A fusion protein containing a ligand-receptor pair and a biologically functional protein.

Abstract

The present disclosure provides fusion proteins with multifunctional biological designs for programmed target engagement. In certain embodiments, the fusion proteins described herein provide simultaneous target antigen engagement and immune checkpoint or costimulatory receptor targeting. In certain aspects, the fusion proteins are masked from exhibiting any on-target / off-tissue effects (i.e., toxicity) associated with target engagement. In certain embodiments, the fusion proteins provide a masked antigen-binding domain and a masked immunomodulatory target-binding domain, such that programmed activation of one binding functionality also results in activation of the other binding functionality, thereby resulting in a bispecific molecule. Thus, the present disclosure also provides methods for masking and conditional activation of antigen-binding domains in specific target tissue environments, as well as targeting and activating immunomodulatory targets without adverse toxic effects.

Description

Background Art

[0001] Background With the development of monoclonal antibodies and other biological agents as drugs, it is possible to design highly specific and targeted therapeutic agents. However, the use of these agents is often hampered by the fact that most molecular targets that can serve as markers for diseased cells such as cancer are also found in non-diseased (normal) cells in the patient's body, albeit with some difference in expression. As a result, when an active targeted biomolecule is used as a therapeutic agent, it may exhibit unintended activity outside the sites where therapeutic benefit is expected, potentially causing toxicity and unwanted side effects. This is also called on-target / off-tumor (also known as on-target / off-tissue) action, which affects not only the balance between drug efficacy and toxicity but also the dosing regimen. On-target / off-tumor action can cause unintended uptake of the therapeutic agent by non-diseased cells and accelerated clearance, resulting in an unfavorable pharmacokinetic profile of the therapeutic agent, also called target-mediated drug disposition (TMDD). Therefore, due to these issues, there is a need for a therapeutic design feature that can not only provide high specificity for the molecular target but also conditionally and locally act on diseased cells / tissues while avoiding drug action on the same target expressed in tissues other than the tumor.

[0002] By targeting the immune checkpoint pathway, a sustained therapeutic response can be provided through the active engagement of the patient's immune system via either positive or negative costimulatory molecules. Unfortunately, therapies targeting the checkpoint pathway also have difficulties with drug toxicity problems mediated by the target and clearance issues. There is also a growing recognition that simultaneous targeting of multiple of these checkpoints and / or costimulatory pathways, or the combination of these checkpoint targets with other non-immune-related targets and therapies, can reactivate a more effective immune response. Therefore, there is great interest in therapeutic strategies related to targeting checkpoints, but immune-related adverse events (irAEs), namely, issues related to toxicity and clearance, still remain. Designs that result in conditional engagement of therapeutic agents may provide a less toxic and more effective solution to targeting immunomodulatory molecules.

Summary of the Invention

[0003] A fusion protein comprising a biologically functional protein, a ligand-receptor pair, a first peptide linker and a second peptide linker, wherein the biologically functional protein comprises at least a first polypeptide and a second polypeptide, the ligand-receptor pair comprises an extracellular portion of an immunoglobulin superfamily (IgSF) receptor and its cognate ligand or a receptor-binding fragment thereof, the ligand is fused via the first peptide linker to the end of the first polypeptide, the receptor is fused via the second peptide linker to the same end of each of the second polypeptides, and the first and second peptide linkers are of sufficient length to allow pairing of the ligand and the receptor, is described herein. In some embodiments, at least one of the first and second peptide linkers comprises a protease cleavage site. In certain embodiments, the ligand is fused via the first peptide linker to the N-terminus of the first polypeptide and the receptor is fused via the second peptide linker to the N-terminus of the second polypeptide.

[0004] In certain embodiments, the biologically functional protein comprises an antibody or an antigen-binding antibody fragment. In certain embodiments, the biologically functional protein consists of a polypeptide backbone. In certain embodiments, the polypeptide backbone is a dimeric Fc region, where the first polypeptide consists of a first Fc polypeptide, the second polypeptide consists of a second Fc polypeptide, and the first and second Fc polypeptides form a dimeric Fc region. In certain embodiments, the biologically functional protein comprises a polypeptide backbone.

[0005] In certain embodiments, the polypeptide backbone includes a dimeric Fc region. In certain embodiments, the dimeric Fc region is a heterodimeric Fc. In certain embodiments, at least one of the ligand or receptor of the ligand-receptor pair is capable of binding to an immunomodulatory target.

[0006] In some embodiments, ligand-receptor pairs are involved in cellular responses selected from the group consisting of modulation of immune checkpoints, modulation of immune cell activity, modulation of T cell receptor signaling, modulation of T cell-dependent cell-mediated cytotoxicity (TDCC), modulation of antibody-dependent cell phagocytosis (ADCP), and modulation of antibody-dependent cell-mediated cytotoxicity (ADCC). In some embodiments, the receptor contains one or more mutations that increase or decrease the receptor's binding affinity to its homologous ligand compared to the wild-type receptor.

[0007] In some embodiments, the ligand includes one or more mutations that increase or decrease the ligand's binding affinity to its homologous receptor compared to the wild-type ligand. In certain embodiments, the ligand-receptor pair is selected from the group consisting of PD1-PDL1, PD1-PDL2, CTLA4-CD80, CD28-CD80, CD28-CD86, CTLA4-CD86, PDL1-CD80, ICOS-ICOSL, NCRSRLG1-NKp30, and CD47-SIRPa. In certain embodiments, the ligand-receptor pair is PD1-PDL1. In certain embodiments, the ligand PDL1 includes the amino acid sequence described in SEQ ID NO: 8. In certain embodiments, the receptor PD1 includes the amino acid sequence described in SEQ ID NO: 9.

[0008] In certain embodiments, the ligand-receptor pair is CTLA4-CD80. In certain embodiments, the ligand CD80 comprises the amino acid sequence described in SEQ ID NO: 25, SEQ ID NO: 185, SEQ ID NO: 187, or SEQ ID NO: 189. In certain embodiments, the receptor CTLA4 comprises the amino acid sequence described in SEQ ID NO: 26.

[0009] In certain embodiments, the receptor and ligand are fused to the N-terminuses of the first and second polypeptides, respectively. In certain embodiments, one of the first or second peptide linkers includes a plurality of protease cleavage sites. In certain embodiments, one of the peptide linkers fused to the ligand or receptor is engineered to include one or more additional protease cleavage sites, and one or more protease cleavage sites of the ligand or receptor and the protease cleavage sites of the first or second peptide linker are cleavable by the same protease or different proteases.

[0010] In a particular embodiment, the proteases are serine protease, MMP1, MMP2, MMP3, MMP7, MMP8, MMP9, MMP10, MMP11, MMP12, MMP13, MMP14, MMP15, MMP16, MMP17, MMP18 (collagenase 4), MMP19, MMP20, MMP21, adamaricin, ceralicin, astacin, caspase 1, caspase 2, caspase 3, caspase 4, caspase 5, caspase 6, caspase 7, caspase 8, caspase 9, caspase 1 The group is selected from 0, caspase 11, caspase 12, caspase 13, caspase 14, cathepsin A, cathepsin B, cathepsin D, cathepsin E, cathepsin K, cathepsin S, granzyme B, guanidinobenzoatase (GB), hepsin, elastase, regmine, matryptase, matryptase 2, meprin, neurosin, MT-SP1, neprilysin, plasmin, PSA, PSMA, TACE, TMPRSS3, TMPRSS4, uPA, calpain, FAP, and KLK. In certain embodiments, the protease is uPA or matryptase.

[0011] In certain embodiments, the peptide linker is 3-50 or 5-20 amino acid long. In certain embodiments, one of the first or second peptide linkers does not have a protease cleavage site. In certain embodiments, the peptide linker is (Gly n Ser) is a linker, and here (Gly n Ser) Linker is (Gly3Ser) n (Gly4Ser)1, (Gly3Ser)1(Gly4Ser) n (Gly3Ser) n (Gly4Ser) n , and (Gly4Ser) n The formula includes an amino acid sequence selected from the group consisting of (wherein n is an integer from 1 to 5). In a particular embodiment, the peptide linker is (EAAAK) nThe linker is (wherein n is an integer from 1 to 5). In certain embodiments, the peptide linker includes the amino acid sequence EAAAKEAAAK (SEQ ID NO: 38). In certain embodiments, the peptide linker is a polyproline linker, optionally PPP or PPPP. In certain embodiments, the peptide linker includes an immunoglobulin hinge region sequence containing an amino acid sequence having up to 30 percent difference in amino acid sequence identity compared to the amino acid sequence of the wild-type immunoglobulin hinge region. In certain embodiments, the peptide linker includes a protease cleavage site containing the amino acid sequence MSGRSANA (SEQ ID NO: 28).

[0012] Also described herein is a fusion protein comprising a Fab region and an Fc region, wherein the Fab region comprises a VH polypeptide and a VL polypeptide forming an antigen-binding domain, and a ligand-receptor pair comprising the extracellular portion of an immunoglobulin superfamily receptor and its congener ligand or receptor-binding fragment thereof, wherein the ligand is fused to the N-terminus of one of the VH or VL polypeptides via a first peptide linker, and the receptor is fused to the N-terminus of the other VH or VL polypeptide via a second peptide linker, wherein the first and second peptide linkers are long enough to enable ligand-receptor pairing, and at least one of the first and second peptide linkers comprises a protease cleavage site, and the ligand-receptor pair sterically inhibits the binding of the antigen-binding domain to its congener antigen.

[0013] In some embodiments, at least one of the first and second polypeptides comprises a first VH polypeptide and a first VL polypeptide, the first VH and VL polypeptides forming a first antigen-binding domain of the antibody, the ligand being fused to one of the first VH or VL polypeptides via a first peptide linker, and the receptor being fused to the other of the first VH or VL polypeptides via a second peptide linker, the ligand-receptor pair sterically inhibiting the binding of the first antigen-binding domain to its homologous antigen. In certain embodiments, the first and second polypeptides further comprise a dimer Fc. In certain embodiments, the dimer Fc region is a heterodimer Fc.

[0014] In certain embodiments, the fusion protein comprises ligand-linker-VL, receptor-linker-VL, ligand-linker-VH, or receptor-linker-VH from the N-terminus to the C-terminus.

[0015] In certain embodiments, the fusion protein comprises, from N-terminus to C-terminus, ligand-cleaving linker-VL, receptor-cleaving linker-VL, ligand-cleaving linker-VH, or receptor-cleaving linker-VH.

[0016] In certain embodiments, the fusion protein comprises, from N-terminus to C-terminus, ligand-linker(sequence number 114)-VL, receptor-linker(sequence number 114)-VL, ligand-linker(sequence number 14)-VH, or receptor-linker(sequence number 14)-VH.

[0017] In certain embodiments, the fusion protein comprises, from N-terminus to C-terminus, ligand-linker(sequence ID 145)-VL, receptor-linker(sequence ID 145)-VL, ligand-linker(sequence ID 145)-VH, or receptor-linker(sequence ID 145)-VH.

[0018] In certain embodiments, the fusion protein comprises ligand-linker(sequence number 147)-VL, receptor-linker(sequence number 147)-VL, ligand-linker(sequence number 147)-VH, or receptor-linker(sequence number 147)-VH from the N-terminus to the C-terminus.

[0019] In certain embodiments, the fusion protein comprises ligand-linker(sequence number 154)-VL, receptor-linker(sequence number 154)-VL, ligand-linker(sequence number 154)-VH, or receptor-linker(sequence number 154)-VH from the N-terminus to the C-terminus.

[0020] In certain embodiments, the fusion protein comprises ligand-linker(sequence number 203)-VL, receptor-linker(sequence number 203)-VL, ligand-linker(sequence number 203)-VH, or receptor-linker(sequence number 203)-VH, from the N-terminus to the C-terminus.

[0021] In certain embodiments, at least one of the ligand or receptor in a ligand-receptor pair is capable of binding to an immunomodulatory target. In certain embodiments, the ligand-receptor pair is involved in cellular responses selected from the group consisting of modulation of immune checkpoints, modulation of immune cell activity, modulation of T cell receptor signaling, modulation of T cell-dependent cell-mediated cytotoxicity (TDCC), modulation of antibody-dependent cell phagocytosis (ADCP), and modulation of antibody-dependent cell-mediated cytotoxicity (ADCC).

[0022] In certain embodiments, the receptor includes one or more mutations that increase or decrease the receptor's binding affinity to its homologous ligand compared to the wild-type receptor. In certain embodiments, the ligand includes one or more mutations that increase or decrease the ligand's binding affinity to its homologous receptor compared to the wild-type ligand. In certain embodiments, the ligand-receptor pair is selected from the group consisting of PD1-PDL1, PD1-PDL2, CTLA4-CD80, CD28-CD80, CD28-CD86, CTLA4-CD86, PDL1-CD80, ICOS-ICOSL, NCRSRLG1-NKp30, and CD47-SIRPa. In certain embodiments, the ligand-receptor pair is PD1-PDL1. In certain embodiments, the ligand PDL1 includes the amino acid sequence described in SEQ ID NO: 8. In certain embodiments, the receptor PD1 includes the amino acid sequence described in SEQ ID NO: 9. In certain embodiments, the ligand-receptor pair is CTLA4-CD80. In certain embodiments, the ligand CD80 includes the amino acid sequence described in SEQ ID NO: 25. In a particular embodiment, the receptor CTLA4 comprises the amino acid sequence described in SEQ ID NO: 26.

[0023] In some embodiments, the receptor and ligand are fused to the N-terminus of the first and second polypeptides, respectively. In certain embodiments, one of the first or second peptide linkers includes multiple protease cleavage sites. In certain embodiments, one of the ligand or receptor is engineered to include one or more additional protease cleavage sites, and one or more protease cleavage sites of the ligand or receptor and the protease cleavage sites of the first or second peptide linker are cleavable by the same protease or different proteases.

[0024] In certain embodiments, the protease is selected from the group consisting of serine protease, MMP1, MMP2, MMP3, MMP7, MMP8, MMP9, MMP10, MMP11, MMP12, MMP13, MMP14, MMP15, MMP16, MMP17, MMP18 (collagenase 4), MMP19, MMP20, MMP21, adamalysin, serralysin, astacin, caspase 1, caspase 2, caspase 3, caspase 4, caspase 5, caspase 6, caspase 7, caspase 8, caspase 9, caspase 10, caspase 11, caspase 12, caspase 13, caspase 14, cathepsin A, cathepsin B, cathepsin D, cathepsin E, cathepsin K, cathepsin S, granzyme B, guanidino benzoatase (GB), hepsin, elastase, legumain, matriptase, matriptase 2, meprin, neurosin, MT-SP1, neprilysin, plasmin, PSA, PSMA, TACE, TMPRSS3, TMPRSS4, uPA, calpain, FAP and KLK. In certain embodiments, the protease is uPA or matriptase. In certain embodiments, the peptide linker is 3-50 or 5-20 amino acids in length. In certain embodiments, one of the first or second peptide linkers does not have a protease cleavage site. In certain embodiments, the peptide linker is a (Gly n Ser) linker, where the (Gly n Ser) linker comprises an amino acid sequence selected from the group consisting of (Gly3Ser) n (Gly4Ser)1, (Gly3Ser)1(Gly4Ser) n , (Gly3Ser) n (Gly4Ser) n , and (Gly4Ser) n (where n is an integer from 1 to 5). In certain embodiments, the peptide linker is (EAAAK) nThe linker is (wherein n is an integer from 1 to 5). In certain embodiments, the peptide linker without a protease cleavage site comprises the amino acid sequence EAAAKEAAAK (SEQ ID NO: 38). In certain embodiments, the peptide linker is a polyproline linker, optionally PPP or PPPP. In certain embodiments, the linker is a glycine (G) proline (P) polypeptide linker, optionally GPPPG, GGPPPGG, GPPPPG, or GGPPP P In certain embodiments, the peptide linker includes an immunoglobulin hinge region sequence containing an amino acid sequence having up to 30 percent difference in amino acid sequence identity compared to the amino acid sequence of the wild-type immunoglobulin hinge region. In certain embodiments, the peptide linker containing a protease cleavage site includes the amino acid sequence MSGRSANA (SEQ ID NO: 28).

[0025] In certain embodiments, the binding of the first antigen-binding domain to its congener antigen is reduced by more than 10 times compared to the parent antigen-binding domain that is not fused to the ligand-receptor pair. In certain embodiments, cleavage of the protease cleavage site in the cellular environment releases one member of the ligand-receptor pair from the fusion protein, thereby enabling the antigen-binding domain to bind to its congener antigen.

[0026] In certain embodiments, the first antigen-binding domain is Fab. In certain embodiments, the first antigen-binding domain binds to an antigen expressed on cancer cells or immune cells. In certain embodiments, the first antigen-binding domain binds to an antigen expressed on T cells. In certain embodiments, the first antigen-binding domain binds to tumor-associated antigens (TAAs). In certain embodiments, the first antigen-binding domain binds to an antigen selected from the group consisting of differentiation antigen group 3 (CD3), human epidermal growth factor receptor 2 (HER2), epidermal growth factor receptor (EGFR), mesothelin (MSLN), tissue factor (TF), differentiation antigen group 19 (CD19), tyrosine protein kinase Met (c-Met), differentiation antigen group 40 (CD40), and cadherin 3 (CDH3).

[0027] In certain embodiments, the antibody or antibody fragment comprises a second antigen-binding domain comprising a second VH polypeptide and a second VL polypeptide. In certain embodiments, the fusion protein comprises a second ligand-receptor pair, the ligand of the second ligand-receptor pair being fused to one of the second VH or VL polypeptides via a third peptide linker, and the receptor of the second ligand-receptor pair being fused to the other of the second VH or VL polypeptide via a fourth peptide linker, at least one of the third and fourth peptide linkers comprising a protease cleavage site, and the ligand-receptor pair sterically inhibits the binding of the second antigen-binding domain to its congener antigen. In certain embodiments, the fusion protein binds to two different antigens. In certain embodiments, one antigen is expressed by T cells and the other antigen is expressed by cancer cells. In certain embodiments, the fusion protein binds to CD3 and HER2.

[0028] Also described herein are fusion proteins comprising an Fc region containing a first Fc polypeptide and a second Fc polypeptide, and a ligand-receptor pair containing the extracellular portion of an immunoglobulin superfamily receptor and its homologous ligand or receptor-binding fragment, wherein the ligand is fused to the terminal of the first Fc polypeptide via a first peptide linker, and the receptor is fused to each of the similar terminals of the second Fc polypeptide via a second peptide linker, wherein the first and second peptide linkers are long enough to allow ligand-receptor pairing, and at least one of the first and second peptide linkers contains a protease cleavage site.

