a viral infection.