[0029] [Invention 1001] A fusion protein comprising a biologically functional protein, a ligand-receptor pair, a first peptide linker, and a second peptide linker, The biologically functional protein comprises at least a first polypeptide and a second polypeptide, and The ligand-receptor pair comprises the extracellular component of an immunoglobulin superfamily receptor and its homologous ligand or its receptor-binding fragment. The ligand is fused to the terminal end of the first polypeptide via the first peptide linker. The receptor is fused to the same end of each of the second polypeptides via the second peptide linker, the first and second peptide linkers being long enough to allow ligand-receptor pairing, and at least one of the first and second peptide linkers containing a protease cleavage site. The aforementioned fusion protein. [Invention 1002] The fusion protein of the present invention 1001, comprising the ligand and the receptor, the extracellular component of an immunoglobulin superfamily (IgSF) polypeptide. [Invention 1003] The fusion protein of the present invention 1001, wherein the ligand and the receptor comprise the extracellular portion of an immunoglobulin variable (IgV) polypeptide. [Invention 1004] A fusion protein according to any one of the invention 1001 to 1003, wherein the biologically functional protein comprises an antibody or an antigen-binding antibody fragment. [Invention 1005] The fusion protein of the present invention 1001, wherein the biologically functional protein comprises a polypeptide backbone. [Invention 1006] The fusion protein of the present invention 1005, wherein the polypeptide backbone is a dimeric Fc region, the first polypeptide consists of a first Fc polypeptide, and the second polypeptide consists of a second Fc polypeptide, and the first and second Fc polypeptides form a dimeric Fc region. [Invention 1007] The fusion protein of the present invention 1001, wherein the biologically functional protein comprises a polypeptide backbone. [Invention 1008] The fusion protein of the present invention 1007, wherein the polypeptide backbone includes a dimeric Fc region. [Invention 1009] A fusion protein according to the present invention 1006 or 1008, wherein the dimeric Fc region is heterodimeric Fc. [Invention 1010] A fusion protein according to any of the present invention, wherein at least one of the ligand or receptor of the ligand-receptor pair is capable of binding to an immunomodulatory target. [Invention 1011] The fusion protein of the present invention, wherein the ligand receptor pair is involved in a cellular response selected from the group consisting of regulation of immune checkpoints, regulation of immune cell activity, regulation of T cell receptor signaling, regulation of T cell-dependent cell-mediated cytotoxicity (TDCC), regulation of antibody-dependent cell phagocytosis (ADCP), and regulation of antibody-dependent cell-mediated cytotoxicity (ADCC). [Invention 1012] A fusion protein according to the present invention, wherein the receptor comprises one or more mutations that increase or decrease the receptor's binding affinity to its homologous ligand compared to the wild-type receptor. [Invention 1013] A fusion protein according to the present invention, wherein the ligand contains one or more mutations that increase or decrease the binding affinity of the ligand to its homologous receptor compared to the wild-type ligand. [Invention 1014] The fusion protein of the present invention, wherein the ligand-receptor pair is selected from the group consisting of PD1-PDL1, PD1-PDL2, CTLA4-CD80, CD28-CD80, CD28-CD86, CTLA4-CD86, PDL1-CD80, ICOS-ICOSL, NCRSRLG1-NKp30, and CD47-SIRPa. [Invention 1015] The fusion protein of the present invention 1014, wherein the ligand-receptor pair is PD1-PDL1. [Invention 1016] The fusion protein of the present invention 1015, wherein the ligand PDL1 contains the amino acid sequence described in SEQ ID NO: 8. [Invention 1017] The fusion protein of the present invention 1015 or 1016, wherein the receptor PD1 contains the amino acid sequence described in SEQ ID NO: 9. [Invention 1018] The fusion protein of the present invention 1014, wherein the ligand-receptor pair is CTLA4-CD80. [Invention 1019] The fusion protein of the present invention 1018, wherein the ligand CD80 comprises the amino acid sequence described in SEQ ID NO: 25, SEQ ID NO: 185, SEQ ID NO: 187, or SEQ ID NO: 189. [Invention 1020] The fusion protein of the present invention 1018 or 1019, wherein the receptor CTLA4 comprises the amino acid sequence described in SEQ ID NO: 26. [Invention 1021] The fusion protein of the present invention 1014, wherein the ligand-receptor pair is selected from the group consisting of CTLA4-CD80, PDL1-CD80, and CD28-CD80, and the ligand CD80 contains the amino acid sequence described in SEQ ID NO: 25, having a mutation selected from the group consisting of (a) H18Y, A26E, E35D, M47S, I61S, and D90G; (b) E35D, M47S, N48K, I61S, K89N; (c) E35D, D46V, M47S, I61S, D90G, K93E; or (d) H18Y, A26E, E35D, M47S, I61S, V68M, A71G, D90G; (e) I58S, V68S, L70S; (f) M47S, I61S, or (g) V22S. [Invention 1022] A fusion protein according to the present invention, wherein the receptor and the ligand are fused to the N-terminus of the first and second polypeptides, respectively. [Invention 1023] A fusion protein according to the present invention, wherein one of the first or second peptide linkers includes a plurality of protease cleavage sites. [Invention 1024] A fusion protein according to the present invention, wherein one of the peptide linkers fused to the ligand or the receptor is manipulated to include one or more additional protease cleavage sites, and the one or more protease cleavage sites of the ligand or the receptor and the protease cleavage sites of the first or second peptide linker are cleavable by the same protease or different proteases. [Invention 1025] The aforementioned proteases are serine protease, MMP1, MMP2, MMP3, MMP7, MMP8, MMP9, MMP10, MMP11, MMP12, MMP13, MMP14, MMP15, MMP16, MMP17, MMP18 (collagenase 4), MMP19, MMP20, MMP21, adamaricin, ceralicin, astacin, caspase 1, caspase 2, caspase 3, caspase 4, caspase 5, caspase 6, caspase 7, caspase 8, caspase 9, caspase 10, caspase 11, caspase A fusion protein of the present invention selected from the group consisting of 12, caspase 13, caspase 14, cathepsin A, cathepsin B, cathepsin D, cathepsin E, cathepsin K, cathepsin S, granzyme B, guanidinobenzoatase (GB), hepsin, elastase, regmine, matryptase, matryptase 2, meprin, neurosin, MT-SP1, neprilysin, plasmin, PSA, PSMA, TACE, TMPRSS3, TMPRSS4, uPA, calpain, FAP, and KLK. [Invention 1026] The fusion protein of the present invention 1025, wherein the protease is uPA or matryptase. [Invention 1027] The fusion protein of the present invention, wherein the peptide linker has a length of 3 to 50 or 5 to 20 amino acids. [Invention 1028] A fusion protein according to the present invention, wherein one of the first or second peptide linkers does not have a protease cleavage site. [Invention 1029] The aforementioned peptide linker, (Gly n Ser) is a linker, and here (Glyn Ser) Linker is (Gly 3 Ser) n (Gly 4 Ser) 1 , (Gly 3 Ser) 1 (Gly 4 Ser) n , (Gly 3 Ser) n (Gly 4 Ser) n , and (Gly 4 Ser) n A fusion protein according to any of the above-described inventions, comprising an amino acid sequence selected from the group consisting of (wherein n is an integer from 1 to 5). [Invention 1030] The aforementioned peptide linker is (EAAAK) n A fusion protein according to any of the above-described inventions, which is a linker (wherein n is an integer from 1 to 5). [Invention 1031] The fusion protein of the present invention 1030, wherein the peptide linker contains the amino acid sequence EAAAKEAAAK (SEQ ID NO: 38). [Invention 1032] The fusion protein according to any of the present inventions, wherein the peptide linker is a polyproline linker, optionally PPP or PPPP, or a glycine-proline linker, optionally GPPPG, GGPPPGG, GPPPPG, or GGPPPPGG. [Invention 1033] A fusion protein according to the present invention, wherein the peptide linker comprises an immunoglobulin hinge region sequence having an amino acid sequence identity difference of up to 30 percent compared to the amino acid sequence of the wild-type immunoglobulin hinge region. [Invention 1034] The fusion protein of the present invention, wherein the peptide linker includes a protease cleavage site containing the amino acid sequence MSGRSANA (SEQ ID NO: 28). [Invention 1035] A fusion protein according to any one of the present invention 1001 to 1004, wherein at least one of the first and second polypeptides comprises a first VH polypeptide and a first VL polypeptide, the first VH and VL polypeptides form a first antigen-binding domain of the antibody, the ligand is fused to one of the first VH or VL polypeptides via a first peptide linker, and the receptor is fused to the other of the first VH or VL polypeptide via a second peptide linker, and the ligand-receptor pair sterically inhibits the binding of the first antigen-binding domain to its homologous antigen. [Invention 1036] The fusion protein of the present invention 1035, wherein the first and second polypeptides further comprise a dimer Fc. [Invention 1037] The fusion protein of the present invention 1036, wherein the aforementioned dimeric Fc region is a heterodimeric Fc. [Invention 1038] A fusion protein according to any one of the present invention 1035 to 1037, wherein at least one of the ligand or receptor of the ligand-receptor pair is capable of binding to an immunomodulatory target. [Invention 1039] A fusion protein according to any of items 1035 to 1038 of the present invention, wherein the ligand receptor pair is involved in a cellular response selected from the group consisting of regulation of immune checkpoints, regulation of immune cell activity, regulation of T cell receptor signaling, regulation of T cell-dependent cell-mediated cytotoxicity (TDCC), regulation of antibody-dependent cell phagocytosis (ADCP), and regulation of antibody-dependent cell-mediated cytotoxicity (ADCC). [Invention 1040] A fusion protein according to any one of the present invention 1035 to 1038, wherein the receptor includes one or more mutations that increase or decrease the binding affinity of the receptor to its homologous ligand compared to the wild-type receptor. [Invention 1041] A fusion protein according to any of the present invention 1035 to 1040, wherein the ligand contains one or more mutations that increase or decrease the binding affinity of the ligand to its homologous receptor compared to the wild-type ligand. [Invention 1042] A fusion protein according to any of the invention items 1035 to 1041, wherein the ligand-receptor pair is selected from the group consisting of PD1-PDL1, PD1-PDL2, CTLA4-CD80, CD28-CD80, CD28-PDL1, CD28-CD86, CTLA4-CD86, PDL1-CD80, ICOS-ICOSL, NCRSRLG1-NKp30, and CD47-SIRPa. [Invention 1043] The fusion protein of the present invention 1042, wherein the ligand-receptor pair is PD1-PDL1. [Invention 1044] The fusion protein of the present invention 1043, wherein the ligand PD-L1 contains the amino acid sequence described in SEQ ID NO: 8. [Invention 1045] The fusion protein of the present invention 1043 or 1044, wherein the receptor PD1 comprises the amino acid sequence described in SEQ ID NO: 9. [Invention 1046] The fusion protein of the present invention 1042, wherein the ligand-receptor pair is CTLA4-CD80. [Invention 1047] The fusion protein of the present invention 1046, wherein the ligand CD80 contains the amino acid sequence described in SEQ ID NO: 25. [Invention 1048] The fusion protein of the present invention 1042, wherein the ligand-receptor pair is selected from the group consisting of CTLA4-CD80, PDL1-CD80, and CD28-CD80, and the ligand CD80 contains the amino acid sequence described in SEQ ID NO: 25, having a mutation selected from the group consisting of (a) H18Y, A26E, E35D, M47S, I61S, and D90G; (b) E35D, M47S, N48K, I61S, K89N; (c) E35D, D46V, M47S, I61S, D90G, K93E; (d) H18Y, A26E, E35D, M47S, I61S, V68M, A71G, D90G; (e) I58S, V68S, L70S; (f) M47S, I61S, or (g) V22S. [Invention 1049] The receptor CTLA4 is a fusion protein according to any of the present inventions 1046 to 1048, wherein the receptor CTLA4 contains the amino acid sequence described in SEQ ID NO: 26. [Invention 1050] The fusion protein of the present invention 1048, wherein the ligand-receptor pair is PDL1-CD80, and PDL1 contains the amino acid sequence described in Sequence ID No. 8. [Invention 1051] The fusion protein of the present invention 1048, wherein the ligand-receptor pair is CD28-CD80, and the CD28 contains the amino acid sequence described in SEQ ID NO: 254. [Invention 1052] The fusion protein of the present invention 1043, wherein the ligand-receptor pair is CD28-PDL1. [Invention 1053] The fusion protein of the present invention 1052, wherein the CD28 contains the amino acid sequence described in Sequence ID No. 254. [Invention 1054] The PDL1 is a fusion protein according to the present invention 1052 or 1053, wherein the PDL1 contains the amino acid sequence described in SEQ ID NO: 8. [Invention 1055] The fusion protein of the present invention 1042, wherein the ligand-receptor pair is CD47-SIRPa. [Invention 1056] The fusion protein of the present invention 1055, wherein the SIRPa contains the amino acid sequence described in SEQ ID NO: 255. [Invention 1057] The fusion protein of Invention 1055 or Invention 1056, wherein the CD47 comprises the amino acid sequence described in Sequence ID No. 254. [Invention 1058] A fusion protein according to any one of the present invention 1035 to 1057, wherein the receptor and the ligand are fused to the N-terminus of the first and second polypeptides, respectively. [Invention 1059] A fusion protein according to any of the invention 1035 to 1038, wherein one of the first or second peptide linkers includes multiple protease cleavage sites. [Invention 1060] A fusion protein according to any one of the present invention 1035 to 1039, wherein one of the ligands or receptors is manipulated to include one or more additional protease cleavage sites, and the one or more protease cleavage sites of the ligand or receptor and the protease cleavage sites of the first or second peptide linker are cleavable by the same protease or different proteases. [Invention 1061] The aforementioned proteases are serine protease, MMP1, MMP2, MMP3, MMP7, MMP8, MMP9, MMP10, MMP11, MMP12, MMP13, MMP14, MMP15, MMP16, MMP17, MMP18 (collagenase 4), MMP19, MMP20, MMP21, adamaricin, ceralicin, astacin, caspase 1, caspase 2, caspase 3, caspase 4, caspase 5, caspase 6, caspase 7, caspase 8, caspase 9, caspase 10, caspase 11, caspase 12, A fusion protein according to any of the present invention 1035 to 1060, selected from the group consisting of caspase 13, caspase 14, cathepsin A, cathepsin B, cathepsin D, cathepsin E, cathepsin K, cathepsin S, granzyme B, guanidinobenzoatase (GB), hepsin, elastase, regmine, matryptase, matryptase 2, meprin, neurosin, MT-SP1, neprilysin, plasmin, PSA, PSMA, TACE, TMPRSS3, TMPRSS4, uPA, calpain, FAP, and KLK. [Invention 1062] The fusion protein of the present invention 1061, wherein the protease is uPA or matryptase. [Invention 1063] A fusion protein according to any of the invention 1035 to 1062, wherein the peptide linker has a length of 3 to 50 or 5 to 20 amino acids. [Invention 1064] A fusion protein according to any of the invention 1035 to 1063, wherein one of the first or second peptide linkers does not have a protease cleavage site. [Invention 1065] The aforementioned peptide linker, (Gly n Ser) is a linker, and here (Gly n Ser) Linker is (Gly 3 Ser) n (Gly 4 Ser) 1 , (Gly 3 Ser) 1 (Gly 4 Ser) n , (Gly 3 Ser) n (Gly 4 Ser) n , and (Gly 4 Ser) n A fusion protein according to any of the present invention 1035 to 1064, comprising an amino acid sequence selected from the group consisting of (wherein n is an integer from 1 to 5). [Invention 1066] The aforementioned peptide linker is (EAAAK) n A fusion protein according to any of the invention 1035 to 1064, which is a linker (wherein n is an integer from 1 to 5). [Invention 1067] The fusion protein of the present invention 1064, wherein the peptide linker that does not have the aforementioned protease cleavage site contains the amino acid sequence EAAAKEAAAK (SEQ ID NO: 38). [Invention 1068] The fusion protein of the present invention 1051 or 1052, wherein the peptide linker is a polyproline linker, optionally PPP or PPPP, or a glycine-proline linker, optionally GPPPG, GGPPPGG, GPPPPG or GGPPPPGG. [Invention 1069] A fusion protein according to any one of the invention 1035 to 1064, wherein the peptide linker comprises an immunoglobulin hinge region sequence having an amino acid sequence identity difference of up to 30 percent compared to the amino acid sequence of the wild-type immunoglobulin hinge region. [Invention 1070] A fusion protein according to any of the present invention 1035 to 1069, wherein the peptide linker containing the protease cleavage site contains the amino acid sequence MSGRSANA (SEQ ID NO: 28). [Invention 1071] A fusion protein according to any one of the present invention 1035 to 1070, wherein the binding of the first antigen-binding domain to its congener antigen is reduced by 10 times or more compared to the parent antigen-binding domain that is not fused to the ligand-receptor pair. [Invention 1072] A fusion protein according to any one of the present invention 1035 to 1071, wherein cleavage of the protease cleavage site in the cellular environment releases one member of the ligand-receptor pair from the fusion protein, thereby enabling the antigen-binding domain to bind to its homologous antigen. [Invention 1073] A fusion protein according to any of the invention 1035 to 1072, wherein the first antigen-binding domain is Fab. [Invention 1074] A fusion protein according to any one of the present invention 1035 to 1073, wherein the first antigen-binding domain binds to an antigen expressed on cancer cells or immune cells. [Invention 1075] A fusion protein according to any of the present invention 1035 to 1074, wherein the first antigen-binding domain binds to an antigen expressed on a T cell. [Invention 1076] A fusion protein according to any of the present invention 1035 to 1074, wherein the first antigen-binding domain binds to a tumor-associated antigen (TAA). [Invention 1077] A fusion protein according to any one of the present invention 1035 to 1074, wherein the first antigen-binding domain binds to TAA, and at least one of the ligand or receptor of the ligand-receptor pair is capable of binding to an immunomodulatory target. [Invention 1078] A fusion protein according to any of the invention 1035 to 1077, wherein the first antigen-binding domain binds to an antigen selected from the group consisting of differentiation antigen group 3 (CD3), human epidermal growth factor receptor 2 (HER2), epidermal growth factor receptor (EGFR), mesothelin (MSLN), tissue factor (TF), differentiation antigen group 19 (CD19), tyrosine protein kinase Met (c-Met), and cadherin 3 (CDH3). [Invention 1079] A fusion protein according to any of invention 1032 to 1078, wherein the antibody or antibody fragment comprises a second antigen-binding domain comprising a second VH polypeptide and a second VL polypeptide. [Invention 1080] The fusion protein of the present invention 1079 comprises a second ligand-receptor pair, wherein the ligand of the second ligand-receptor pair is fused to one of the second VH or VL polypeptides via a third peptide linker, and the receptor of the second ligand-receptor pair is fused to the other of the second VH or VL polypeptide via a fourth peptide linker, wherein at least one of the third and fourth peptide linkers comprises a protease cleavage site, and the ligand-receptor pair sterically inhibits the binding of the second antigen-binding domain to its homologous antigen. [Invention 1081] A fusion protein according to Invention 1079 or Invention 1080 that binds to two different antigens. [Invention 1082] The fusion protein of the present invention 1081, wherein one antigen is an antigen expressed by T cells, and the other antigen is an antigen expressed by cancer cells. [Invention 1083] The fusion protein of the present invention 1082, wherein the antigen expressed by the T cell is CD3. [Invention 1084] below: (a) An anti-CD3 paratope comprising VH containing three CDRs HCDR1, HCDR2 and HCDR3, and VL containing three CDRs LCDR1, LCDR2 and LCDR3, (a) HCDR1, HCDR2, and HCDR3 are sequence numbers 207, 208, and 209, respectively, and LCDR1, LCDR2, and LCDR3 are 211, 212, and 214, respectively. (b) HCDR1, HCDR2, and HCDR3 are sequence numbers 224, 225, and 226, respectively, and LCDR1, LCDR2, and LCDR3 are 228, 229, and 230, respectively. (c) HCDR1, HCDR2, and HCDR3 are sequence numbers 232, 233, and 234, respectively, and LCDR1, LCDR2, and LCDR3 are 236, 237, and 238, respectively, or (d) HCDR1, HCDR2, and HCDR3 are sequence numbers 240, 241, and 242, respectively, and LCDR1, LCDR2, and LCDR3 are 244, 245, and 246, respectively. The aforementioned anti-CD3 paratope A fusion protein of the present invention 1083, comprising the above. [Invention 1085] A fusion protein according to any of the invention's 1081-1084, which binds to CD3 and HER2. [Invention 1086] An Fc region comprising a first Fc polypeptide and a second Fc polypeptide, Ligand-receptor pairs comprising the extracellular component of an immunoglobulin superfamily receptor and its related ligand or receptor-binding fragment A fusion protein containing, The ligand is fused to the terminal end of the first polypeptide via a first peptide linker, and the receptor is fused to the similar terminal end of each of the second polypeptides via a second peptide linker. The first and second peptide linkers are of sufficient length to enable the pairing of the ligand and the receptor, and At least one of the first and second peptide linkers includes a protease cleavage site. The aforementioned fusion protein. [Invention 1087] A fusion protein comprising a biologically functional protein, a ligand-receptor pair, a first peptide linker, and a second peptide linker, The biologically functional protein comprises at least a first polypeptide and a second polypeptide, and The ligand-receptor pair comprises the extracellular component of an immunoglobulin superfamily receptor and its homologous ligand or its receptor-binding fragment. The ligand is fused to the terminal end of the first polypeptide via the first peptide linker. The receptor is fused to the same end of each of the second polypeptides via the second peptide linker, and the first and second peptide linkers are long enough to allow the ligand-receptor pairing. The aforementioned fusion protein. [Invention 1088] The fusion protein of the present invention 1086, wherein the ligand and receptor are fused to the N-terminuses of the first and second Fc polypeptides, respectively. [Invention 1089] below: Fab region and Fc region, The Fab region includes a VH polypeptide and a VL polypeptide that form an antigen-binding domain. The aforementioned Fab region and Fc region; and A ligand-receptor pair comprising the extracellular component of an immunoglobulin superfamily receptor and its related ligand or receptor-binding fragment, The ligand is fused to the N-terminus of one of the VH or VL polypeptides via a first peptide linker, and the receptor is fused to the N-terminus of the other VH or VL polypeptide via a second peptide linker. The first and second peptide linkers are of sufficient length to enable the pairing of the ligand and the receptor. At least one of the first and second peptide linkers includes a protease cleavage site, and The ligand-receptor pair sterically inhibits the binding of the antigen-binding domain to its homologous antigen. The aforementioned ligand-receptor pair A fusion protein containing [the specified ingredient]. [Invention 1090] A fusion protein of the present invention 1089, further comprising an additional Fab region or scFv. [Invention 1091] A method for treating cancer, comprising administering a sufficient amount of any of the fusion proteins of the present invention described above to a patient in need thereof. [Invention 1092] A method for modulating an immune response, comprising administering a sufficient amount of any of the fusion proteins of the present invention described above to a patient in need thereof. [Invention 1093] The method of the present invention 1092, wherein the immune response is selected from the group consisting of inhibition of immune checkpoints, stimulation of immune checkpoints, activation of immune cells, stimulation of T cell receptor signaling, T cell-dependent cell-mediated cytotoxicity (TDCC), antibody-dependent cell phagocytosis (ADCP), and stimulation of antibody-dependent cell-mediated cytotoxicity (ADCC). [Invention 1094] Any method 1091 to 1093 of the present invention, wherein the fusion protein is administered intravenously. [Invention 1095] A vector encoding an amino acid sequence containing at least one polypeptide of any fusion protein according to invention 1001 to 1090. [Invention 1096] A cell containing the vector of the present invention 1095. [Invention 1097] A kit comprising a vector according to Invention 1095, cells according to Invention 1096, a purified fusion protein according to any of Inventions 1001-1090, or a combination thereof, and instructions for use. [Invention 1098] A fusion protein according to any one of the present invention 1001 to 1090, wherein cleavage of the protease cleavage site in the cellular environment releases one member of the ligand-receptor pair from the fusion protein, thereby enabling the other member of the ligand-receptor pair to bind to its genus partner on the cell surface. These and other features, aspects, and advantages of the present invention will be better understood by referring to the following description and accompanying drawings. [Brief explanation of the drawing]

[0030] [Figure 1A] This specification shows a schematic diagram of the structure of a particular fusion protein. By fusing PD-1 (checkerboard pattern) and PD-L1 (striped pattern) to the N-terminuses of the heavy and light chains, respectively, the paratope of Fab (gray) can be sterically blocked by an Ig superfamily heterodimer formed between the two. Removal of one side of this mask via TME-specific proteolytic cleavage (release) of one of the linkers introduced between the masking domain and Fab can release a portion of the mask, thereby restoring binding to the target. Furthermore, the portion of the mask that remains covalently bound to Fab confers functionality by binding to its immunomodulatory partner. [Figure 1B] A schematic diagram of an antibody having two Fab arms masked with an IgSF domain pair conjugated to the N-terminus using a TME protease-cleaving or non-cleaving linker is shown. The Fab paratopes a-TAA1 and a-TAA2 may be the same or different, and the IgSF pairs 1:2 and 3:4 may also be the same or different. [Figure 1C] A schematic diagram of a Fab×scFv construct having a Fab arm specific to target 1 and an scFv arm specific to target 2 is shown. Binding to the Fab arm and target 1 is masked by a pair of IgSF domains attached to the N-terminus using a TME protease-cleavable or non-cleavable linker. [Figure 2] This diagram shows a schematic representation of the modified bispecific CD3×Her2 Fab×scFv Fc fusion protein described herein. One arm of the antibody-like molecule contains anti-CD3 Fab blocked by a PD-1 / PD-L1 mask, while the other arm contains anti-Her2 scFv. [Figure 3A]The UPLC-SEC chromatograms and non-reducible and reduced CE-SDS profiles of a representative bispecific CD3×Her2 Fab×scFv Fc variant are shown. The UPLC-SEC chromatogram of the unmasked variant 30421 is also shown. [Figure 3B] The UPLC-SEC chromatograms and non-reducible and reduced CE-SDS profiles of representative bispecific CD3×Her2 Fab×scFv Fc variants are shown. The non-reducible (left) and reduced (right) CE-SDS profiles of the unmasked variant 30421 are also shown. [Figure 3C] The UPLC-SEC chromatograms and non-reducible and reduced CE-SDS profiles of a representative bispecific CD3×Her2 Fab×scFv Fc variant are shown. The UPLC-SEC chromatogram of the masked non-cleavable variant 30423 is also shown. [Figure 3D] The UPLC-SEC chromatograms and non-reducible and reduced CE-SDS profiles of representative bispecific CD3×Her2 Fab×scFv Fc variants are shown. The non-reducible (left) and reduced (right) CE-SDS profiles of the masked non-cleavable variant 30423 are also shown. [Figure 3E] The UPLC-SEC chromatograms and non-reducible and reduced CE-SDS profiles of a representative bispecific CD3×Her2 Fab×scFv Fc variant are shown. The UPLC-SEC chromatogram of the masked light-chain cleavable variant 30430 is also shown. [Figure 3F] The UPLC-SEC chromatograms and non-reducible and reduced CE-SDS profiles of a representative bispecific CD3×Her2 Fab×scFv Fc variant are shown. The non-reducible (left) and reduced (right) CE-SDS profiles of the masked light-chain cleavage variant 30430 are also shown. [Figure 3G]The UPLC-SEC chromatograms and non-reducible and reduced CE-SDS profiles of a representative bispecific CD3×Her2 Fab×scFv Fc variant are shown. The UPLC-SEC chromatogram of the masked heavy-chain cleavable variant 30436 is also shown. [Figure 3H] The UPLC-SEC chromatograms and non-reducible and reduced CE-SDS profiles of representative bispecific CD3×Her2 Fab×scFv Fc variants are shown. The non-reducible (left) and reduced (right) CE-SDS profiles of the masked heavy-chain cleavable variant 30436 are also shown. [Figure 4] This shows an overlay of DSC thermograms of the unmodified variant (30421) and variants masked with PD-1:PD-L1 (30430, 30436) of the investigated CD3×Her2 Fab×scFv Fc system. [Figure 5] The reduced CE-SDS profiles of representative variants, both untreated (-uPa) and treated (+uPa), are shown, performed at 37°C for 24 hours with a uPa:variant ratio of 1:50. Profiles are also shown for unmasked variants (30421), masked but non-cleaving variants (30423), and masked cleaving variants (30430, 30436, 31934). [Figure 6] This shows the results of native binding of CD3 target-directed variants to Jurkat cells, as determined by ELISA. Results are shown for unmasked variants (30421), constructs with only PD-L1 or PD-1 sites bound (31929, 31931), and variants with a complete, non-cleaving mask (30423) or variants with a complete mask and cleaving PD-L1 or PD-1 sites (30430, 30436). For variants 30423, 30430, and 30436, both untreated (-uPa) and treated (+uPa) samples were tested. [Figure 7]The results show cytotoxicity of JIMT-1 tumor cells by Pan T cells determined by TDCC assay after treatment with engineered variants that cross-link T cells and tumor cells. Results are shown for an unmasked variant (30421), a variant with only the PD-1 site bound to the heavy chain (31929), and a variant with a complete non-cleavable mask (30423) or a variant with a complete mask and a cleavable PD-L1 site on the light chain (30430). For variant 30430, both untreated (-uPa) and treated (+uPa) samples were tested. An unrelated anti-RSV antibody (22277) was used as a negative control. [Figure 8A] This report presents the results of flow cytometry-based native binding assays of selected CD3 target-directed variants to PD-L1 transfected CHO-S cells. Results are shown for unmasked variants (30421), constructs with binding only to PD-L1 or PD-1 sites (31929, 31931), and variants with a complete, non-cleaving mask (30423, 30426) or variants with a complete mask and cleaving PD-L1 or PD-1 sites (30430, 30436). Fc fusions of affinity-matured PD-1 sites are also included (31829). For variants 30423, 30426, 30430, and 30436, both untreated (-uPa) and treated (+uPa) samples were tested. [Figure 8B]This report presents the results of flow cytometry-based native binding assays of selected CD3 target-directed variants to PD-1 transfected CHO-S cells. Results are shown for unmasked variants (30421), constructs with binding only to PD-L1 or PD-1 sites (31929, 31931), and variants with a complete, non-cleaving mask (30423, 30426) or variants with a complete mask and cleaving PD-L1 or PD-1 sites (30430, 30436). Fc fusions of affinity-matured PD-1 sites are also included (31829). For variants 30423, 30426, 30430, and 30436, both untreated (-uPa) and treated (+uPa) samples were tested. [Figure 9A] This diagram shows a schematic of a hybrid PD-1 / PD-L1 reporter gene assay that investigates cross-linking between T cells and JIMT-1 cells and the blockade of PD-1:PD-L1 checkpoint engagement. [Figure 9B] This report presents an analysis of a hybrid PD-1 / PD-L1 reporter gene assay investigating crosslinking between T cells and JIMT-1 cells and blockade of PD-1:PD-L1 checkpoint engagement. Results are shown for an unmasked variant (30421) and the same unmasked variant combined with excess anti-PD-L1 antibody (30421 + 150 nM anti-PD-L1). Constructs with only the PD-1 site bound to the heavy chain (31929), as well as variants with a complete non-cleavable mask (30423) or a complete mask with a cleavable PD-L1 site on the light chain (30430), were also investigated. For variant 30430, untreated (-uPa) and treated (+uPa) samples were tested. An unrelated anti-RSV antibody (22277) was used as a negative control. Measurements were performed in triplicates, and error bars reflecting the standard deviation are shown. [Figure 10] This figure represents a modified, monospecific bivalent fusion protein targeted at tumor-associated antigens (TAAs). The Fab paratope is sterically blocked by a PD-1 / PD-L1 mask. [Figure 11A] ULC-SEC chromatograms of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11B] ULC-SEC chromatograms of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11C] ULC-SEC chromatograms of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11D] ULC-SEC chromatograms of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11E]ULC-SEC chromatograms of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11F] ULC-SEC chromatograms of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11G] ULC-SEC chromatograms of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11H] ULC-SEC chromatograms of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11I]ULC-SEC chromatograms of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11J] ULC-SEC chromatograms of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11K] The unreduced SDS-PAGE profiles of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3 are shown. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 11L] This report shows non-reducible and reduced CE-SDS profiles of masked fusion proteins targeted at EGFR, MSLN, TF, CD19, cMet, and CDH3. Data for non-cleavable variants are shown for all fusion proteins (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleavable variants are also included (31723, 31729, 31737, 31733). [Figure 12A]This shows reduced SDS-PAGE profiles of representative EGFR-targeted fusion proteins. Both untreated (-uPa) and treated (+uPa) samples were tested. For each system, data are shown for uPa non-cleaving variants (31722, 31728, 31736, 31732) and variants with a u-Pa cleavage sequence between the VL and PD-L1 sites (31723, 31729, 31737, 31733). [Figure 12B] This shows reduced SDS-PAGE profiles of representative fusion proteins targeted at MSLNs. Both untreated (-uPa) and treated (+uPa) samples were tested. For each system, data are shown for uPa non-cleaving variants (31722, 31728, 31736, 31732) and variants with a u-Pa cleavage sequence between the VL and PD-L1 sites (31723, 31729, 31737, 31733). [Figure 12C] The reduced SDS-PAGE profiles of representative fusion proteins targeted to TF are shown. Both untreated (-uPa) and treated (+uPa) samples were tested. For each system, data are shown for uPa non-cleaving variants (31722, 31728, 31736, 31732) and variants with a u-Pa cleavage sequence between the VL and PD-L1 sites (31723, 31729, 31737, 31733). [Figure 12D] This shows reduced SDS-PAGE profiles of representative fusion proteins targeted at CD19. Both untreated (-uPa) and treated (+uPa) samples were tested. For each system, data are shown for uPa non-cleaving variants (31722, 31728, 31736, 31732) and variants with a u-Pa cleavage sequence between the VL and PD-L1 sites (31723, 31729, 31737, 31733). [Figure 13A]This shows the results of native binding by flow cytometry of selected fusion proteins targeted at different antigens to the following cell lines expressing those antigens: EGFR on MDA-MB-468. For all systems, data for non-cleaved variants are shown (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples with cleaved variants are also included (31723, 31729, 31737, 31733), tested with and without uPa treatment (-uPa). For all systems, unmodified controls (32474, 16427, 16417, 6323, 4372, 17606, 17214) and controls unrelated to cMet and CDH3 (22277) are also included. Where available (EGFR, MSLN, TF), SPR data is also included for comparison. [Figure 13B] This shows the results of native binding by flow cytometry of selected fusion proteins targeted at different antigens to the following cell lines expressing those antigens: MSLN on OVCAR3. For all systems, data for non-cleaved variants are shown (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleaved variants are also included (31723, 31729, 31737, 31733), tested with and without uPa treatment (-uPa). For all systems, unmodified controls (32474, 16427, 16417, 6323, 4372, 17606, 17214) and controls unrelated to cMet and CDH3 (22277) are also included. Where available (EGFR, MSLN, TF), SPR data is also included for comparison. [Figure 13C]This shows the results of native binding by flow cytometry of selected fusion proteins targeted at different antigens to the following cell lines expressing those antigens: TF on MDA-MB-231. For all systems, data for non-cleaved variants are shown (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples with cleaved variants are also included (31723, 31729, 31737, 31733), tested with and without uPa treatment (-uPa). For all systems, unmodified controls (32474, 16427, 16417, 6323, 4372, 17606, 17214) and controls unrelated to cMet and CDH3 (22277) are also included. Where available (EGFR, MSLN, TF), SPR data is also included for comparison. [Figure 13D] This shows the results of flow cytometry-based native binding of selected fusion proteins targeted at different antigens to the following cell lines expressing the antigen: CD19 on Raji. For all systems, data for non-cleaved variants are shown (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples with cleaved variants are also included (31723, 31729, 31737, 31733), tested with and without uPa treatment (-uPa). For all systems, unmodified controls (32474, 16427, 16417, 6323, 4372, 17606, 17214) and controls unrelated to cMet and CDH3 (22277) are also included. Where available (EGFR, MSLN, TF), SPR data is also included for comparison. [Figure 13E]This shows the results of native binding by flow cytometry of selected fusion proteins targeted at different antigens to the following cell lines expressing those antigens: cMet on EBC1. For all systems, data for non-cleaved variants are shown (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples of cleaved variants are also included (31723, 31729, 31737, 31733), tested with and without uPa treatment (-uPa). For all systems, unmodified controls (32474, 16427, 16417, 6323, 4372, 17606, 17214) and controls unrelated to cMet and CDH3 (22277) are also included. Where available (EGFR, MSLN, TF), SPR data is also included for comparison. [Figure 13F] This shows the results of native binding by flow cytometry of selected fusion proteins targeted at different antigens to the following cell lines expressing those antigens: CDH3 on JIMT1. For all systems, data for non-cleaved variants are shown (31722, 31728, 31736, 31732, 28647, 28662), and for EGFR, MSLN, TF, and CD19, samples with cleaved variants are also included (31723, 31729, 31737, 31733), tested with and without uPa treatment (-uPa). For all systems, unmodified controls (32474, 16427, 16417, 6323, 4372, 17606, 17214) and controls unrelated to cMet and CDH3 (22277) are also included. Where available (EGFR, MSLN, TF), SPR data is also included for comparison. [Figure 14]This report shows the results of growth inhibition tests on NCI-H292 cells treated with EGFR-targeted variants. Data are presented for unmasked variants (32474) and PD-1:PD-L masked variants. Masked variants include non-cleaved morphs (31722) and morphs with cleavable PD-L1 sites on the light chain (31723). An unrelated control (22277) is also included. For all variants, samples were tested with and without treatment (-uPa). Error bars reflect the standard deviation of the triple measurements. [Figure 15] This specification shows a schematic diagram of the modified bispecific CD3×Her2 Fab×scFv Fc variant investigated herein. One arm of the fusion protein contains anti-CD3 Fab blocked by a CD80 / CTLA4 mask, while the other arm contains anti-Her2 scFv. [Figure 16A] The UPLC-SEC chromatogram and non-reducible and reduced CE-SDS profiles of variant 30444 are shown. The UPLC-SEC chromatogram of the masked light-chain cleavable variant 30444 is also shown. [Figure 16B] The UPLC-SEC chromatogram and non-reducible and reduced CE-SDS profiles of variant 30444 are shown. Non-reducible (left) and reduced (right) CE-SDS profiles of masked light-chain cleavage variant 30444. [Figure 16C] The UPLC-SEC chromatogram and non-reducible and reduced CE-SDS profiles of variant 30444 are shown. Non-reducible (left) and reduced (right) CE-SDS profiles of masked light-chain cleavage variant 30444. [Figure 16D] The UPLC-SEC chromatogram and non-reducible and reduced CE-SDS profiles of variant 30444 are shown. The UPLC-SEC chromatogram of masked light-chain-cleavable variant 33525 after protein A purification is also shown. [Figure 16E]The UPLC-SEC chromatogram and non-reducible and reduced CE-SDS profiles of variant 30444 are shown. The UPLC-SEC chromatogram of masked light-chain-cleavable variant 33526 after protein A purification is also shown. [Figure 16F] The UPLC-SEC chromatogram and non-reducible and reduced CE-SDS profiles of variant 30444 are shown. The UPLC-SEC chromatogram of masked light-chain cleavage variant 33527 after protein A purification is also shown. [Figure 17] The reduced CE-SDS profiles of variant 30444, both untreated (-uPa) and treated (+uPa), are shown. [Figure 18] This report shows the results of native binding of CD3 target-directed variants to Jurkat cells, as determined by ELISA. Results are presented for an unmasked variant (30421), a variant with a complete PD-1 / PD-L1-based mask and a cleaved PD-L1 site (30430), and a variant with a complete CD80 / CTLA4-based mask and a cleaved CTLA4 site (30444). For variants 30430 and 30444, both untreated (-uPa) and treated (+uPa) samples were tested. [Figure 19] This diagram shows a schematic representation of IgV with an immunomodulatory pair (e.g., PD-1:PD-L1) fused to a heterodimer IgG Fc via a hinge. When one of the two linkers is cleaved by a TME-associated protease such as uPa, one site (e.g., PD-L1) is released, while the site with the desired function (e.g., PD-1) remains bound to the Fc and can bind to its partner on the cell. In the case of PD-1, it can bind to PD-L1 on target cells and inhibit checkpoint function. [Figure 20A]UPLC-SEC chromatograms (A-C) and non-reducible and reduced CE-SDS profiles (D) of CD40-targeted variants are shown. Reduced CE-SDS (E) with and without uPa treatment (-uPa) and untreated (+uPa) of the same variants, flow cytometry-bound data (F), and CD40 RGA assay (G) results are also shown. The test samples include an unmasked variant (32477), a variant with a non-cleavable PD-1 / PD-L1-based mask (32478), and a variant with a PD-1 / PD-L1-based mask that can remove the PD-L1 site by uPa cleavage (32479). Functional studies via RGA assay (G) also include the native CD40-binding partner CD40L and an unrelated control (v22277). CD40 RGA assay data are summarized in Table (H). [Figure 20B] UPLC-SEC chromatograms (A-C) and non-reducible and reduced CE-SDS profiles (D) of CD40-targeted variants are shown. Reduced CE-SDS (E) with and without uPa treatment (-uPa) and untreated (+uPa) of the same variants, flow cytometry-bound data (F), and CD40 RGA assay (G) results are also shown. The test samples include an unmasked variant (32477), a variant with a non-cleavable PD-1 / PD-L1-based mask (32478), and a variant with a PD-1 / PD-L1-based mask that can remove the PD-L1 site by uPa cleavage (32479). Functional studies via RGA assay (G) also include the native CD40-binding partner CD40L and an unrelated control (v22277). CD40 RGA assay data are summarized in Table (H). [Figure 20C]UPLC-SEC chromatograms (A-C) and non-reducible and reduced CE-SDS profiles (D) of CD40-targeted variants are shown. Reduced CE-SDS (E) with and without uPa treatment (-uPa) and untreated (+uPa) of the same variants, flow cytometry-bound data (F), and CD40 RGA assay (G) results are also shown. The test samples include an unmasked variant (32477), a variant with a non-cleavable PD-1 / PD-L1-based mask (32478), and a variant with a PD-1 / PD-L1-based mask that can remove the PD-L1 site by uPa cleavage (32479). Functional studies via RGA assay (G) also include the native CD40-binding partner CD40L and an unrelated control (v22277). CD40 RGA assay data are summarized in Table (H). [Figure 20D] UPLC-SEC chromatograms (A-C) and non-reducible and reduced CE-SDS profiles (D) of CD40-targeted variants are shown. Reduced CE-SDS (E) with and without uPa treatment (-uPa) and untreated (+uPa) of the same variants, flow cytometry-bound data (F), and CD40 RGA assay (G) results are also shown. The test samples include an unmasked variant (32477), a variant with a non-cleavable PD-1 / PD-L1-based mask (32478), and a variant with a PD-1 / PD-L1-based mask that can remove the PD-L1 site by uPa cleavage (32479). Functional studies via RGA assay (G) also include the native CD40-binding partner CD40L and an unrelated control (v22277). CD40 RGA assay data are summarized in Table (H). [Figure 20E]UPLC-SEC chromatograms (A-C) and non-reducible and reduced CE-SDS profiles (D) of CD40-targeted variants are shown. Reduced CE-SDS (E) with and without uPa treatment (-uPa) and untreated (+uPa) of the same variants, flow cytometry-bound data (F), and CD40 RGA assay (G) results are also shown. The test samples include an unmasked variant (32477), a variant with a non-cleavable PD-1 / PD-L1-based mask (32478), and a variant with a PD-1 / PD-L1-based mask that can remove the PD-L1 site by uPa cleavage (32479). Functional studies via RGA assay (G) also include the native CD40-binding partner CD40L and an unrelated control (v22277). CD40 RGA assay data are summarized in Table (H). [Figure 20F] UPLC-SEC chromatograms (A-C) and non-reducible and reduced CE-SDS profiles (D) of CD40-targeted variants are shown. Reduced CE-SDS (E) with and without uPa treatment (-uPa) and untreated (+uPa) of the same variants, flow cytometry-bound data (F), and CD40 RGA assay (G) results are also shown. The test samples include an unmasked variant (32477), a variant with a non-cleavable PD-1 / PD-L1-based mask (32478), and a variant with a PD-1 / PD-L1-based mask that can remove the PD-L1 site by uPa cleavage (32479). Functional studies via RGA assay (G) also include the native CD40-binding partner CD40L and an unrelated control (v22277). CD40 RGA assay data are summarized in Table (H). [Figure 20G]UPLC-SEC chromatograms (A-C) and non-reducible and reduced CE-SDS profiles (D) of CD40-targeted variants are shown. Reduced CE-SDS (E) with and without uPa treatment (-uPa) and untreated (+uPa) of the same variants, flow cytometry-bound data (F), and CD40 RGA assay (G) results are also shown. The test samples include an unmasked variant (32477), a variant with a non-cleavable PD-1 / PD-L1-based mask (32478), and a variant with a PD-1 / PD-L1-based mask that can remove the PD-L1 site by uPa cleavage (32479). Functional studies via RGA assay (G) also include the native CD40-binding partner CD40L and an unrelated control (v22277). CD40 RGA assay data are summarized in Table (H). [Figure 20H] UPLC-SEC chromatograms (A-C) and non-reducible and reduced CE-SDS profiles (D) of CD40-targeted variants are shown. Reduced CE-SDS (E) with and without uPa treatment (-uPa) and untreated (+uPa) of the same variants, flow cytometry-bound data (F), and CD40 RGA assay (G) results are also shown. The test samples include an unmasked variant (32477), a variant with a non-cleavable PD-1 / PD-L1-based mask (32478), and a variant with a PD-1 / PD-L1-based mask that can remove the PD-L1 site by uPa cleavage (32479). Functional studies via RGA assay (G) also include the native CD40-binding partner CD40L and an unrelated control (v22277). CD40 RGA assay data are summarized in Table (H). [Figure 21A] PD1 and PDL1 contain immunoglobulin domains and form a complex. In the image, the binding Fab is docked with the PD1-PDL1 complex at its paratope terminus. Linking PD1 and PDL1 to the VH and VL chains with appropriate linkers may block antigen binding. [Figure 21B-1]Other exemplary immunomodulatory pair structures that can function as masks include PD-1 / PD-L1 (PDB:4ZQK), PD-1 / PD-L2 (PDB:3BP5), CTLA4 / CD86 (PDB:1I85), NCRSRLG1 / NKp30 (PDB:3PV6), SIRPa / CD47 (PDB:4KJY), and CTLA4 / CD80 (PDB:1I8L). [Figure 21B-2] This is a continuation of Figure 21B-1. [Figure 22] This report shows the results of native binding of CD3-targeted variants to Pan T cells, as determined by flow cytometry. Results are presented for unmasked variants (30421), anti-CD3 one-arm antibody (18560), constructs bound only to the PD-1 site (31929), and variants with a complete, non-cleaving mask (30423) or variants with a complete mask and a cleaving PD-L1 site (30430, 30436). For variants 30423 and 30430, both untreated (-uPa) and treated (+uPa) samples were tested. Data are also presented for an unrelated control (22277). [Figure 23A-1] The results show cytotoxicity of HCC1954, JIMT-1, HCC827, and MCF-7 tumor cells by Pan T cells, determined in two repeats of the TDCC assay after treatment with engineered variants that cross-link T cells and tumor cells. Results are shown for the unmasked variant (30421), the unmasked variant combined with a saturated dose of anti-PD-L1 antibody (30421 + 120 nM atezolizumab), the variant with only the PD-1 site bound to the heavy chain (31929), and the variant with a complete non-cleavable mask (30423) or the variant with a complete mask and a cleavable PD-L1 site on the light chain (30430). For variants 30430 and 30423, untreated (-uPa) and treated (+uPa) samples were tested. An unrelated anti-RSV antibody (22277) was used as a negative control. [Figure 23A-2] This is a continuation of Figure 23A-1. [Figure 23B-1] The results show cytotoxicity of HCC1954, JIMT-1, HCC827, and MCF-7 tumor cells by Pan T cells, determined in two repeats of the TDCC assay after treatment with engineered variants that cross-link T cells and tumor cells. Results are shown for the unmasked variant (30421), the unmasked variant combined with a saturated dose of anti-PD-L1 antibody (30421 + 120 nM atezolizumab), the variant with only the PD-1 site bound to the heavy chain (31929), and the variant with a complete non-cleavable mask (30423) or the variant with a complete mask and a cleavable PD-L1 site on the light chain (30430). For variants 30430 and 30423, untreated (-uPa) and treated (+uPa) samples were tested. An unrelated anti-RSV antibody (22277) was used as a negative control. [Figure 23B-2] This is a continuation of Figure 23B-1. [Figure 24-1] The results show IFNγ release from Pan T cells determined in two repeats of TDCC assays with HCC1954, JIMT-1, HCC827, and MCF-7 cancer cells after treatment with engineered variants that cross-link T cells and tumor cells. Results are shown for an unmasked variant (30421), an unmasked variant combined with a saturated amount of anti-PD-L1 antibody (30421 + 120 nM atezolizumab), a variant with only the PD-1 site bound to the heavy chain (31929), and a variant with a complete non-cleavable mask (30423) or a variant with a complete mask and a cleavable PD-L1 site on the light chain (30430). For variants 30430 and 30423, untreated (-uPa) and treated (+uPa) samples were tested. An unrelated anti-RSV antibody (22277) was used as a negative control. [Figure 24-2] This is a continuation of Figure 24-1. [Figure 25]This shows the number of Her2 and PD-L1 receptors per cell determined by flow cytometry for a series of cancer cell lines used in the TDCC and RGA assays. [Figure 26A] This report presents the results of a hybrid PD-1 / PD-L1 reporter gene assay investigating crosslinking and PD-1:PD-L1 checkpoint engagement blockade between T cells and four different cancer cell lines (HCC1954, JIMT-1, HCC827, MCF-7). Results are shown for an unmasked variant (30421), a combination of the unmasked variant with a saturated dose of anti-PD-L1 antibody (30421 + 150 nM atezolizumab), a variant with only the PD-1 site bound to the heavy chain (31929), and a variant with a complete non-cleaving mask (30423) or a variant with a complete mask and a cleaving PD-L1 site on the light chain (30430). For variant 30430, both untreated (-uPa) and treated (+uPa) samples were tested. An unrelated anti-RSV antibody (22277) was used as a negative control. [Figure 26B] This report presents the results of a hybrid PD-1 / PD-L1 reporter gene assay investigating crosslinking and PD-1:PD-L1 checkpoint engagement blockade between T cells and four different cancer cell lines (HCC1954, JIMT-1, HCC827, MCF-7). Results are shown for an unmasked variant (30421), a combination of the unmasked variant with a saturated dose of anti-PD-L1 antibody (30421 + 150 nM atezolizumab), a variant with only the PD-1 site bound to the heavy chain (31929), and a variant with a complete non-cleaving mask (30423) or a variant with a complete mask and a cleaving PD-L1 site on the light chain (30430). For variant 30430, both untreated (-uPa) and treated (+uPa) samples were tested. An unrelated anti-RSV antibody (22277) was used as a negative control. [Figure 26C]This report presents the results of a hybrid PD-1 / PD-L1 reporter gene assay investigating crosslinking and PD-1:PD-L1 checkpoint engagement blockade between T cells and four different cancer cell lines (HCC1954, JIMT-1, HCC827, MCF-7). Results are shown for an unmasked variant (30421), a combination of the unmasked variant with a saturated dose of anti-PD-L1 antibody (30421 + 150 nM atezolizumab), a variant with only the PD-1 site bound to the heavy chain (31929), and a variant with a complete non-cleaving mask (30423) or a variant with a complete mask and a cleaving PD-L1 site on the light chain (30430). For variant 30430, both untreated (-uPa) and treated (+uPa) samples were tested. An unrelated anti-RSV antibody (22277) was used as a negative control. [Figure 26D] This report presents the results of a hybrid PD-1 / PD-L1 reporter gene assay investigating crosslinking and PD-1:PD-L1 checkpoint engagement blockade between T cells and four different cancer cell lines (HCC1954, JIMT-1, HCC827, MCF-7). Results are shown for an unmasked variant (30421), a combination of the unmasked variant with a saturated dose of anti-PD-L1 antibody (30421 + 150 nM atezolizumab), a variant with only the PD-1 site bound to the heavy chain (31929), and a variant with a complete non-cleaving mask (30423) or a variant with a complete mask and a cleaving PD-L1 site on the light chain (30430). For variant 30430, both untreated (-uPa) and treated (+uPa) samples were tested. An unrelated anti-RSV antibody (22277) was used as a negative control. [Figure 27] This figure represents a modified monospecific bivalent fusion protein targeting EGFR(α-EGFR). The Fab paratope is sterically blocked by a SIRPa / CD47 mask. [Figure 28A]This shows the ULC-SEC chromatogram of the EGFR-targeted SIRPa / CD47 masked complete cleavage variant (34164). [Figure 28B] This shows the non-reducible and reduced CE-SDS profiles of the EGFR-targeted SIRPa / CD47 masked fully cleavable variant (34164). [Figure 28C] The reduced CE-SDS for the EGFR-targeted SIRPa / CD47 masked fully cleavable variant (34164) is also shown for the same variant with and without uPa treatment (-uPa). [Figure 29] The results of a native binding assay using high-content analysis on EGFR-positive H292 cells are shown. The test samples include an unmasked EGFR-targeted control (v32474), untreated (-uPa) and treated (+uPa) EGFR-targeted SIRPa / CD47 masked fully cleaved variants (34164), and an unrelated control (v22277). [Figure 30A] This shows single titration point (1 nM) data from flow cytometry binding experiments on Her2+ / PD-L1+ JIMT-1 cells. Data are shown for a triplicate variant (v31929) where only the PD-1 site binds to the heavy chain, and for bispecific variants (v32497 and v33551, respectively) that follow the same format but cannot bind to either PD-L1 or Her2. [Figure 30B] This document presents data from crosslinking experiments using human Pan T cells and Her2+ / PD-L1+ JIMT-1 cells. Data are shown for a triplicate variant (v31929) in which only the PD-1 site bound to the heavy chain, and for a bispecific variant (v32497 and v33551, respectively) with the same format but unable to bind to either PD-L1 or Her2. The crosslinking assay also includes data from an unrelated control (v22277). [Figure 31A]This study demonstrates the mechanism of T cell recruitment and activation of the CD3×Her2 Fab×scFv Fc variant masked with PD-1:PD-L1. The therapeutic antibody is induced into the tumor microenvironment (TME) via TAA binding. [Figure 31B] This study demonstrates the mechanism of T cell recruitment and activation of the CD3×Her2 Fab×scFv Fc variant masked with PD-1:PD-L1. The PD-L1 site of the mask is released via cleavage by a TME-specific protease. [Figure 31C] This describes the mechanism of T cell recruitment and activation of the PD-1:PD-L1 masked CD3×Her2 Fab×scFv Fc variant. The activated therapeutic agent is activated to engage T cells via the unmasked α-CD3 paratope to kill tumor cells and inhibits checkpoint activity by binding to PD-L1 on tumor cells. [Figure 32] This shows the results of native binding of CD3 target-directed variants to Pan T cells, as determined by flow cytometry. Results are shown for an unmasked variant (30421), a construct with only the PD-1 site bound (31929), and a variant with a non-functional PD-1 domain attached to the heavy chain (32497). Data are also shown for an unrelated control (22277). [Figure 33] The results show cytotoxicity of JIMT-1 tumor cells by Pan T cells determined by a TDCC assay after treatment with engineered variants that cross-link T cells and tumor cells. Results are shown for an unmasked variant (30421), a variant with only the PD-1 site bound to the heavy chain (31929), and a variant with a non-functional PD-1 domain attached to the heavy chain (32497). [Figure 34A] This is a schematic diagram of IgSF core Ig folding, showing a beta sandwich consisting of seven antiparallel beta strands arranged on two beta sheets of three strands and four strands. [Figure 34B]This is a schematic diagram showing the IgC1 subgroup domain (top) and IgC2 subgroup domain (bottom) of IgSF, which have different strand configurations. [Figure 34C] This is a schematic diagram of an IgV domain, containing nine beta strands arranged in two sheets of four and five strands. [Figure 34D] This is a schematic diagram of an IgV domain, containing nine beta strands arranged in two sheets of four and five strands. [Modes for carrying out the invention]

[0031] Detailed explanation definition Terms used in the claims and specification are briefly defined here and more specifically below.

[0032] A “fusion protein” refers to a protein containing multiple polypeptide regions or domains linked to one another, for example, by peptide bonds. Therefore, as used herein, “fused” refers to polypeptide sequences linked to one another via peptide bonds. Examples include antibodies or backbones fused to immunomodulatory ligand / receptor pairs. Fusion proteins described herein may also be referred to as “variants” or “constructs.”

[0033] In a broad sense, "biologically functional proteins" refer to polypeptides or proteins that have biological function, such as antibodies, or dimers (Fc).

[0034] A "ligand-receptor pair" refers to a receptor polypeptide and a ligand polypeptide that specifically bind to each other. Examples include PD-1-PD-L1, CTLA4-CD80, or CD28-CD80.

[0035] "Receptor-binding fragment" refers to any polypeptide that specifically binds to the receptor of a ligand-receptor pair. The receptor-binding fragment may be either naturally occurring or non-naturally occurring.

[0036] An "immunomodulatory" molecule refers to a molecule that has the ability to regulate an immune response, such as upregulation or downregulation of an immune response, and / or immune cell activity, either directly or indirectly.

[0037] "Peptide linker" refers to a peptide that connects or links other peptides or polypeptides.

[0038] The terms "Fc region", "Fc", and "Fc domain" are used interchangeably herein and refer to the C-terminal region of an immunoglobulin heavy chain that contains at least a portion of the constant region.

[0039] "Bispecific" refers to a biologically functional protein that can specifically bind to two different epitopes.

[0040] "Multispecific" refers to a biologically functional protein that can specifically bind to two or more different target molecules or epitopes.

[0041] "Masked" means that a polypeptide domain, such as the antigen-binding domain of an antibody, is sterically inhibited from binding to a target sequence, or that a ligand is sterically inhibited from binding to its cognate binding partner, such as its receptor.

[0042] "Protease-activated" or "protease-cleaved" or "cleaved" refers to a fusion protein that contains a protease cleavage site and refers to the fusion protein after being cleaved by a protease.

[0043] "Protease cleavage site" refers to a site within the amino acid sequence of a fusion protein that contains a protease recognition sequence and is cleaved by a protease.

[0044] An "immune checkpoint" refers to an immune system control pathway that controls the activation of the immune system.

[0045] "Specifically binds" (and grammatical variations thereof) means a binding that is measurably different from non-specific interactions when referring to the binding of a particular antigen, epitope, ligand or receptor.

[0046] As detailed below, "mammal" includes both human and non-human, including but not limited to humans, non-human primates, dogs, cats, mice, cows, horses, and pigs.

[0047] It should be noted that, as used in this specification and the appended claims, the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise.

[0048] Abbreviations used in this application include: PD-1 (Programmed Cell Death Protein 1); PDL-1 (Programmed Cell Death Ligand 1); CD3 (Differentiation Antigen Group 3); CTLA4 (Cytotoxic T Lymphocyte-Associated Protein 4 or Differentiation Antigen Group 152); CD80 (Differentiation Antigen Group 80); CD28 (Differentiation Antigen Group 28); CD86 (Differentiation Antigen Group 86); ICOS (Inducible T Cell Costimulator); ICOSL (Inducible T Cell Costimulator Ligand); CD47 (Differentiation Antigen Group 47); SIRPA (Signal Regulatory Protein Alpha), HHLA2 (Human Endogenous Retrovirus H Long Terminal Repeat-Associated 2), NKp30 (Natural Killer Cell Receptor 3), NCR3LG1 (Natural Killer Cell-Inducing Receptor 3 Ligand 1), HHLA2 (HERV-H LTR-related factors 2), VISTA (V-domain Ig inhibitor of T cell activation), VTCN1 (V-set domain-containing T cell activation inhibitor 1), CD276 (differentiation antigen group 276), human epidermal growth factor receptor 2 (HER2), epidermal growth factor receptor (EGFR), mesothelin (MSLN), tissue factor (TF), differentiation antigen group 19 (CD19), tyrosine protein kinase Met (c-Met), and cadherin 3 (CDH3).

[0049] Where used herein, the term “approximately” refers to a variation of approximately ±10% from a given value. It should be understood that such variation is always included in any given value provided herein, whether or not it is specifically mentioned.

[0050] Where used herein, the terms “comprising,” “having,” “including,” and “containing,” and their grammatical variations, are inclusive or open-ended and do not exclude additional unlisted elements and / or process steps. Where used herein in relation to a composition, use, or method, the term “essentially consisting of” indicates that additional elements and / or process steps may exist, but these additions do not substantially affect the manner in which the listed composition, method, or use functions. Where used herein in relation to a composition, use, or method, the term “consisting of” excludes the existence of additional elements and / or process steps. A composition, use, or method described herein as containing certain elements and / or steps may also, in certain embodiments, essentially consist of those elements and / or steps, and in other embodiments, whether these embodiments are specifically referenced or not.

[0051] Any embodiment discussed herein is intended to be carried out with respect to any method, use or composition disclosed herein, and vice versa.

[0052] It should be understood that the active enumeration of features in one embodiment may serve as a basis for excluding features in another embodiment. In particular, when a list of options is presented for a given embodiment or claim, it should be understood that one or more options may be removed from the list, and the shortened list may form an alternative embodiment, whether or not such an alternative embodiment is specifically referred to.

[0053] The various amino acid sequences and clonal sequences referred to herein are listed in Table AA.

[0054] Fusion protein Disclosed herein are fusion proteins comprising a biologically functional protein, such as an antibody or polypeptide backbone, fused to a ligand-receptor pair. In the fusion proteins according to this disclosure, the biologically functional protein comprises at least a first polypeptide and a second polypeptide, wherein the ligand is fused to one end of the polypeptide via a first peptide linker, and the receptor is fused to each of the other polypeptides via a second peptide linker. In some embodiments, at least one of the first and second peptide linkers includes a protease cleavage site that occurs naturally in the target cell environment, such as the tumor microenvironment. Methods of using the fusion proteins disclosed herein are also disclosed.

[0055] The fusion proteins according to this disclosure are masked to reduce any on-target / off-tissue (e.g., off-tumor) effects (i.e., toxicity) associated with target engagement. In the target cell environment, the fusion protein is unmasked when a peptide linker(s) containing a protease cleavage site is cleaved. In certain embodiments, the fusion protein according to this disclosure comprises a polypeptide backbone fused to a ligand-receptor pair. In this context, the fusion protein is masked in such a way that the ligand and receptor of the ligand-receptor pair inhibit engagement with native congener receptors or ligands through their association with each other. In the target cell environment, when a peptide linker(s) containing a protease cleavage site is cleaved, the fusion protein is unmasked by the release of one member of the ligand-receptor pair from the fusion protein, thereby allowing the other member of the ligand-receptor pair to bind to its congener partner. Thus, in certain embodiments, this disclosure provides a biological design for programmed checkpoint or co-stimulatory receptor targeting.

[0056] In certain embodiments, the fusion protein according to this disclosure comprises an antibody or antigen-binding antibody fragment containing an antigen-binding domain fused to a ligand-receptor pair. In this context, the fusion protein is masked in that the ligand-receptor pair sterically inhibits the binding of the antigen-binding domain to its congener antigen. The fusion protein is further masked in that the ligand and receptor of the ligand-receptor pair each inhibit engagement with a native congener receptor or ligand through association with each other. In the target cell environment, when the peptide linker(s) containing the protease cleavage site is cleaved, the masking of the fusion protein is unmasked by the release of one member of the ligand-receptor pair from the fusion protein, thereby allowing both the other member of the ligand-receptor pair to bind to its congener partner and the antigen-binding domain to bind to its congener antigen. Thus, in certain embodiments, this disclosure provides a multifunctional biological design for programmed target antigen engagement and simultaneous checkpoint or co-stimulatory receptor targeting. In certain embodiments, the fusion protein design described herein reduces target-mediated drug diathesis. In certain embodiments, the fusion protein provides a masked antigen-binding domain, e.g., a biologically functional protein, and a masked immunomodulatory target-binding domain, e.g., a ligand-receptor pair, where programmed activation of one binding function further leads to activation of the other binding function, thereby yielding a bifunctional molecule. Thus, in certain embodiments, the disclosure provides methods for masking and conditionally activating antigen-binding domains in specific target tissue environments, as well as for targeting and activating immunomodulatory targets to reduce harmful toxic effects.

[0057] Ligand-receptor pairs Fusion proteins comprising ligand-receptor pairs are described herein. In certain embodiments, the ligand-receptor pair is an immunomodulatory pair of ligand-receptor domains belonging to the immunoglobulin superfamily (IgSF) (Natarajan, Kannan; Mage, Michael G; and Margulies, David H (April 2015) Immunoglobulin Superfamily. In: eLS. John Wiley & Sons, Ltd: Chichester., AF Williams 1, AN Barclay (1988) The Immunoglobulin Superfamily--Domains for Cell Surface Recognition Annu Rev Immunol 6:381-405).

[0058] The immunoglobulin superfamily (IgSF) is classified into commonly found protein domains based on the core folding of immunoglobulin (Ig). This Ig folding consists of a beta sandwich composed of a total of seven antiparallel beta strands arranged in two beta sheets of three and four strands (Figure 34A). The two beta sandwiches are internally connected via a disulfide bridge between strand B and strand F. A structural motif commonly recognized in Ig folding is the "Greek key" motif. Common subgroups of IgSF are the IgV, IgC1, and IgC2 domains. Each member is identified based on common structural features and the arrangement of beta strands. The IgC domain contains seven beta strands arranged in two sheets of three and four strands (Figure 34B), while the IgV domain contains nine beta strands arranged in two sheets of four and five strands (Figure 34C, D). IgC1 and IgC2 differ in the structural arrangement of their strands. IgSF domains can be present in a wide variety of biologically important proteins, including antigen receptors, immunoglobulins, and immunomodulatory receptors. The surface-exposed residues of the core beta sandwich and the loops connecting the beta strands can function as interaction interfaces for antigen recognition, other structural domains of the tertiary / quaternary assembly, or receptor / ligand pairs. Since the antigen recognition site of immunoglobulins (the VH-VL pair in antibodies such as IgG1) contains a dimer of two IgV domains, either an IgSF or IgV domain dimer is structurally compatible to form a stereomask at the antigen recognition site if covalently bound to the N-terminus of an antibody (Figure 21).

[0059] In certain embodiments, the ligand-receptor pair is immunomodulatory, for example, an immune checkpoint, resulting in the modulation of immune cell effector function, the modulation of T cell receptor signaling, the modulation of interactions between antigen-presenting cells and effector cells, or a combination thereof. In certain embodiments, the ligand-receptor pair comprises the extracellular component of the IgSF receptor and its homologous ligand or its receptor-binding fragment. The receptor-binding fragment refers to any polypeptide that specifically binds to the receptor of the ligand-receptor pair, and may be naturally occurring or non-natural. "Naturally occurring," as used herein and applied to an object, refers to the fact that the object is found in nature. For example, a polypeptide sequence or polynucleotide sequence present in a living organism that can be isolated from a natural source and has not been artificially modified in the laboratory is naturally occurring. In certain embodiments, the ligand-receptor pair may be two interacting protein domains belonging to the immunoglobulin domain superfamily. "Non-natural," as used herein, refers to an engineered polypeptide sequence that is structurally similar to IgSF, such as a variant of a naturally occurring protein.

[0060] In certain embodiments, the disclosure herein relates to using immunomodulatory pairs of ligand-receptor domains belonging to IgSF as a mask for an antibody or antibody fragment to inhibit target antigen binding. Examples of immunomodulatory pairs of ligand-receptor domains belonging to the immunoglobulin superfamily include, but are not limited to, pairs of the B7 / CD28 family (e.g., PD1-PDL1, PD1-PDL2, CTLA4-CD80, CD28-CD80, CD28-CD86, CTLA4-CD86, PDL1-CD80, as well as ICOS-ICOSL, NCR3LG1-NKp30, HHLA2-CD28H and CD47-SIRPα). CD80 (also known as B7-1) The following proteins belong to the B7 family: CD86 (B7-2), PDL1 (B7-H1), ICOSL (B7-H2), PDL2 (B7-DC), CD276 (B7-H3), VTCN1 (B7-H4), VISTA (B7-H5), NCR3LG1 (B7-H6), and HHLA2 (B7-H7). The B7 family of proteins is typically considered ligands and pairs with members of the CD28 family, including CD28, CTLA4, CD28H, NKp30, PD1, and ICOS. (SMWest and XADeng. Considering B7-CD28 as a family through sequence and structure. Exp Biol Med (Maywood) 2019;244(17):1577-1583;doi:10.1177 / 1535370219855970).

[0061] In certain embodiments, the ligand-receptor pair comprises a member of the IgSF B7 / CD28 family. In certain embodiments, the ligand and receptor comprise the extracellular portion of an immunoglobulin superfamily (IgSF) polypeptide. In certain embodiments, the ligand and receptor comprise the extracellular portion of an IgSF immunoglobulin variable (IgV) polypeptide. In certain embodiments, the ligand is a member of the IgSF B7 family and the receptor is a member of the IgSF CD28 family.

[0062] In certain embodiments, the ligand-receptor pairs include leukocyte co-stimulatory receptors. Examples of leukocyte co-stimulatory receptors belonging to the B7 / CD28 family include ICOS (also known as CD278) and CD28. Examples of co-stimulatory ligand-receptor pairs include CD80:CD28, CD86:CD28, and ICOS:ICOSL (ICOS ligand). Examples of co-inhibitory ligand-receptor pairs include PD1-PDL1, PD1-PDL2, CTLA4-CD80, CTLA4-CD86, PDL1-CD80, and CD47-SIRPα. When these are linked to the N-terminus of the Fab, our results described herein show that they block access to the CDRs and block binding to the antigen (Figure 21A).

[0063] Other members of this large IgSF can also be used in a similar manner and can perform immunomodulatory functions. Figure 21B shows the known structural diagrams of known B7-CD28 members. The sizes and orientations of the domains of other pairs are very similar to those of PD-1 and PD-L1, and from this, like the PD-1 / PD-L1 receptor-ligand pair, they can be used for binding or functional blocking.

[0064] The concept of functional masking extends beyond members of the B7 family. For example, Figure 21B shows the structural diagram of SIRPα / CD47, another ligand-receptor pair that includes domains belonging to the IgSF, which is located at the N-terminus of the Fab and shows good spatial compatibility for blocking binding. A number of candidate therapeutic agents have been evaluated for increasing the phagocytosis of cancer cells by the use of antagonists in this direction and are good candidates for functional masks (Murata Y, Saito Y, Kotani T, Matozaki T. (2018) CD47-signal regulatory protein α signaling system and its application to cancer immunotherapy. Cancer Sci. 2018 Aug;109(8):2349-2357).

[0065] In certain embodiments, the affinity of the ligand-receptor domain of the ligand-receptor pair of the fusion protein is altered compared to the wild-type ligand and receptor. In certain embodiments, one or both of the ligand-receptor domains of the masking pair are manipulated so that the ligand and receptor contain sequences different from those of the wild-type ligand or receptor. In certain embodiments, the ligand contains one or more mutations that increase the binding affinity of the ligand to its congener receptor. In certain embodiments, the relative binding affinity of the ligand of the ligand-receptor pair compared to the wild-type ligand is greater than 1, 1.5, 2, 2.53, 5, 10, 20, 30, 40, 50, 100, 500, 1000, 5,000, 10,000, 50,000, or 100,000 times greater than the binding affinity of the wild-type ligand to its naturally occurring congener receptor.

[0066] In certain embodiments, the receptor includes one or more mutations that increase the receptor's binding affinity to its congener ligand. In certain embodiments, the relative binding affinity of the ligand-receptor pair of the receptor compared to the wild-type receptor is greater than 1, 1.5, 2, 2.5, 3, 5, 10, 20, 30, 40, 50, 100, 500, 1000, 5,000, 10,000, or 100,000 times greater than the binding affinity of the wild-type receptor to its naturally occurring congener ligand.

[0067] In certain embodiments, the ligand includes one or more mutations that reduce the ligand's binding affinity to its congener receptor. In certain embodiments, the ligand-receptor pair's relative binding affinity to the ligand is more than 1, 1.5, 2, 2.5, 3, 5, 10, 20, 30, 40, 50, 100, 500, 1000, 5,000, 10,000, 50,000, or 100,000 times smaller than the wild-type ligand's binding affinity to its naturally occurring congener receptor.

[0068] In certain embodiments, the receptor includes one or more mutations that reduce the receptor's binding affinity to its congener ligand. In certain embodiments, the relative binding affinity of the ligand-receptor pair of the receptor compared to the wild-type receptor is more than 1, 1.5, 2, 2.5, 3, 5, 10, 20, 30, 40, 50, 100, 500, 1000, 5,000, 10,000, or 100,000 times smaller than the binding affinity of the wild-type receptor to its naturally occurring congener ligand.

[0069] The ligand-receptor pair may be, for example, the IgV domains of PD-L1 (Uniprot ID Q9NZQ7, 33-146) and PD-1 (Uniprot ID Q15116, 18-132). In some embodiments, the ligand is PD-L1 and has an amino acid sequence corresponding to, for example, SEQ ID NO: 8 or SEQ ID NO: 10. In certain embodiments, PD-L1 has an amino acid sequence substantially identical to SEQ ID NO: 8. In certain embodiments, PD-L1 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 8. In certain embodiments, PD-L1 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 8. Any PD-L1 variant, for example, a high-affinity variant known in the art, such as those provided in Z. Laing et al., High-affinity human PD-L1 variants attenuate the suppression of T cell activation; Oncotarget 8, 88360-88375 (2017) or WO2018 / 170021A1, can be used. In certain embodiments, the receptor is a high-affinity PD-L1 variant. In some embodiments, the receptor is a high-affinity PD-L1 variant having an amino acid sequence corresponding to SEQ ID NO: 10 or an amino acid sequence substantially identical to SEQ ID NO: 10.

[0070] In some embodiments, the receptor is PD-1, having an amino acid sequence corresponding to, for example, SEQ ID NO: 7 or SEQ ID NO: 11. In certain embodiments, PD-1 has an amino acid sequence substantially identical to SEQ ID NO: 7 or SEQ ID NO: 11. In certain embodiments, PD-1 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 7 or SEQ ID NO: 11. In certain embodiments, PD-1 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 7 or SEQ ID NO: 11. Any PD-1 variant, for example, a high-affinity variant known in the art, such as those described in RLMaute et al., Engineering high-affinity PD-1 variants for optimized immunotherapy and immuno-PET imaging. Proc Natl Acad Sci USA 112, E6506-6514 (2015), WO2016 / 022994A2 or E. Lazar-Molnar et al., Structure-guided development of a high-affinity human Programmed Cell Death-1: Implications for tumor immunotherapy EBIOMedicine 17.30-44 (2017) and WO2019 / 241758A1, may be used.

[0071] In certain embodiments, the receptor is a high-affinity PD-1 variant. In some embodiments, the receptor is a high-affinity PD-1 variant having an amino acid sequence corresponding to or substantially identical to SEQ ID NO: 9.

[0072] In certain embodiments, the ligand is CD80, having, for example, an amino acid sequence corresponding to SEQ ID NO: 25. In certain embodiments, CD80 has an amino acid sequence substantially identical to SEQ ID NO: 25. In certain embodiments, CD80 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 25. In certain embodiments, CD80 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 25. In some embodiments, CD80 has an amino acid sequence substantially identical to SEQ ID NO: 185, SEQ ID NO: 187, or SEQ ID NO: 189. In certain embodiments, CD80 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 185, SEQ ID NO: 187, or SEQ ID NO: 189. In certain embodiments, CD80 has a mutation that increases its affinity for its receptor or reduces its tendency to form homodimers during preparation. In certain embodiments, CD80 has an amino acid sequence corresponding to SEQ ID NO: 25 and comprises one of the following mutation sets: (a) H18Y, A26E, E35D, M47S, I61S and D90G; (b) E35D, M47S, N48K, I61S, K89N; (c) E35D, D46V, M47S, I61S, D90G, K93E; or (d) H18Y, A26E, E35D, M47S, I61S, V68M, A71G, D90G.

[0073] In certain embodiments, the ligand is PD-L2, which has, for example, an amino acid sequence corresponding to SEQ ID NO: 250. In certain embodiments, PD-L2 has an amino acid sequence substantially identical to SEQ ID NO: 250. In certain embodiments, PD-L2 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 250. In certain embodiments, PD-L2 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 250.

[0074] In certain embodiments, the ligand is CD86, which has, for example, an amino acid sequence corresponding to SEQ ID NO: 248. In certain embodiments, CD86 has an amino acid sequence substantially identical to SEQ ID NO: 248. In certain embodiments, CD86 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 248. In certain embodiments, CD86 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 248.

[0075] In certain embodiments, the ligand is ICOSL, which has, for example, an amino acid sequence corresponding to SEQ ID NO: 256. In certain embodiments, ICOSL has an amino acid sequence substantially identical to SEQ ID NO: 256. In certain embodiments, ICOSL has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 256. In certain embodiments, ICOSL has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 256.

[0076] In certain embodiments, the ligand is CD276, which has, for example, an amino acid sequence corresponding to SEQ ID NO: 258. In certain embodiments, CD276 has an amino acid sequence substantially identical to SEQ ID NO: 258. In certain embodiments, CD276 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 258. In certain embodiments, CD276 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 258.

[0077] In certain embodiments, the ligand is VTCN1, which has, for example, an amino acid sequence corresponding to SEQ ID NO: 259. In certain embodiments, VTCN1 has an amino acid sequence substantially identical to SEQ ID NO: 259. In certain embodiments, VTCN1 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 259. In certain embodiments, VTCN1 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 259.

[0078] In certain embodiments, the ligand is VISTA, which has, for example, an amino acid sequence corresponding to SEQ ID NO: 260. In certain embodiments, VISTA has an amino acid sequence substantially identical to SEQ ID NO: 260. In certain embodiments, VISTA has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 260. In certain embodiments, VISTA has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 260.

[0079] In certain embodiments, the ligand is HHLA2, which has, for example, an amino acid sequence corresponding to SEQ ID NO: 262. In certain embodiments, HHLA2 has an amino acid sequence substantially identical to SEQ ID NO: 262. In certain embodiments, HHLA2 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 262. In certain embodiments, HHLA2 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 262.

[0080] In certain embodiments, the ligand is SIRPα, which has, for example, an amino acid sequence corresponding to SEQ ID NO: 255. In certain embodiments, SIRPα has an amino acid sequence substantially identical to SEQ ID NO: 255. In certain embodiments, SIRPα has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 255. In certain embodiments, SIRPα has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 255.

[0081] In some embodiments, the receptor is CTLA4, having, for example, an amino acid sequence corresponding to SEQ ID NO: 26. In certain embodiments, CTLA4 has an amino acid sequence substantially identical to SEQ ID NO: 26. In certain embodiments, CTLA4 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 26. In certain embodiments, CTLA4 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 26.

[0082] In some embodiments, the receptor is CD28, having, for example, an amino acid sequence corresponding to SEQ ID NO: 253. In certain embodiments, CD28 has an amino acid sequence substantially identical to SEQ ID NO: 253. In certain embodiments, CD28 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 253. In certain embodiments, CD28 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 253.

[0083] In some embodiments, the receptor is CD28H, having, for example, an amino acid sequence corresponding to SEQ ID NO: 263. In certain embodiments, CD28H has an amino acid sequence substantially identical to SEQ ID NO: 263. In certain embodiments, CD28H has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 263. In certain embodiments, CD28H has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 263.

[0084] In some embodiments, the receptor is NKp30, having, for example, an amino acid sequence corresponding to SEQ ID NO: 264. In certain embodiments, NKp30 has an amino acid sequence substantially identical to SEQ ID NO: 264. In certain embodiments, NKp30 has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 264. In certain embodiments, NKp30 has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 264.

[0085] In some embodiments, the receptor is ICOS, having, for example, an amino acid sequence corresponding to SEQ ID NO: 257. In certain embodiments, ICOS has an amino acid sequence substantially identical to SEQ ID NO: 257. In certain embodiments, ICOS has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 257. In certain embodiments, ICOS has an amino acid sequence that is approximately 96%, 97%, 98%, or 99% identical to SEQ ID NO: 257.

[0086] In certain embodiments, the IgSF ligand and / or receptor has an immunoglobulin variable domain (IgV)-like structure. The amino acid sequences of some exemplary naturally occurring IgV domain receptors and ligands described herein are shown in Table CC.

[0087] In certain embodiments, the ligand and / or receptor of a manipulated ligand-receptor pair that does not exist naturally but pairs together, comprises an immunoglobulin domain having at least one of the domains having affinity for naturally occurring immunomodulatory receptors.

[0088] In certain embodiments, immunomodulatory ligand-receptor pairs are selected to function as antagonists or agonists of their congeneral target pairs. In certain embodiments, immunomodulatory ligand-receptor pairs are selected to function as antagonists or agonists of their congeneral target pairs in a tumor environment. In certain embodiments, one or both of the ligand or receptor of the ligand-receptor pair are designed to play a functional role after activation by protease cleavage.

[0089] Fusion protein format The fusion proteins described herein can be in many different formats. A fusion protein can be thought of as having a modular structure comprising at least a ligand-receptor pair, where each of the ligand and receptor is fused to a biologically functional protein via a peptide linker. The biologically functional protein then comprises at least a first and a second polypeptide. For example, either the N-terminus or C-terminus of the ligand-receptor pair may be fused to the first and second polypeptides of the biologically functional protein, for example, via a peptide linker. The ligand is fused to the first polypeptide, and the receptor is fused to each of the similar ends of the second polypeptide. The term “each of the similar ends” means, when describing the fusion of the ligand-receptor pair to polypeptides, that the ligand and receptor are fused to either the N-terminus or the C-terminus of the first and second polypeptides, respectively. Thus, in a particular embodiment, the ligand is fused to the N-terminus of the first polypeptide via a first peptide linker, and the receptor is fused to the N-terminus of the second polypeptide via a second peptide linker. In certain embodiments, the ligand is fused to the C-terminus of a first polypeptide via a first peptide linker, and the receptor is fused to the C-terminus of a second polypeptide via a second peptide linker. The ligand and receptor may be fused via their C-terminus or their N-terminus. Both the ligand and receptor may be fused via their N-terminus or C-terminus, or one of the ligand or receptor may be fused via its N-terminus and the other via its C-terminus.

[0090] In certain embodiments, the N-terminus of the ligand is fused to the N-terminus of the first polypeptide via a first peptide linker, and the N-terminus of the receptor is fused to the N-terminus of the second polypeptide via a second peptide linker. In certain embodiments, the C-terminus of the ligand is fused to the C-terminus of the first polypeptide via a first peptide linker, and the C-terminus of the receptor is fused to the second polypeptide via a second peptide linker.

[0091] In certain embodiments, the ligand is fused to the terminal end of a first polypeptide of a biologically functional protein via a first peptide linker containing a protease cleavage site. In certain embodiments, the receptor is fused to the terminal end of a second polypeptide of a biologically functional protein via a second peptide linker containing a protease cleavage site. In certain embodiments, the ligand is fused to the terminal end of a first polypeptide of a biologically functional protein via a first peptide linker containing a protease cleavage site, and the receptor is fused to the terminal end of a second polypeptide of a biologically functional protein via a second peptide linker containing a protease cleavage site. If both the first and second peptide linkers contain protease cleavage sites, the protease cleavage sites may be cleavable by the same protease or by different proteases.

[0092] In certain embodiments, the ligand is fused to the terminus of a first polypeptide of a biologically functional protein via a first peptide linker containing a protease cleavage site, and the ligand is engineered to include an internal protease cleavage site which may be the same as or different from the cleavage site of the first peptide linker. In certain embodiments, the receptor is fused to the terminus of a second polypeptide of a biologically functional protein via a second peptide linker containing a protease cleavage site, and the receptor is engineered to include an internal protease cleavage site which may be the same as or different from the cleavage site of the second peptide linker. By including protease cleavage sites in both the peptide linker and the member of the ligand-receptor pair connected to the biologically functional protein by the linker, it becomes possible to cleave and inactivate that member of the ligand-receptor pair in the target cell environment, and the masking of the ligand-receptor pair member that remains fused to the biologically active protein is unmasked (i.e., conditional activation).

[0093] In certain embodiments, the fusion protein is conjugated with another therapeutic and / or diagnostic portion, such as a chemotherapeutic agent or a radioisotope.

[0094] Biologically functional proteins A biologically functional protein may function as a backbone and / or contain binding domains. Examples of polypeptide backbones include immunoglobulin Fc domains, albumins, albumin analogs and derivatives, toxins, cytokines, chemokines, growth factors, and protein pairs such as leucine zipper domains. In certain embodiments, the biologically functional protein may include labels, drugs, or combinations thereof. Any label known in the art that is suitable for detecting the fusion proteins described herein may be used. The biologically functional protein may include any drug, toxin, or chemical substance known in the art that can be conjugated to the protein to achieve a desired biological outcome.

[0095] In certain embodiments, the biologically functional protein of the fusion protein described herein comprises at least one antigen-binding domain. The binding domain may be, for example, an immunoglobulin-based binding domain or a non-immunoglobulin-based antibody mimetic, or other polypeptides or small molecules capable of specifically binding to their targets, such as natural or artificial ligands. Non-immunoglobulin-based antibody mimetic formats include, for example, antikalin, finomer, affimer, alpha-body, DARPin, and avimer.

[0096] The fusion proteins described herein include biologically functional proteins. Examples of biologically functional proteins include, but are not limited to, antibodies, e.g., antigen-binding domains and polypeptide backbones, e.g., polypeptides containing a dimer Fc. Thus, in certain embodiments, the first and second polypeptides of the biologically functional protein are polypeptides containing variable domains and / or constant domains of an antibody, or other domains that confer antigen-binding or backbone function to the fusion protein.

[0097] antibody In certain embodiments, the biologically functional protein is an antibody, i.e., an immunoglobulin. The antibodies according to this disclosure may take various forms described herein, including antibody fragments. Thus, in certain embodiments, the biologically functional protein is an antibody fragment. The terms “antibody” and “immunoglobulin” are used interchangeably herein to refer to polypeptides encoded by one or more immunoglobulin genes or modified immunoglobulin genes, which bind specifically to an antigen.

[0098] Specific binding can be measured, for example, by enzyme-linked immunosorbent assay (ELISA), surface plasmon resonance (SPR) technology (e.g., using a BIAcore instrument) (Liljeblad et al., 2000, Glyco J, 17:323-329), or by conventional binding assays (Heeley, 2002, Endocr Res, 28:217-229). In certain embodiments, specific binding is defined as the degree of binding to unrelated proteins being less than approximately 10% of the binding to the target antigen, as measured, for example, by SPR. In certain embodiments, specific binding of an antibody or antibody fragment to a particular antigen or epitope is defined as a dissociation constant (K) of ≤1 μM, e.g., ≤100 nM, ≤10 nM, ≤1 nM, ≤0.1 nM, ≤0.01 nM, or ≤0.001 nM. D ) is defined by. In a particular embodiment, the specific binding of an antibody or antibody fragment to a particular antigen or epitope is defined as 10 -6 Less than M, for example, 10 -7 Less than M, or 10 -8 Dissociation constant less than M (K D ) is defined by. In some embodiments, the specific binding of an antibody or antibody fragment to a particular antigen or epitope is defined by 10 -6 M~10 -13 M, for example, 10 -7 M~10 -13 M, 10 -8 M~10 -13 M, or 10 -9 M~10 -13 The dissociation constant of M (K D ) is defined by.

[0099] Conventional immunoglobulin structural units typically consist of two pairs of polypeptide chains, each pair having one "light" chain (approximately 25 kD) and one "heavy" chain (approximately 50-70 kD). The light chain is classified as either kappa or lambda. The "class" of an immunoglobulin refers to the type of constant domain in its heavy chain. Antibodies have five major classes: IgA, IgD, IgE, IgG, and IgM, some of which can be further divided into subclasses (isotypes), such as IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2. The heavy chain constant domains corresponding to different classes of immunoglobulins are called alpha (α), delta (δ), epsilon (ε), gamma (γ), and mu (μ), respectively.

[0100] In certain embodiments, the antibodies described herein are based on IgG class immunoglobulins, e.g., IgG1, IgG2, IgG3, or IgG4 immunoglobulins. In some embodiments, the antibodies described herein are based on IgG1, IgG2, or IgG4 immunoglobulins. In some embodiments, the antibodies described herein are based on IgG1 immunoglobulin. In the context of this disclosure, when an antibody is based on a particular immunoglobulin isotype, it means that the antibody includes all or part of the constant region of that particular immunoglobulin isotype. It is also understood that in some embodiments, the antibody may include hybrids of isotypes and / or subclasses.

[0101] The N-terminal domain of each polypeptide chain of immunoglobulin defines a variable region of approximately 100-110 amino acids or longer, primarily involved in antigen recognition. The terms variable light chain (VL) and variable heavy chain (VH) refer to these domains located in the light and heavy chains, respectively.

[0102] Therefore, it can be understood that immunoglobulins contain different domains within their heavy and light chains. Such domains may overlap and may include Fc domains (or Fc regions), CH1 domains, CH2 domains, CH3 domains, hinge domains, heavy chain constant domains (CH1-hinge-Fc or CH1-hinge-CH2-CH3), variable heavy chain domains (VH), variable light chain domains (VL), and light chain constant domains (CL). The “Fc domain” includes the CH2 and CH3 domains and, optionally, the hinge domain (or hinge region).

[0103] Each of the VH and VL domains of immunoglobulins contains three loops that are hypervariable in sequence and form antigen-binding sites. Each of these loops is called a “hypervariable region” or “HVR.” The terms hypervariable region (HVR) and complementarity-determining region (CDR) are used interchangeably herein with respect to the portion of the variable region that forms the antigen-binding domain. Except for CDR1 of VH, CDRs generally contain amino acid residues that form the hypervariable loop. The VH and VL domains consist of several relatively invariant stretches called framework regions (FRs) that are about 15–30 amino acids long, separated by shorter CDRs, each of which is typically about 5–15 amino acids long, but may be longer or shorter in some cases. The three CDRs and four FRs that make up each of the VH and VL domains are arranged from the N-terminus to the C-terminus as FR1–CDR1–FR2–CDR2–FR3–CDR3–FR4.

[0104] Many different definitions of the CDR region are commonly used, including those described in Kabat et al. (1983, Sequences of Proteins of Immunological Interest, NIH Publication No. 369-847, Bethesda, MD), Chothia et al. (1987, J Mol Biol, 196:901-917), as well as the IMGT, AbM, and Contact definitions. These different definitions involve overlaps or subsets of amino acid residues when compared to one another. For example, the definitions according to Kabat, Chothia, IMGT, AbM, and Contact are listed in Table 1 below. Thus, as will be readily apparent to those skilled in the art, the exact numbering and arrangement of CDRs may vary depending on the numbering system employed. However, it is understood that the disclosure herein of variable heavy chain domains (VHs) includes the disclosure of the relevant (inherent) heavy chain CDRs (HCDRs) as defined by any known numbering system. Similarly, the disclosure herein of variable light chain domains (VLs) includes the disclosure of the associated (unique) heavy chain CDR (HCDR) as defined by one of the known numbering systems.

[0105] (Table 1) TIFF0007846668000001.tif138165

[0106] Those skilled in the art will understand that a limited number of amino acid substitutions can be introduced into the CDR, VH, or VL sequences of known antibodies without the antibody losing its ability to bind to its target. Candidate amino acid substitutions can be identified by techniques such as computer modeling or alanine scanning as described above, and the resulting variants are tested for binding activity by standard techniques. For example, in a particular embodiment, the EGFR-binding domain contained in the fusion protein comprises a set of CDRs (i.e., heavy chain CDR1, CDR2, and CDR3, and light chain CDR1, CDR2, and CDR3) having 90%, 95%, 98%, 99%, or 100% sequence identity with a set of CDRs derived from cetuximab or panitumumab, and the binding domain retains its ability to bind to EGFR. In certain embodiments, the EGFR-binding domain in the fusion protein contains variants of these CDR sequences, including 1 to 10 amino acid substitutions across the three CDRs, e.g., 1 to 7 amino acid substitutions, 1 to 5 amino acid substitutions, 1 to 4 amino acid substitutions, 1 to 3 amino acid substitutions, 1 to 2 amino acid substitutions, or 1 amino acid substitution across the three CDRs (i.e., a CDR can be modified by containing up to 10 amino acid substitutions in any combination of the modified CDRs), and the variant retains the ability to bind to EGFR. Typically, such amino acid substitutions may be conserved amino acid substitutions outlined in column 1 or column 2 of Table 4 below.

[0107] In certain embodiments, the antibodies described herein include at least one immunoglobulin domain derived from a mammalian immunoglobulin such as bovine immunoglobulin, human immunoglobulin, camel immunoglobulin, rat immunoglobulin, or mouse immunoglobulin. In some embodiments, the biologically functional protein may be a chimeric antibody comprising two or more immunoglobulin domains, at least one domain derived from a first mammalian immunoglobulin, e.g., human immunoglobulin, and at least a second domain derived from a second mammalian immunoglobulin, e.g., mouse or rat immunoglobulin. In some embodiments, the biologically functional protein includes at least one constant immunoglobulin domain derived from human immunoglobulin.

[0108] Those skilled in the art will understand that these domains can be combined in various ways to provide antibodies in different formats, including multispecific antibodies in different formats. These formats are generally based on antibody formats known in the art (see, for example, Brinkmann & Kontermann, Overview, 2017, MABS, 9(2):182-212, and Muller & Kontermann, “Bispecific Antibodies” in Handbook of Therapeutic Antibodies, Wiley-VCH Verlag GmbH & Co. (2014)).

[0109] Antibodies of biologically functional proteins described herein may have different valencies. In certain embodiments, the biologically functional protein contains a single antigen-binding domain. In certain embodiments, the biologically functional protein contains two or more antigen-binding domains. In certain embodiments, the biologically functional protein contains antibodies having different valencies and specificities. As used herein, a “bispecific antibody” contains two binding domains. In certain embodiments, each of the two binding domains has a unique binding specificity. As used herein, a “multispecific antibody” contains two or more binding domains. In certain embodiments, each of the two or more binding domains has a unique binding specificity. In some embodiments, at least two of the two or more binding domains have a unique binding specificity. For example, an antibody may be bivalent and bispecific, or bivalent and have single specificity. Alternatively, an antibody may be trivalent and bispecific, i.e., the antibody contains three binding domains. An antibody may also be bispecific and tetravalent, i.e., the antibody contains four binding domains. Other valencies are also possible.

[0110] When an antibody contains two binding domains that bind to the same target molecule, the binding domains can either bind to the same epitope on the target molecule or to different epitopes on the target molecule. In some embodiments, the antibody contains two binding domains that bind to different epitopes on the target molecule. The term "biparatope" can be used to refer to an antibody that contains two binding domains that bind to different epitopes on the same target molecule (antigen). A biparatope antibody can bind to a single antigen molecule via two different epitopes or to two distinct antigen molecules via two different epitopes.

[0111] In certain embodiments, the antibody is biparatope and bispecific, in that each comprises a first binding domain and a second binding domain that bind to different epitopes on a first target molecule, and a third binding domain that binds to a second target molecule. Alternatively, a bispecific biparatope antibody may comprise a first binding domain and a second binding domain that bind to different epitopes on a first target molecule, and a third binding domain and a fourth binding domain that bind to different epitopes on a second target molecule.

[0112] In some embodiments, the antibody further comprises a backbone, and the binding domain is operably linked to the backbone. “Operatally linked,” as used herein, means that the components described are in a relationship that allows each to function in the manner intended. The binding domain can be linked directly or indirectly to the backbone. “Indirectly linked” means that a given binding domain is linked to the backbone via another component, such as a linker or one of other binding domains. Various formats of fusion proteins including a backbone are described in more detail below.

[0113] Antigen-binding domain format In some embodiments, the fusion proteins described herein include an antibody having at least one antigen-binding domain, which is an antibody fragment such as Fab, Fab', single-chain Fab (scFab), single-chain Fv (scFv), or a single-domain antibody (sdAb).

[0114] "Fab" or "Fab fragment" contains the constant domain (CL) of the light chain and the first constant domain (CH1) of the heavy chain, along with the variable domains VL and VH of the light chain and heavy chain, respectively, which include the CDR. Fab' or Fab' fragment differs from the Fab fragment in that several amino acid residues, including one or more cysteine ​​residues derived from the hinge region, are added to the C-terminus of the heavy chain CH1 domain.

[0115] A Fab fragment may consist of two separate polypeptide chains (a light chain and a heavy chain) or it may be a single-stranded Fab. A single-stranded Fab is a Fab molecule in which the Fab light chain and Fab heavy chain are linked by a peptide linker to form a single peptide chain. Typically, in a single-stranded Fab molecule, the C-terminus of the Fab light chain is connected to the N-terminus of the Fab heavy chain, but other formats are also possible.

[0116] An "scFv" is a single polypeptide chain containing the heavy chain variable domain (VH) and light chain variable domain (VL) of an antibody. Optionally, an scFv may contain a polypeptide linker between the VH and VL domains, which can help the scFv form a structure desirable for antigen binding. An scFv may contain VL-linker-VH, where VL is linked from its C-terminus to the N-terminus of VH by a linker, or VH-linker-VL, where VH is linked via its C-terminus to the N-terminus of VL by a linker. For an overview of scFvs, see Pluckthun in The Pharmacology of Monoclonal Antibodies, vol. 113, Rosenburg and Moore eds., Springer-Verlag, New York, pp. 269-315 (1994).

[0117] The term "sdAb" refers to a single immunoglobulin domain. An sdAb may be derived, for example, from a camel. Camel antibodies lack a light chain, and the antigen-binding site consists of a single domain called "VHH". An sdAb contains three CDR / hypervariable loops that form the antigen-binding sites: CDR1, CDR2, and CDR3. sdAbs are fairly stable and readily expressed, for example, as fusions containing the antibody's Fc chain (see, e.g., Harmsen & De Haard, 2007, Appl. Microbiol Biotechnol. 77(1):13-22).

[0118] In some embodiments, one or more binding domains in the antibody may be a natural or artificial ligand for the target receptor, or a functional fragment of such ligand, i.e., a fragment capable of specifically binding to the target receptor.

[0119] The antigen-binding domain may be in the form of individual combinations of scFv, Fab, and sdAb. For example, if the binding domain is in the form of scFv, formats such as tandem scFv ((scFv)2 or taFv) or triple-bodied (three scFv) can be constructed, in which case the scFvs are linked together by a flexible linker. scFv can also be used to construct dia-bodied, tria-bodied, and tetra-bodied (tandem dia-bodied or TandAb) formats, which contain two, three, and four scFvs, respectively, connected by short linkers. Due to the limited length of the linkers (usually about 5 amino acids long), scFv dimerize from head to tail. In any of the aforementioned formats, scFv can be further stabilized by including interdomain disulfide bonds. For example, a disulfide bond can be introduced between VL and VH by introducing additional cysteine ​​residues into each chain (e.g., at position 44 of VH and position 100 of VL) (see, e.g., Fitzgerald et al., 1997, Protein Engineering, 10:1221-1225), or by introducing a disulfide bond between two VHs to provide an antigen-binding domain having a DART format (see, e.g., Johnson et al., 2010, J Mol. Biol., 399:436-449).

[0120] Similarly, formats containing two or more sdAbs, such as VH or VHH, linked to each other by a suitable linker, can also be used for biologically functional proteins. Other examples of antibody formats lacking a backbone include those based on Fab fragments, such as the Fab2, F(ab')2, and F(ab')3 formats, in which the Fab fragments are linked by a linker or IgG hinge region.

[0121] Furthermore, alternative formats can be generated by employing combinations of different forms of antigen-binding domains. For example, fusing scFv or sdAb to the C-terminus of either or both of the light and heavy chains of a Fab fragment can yield a bivalent (Fab-scFv) or (Fab-sdAb) or a trivalent (Fab-(scFv)2 or Fab-(sdAb)2). Similarly, fusing one or two scFv or sdAb to the hinge region of an F(ab') fragment can yield a trivalent or tetravalent F(ab')2-scFv / sdAb. The binding domain can be one or a combination of the above forms (e.g., scFv, Fab and / or sdAb, or ligand-based binding domains).

[0122] In certain specific embodiments, the biologically functional protein comprises a bispecific antibody that binds to an immune cell antigen, e.g., CD3, and a tumor-associated antigen (TAA), e.g., HER2. In certain more specific embodiments, the biologically functional protein comprises a bispecific antibody in Fab-scFv format, where Fab binds to the immune cell antigen and scFv binds to TAA. In certain more specific embodiments, the biologically functional protein comprises a bispecific antibody in Fab-scFv format, where Fab binds to CD3 and scFv binds to HER2. In some embodiments, the biologically functional protein comprises a bispecific antibody in Fab-Fab format, where one Fab binds to CD3 and the other Fab binds to HER2.

[0123] In certain embodiments, a biologically functional protein comprises two or more antigen-binding domains operably linked to a heterodimer Fc. In this context, the biologically functional protein may be bivalent, trivalent, or tetravalent. Non-limiting examples of the format are described below. Other structures are known in the art (see, for example, Spiess et al., 2015, Mol Immunol., 67:95-106).

[0124] Exemplary structures of biologically functional proteins, i.e., bivalent antibodies, that contain two binding domains operably linked to a heterodimer Fc include, but are not limited to, the following: a) an mAb format in which the first binding domain is a Fab operably linked to the N-terminus of the first Fc polypeptide of the heterodimer Fc, and the second binding domain is a Fab operably linked to the N-terminus of the second Fc polypeptide; b) a hybrid format in which the first binding domain is an scFv operably linked to the N-terminus of one Fc polypeptide of the heterodimer Fc, and the second binding domain is a Fab operably linked to the N-terminus of the other Fc polypeptide; and c) a dual scFv format in which the first binding domain is an scFv operably linked to the N-terminus of the first Fc polypeptide of the heterodimer Fc, and the second binding domain is an scFv operably linked to the N-terminus of the second Fc polypeptide.

[0125] Another example is an antibody comprising one binding domain (either the first or second) which is operably ligated to the N-terminus of the first Fc polypeptide and the other binding domain which is operably ligated to the C-terminus of the second Fc polypeptide and is operably ligated to the C-terminus of the second Fc polypeptide.

[0126] Exemplary structures of multispecific antibodies (i.e., trivalent antibodies) containing three binding domains operably linked to a heterodimer Fc include, but are not limited to, the following: A) mAb-Fv format in which the first binding domain is a Fab operably ligated to the N-terminus of the first Fc polypeptide of the heterodimer Fc, the second binding domain is a Fab operably ligated to the N-terminus of the second Fc polypeptide, and the third binding domain consists of a VH domain ligated to the C-terminus of one Fc polypeptide and a VL domain ligated to the C-terminus of the other Fc polypeptide; B) mAb-scFv format, where the first binding domain is a Fab operably ligated to the N-terminus of the first Fc polypeptide of the heterodimer Fc, the second binding domain is a Fab operably ligated to the N-terminus of the second Fc polypeptide, and the third binding domain is an scFv operably ligated to the C-terminus of either the first or second Fc polypeptide; C) scFv-mAb format in which the first binding domain is a Fab operably ligated to the N-terminus of the first Fc polypeptide of the heterodimer Fc, the second binding domain is a Fab operably ligated to the N-terminus of the second Fc polypeptide, and the third binding domain is an scFv operably ligated to the N-terminus of either the first or second binding domain; D) A central scFv format in which the first binding domain is an scFv operably ligated to the N-terminus of one Fc polypeptide of the heterodimer Fc, the second binding domain is a Fab operably ligated to the N-terminus of the other Fc polypeptide, and the third binding domain is a Fab operably ligated to the first binding domain (scFv); E) A Fab hybrid format in which the first binding domain is an scFv operably ligated to the N-terminus of one Fc polypeptide of a heterodimer Fc, the second binding domain is a Fab operably ligated to the N-terminus of the other Fc polypeptide, and the third binding domain is a Fab operably ligated to the N-terminus of either the first or second binding domain; F) scFv hybrid format in which the first binding domain is an scFv operably ligated to the N-terminus of one Fc polypeptide of heterodimer Fc, the second binding domain is a Fab operably ligated to the N-terminus of the other Fc polypeptide, and the third binding domain is an scFv operably ligated to the N-terminus of either the first or second binding domain; G) A hybrid scFv format in which the first binding domain is an scFv operably ligated to the N-terminus of one Fc polypeptide of the heterodimer Fc, the second binding domain is a Fab operably ligated to the N-terminus of the other Fc polypeptide, and the third binding domain is an scFv operably ligated to the C-terminus of either the first or second Fc polypeptide; H) A hybrid Fab format in which the first binding domain is an scFv operably ligated to the N-terminus of one Fc polypeptide of the heterodimer Fc, the second binding domain is a Fab operably ligated to the N-terminus of the other Fc polypeptide, and the third binding domain is a Fab operably ligated to the C-terminus of the first or second Fc polypeptide; and I) A Fab-mAb format in which the first binding domain is a Fab operably ligated to the N-terminus of the first Fc polypeptide of the heterodimer Fc, the second binding domain is a Fab operably ligated to the N-terminus of the second Fc polypeptide, and the third binding domain is a Fab operably ligated to the N-terminus of either the first or second binding domain.

[0127] Exemplary structures of multispecific antibodies, i.e., tetravalent antibodies, containing four binding domains operably linked to a heterodimer Fc include, but are not limited to, the following: i) a central scFv2 format in which the first binding domain is an scFv operably linked to the N-terminus of one Fc polypeptide of the heterodimer Fc, the second binding domain is an scFv operably linked to the N-terminus of the other Fc polypeptide, the third binding domain is a Fab operably linked to one of the scFvs, and the fourth binding domain is a Fab operably linked to the other scFv; and ii) a dual variable domain format in which the first binding domain is an Fab operably linked to the N-terminus of one Fc polypeptide of the heterodimer Fc, the second binding domain is a Fab operably linked to the N-terminus of the other Fc polypeptide, the third binding domain is an scFv operably linked to one of the Fabs, and the fourth binding domain is an scFv operably linked to the other Fab.

[0128] Antibodies of biologically functional proteins described herein may include labels, drugs, or combinations thereof. Any label known in the art that is suitable for detecting the fusion proteins described herein may be used. Antibody-drug conjugates are described in more detail below.

[0129] In certain embodiments, the antigen-binding domain of an antibody of a biologically functional protein described herein binds to the same antigen on the same cell. In certain embodiments, the antigen-binding domain binds to multiple antigens on the same cell. In certain embodiments, the antigen-binding domain binds to multiple antigens, with at least one antigen located on a different cell from the others. In certain embodiments, the antigen-binding domain(s) of the antibody bind to tumor cells or immune cells. In certain embodiments, the antigen-binding domain(s) of the antibody bind to both tumor cells and immune cells.

[0130] Chimeric antibodies, humanized antibodies, and variant antibodies In some embodiments, antibodies can be derived from immunoglobulins of different species; for example, the antibody may be a chimeric antibody or a humanized antibody. A “chimeric antibody” typically refers to an antibody that contains at least one variable domain derived from a rodent antibody (usually a mouse antibody) and at least one constant domain derived from a human antibody. A “humanized antibody” is a type of chimeric antibody that contains a minimal amount of sequence derived from a non-human antibody.

[0131] The human constant domain of a chimeric antibody does not need to be of the same isotype as the non-human constant domain being replaced. Chimeric antibodies are discussed, for example, in Morrison et al., 1984, Proc. Natl. Acad. Sci. USA, 81:6851-55, and U.S. Patent No. 4,816,567. Generally, a humanized antibody is a human immunoglobulin (recipient antibody) in which residues derived from the hypervariable region of the recipient are replaced with residues derived from the hypervariable region of a non-human species (donor antibody), such as mouse, rat, rabbit, or non-human primate, which have desirable specificity and affinity for the target antigen. The technique for producing these humanized antibodies is often called "CDR grafting." Both "chimeric antibody" and "humanized antibody" generally refer to antibodies that combine immunoglobulin regions or domains derived from multiple species.

[0132] In some cases, additional modifications are made to further improve the performance of the antibody. For example, framework region (FR) residues of human immunoglobulin may be replaced with corresponding non-human residues, or residues not found in either the recipient antibody or the donor antibody may be included in the humanized antibody. Generally, the variable domain of a humanized antibody includes all or substantially all of the hypervariable region derived from the non-human immunoglobulin and all or substantially all of the FR derived from the human immunoglobulin sequence. Humanized antibodies are described in detail, for example, Jones, et al., 1986, Nature, 321:522-525; Riechmann, et al., 1988, Nature, 332:323-329 and Presta, 1992, Curr. Op. Struct. Biol., 2:593-596.

[0133] Numerous approaches are known in the art for selecting the optimal human framework for grafting non-human CDRs. Early approaches used a limited subset of well-characterized human antibodies, regardless of sequence identity with the non-human antibody providing the CDR ("fixed framework" approach). More recent approaches employ variable regions of non-human antibodies that have high amino acid sequence identity with the variable region providing the CDR ("homologous matching" or "best-fit" approach). Another approach is to select a fragment of the framework sequence within the light or heavy chain variable region derived from several different human antibodies. CDR grafting can, in some cases, result in a partial or complete loss of affinity of the grafted molecule to its target antigen. In such cases, affinity can be restored by reverting some of the human-derived residues to the corresponding non-human-derived residues. Methods for preparing humanized antibodies using these approaches are well known in the art (see, for example, Tsurushita & Vasquez, 2004, Humanization of Monoclonal Antibodies, Molecular Biology of B Cells, 533-545, Elsevier Science (USA); Jones et al., 1986, Nature, 321:522-525; Riechmann et al., 1988, Nature, 332:323-329; Presta et al., 1997, Cancer Res, 57(20):4593-4599).

[0134] In addition to or instead of these conventional approaches, more recent techniques can be employed to further reduce the immunogenicity of CDR-grafted humanized antibodies. For example, a framework based on human germline sequences or consensus sequences can be used as the acceptor human framework, rather than a human framework with somatic mutations (or multiple mutations). Another technique aimed at reducing the potential immunogenicity of non-human CDRs is to graft only specificity-determining residues (SDRs). In this approach, only the minimum number of CDR residues ("SDRs") necessary for antigen-binding activity are grafted onto a human germline framework. This method improves the "human-ness" (i.e., similarity to human germline sequences) of the humanized antibody, thus helping to reduce the risk of immunogenicity in the variable region. These techniques are described in various publications (see, for example, Almagro & Fransson, 2008, Front Biosci, 13:1619-1633; Tan, et al., 2002, J Immunol, 169:1119-1125; Hwang, et al., 2005, Methods, 36:35-42; Pelat, et al., 2008, J Mol Biol, 384:1400-1407; Tamura, et al., 2000, J Immunol, 164:1432-1441; Gonzales, et al., 2004, Mol Immunol, 1:863-872; and Kashmiri, et al., 2005, Methods, 36:25-34).

[0135] In certain embodiments, the antibody comprises a humanized antibody sequence, for example, one or more humanized variable domains. In some embodiments, the antibody is a humanized antibody.

[0136] In certain embodiments, the antigen-binding domain in the fusion protein is a substituted variant of a known antibody, containing one or more amino acid substitutions in the CDR of the parent antibody. In certain embodiments, the substituted variant has a change (e.g., improvement) in certain biological properties compared to the parent antibody. For example, the substituted variant may have increased affinity for the target protein or decreased immunogenicity. In some embodiments, the substituted variant substantially retains certain biological properties of the parent antibody.

[0137] CDR hotspots are residues encoded by codons that undergo high mutation rates during the somatic cell maturation process (see, e.g., Chowdhury, 2008, Methods Mol. Biol., 207:179-196). Affinity maturation by secondary library construction and reselection has been described (see, e.g., Hoogenboom et al. in Methods in Molecular Biology, 178:1-37, O'Brien et al., ed., Human Press, Totowa, NJ (2001)).

[0138] Methods for affinity maturation can introduce diversity into the variable genes selected for maturation by various techniques well known in the art, such as error-prone PCR, chain shuffling, or oligonucleotide-directed mutagenesis. A secondary library is then constructed and screened to identify any antibody variant with the desired affinity. Another method for introducing diversity involves a CDR-directed approach that randomizes several CDR residues (e.g., two, three, four, or more residues at once). Often, either or both of the heavy chain or light chain CDR3s are targeted in the CDR-directed approach. CDR residues involved in antigen binding can be identified, for example, using alanine scanning mutagenesis (see, e.g., Cunningham and Wells, 1989, Science, 244:1081-1085) or by computer modeling that identifies the contact points between the antibody and antigen using the crystal structure of the antigen-antibody complex.

[0139] In certain embodiments, a substituted variant includes one or more substitutions within one or more CDRs, provided that the substitutions do not substantially reduce the binding domain's ability to bind to its target antigen. For example, a substituted variant may include one or more conservative substitutions described herein within one or more CDRs that do not substantially reduce binding affinity. In some embodiments, a substituted variant includes one or more amino acid substitutions within a CDR that are not involved in the antigen-contact amino acid. In some embodiments, a substituted variant includes a variant VH sequence or VL sequence in which each CDR is either unchanged or contains no more than one, two, or three amino acid substitutions.

[0140] Glycosylated variant In certain embodiments, the fusion proteins described herein include biologically functional proteins based on IgG Fc with modified native glycosylation. As is known in the art, the glycosylation of Fc can be modified to increase or decrease effector function.

[0141] For example, mutations in the conserved asparagine residue at position 297 to alanine, glutamine, lysine, or histidine (i.e., N297A, Q, K, or H) result in a deglycosylated Fc lacking all effector functions (Bolt et al., 1993, Eur. J. Immunol., 23:403-411; Tao & Morrison, 1989, J. Immunol., 143:2595-2601).

[0142] Conversely, removing fucose from the oligosaccharide linked to the heavy chain N297 has been shown to improve ADCC due to improved binding to FcγRIIIa (see, for example, Shields et al., 2002, J Biol Chem., 277:26733-26740, and Niwa et al., 2005, J.Immunol.Methods, 306:151-160). Such low-fucose antibodies can be produced, for example, in knockout Chinese hamster ovary (CHO) cells lacking fucosyltransferase (FUT8) (Yamane-Ohnuki et al., 2004, Biotechnol. Bioeng., 87:614-622), in the variant CHO cell line Lec13 (International Public Publication No. WO03 / 035835) with reduced ability to bind fucose to the glycans linked to N297, or in other cells that produce defucosylated antibodies (e.g., Li et al., 2006, Nat Biotechnol, 24:210-215; Shields et al., 2002, ibid, and Shinkawa et al., 2003, J. Biol. Chem., 278:3466-3473). In addition to carbohydrates, International Publication No. WO2009 / 135181 describes the addition of a fucose analog to the culture medium during antibody production to inhibit the incorporation of fucose into the glycans on the antibody.

[0143] Other methods for acidifying antibodies containing little to no fucose on the Fc glycosylation site (N297) are well known in the art. For example, see the GlymaX® technology (ProBioGen AG) (von Horsten et al., 2010, Glycobiology, 20(12):1607-1618 and U.S. Patent Nos. 8,409,572).

[0144] Other glycosylated variants include those containing bifid oligosaccharides, for example, variants in which the branched oligosaccharide bound to the antibody's Fc region is bifid by N-acetylglucosamine (GlcNAc). Such glycosylated variants have reduced fucosylation and / or improved ADCC function. See, for example, International Publication No. WO2003 / 011878, U.S. Patent No. 6,602,684, and U.S. Patent Application Publication No. US2005 / 0123546. Useful glycosylated variants also include those having at least one galactose residue in the oligosaccharide bound to the Fc region, which may have improved CDC function (see, for example, International Publication Nos. WO1997 / 030087, WO1998 / 58964, and WO1999 / 22764).

[0145] Polypeptide skeleton In certain embodiments, the biologically functional protein of the fusion protein described herein is a polypeptide backbone, which can function, for example, to stabilize or extend the in vivo half-life of a ligand-receptor pair.

[0146] In certain embodiments, the biologically functional protein consists of a dimeric Fc region. In certain embodiments, the first and second polypeptides of the biologically functional protein consist of a dimeric Fc, the first polypeptide consists of a first Fc polypeptide, the second polypeptide consists of a second Fc polypeptide, and the first and second Fc polypeptides form a dimeric Fc region. In certain embodiments, the dimeric Fc region i is a heterodimer Fc. The heterodimer Fc region is described in more detail herein. In certain embodiments, the polypeptide backbone comprises a first and a second polypeptide. In certain embodiments, the ligand of a ligand-receptor pair is fused to the first polypeptide via a peptide linker, and the receptor is fused to the same end of each of the second polypeptides via a peptide linker. Thus, in certain embodiments, the ligand is fused to the N-terminus of the first polypeptide via a peptide linker, and the receptor is fused to the N-terminus of the second polypeptide via a second peptide linker. Conversely, in certain embodiments, the ligand is fused to the C-terminus of the first polypeptide via a peptide linker, and the receptor is fused to the second polypeptide via a second peptide linker.

[0147] In certain more specific embodiments, the biologically functional protein comprises a polypeptide backbone consisting of a dimeric Fc region and a ligand-receptor pair which is PDL-1 and PD-1. In certain embodiments, the fusion protein comprises a biologically functional protein consisting of a dimeric Fc region and a ligand-receptor pair which is CD80 and CTLA4. In certain embodiments, the Fc domain of the polypeptide backbone comprises amino acid sequences corresponding to SEQ ID NOs: 4 and 5, and optionally, amino acid sequences corresponding to SEQ ID NOs: 6. In certain embodiments, the polypeptide backbone consists of a heterodimeric Fc comprising SEQ ID NOs: 4 and 5, where the first Fc polypeptide comprises SEQ ID NOs: 4 and the second Fc polypeptide comprises SEQ ID NOs: 5. In some embodiments, the polypeptide backbone consisting of a heterodimeric Fc comprises modified CH3 and / or CH2 domains as shown in Tables 2 and 3, respectively.

[0148] FC Domain In certain embodiments, the fusion proteins described herein include a biologically functional protein, such as an antibody or polypeptide backbone containing a dimeric immunoglobulin Fc region. The term “Fc region” includes native sequence Fc regions and variant Fc regions. Unless otherwise specified herein, the numbering of amino acid residues in the Fc region or constant region follows the EU numbering system, also known as the EU index, as described in Kabat et al., Sequences of Proteins of Immunological Interest, 5th Ed. Public Health Service, National Institutes of Health, Bethesda, MD (1991). The “Fc polypeptide” of dimeric Fc refers to one of the two polypeptides that form the dimeric Fc region, i.e., the polypeptide containing the C-terminal constant region of an immunoglobulin heavy chain capable of stable self-assembly.

[0149] The Fc region may contain either a CH3 domain or a CH3 and CH2 domain. The CH3 domain contains two CH3 sequences, each containing one of the two Fc polypeptides of the dimer Fc. Similarly, the CH2 domain contains two CH2 sequences, each containing one of the two Fc polypeptides of the dimer Fc.

[0150] In certain embodiments, the fusion protein comprises an Fc based on human IgG Fc. In some embodiments, the fusion protein comprises an Fc based on human IgG1 Fc. In some embodiments, the fusion protein comprises an Fc based on a heterodimer Fc containing two different Fc polypeptides.

[0151] In certain embodiments, the fusion protein includes Fc based on modified IgG Fc, and the CH3 domain includes one or more amino acid modifications. In some embodiments, the fusion protein includes Fc based on modified IgG Fc, and the CH2 domain includes one or more amino acid modifications. In some embodiments, the fusion protein includes Fc based on modified IgG Fc, and the CH3 domain includes one or more amino acid modifications, and the CH2 domain includes one or more amino acid modifications.

[0152] Modified Fc CH3 domain In certain embodiments, the fusion protein comprises a heterodimeric immunoglobulin Fc containing a modified CH3 domain, the modified CH3 domain containing one or more asymmetric amino acid modifications. As used herein, “asymmetric amino acid modification” means that an amino acid at a particular position on the first Fc polypeptide is modified differently from an amino acid at a corresponding position on the second Fc polypeptide. These asymmetric amino acid modifications may include modifications of only one of two amino acids at the corresponding positions on each Fc polypeptide, or may include modifications of both amino acids at the corresponding positions on the first and second Fc polypeptides, respectively.

[0153] In certain embodiments, the fusion protein comprises a heterodimer Fc containing a modified CH3 domain, the modified CH3 domain containing one or more asymmetric amino acid modifications that promote the formation of the heterodimer Fc rather than the homodimer Fc. Amino acid modifications that can be made to the CH3 domain of Fc to promote the formation of the heterodimer Fc are known in the art and include, for example, those described in International Publication No. WO96 / 027011 ("Knob-into-Hole"), Gunasekaran et al., 2010, J Biol Chem, 285, 19637-46 ("Electrostatic Steering"), Davis et al., 2010, Prot Eng Des Sel, 23(4):195-202 (Chain Exchange Domain (SEED) Technique), and Labrijn et al., 2013, Proc Natl Acad Sci USA, 110(13):5145-50 (Fab Arm Exchange). Another example is the approach described in international publication numbers WO2012 / 058768 and WO2013 / 063702, which combines positive and negative design strategies to produce an asymmetrically modified stable Fc region.

[0154] In certain embodiments, the fusion protein comprises a heterodimer Fc having a modified CH3 domain as described in international publication number WO2012 / 058768 or international patent publication number WO2013 / 063702.

[0155] In some embodiments, the fusion protein comprises a heterodimer human IgG1 Fc having a modified CH3 domain. Table 2 below provides amino acid sequences of human IgG1 Fc sequences corresponding to amino acids 231–447 of the full-length human IgG1 heavy chain. The CH2 domain is typically defined as containing amino acids 231–340 of the full-length human IgG1 heavy chain, and the CH3 domain is typically defined as containing amino acids 341–447 of the full-length human IgG1 heavy chain.

[0156] In certain embodiments, the fusion protein comprises a heterodimer Fc having a modified CH3 domain that includes one or more asymmetric amino acid modifications that promote the formation of a heterodimer Fc rather than a homodimer Fc, the modified CH3 domain comprising a first Fc polypeptide with amino acid modifications at positions F405 and Y407, and a second Fc polypeptide with amino acid modifications at positions T366 and T394. In some embodiments, the amino acid modification at position F405 of the first Fc polypeptide of the modified CH3 domain is F405A, F405I, F405M, F405S, F405T, or F405V. In some embodiments, the amino acid modification at position Y407 of the first Fc polypeptide of the modified CH3 domain is Y407I or Y407V. In some embodiments, the amino acid modification at position T366 of the second Fc polypeptide of the modified CH3 domain is T366I, T366L, or T366M. In some embodiments, the amino acid modification at position T394 of the modified CH3 domain second Fc polypeptide is T394W. In some embodiments, the modified CH3 domain first Fc polypeptide further includes an amino acid modification at position L351. In some embodiments, the amino acid modification at position L351 in the modified CH3 domain first Fc polypeptide is L351Y. In some embodiments, the modified CH3 domain second Fc polypeptide further includes an amino acid modification at position K392. In some embodiments, the amino acid modification at position K392 in the modified CH3 domain second Fc polypeptide is K392F, K392L, or K392M. In some embodiments, one or both of the modified CH3 domain first and second Fc polypeptides further include an amino acid modification T350V.

[0157] In certain embodiments, the fusion protein comprises a heterodimer Fc having a modified CH3 domain containing one or more asymmetric amino acid modifications that promote the formation of a heterodimer Fc rather than a homodimer Fc, wherein the modified CH3 domain comprises a first Fc polypeptide containing amino acid modifications F405A, F405I, F405M, F405S, F405T, or F405V together with amino acid modifications Y407I or Y407V, and a second Fc polypeptide containing amino acid modifications T366I, T366L, or T366M together with amino acid modifications T394W. In some embodiments, the first Fc polypeptide of the modified CH3 domain further comprises amino acid modification L351Y. In some embodiments, the second Fc polypeptide of the modified CH3 domain further comprises amino acid modifications K392F, K392L, or K392M. In some embodiments, one or both of the first and second Fc polypeptides of the modified CH3 domain further comprises the amino acid modification T350V.

[0158] In a particular embodiment, the fusion protein comprises a heterodimer Fc having a modified CH3 domain, having a first Fc polypeptide having amino acid modifications at positions F405 and Y407 and optionally further having an amino acid modification at position L351, and a second Fc polypeptide having amino acid modifications at positions T366 and T394 and optionally further having an amino acid modification at position K392, wherein the first Fc polypeptide has amino acids at either or both of positions S400 or Q347. The second Fc polypeptide further includes modifications and / or further includes amino acid modifications at either or both of positions K360 or N390, wherein the amino acid modification at position S400 is S400E, S400D, S400R, or S400K, the amino acid modification at position Q347 is Q347R, Q347E, or Q347K, the amino acid modification at position K360 is K360D, or K360E, and the amino acid modification at position N390 is N390R, N390K, or N390D.

[0159] In certain embodiments, the fusion protein comprises a heterodimer Fc containing a modified CH3 domain having one of the modifications shown in Table 2: variant 1, variant 2, variant 3, variant 4, or variant 5. In certain embodiments, the CH3 domain has an amino acid sequence corresponding to SEQ ID NO: 4 or SEQ ID NO: 5. In certain embodiments, CH3 has an amino acid sequence substantially identical to SEQ ID NO: 4 or SEQ ID NO: 5. In certain embodiments, the CH3 domain has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 4 or SEQ ID NO: 5.

[0160] (Table 2) TIFF0007846668000002.tif104154

[0161] Modified Fc CH2 domain In certain embodiments, the fusion protein comprises an IgG Fc-based Fc having a modified CH2 domain. In some embodiments, the fusion protein comprises an IgG Fc-based Fc having a modified CH2 domain, and the modification of the CH2 domain results in altered binding to one or more Fc receptors (FcRs), such as receptors of the FcγRI, FcγRII, and FcγRIII subclasses.

[0162] Numerous amino acid modifications to the CH2 domain that selectively alter the affinity of Fc to different Fcγ receptors are known in the art. Both amino acid modifications that result in increased binding and those that result in decreased binding may be useful in certain indications. For example, increased binding affinity of Fc to FcγRIIIa (activating receptor) results in increased antibody-dependent cell-mediated cytotoxicity (ADCC), and consequently, increased lysis of target cells. Decreased binding to FcγRIIb (inhibitory receptor) may similarly be beneficial in some situations. In certain indications, a reduction or elimination of ADCC and complement-mediated cytotoxicity (CDC) may be desired. In such cases, modified CH2 domains containing amino acid modifications that result in increased binding to FcγRIIb or amino acid modifications that reduce or eliminate the binding of the Fc region to all Fcγ receptors ("knockout" variants) may be useful.

[0163] Examples of amino acid modifications to the CH2 domain that alter Fc binding by the Fcγ receptor include, but are not limited to, the following: S298A / E333A / K334A and S298A / E333A / K334A / K326A (increased affinity for FcγRIIIa) (Lu, et al., 2011, J Immunol Methods, 365(1-2):132-41); F243L / R292P / Y300L / V305I / P396L (increased affinity for FcγRIIIa) (Stavenhagen, et al., 2007, Cancer Res, 67(18):8882-90); F243L / R292P / Y300L / L235V / P396L (increased affinity for FcγRIIIa) (Nordstrom JL, et al.) al., 2011, Breast Cancer Res, 13(6):R123); F243L (increased affinity for FcγRIIIa) (Stewart, et al., 2011, Protein Eng Des Sel., 24(9):671-8); S298A / E333A / K334A (increased affinity for FcγRIIIa) (Shields, et al., 2001, J Biol Chem, 276(9):6591-604); S239D / I332E / A330L and S239D / I332E (increased affinity for FcγRIIIa) ((Lazar, et al., 2006, Proc Natl Acad Sci USA, 103(11):4005-10), and S239D / S267E and S267E / L328F (increased affinity for FcγRIIb) (Chu, et al., 2008, Mol Immunol, 45(15):3926-33).

[0164] Additional modifications affecting Fc binding to the Fcγ receptor are described in Therapeutic Antibody Engineering (Strohl & Strohl, Woodhead Publishing series in Biomedicine No. 11, ISBN 1 907568 37 9, Oct 2012, page 283).

[0165] In certain embodiments, the fusion protein comprises an IgG Fc-based Fc having a modified CH2 domain, the modified CH2 domain comprising one or more amino acid modifications (i.e., a "knockout" variant) resulting in reduced or excluded binding of the Fc region to all Fcγ receptors.

[0166] Various publications describe strategies used to manipulate antibodies to generate "knockout" variants (see, for example, Strohl, 2009, Curr Opin Biotech 20:685-691, and Strohl & Strohl, “Antibody Fc engineering for optimal antibody performance” In Therapeutic Antibody Engineering, Cambridge: Woodhead Publishing, 2012, pp 225-249). These strategies include reducing effector function through glycosylation modification (described in more detail below), using the IgG2 / IgG4 skeleton, or introducing mutations in the Fc hinge or CH2 domain (see also U.S. Patent Publication No. 2011 / 0212087, International Publication No. WO2006 / 105338, U.S. Patent Publication No. 2012 / 0225058, U.S. Patent Publication No. 2012 / 0251531 and Strop et al., 2012, J.Mol.Biol., 420:204-219).

[0167] Specific, non-limiting examples of known amino acid modifications for reducing FcγR and / or complement binding to Fc include those identified in Table 3.

[0168] (Table 3) TIFF0007846668000003.tif99150

[0169] Additional examples include Fc regions manipulated to include amino acid modifications L235A / L236A / D265S. Furthermore, asymmetric amino acid modifications in the CH2 domain that reduce Fc binding to all Fcγ receptors are described in international publication number WO2014 / 190441.

[0170] In certain embodiments, the CH2 domain has an amino acid sequence corresponding to SEQ ID NO: 6. In certain embodiments, CH2 has an amino acid sequence substantially identical to SEQ ID NO: 6. In certain embodiments, the CH2 domain has an amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 6.

[0171] Antibody-drug conjugates Certain embodiments of the fusion proteins described herein include a biologically functional protein that is an antibody conjugated with a drug, i.e., an antibody-drug conjugate (ADC). The drug in the ADC may be any therapeutic molecule, e.g., a toxin, a chemotherapeutic agent, or a small molecule inhibitor. The ADC may be conjugated with a drug via a linker that may be a cleavable or non-cleavable linker. A cleavable linker may be readily cleaved under intracellular conditions, for example, by a lysosomal process. Examples of cleavable linkers include those that are protease-sensitive, acid-sensitive, reduction-sensitive, or photocleavable. Drug conjugation may be carried out by any method known in the art, including, but not limited to, lysine or cysteine ​​conjugation, bisthiol linkers, conjugation using glycosylation sites of antibodies, ultraviolet conjugation, and the use of non-natural amino acids.

[0172] Peptide linker, protease, and protease cleavage site The fusion proteins described herein include at least first and second peptide linkers. A peptide linker is a peptide that connects or links other peptides or polypeptides. In certain embodiments, the peptide linker fuses a biologically functional protein, such as an antibody or a polypeptide with a dimeric Fc backbone, to the ligand and / or receptor of a ligand-receptor pair.

[0173] In certain embodiments, if a biologically functional protein contains an Fc region, the Fc polypeptide can be fused to the ligand or receptor of a ligand-receptor pair, or linked to the ligand or receptor of a ligand-receptor pair via a linker. In certain embodiments, the ligand is fused to the terminal of a first polypeptide via a first peptide linker, and the receptor is fused to the similar terminal of each of the second polypeptides via a second peptide linker. In certain embodiments of the fusion proteins described herein, both the receptor and ligand are fused to the N-terminus of the first and second polypeptides via peptide linkers. In certain embodiments of the fusion proteins described herein, both the receptor and ligand are fused to the C-terminus of the first and second polypeptides via peptide linkers.

[0174] The peptide linker is of sufficient length to enable ligand-receptor pairing. In addition to providing spacing functionality, the peptide linker may provide suitable flexibility or rigidity for appropriately oriented one or more domains of the fusion protein as herein, both within the fusion protein and between or within the fusion protein and its target(s). Furthermore, the peptide linker may support the expression of the full-length fusion protein and the stability of the purified protein both in vitro and in vivo after administration to subjects requiring it, e.g., humans, preferably being non-immunogenic or having reduced immunogenicity in those same subjects. In certain embodiments, the peptide linker may comprise a human immunoglobulin hinge, part or all of the stem region of a type C lectin, a family of type II membrane proteins, or a combination thereof.

[0175] In certain embodiments, the peptide linker is long enough to allow ligand-receptor pairing and is approximately 2 to approximately 150 amino acids. In certain embodiments, the peptide linker is in the range of approximately 3 to approximately 50 amino acids, or approximately 5 to approximately 20 amino acids, or approximately 10 to approximately 50 amino acids, or approximately 2 to approximately 40 amino acids, or approximately 8 to approximately 20 amino acids, approximately 10 to approximately 60 amino acids, approximately 10 to approximately 30 amino acids, or approximately 15 to approximately 25 amino acids. In some embodiments, the peptide linker is 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, or 60 amino acids.

[0176] At least one of the peptide linkers of the fusion protein described herein includes a protease cleavage site, also called a cleavage sequence. In certain embodiments, the fusion protein comprises at least one peptide linker containing a protease cleavage site and at least one peptide linker not containing a protease cleavage site. When used, the protease cleavage site is located within the peptide linker to maximize recognition and cleavage by the desired protease(s) and minimize recognition and nonspecific cleavage by other proteases. The peptide linker may contain one or more cleavage sites. In this regard, the fusion protein can be cleaved by one, two, three, four, five or more proteases. The protease cleavage site(s) may also be located within (or, in other words, surrounded by) the peptide linker and positioned within the fusion protein as a whole to achieve best desired cleavage and release of the post-cleavage fusion protein fragment (e.g., ligand of a receptor-ligand-ligand pair, receptor of a ligand-receptor pair, or both ligand and receptor). Polypeptide sites that are fused to a fusion protein by a peptide linker and released from the fusion protein after cleavage of the peptide linker may be referred to herein as cleavable sites (CMs). In certain embodiments in which the fusion protein comprises multiple CMs, they may be fused to the fusion protein by the same or different peptide linkers having the same or different cleavage sites.

[0177] The protease cleavage site or sequence may be selected based on a protease co-localized in a tissue where the activity of the fusion protein or a biologically functional protein is desired. The cleavage site may function as a substrate for multiple proteases, e.g., a serine protease and a second different protease, e.g., a matrix metalloproteinase (MMP). In some embodiments, the cleavage site may function as a substrate for multiple serine proteases, e.g., a matryptase and a urokinase-type plasminogen activator (uPA). In some embodiments, the peptide linker may function as a substrate for multiple MMPs, e.g., MMP9 and MMP14.

[0178] In a particular embodiment, the peptide linker is approximately 0.001–1500 × 10 4 M -1 S -1 Or at least 0.001, 0005, 0.01, 0.05, 0.1, 0.5, 1, 2.5, 5, 7.5, 10, 15, 20, 25, 50, 75, 100, 125, 150, 200, 250, 500, 750, 1000, 1250, or 1500 × 10 4 M -1 S -1 It is specifically cleaved by proteases at the following rate.

[0179] For specific enzymatic cleavage, contact occurs between the enzyme and the peptide linker. In certain embodiments, if the fusion protein contains at least a first peptide linker and sufficient enzymatic activity is present, the peptide linker is cleaved. Sufficient enzymatic activity may refer to the ability of the enzyme to contact the peptide linker and result in cleavage. It is readily conceivable that the enzyme may be in the vicinity of the peptide linker but be unable to cleave it due to other cellular factors or protein modifications of the enzyme.

[0180] In certain embodiments, the peptide linker includes a protease cleavage site of 5–10 amino acids, 7–10 amino acids, or 8–10 amino acids. In other embodiments, the peptide linker consists of a protease cleavage site of 5–10 amino acids, 7–10 amino acids, or 8–10 amino acids. In one embodiment, the protease cleavage site is preceded at the N-terminus by a linker sequence of approximately 1–20 amino acids, 2–5 amino acids, 5–10 amino acids, 10–15 amino acids, 10–20 amino acids, 12–16 amino acids, or approximately 5 or approximately 10 amino acids. In other embodiments, the protease cleavage site is followed at the C-terminus by a linker sequence of approximately 1–20 amino acids, 2–5 amino acids, 5–10 amino acids, 10–15 amino acids, 10–20 amino acids, 12–16 amino acids, or in some cases, approximately 5 or approximately 10 amino acids. In yet another embodiment, the protease cleavage site is preceded by a linker sequence at the N-terminus and followed by a linker sequence at the C-terminus. Thus, in a particular embodiment, the protease cleavage site is located between two linkers. Either the linker on the N-terminus or C-terminus of the protease cleavage site can be of various lengths, such as approximately 2–20, 6–20, 8–15, 8–10, 10–18, or 12–16 amino acids. In a particular embodiment, the N-terminal or C-terminal linker sequence is approximately 3 amino acids long or approximately 5 amino acids long.

[0181] The exemplary peptide linkers of this disclosure include a variety of proteases, for example, but not limited to serine proteases, MMPs (MMP1, MMP2, MMP3, MMP7, MMP8, MMP9, MMP10, MMP11, MMP12, MMP13, MMP14, MMP15, MMP16, MMP17, MMP18 (collagenase 4), MMP19, MMP20, MMP21, etc.), adamaricin, ceralisin, astacin, caspases (e.g., caspase 1, caspase 2, caspase 3, caspase 4, caspase 5, caspase 6, caspase 7, caspase 8, caspase 9, caspase The protease comprises one or more protease cleavage sites recognized by caspase 10, caspase 11, caspase 12, caspase 13, caspase 14), cathepsin (e.g., cathepsin A, cathepsin B, cathepsin D, cathepsin E, cathepsin K, cathepsin S), granzyme B, guanidinobenzoatase (GB), hepsin, elastase, regmine, matryptase, matryptase 2, meprin, neurosin, MT-SP1, neprilysin, plasmin, PSA, PSMA, TACE, TMPRSS3 / 4, uPA, and calpain, FAP, and KLK. In some embodiments, the protease is uPA or matryptase.

[0182] In certain embodiments, the peptide linker includes cleavage sites that are cleaved by multiple proteases. In this regard, each cleavage site can be cleaved by one, two, three, four, five or more proteases. In other embodiments, the peptide linker may include cleavage sites that are substantially cleaved by one enzyme but not substantially cleaved by another. Thus, in some embodiments, the peptide linker includes cleavage sites with high specificity. "High specificity" means >90% cleavage observed by a particular protease and less than 50% cleavage observed by other proteases. In certain embodiments, the peptide linker includes cleavage sites that exhibit >80% cleavage by one protease but less than 50% cleavage by other proteases. In certain embodiments, the peptide linker includes cleavage sites that exhibit >70%, 75%, 76%, 77%, 78%, or 79% cleavage by one protease, but less than 65%, 60%, 55%, 54%, 53%, 52%, 51%, 50%, 49%, 48%, 47%, 46%, or 45% cleavage by another protease. For example, in one embodiment, a cleavage site may be >90% cleaved by matryptase and about 75% cleaved by uPa and plasmin. In another embodiment, a cleavage site may be cleaved by uPa and matryptase, but specific cleavage by plasmin is not observed. In yet another embodiment, a cleavage site may be cleaved by uPa but not by matryptase or plasmin. In one embodiment, a cleavage site may exhibit some level of resistance to nonspecific protease cleavage (e.g., cleavage by plasmin or other nonspecific proteases). In this regard, the protease cleavage site may exhibit "high nonspecific protease resistance" (<25% cleavage by plasmin or an equivalent nonspecific protease), "moderate nonspecific protease resistance" (<75% cleavage by plasmin or an equivalent nonspecific protease), or "low nonspecific protease resistance" (up to approximately 90% cleavage by plasmin or an equivalent nonspecific protease).Such cleavage activity can be measured using assays known in the art, for example, by incubation with a suitable protease and subsequent SDS-PAGE or other analysis. In certain embodiments, the protease cleavage site may exhibit maximum complete resistance to protease cleavage up to 24 hours of contact with the protease. In other embodiments, the protease cleavage sequence may exhibit maximum complete resistance to nonspecific protease cleavage after 0.5 to 36 hours of contact with the protease. In yet another embodiment, the protease cleavage sequence may exhibit maximum complete resistance to nonspecific protease cleavage after 0.5, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 12, 20, 24, 36, 48, or 72 hours of contact with a suitable protease.

[0183] Therefore, in certain embodiments, the cleavage sites are selected based on their preference for various desired proteases. Thus, a desired cleavage profile for a particular peptide linker, including the cleavage sites, can be selected for a desired purpose (e.g., high-specific cleavage in a particular tumor microenvironment or a particular organ) in which a particular protease or set of proteases can demonstrate high, specific, elevated, efficient, moderate, low, or no cleavage of a particular cleavage site within the peptide linker. Methods for determining cleavage are known in the art.

[0184] In certain embodiments, the peptide linker may include one or more cleavage sites arranged in tandem, with or without additional linkers between each cleavage site. In certain embodiments, the peptide linker includes a first cleavage site and a second cleavage site, the first cleavage site being cleaved by a first protease and the second cleavage site being cleaved by a second protease. In non-limiting examples, the peptide linker may include a first cleavage site cleaved by a matryptase and uPa and a second cleavage site cleaved by an MMP. In certain embodiments, the peptide linker includes a first cleavage site, a second cleavage site and a third cleavage site, the first cleavage site being cleaved by a first protease, the second cleavage site being cleaved by a second protease and the third cleavage site being cleaved by a third protease.

[0185] Illustrative proteolytic enzymes and their recognition sequences useful in the fusion proteins described herein can be identified by those skilled in the art and are known in the art, including those listed in the MEROPS database (see, for example, Rawlings, et al., Nucleic Acids Research, Volume 46, Issue D1, 4 January 2018, Pages D624-D632) and elsewhere (Hoadley et al., Cell, 2018; GTEX Consortium, Nature, 2017; Robinson et al., Nature, 2017).

[0186] Other methods can also be used to identify the cutting sites used herein, for example, as described in U.S. Patent Nos. 9,453,078, 10,138,272, 9,562,073 and published International Patent Applications WO2015048329;WO2015116933;WO2016118629.

[0187] Accordingly, one embodiment of the present disclosure provides a fusion protein comprising at least two peptide linkers, at least one of which comprises one or more of the cleavage sites shown herein. In one embodiment, the present disclosure provides a fusion protein comprising a peptide linker, wherein the peptide linker comprises a protease cleavage site and is cleavable by uPA. In one embodiment, the present disclosure provides a fusion protein comprising a peptide linker, wherein the peptide linker comprises the amino acid sequence MSGRSANA (SEQ ID NO: 28). In a particular embodiment, the peptide linker sequence comprises at least one protease cleavage site selected from TSGRSANP, LSGRSDNH, GSGRSAQV, GSSRNADV, GTARSDNV, GTARSDNV, GGGRVNNV, MSARILQV, or GKGRSANA (each SEQ ID NO: 30-37).

[0188] In certain embodiments, the fusion protein comprising the peptide linker described herein comprises two heterogeneous polypeptides, namely a first polypeptide located at the amino (N) terminus of the peptide linker and a second polypeptide located at the carboxyl (C) terminus of the peptide linker, thereby separating the two heterogeneous polypeptides by the peptide linker.

[0189] In certain embodiments, the fusion protein includes at least one peptide linker that does not contain a protease cleavage site. In certain embodiments, the peptide linker includes the amino acid sequence (EAAAK)n (wherein n is an integer from 1 to 5). In some embodiments, the peptide linker is EAAAK (SEQ ID NO: 39). In some embodiments, the peptide linker EAAAKEAAAK (SEQ ID NO: 38). In some embodiments, the peptide linker includes a polyproline linker, optionally having the amino acid sequence PPP (SEQ ID NO: 41) or PPPP (SEQ ID NO: 40). In certain embodiments, the linker is a glycine (G)-proline (P) polypeptide linker, optionally having the amino acid sequence GPPPG, GGPPPGG, GPPPPG, or GGPPP P It is GG. In a particular embodiment, the peptide linker is Gly n It is a Ser linker. In certain embodiments, the peptide linker is (Gly3Ser) n (Gly4Ser)1, (Gly3Ser)1(Gly4Ser) n (Gly3Ser) n (Gly4Ser) n , or (Gly4Ser) n The formula includes an amino acid sequence of (wherein n is an integer from 1 to 5). In certain embodiments, a peptide linker suitable for connecting different domains includes a sequence containing a glycine-serine linker, for example, (G) m S) n -GG, (SGn)m, (SEGn)m (where m and n are between 0 and 20) are included.

[0190] In certain embodiments, the peptide linker is an amino acid sequence obtained from, derived from, or designed from an antibody hinge region sequence, a sequence that links the binding domain to a receptor, or a sequence that links the binding domain to a cell surface transmembrane region or membrane anchor. In some embodiments, the peptide linker has at least one cysteine ​​capable of participating in at least one disulfide bond under physiological or other standard peptide conditions (e.g., peptide purification conditions, peptide storage conditions). In certain embodiments, a peptide linker corresponding to or similar to an immunoglobulin hinge peptide holds a cysteine ​​corresponding to a hinge cysteine ​​positioned relative to the amino terminus of the hinge. In further embodiments, the peptide linker is from an IgG1 hinge and is modified to remove any cysteine ​​residue, or is an IgG1 hinge having one or two cysteines corresponding to a hinge cysteine.

[0191] In certain embodiments, the peptide linker for use herein may include a "modified wild-type immunoglobulin hinge region" or a "modified immunoglobulin hinge region." Such altered hinge regions refer to (a) wild-type immunoglobulin hinge regions having up to 30 percent amino acid changes (e.g., up to 25 percent, 20 percent, 15 percent, 10 percent, or 5 percent amino acid substitutions or deletions), (b) portions of wild-type immunoglobulin hinge regions having up to 30 percent amino acid changes (e.g., up to 25 percent, 20 percent, 15 percent, 10 percent, or 5 percent amino acid substitutions or deletions) and being at least 10 amino acids long (e.g., at least 12, 13, 14, or 15 amino acids), or (c) portions of wild-type immunoglobulin hinge regions including core hinge regions (parts of which may be 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, or 15 amino acids long). In certain embodiments, one or more cysteine ​​residues in a wild-type immunoglobulin hinge region, such as the IgG1 hinge including the upper and core regions, may be replaced by one or more other amino acid residues (e.g., one or more serine residues). The altered immunoglobulin hinge region may have, alternatively or additionally, proline residues in the wild-type immunoglobulin hinge region, such as the IgG1 hinge including the upper and core regions, replaced by other amino acid residues (e.g., serine residues).

[0192] Alternative hinge and linker sequences that can be used as linking regions can be made from a portion of a cell surface receptor that links IgV-like or IgC-like domains. Regions between IgV-like domains in which the cell surface receptor contains multiple IgV-like domains in tandem, and regions between IgC-like domains in which the cell surface receptor contains multiple tandem IgC-like domains, can also be used as linking regions or linker peptides. In certain embodiments, the hinge and linker sequences may be 5-60 amino acid long and may be primarily flexible, but may also provide more rigid properties, and may contain primarily helical structures with minimal beta-sheet structures.

[0193] In certain embodiments, the proteases described herein are expressed in higher concentrations in vivo near the tumor microenvironment of specific target cells, e.g., target tumor cells. A variety of different conditions or diseases are known in which the target (e.g., a specific tumor type, a specific tumor expressing a specific tumor-associated antigen) co-localizes with the protease (the substrates of the protease are known in the art). In the case of cancer, the target tissue may be cancerous tissue, particularly cancerous tissue of solid tumors. Increased levels of proteases in numerous cancers, e.g., humoral or solid tumors, have been reported in the literature. See, for example, La Rocca et al, (2004) British J. of Cancer 90(7):1414-1421.

[0194] In certain embodiments, the fusion protein comprises ligand-linker-VL, receptor-linker-VL, ligand-linker-VH, or receptor-linker-VH from the N-terminus to the C-terminus.

[0195] In certain embodiments, the fusion protein comprises, from N-terminus to C-terminus, ligand-cleaving linker-VL, receptor-cleaving linker-VL, ligand-cleaving linker-VH, or receptor-cleaving linker-VH.

[0196] In certain embodiments, the fusion protein comprises, from N-terminus to C-terminus, ligand-linker(sequence number 114)-VL, receptor-linker(sequence number 114)-VL, ligand-linker(sequence number 14)-VH, or receptor-linker(sequence number 14)-VH.

[0197] In certain embodiments, the fusion protein comprises ligand-linker(sequence number 145)-VL, receptor-linker(sequence number 145)-VL, ligand-linker(sequence number 145)-VH, or receptor-linker(sequence number 145)-VH from the N-terminus to the C-terminus.

[0198] In certain embodiments, the fusion protein comprises ligand-linker(sequence number 147)-VL, receptor-linker(sequence number 147)-VL, ligand-linker(sequence number 147)-VH, or receptor-linker(sequence number 147)-VH from the N-terminus to the C-terminus.

[0199] In certain embodiments, the fusion protein comprises ligand-linker(sequence ID 154)-VL, receptor-linker(sequence ID 154)-VL, ligand-linker(sequence ID 154)-VH, or receptor-linker(sequence ID 154)-VH, from the N-terminus to the C-terminus.

[0200] In certain embodiments, the fusion protein comprises ligand-linker(sequence number 203)-VL, receptor-linker(sequence number 203)-VL, ligand-linker(sequence number 203)-VH, or receptor-linker(sequence number 203)-VH, from the N-terminus to the C-terminus.

[0201] In certain embodiments, the fusion protein includes a ligand-linker-Fc or receptor-linker-Fc from the N-terminus to the C-terminus.

[0202] In certain embodiments, the fusion protein comprises a ligand-cleaving linker-Fc or receptor-cleaving linker-Fc from the N-terminus to the C-terminus.

[0203] In certain embodiments, the fusion protein includes a ligand-cleaving linker (SEQ ID NO: 28)-Fc or a receptor-cleaving linker (SEQ ID NO: 28)-Fc from the N-terminus to the C-terminus.

[0204] In certain embodiments, the fusion protein includes ligand-linker-Fc1 or receptor-linker-Fc1 from the N-terminus to the C-terminus.

[0205] In certain embodiments, the fusion protein comprises a ligand-cleaving linker-Fc2 or receptor-cleaving linker-Fc2 from the N-terminus to the C-terminus.

[0206] In certain embodiments, Fc1 and Fc2 can form a heterodimer. In certain embodiments, Fc1 is linked to a ligand and Fc2 is linked to a receptor. In certain embodiments, the linker connecting the ligand and Fc1 is cleavable, and the linker connecting the receptor and Fc2 is non-cleavable. In certain embodiments, the linker connecting the ligand and Fc1 is non-cleavable, and the linker connecting the receptor and Fc2 is cleavable. In certain embodiments, the linker connecting the ligand and Fc1 is cleavable, and the linker connecting the receptor and Fc2 is cleavable. In certain embodiments, the linker connecting the ligand and Fc1 is non-cleavable, and the linker connecting the receptor and Fc2 is non-cleavable.

[0207] target In some embodiments, the antigen-binding domain of the fusion protein described herein specifically binds to cell surface molecules. In certain embodiments, the antigen-binding domain of the fusion protein specifically binds to tumor-associated antigens (TAAs). TAAs are any antigenic substance expressed on the surface of tumor cells. In some embodiments, the antigen-binding domain is fibroblast-activating protein alpha (FAPa), trophoblast glycoprotein (5T4), tumor-associated calcium signaling molecule 2 (Trop2), fibronectin EDB (EDB-FN), fibronectin F.IIIB domain, CGS-2, EpCAM, EGFR, HER-2, HER-3, cMet, CEA, and FOLR1, EpCAM, EGFR, HER-2, HER-3, cMet, CEA, and FOLR1, EpCAM, EGFR, HER-2, HER-3, c-Met, FOLR1, PSMA, CD38, BCMA, and CEA, 5T4, AFP, B7-H3, cadherin-6, CAIX, CD117, CD123, CD138, CD166, C D19, CD20, CD205, CD22, CD30, CD33, CD40, CD352, CD37, CD44, CD52, CD56, CD70, CD71, CD74, CD79b , DLL3, DR5, EphA2, FAP, FGFR2, FGFR3, GPC3, gpA33, FLT-3, gpNMB, HPV-16E6, HPV-16E7, ITGA2, IT It specifically binds to TAAs selected from GA3, SLC39A6, MAGE, mesothelin (MSLN), Mucl, Mucl6, NaPi2b, Nectin-4, P-cadherin, NY-ESO-1, PRLR, PSCA, PTK7, ROR1, SLC44A4, SLTRK5, SLTRK6, STEAP1, TIM1, tissue factor (TF), Trop2, and WT1.

[0208] In some embodiments, the antigen-binding domain specifically binds to immune checkpoint proteins. Examples of immune checkpoint proteins include, but are not limited to, CD27, CD137, 2B4, TIGIT, CD155, ICOS, HVEM, CD40L, LIGHT, TIM-1, 0X40, DNAM-1, PD-L1, PD1, PD-L2, CTLA-4, CD80, CD40, CEACAM1, CD48, CD70, A2AR, CD39, CD73, B7-H3, B7-H4, BTLA, IDOl, ID02, TDO, KIR, LAG-3, TIM-3, VISTA, CD47, or SIRPa.

[0209] In some embodiments, the antigen-binding domain specifically binds to antigens expressed on virus-infected cells, bacterial-infected cells, damaged red blood cells, arterial plaque cells, or inflammatory or fibrous tissue cells.

[0210] In certain embodiments, the antigen-binding domain specifically binds to cytokine receptors. Examples of cytokine receptors include type I cytokine receptors, e.g., GM-CSF receptor, G-CSF receptor, type I IL receptor, Epo receptor, LIF receptor, CNTF receptor, TPO receptor; type II cytokine receptors, e.g., IFN-alpha receptors (IFNAR1, IFNAR2), IFB-beta receptor, IFN-gamma receptors (IFNGR1, IFNGR2), type II IF receptor; and chemokine receptors, e.g., CC chemokine receptor, CXC chemokine receptor, CX3C chemokine receptor, XC chemokine receptor. Cain receptors; tumor necrosis receptor superfamily receptors, e.g., TNFRSF5 / CD40, TNFRSF8 / CD30, TNFRSF7 / CD27, TNFRSFA / TNFR1 / CD120a, TNFRSF1B / TNFR2 / CD120b; TGF-beta receptors, e.g., TGF-beta receptor 1, TGF-beta receptor 2; Ig superfamily receptors, e.g., IF-1 receptor, CSF-1R, PDGFR (PDGFRA, PDGFRB), SCFR, are included but not limited to these.

[0211] In certain embodiments, the antigen-binding domain of the fusion protein described herein specifically binds to at least one target molecule or target in vivo. In certain embodiments, the target is differentiation antigen group 3 (CD3), human epidermal growth factor receptor 2 (HER2), epidermal growth factor receptor (EGFR), mesothelin (MSLN), tissue factor (TF), differentiation antigen group 19 (CD19), tyrosine protein kinase Met (c-Met), differentiation antigen group 40 (CD40), cadherin 3 (CDH3), or a combination thereof. In certain embodiments, the fusion protein comprises an antibody, the antigen-binding domain of which at least one antigen-binding domain binds to an epitope on CD3, HER2, EGFR, MSLN, TF, CD19, c-Met, CD40, CDH3, or a combination thereof.

[0212] In some embodiments, the target is HER2, and the anti-HER2 paratope of the fusion protein has a VH amino acid sequence corresponding to SEQ ID NO: 120 and a VL amino acid sequence corresponding to SEQ ID NO: 124. In certain embodiments, the anti-HER2 paratope has a VH amino acid sequence substantially identical to SEQ ID NO: 120 and a VL amino acid sequence substantially identical to SEQ ID NO: 124. In certain embodiments, the anti-HER2 paratope has a VH amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 120 and a VL amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 124. In certain embodiments, the anti-HER2 paratope has a VH amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 120 and a VL amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 124. In some embodiments, the anti-HER2 paratope includes scFv having an amino acid sequence corresponding to SEQ ID NO: 3. In some embodiments, the anti-HER2 includes VH having three CDRs, HCDR1, HDR2, and HCDR3, having amino acid sequences corresponding to SEQ ID NOs: 121, 122, and 123, respectively, and VL having three CDRs, LCDR1, LCDR2, and LCDR3, having amino acid sequences corresponding to SEQ ID NOs: 125, 126, and 127, respectively.

[0213] In some embodiments, the target is EGFR, and the anti-EGFR paratope of the fusion protein has a VH amino acid sequence corresponding to SEQ ID NO: 14 and a VL amino acid sequence corresponding to SEQ ID NO: 13. In certain embodiments, the anti-EGFR paratope has a VH amino acid sequence substantially identical to SEQ ID NO: 14 and a VL amino acid sequence substantially identical to SEQ ID NO: 13. In certain embodiments, the anti-EGFR paratope has a VH amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 14 and a VL amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 13. In certain embodiments, the anti-EGFR paratope has a VH amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 14 and a VL amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 13. In some embodiments, the anti-EGFR comprises a VH having three CDRs, HDR1, HDR2, and HDR3, each having amino acid sequences corresponding to SEQ ID NOs: 84, 85, and 86, respectively, and a VL having three CDRs, LCDR1, LCDR2, and LCDR3, each having amino acid sequences corresponding to SEQ ID NOs: 59, 60, and 61, respectively.

[0214] In some embodiments, the target is MSLN, and the anti-MSLN paratope of the fusion protein has a VH amino acid sequence corresponding to SEQ ID NO: 16 and a VL amino acid sequence corresponding to SEQ ID NO: 15. In certain embodiments, the anti-MSLN paratope has a VH amino acid sequence substantially identical to SEQ ID NO: 16 and a VL amino acid sequence substantially identical to SEQ ID NO: 15. In certain embodiments, the anti-MSLN paratope has a VH amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 16 and a VL amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 15. In certain embodiments, the anti-MSLN paratope has a VH amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 16 and a VL amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 15. In some embodiments, the anti-MSLN comprises a VH having three CDRs, HDR1, HDR2, and HDR3, each having amino acid sequences corresponding to SEQ ID NOs. 69, 70, and 71, respectively, and a VL having three CDRs, LCDR1, LCDR2, and LCDR3, each having amino acid sequences corresponding to SEQ ID NOs. 74, 75, and 76, respectively.

[0215] In some embodiments, the target is TF (tissue factor), and the anti-TF paratope of the fusion protein has a VH amino acid sequence corresponding to SEQ ID NO: 18 and a VL amino acid sequence corresponding to SEQ ID NO: 17. In certain embodiments, the anti-TF paratope has a VH amino acid sequence substantially identical to SEQ ID NO: 18 and a VL amino acid sequence substantially identical to SEQ ID NO: 17. In certain embodiments, the anti-TF paratope has a VH amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 18 and a VL amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 17. In certain embodiments, the anti-TF paratope has a VH amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 18 and a VL amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 17. In some embodiments, the anti-TF comprises a VH having three CDRs, HDR1, HDR2, and HDR3, each having amino acid sequences corresponding to SEQ ID NOs. 54, 55, and 56, respectively, and a VL having three CDRs, LCDR1, LCDR2, and LCDR3, each having amino acid sequences corresponding to SEQ ID NOs. 48, 49, and 50, respectively.

[0216] In some embodiments, the target is CD19, and the anti-CD19 paratope of the fusion protein has a VH amino acid sequence corresponding to SEQ ID NO: 20 and a VL amino acid sequence corresponding to SEQ ID NO: 19. In certain embodiments, the anti-CD19 paratope has a VH amino acid sequence substantially identical to SEQ ID NO: 20 and a VL amino acid sequence substantially identical to SEQ ID NO: 19. In certain embodiments, the anti-CD19 paratope has a VH amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 20 and a VL amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 19. In certain embodiments, the anti-CD19 paratope has a VH amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 20 and a VL amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 19. In some embodiments, anti-CD19 has a VH having three CDRs, HDR1, HDR2, and HDR3, each having amino acid sequences corresponding to SEQ ID NOs. 64, 65, and 66, respectively, and a VL having three CDRs, LCDR1, LCDR2, and LCDR3, each having amino acid sequences corresponding to SEQ ID NOs. 74, 75, and 165, respectively.

[0217] In some embodiments, the target is c-Met, and the anti-c-Met paratope of the fusion protein has a VH amino acid sequence corresponding to SEQ ID NO: 22 and a VL amino acid sequence corresponding to SEQ ID NO: 21. In certain embodiments, the anti-c-Met paratope has a VH amino acid sequence substantially identical to SEQ ID NO: 22 and a VL amino acid sequence substantially identical to SEQ ID NO: 21. In certain embodiments, the anti-c-Met paratope has a VH amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 22 and a VL amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 21. In certain embodiments, the anti-c-Met paratope has a VH amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 22 and a VL amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 21. In some embodiments, the anti-c-Met comprises a VH having three CDRs, HCDR1, HDR2, and HCDR3, each having amino acid sequences corresponding to SEQ ID NOs. 99, 100, and 101, respectively, and a VL having three CDRs, LCDR1, LCDR2, and LCDR3, each having amino acid sequences corresponding to SEQ ID NOs. 94, 95, and 96, respectively.

[0218] In some embodiments, the target is CDH3, and the anti-CDH3 paratope of the fusion protein has a VH amino acid sequence corresponding to SEQ ID NO: 24 and a VL amino acid sequence corresponding to SEQ ID NO: 23. In certain embodiments, the anti-CDH3 paratope has a VH amino acid sequence substantially identical to SEQ ID NO: 24 and a VL amino acid sequence substantially identical to SEQ ID NO: 23. In certain embodiments, the anti-CDH3 paratope has a VH amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 24 and a VL amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 23. In certain embodiments, the anti-CDH3 paratope has a VH amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 24 and a VL amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 23. In some embodiments, anti-CDH3 comprises VH having three CDRs, HCDR1, HDR2, and HCDR3, each having amino acid sequences corresponding to SEQ ID NOs. 89, 90, and 91, respectively, and VL having three CDRs, LCDR1, LCDR2, and LCDR3, each having amino acid sequences corresponding to SEQ ID NOs. 94, 95, and 96, respectively.

[0219] In some embodiments, the target is CD40, and the anti-CD40 paratope of the fusion protein has a VH amino acid sequence corresponding to SEQ ID NO: 172 and a VL amino acid sequence corresponding to SEQ ID NO: 177. In certain embodiments, the anti-CD40 paratope has a VH amino acid sequence substantially identical to SEQ ID NO: 172 and a VL amino acid sequence substantially identical to SEQ ID NO: 177. In certain embodiments, the anti-CD40 paratope has a VH amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 172 and a VL amino acid sequence that is approximately 80%, approximately 85%, approximately 90%, or approximately 95% identical to SEQ ID NO: 177. In certain embodiments, the anti-CD40 paratope has a VH amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 172 and a VL amino acid sequence that is approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to SEQ ID NO: 177. In some embodiments, anti-CD40 has a VH having three CDRs, HDR1, HDR2, and HDR3, each having amino acid sequences corresponding to SEQ ID NOs. 173, 174, and 175, respectively, and a VL having three CDRs, LCDR1, LCDR2, and LCDR3, each having amino acid sequences corresponding to SEQ ID NOs. 178, 179, and 180, respectively.

[0220] In certain embodiments, the antigen-binding domain of the fusion protein specifically binds to molecules on immune cells, such as polypeptides. In certain embodiments, the fusion protein includes both an antigen-binding domain that specifically binds to TAA and an antigen-binding domain that specifically binds to molecules on immune cells, such as polypeptides. Thus, in certain embodiments, the fusion protein binds to both tumor cells and immune cells. In certain embodiments, the immune cells are T cells. In certain embodiments, the immune cells are macrophages, dendritic cells, neutrophils, B cells, or NK cells.

[0221] In a particular embodiment, the fusion protein binds to the CD3 antigen on T cells and to one or more TAAs on tumor cells.

[0222] Masked T-cell Engagers T-cell engagers (TCEs) are often bispecific antibody polypeptide constructs that simultaneously bind to TAAs on tumor cells and CD3 epitopes on T cells, forming artificial immune synapses independent of the TCR. This activates T cells, allowing them to exert cytotoxic effects against tumor cells. Bispecific antibodies capable of targeting T cells to tumor cells have been identified and their efficacy in cancer treatment has been validated. Blinatumomab is an example of a bispecific anti-CD3-CD19 antibody in a format called BiTE™ (Bispecific T-cell Engager), and is specifically targeted for the treatment of B-cell diseases such as relapsed B-cell non-Hodgkin lymphoma and chronic lymphocytic leukemia (Baeuerle et al (2009) Cancer Research 12:4941-4944) and has received FDA approval. T-cell enhancers targeting other tumor-related antigens have also been developed, and some are in clinical trials: AMG110 / MT110 EpCAM for lung, gastric, and colorectal cancer; AMG211 / MEDI565 CEA for gastrointestinal adenocarcinoma; and AMG 212 / BAY2010112 PSMA for prostate cancer (see Suruadevara, CM et al, Oncoimmunology. 2015 Jun;4(6):e1008339). While these studies showed promising clinical efficacy, they were again hampered by severe dose-limiting toxicity, mainly due to cytokine release syndrome (CRS). This narrowed the treatment window. The use of masked T-cell binding paratopes that are primarily activated in the tumor microenvironment may reduce the toxicity of TCEs.

[0223] In certain embodiments, the fusion protein binds to the CD3 antigen on T cells and TAA on tumor cells. In certain embodiments, the fusion protein binds to the CD3 antigen on T cells, TAA on tumor cells, and the extracellular domain of IgSF on tumor cells. In certain embodiments, the fusion protein binds to the CD3 antigen on T cells, TAA on tumor cells, and the extracellular domain of IgSF on T cells.

[0224] In certain embodiments, the fusion protein is unmasked by a protease in the tumor microenvironment and binds to TAA on tumor cells and CD3 antigen on T cells, thereby resulting in crosslinking between T cells and tumor cells, as shown in Example 20. In certain embodiments, the unmasked fusion protein binds to both CD3 antigen on T cells and TAA and IgSF ligands on tumor cells, as illustrated in Figure 31. In certain embodiments, the binding of IgSF ligand (e.g., PD-L1) on tumor cells blocks checkpoint inhibition by inhibiting the binding of its IgSF receptor (e.g., PD-1) on T cells (Figure 31C).

[0225] In certain embodiments, the fusion protein comprises anti-CD3 paratopes VH and VL that are substantially identical to those of the paratopes shown in Table BB. In certain embodiments, the CD3 paratopes comprise the following VH and VL amino acid sequences: (a) VH containing the amino acid sequence corresponding to SEQ ID NO: 2 and VL containing the amino acid sequence corresponding to SEQ ID NO: 1; (b) VH containing the amino acid sequence corresponding to SEQ ID NO: 206 and VL containing the amino acid sequence corresponding to SEQ ID NO: 210; (c) VH containing the amino acid sequence corresponding to SEQ ID NO: 215 and VL containing the amino acid sequence corresponding to SEQ ID NO: 219; (d) VH containing the amino acid sequence corresponding to SEQ ID NO: 223 and VL containing the amino acid sequence corresponding to SEQ ID NO: 227; (d) VH containing the amino acid sequence corresponding to SEQ ID NO: 231 and VL containing the amino acid sequence corresponding to SEQ ID NO: 235; or (e) VH containing the amino acid sequence corresponding to SEQ ID NO: 239 and VL containing the amino acid sequence corresponding to SEQ ID NO: 243.

[0226] In a particular embodiment, the CD3 paratope includes VH and VL which are approximately 90%, approximately 91%, approximately 92%, approximately 93%, approximately 94%, approximately 95%, approximately 96%, approximately 97%, approximately 98%, or approximately 99% identical to the following: (a) VH containing the amino acid sequence corresponding to SEQ ID NO: 2 and VL containing the amino acid sequence corresponding to SEQ ID NO: 1; (b) VH containing the amino acid sequence corresponding to SEQ ID NO: 206 and VL containing the amino acid sequence corresponding to SEQ ID NO: 210; (c) VH containing the amino acid sequence corresponding to SEQ ID NO: 215 and VL containing the amino acid sequence corresponding to SEQ ID NO: 219; VH containing the amino acid sequence corresponding to SEQ ID NO: 223 and VL containing the amino acid sequence corresponding to SEQ ID NO: 227; VH containing the amino acid sequence corresponding to SEQ ID NO: 231 and VL containing the amino acid sequence corresponding to SEQ ID NO: 235; or VH contains the amino acid sequence corresponding to SEQ ID NO: 239, and VL contains the amino acid sequence corresponding to SEQ ID NO: 243.

[0227] In a particular embodiment, the anti-CD3 paratope comprises a VH containing three heavy-chain CDRs, HCDR1, HCDR2, and HCDR3, having amino acid sequences corresponding to SEQ ID NOs. 207, 208, and 209, and a VL containing three light-chain CDRs, LCDR1, LCDR2, and LCDR3, having amino acid sequences corresponding to SEQ ID NOs. 211, 212, and 214. In a particular embodiment, the anti-CD3 paratope comprises a VH containing three heavy-chain CDRs, HCDR1, HCDR2, and HCDR3, having amino acid sequences corresponding to SEQ ID NOs. 224, 225, and 226, and a VL containing three light-chain CDRs, LCDR1, LCDR2, and LCDR3, having amino acid sequences corresponding to SEQ ID NOs. 228, 229, and 230. In a particular embodiment, the anti-CD3 paratope comprises a VH containing three heavy-chain CDRs, HCDR1, HCDR2, and HCDR3, having amino acid sequences corresponding to SEQ ID NOs. 232, 233, and 234, and a VL containing three light-chain CDRs, LCDR1, LCDR2, and LCDR3, having amino acid sequences corresponding to SEQ ID NOs. 236, 237, and 238. In a particular embodiment, the anti-CD3 paratope comprises a VH containing three heavy-chain CDRs, HCDR1, HCDR2, and HCDR3, having amino acid sequences corresponding to SEQ ID NOs. 240, 241, and 242, and a VL containing three light-chain CDRs, LCDR1, LCDR2, and LCDR3, having amino acid sequences corresponding to SEQ ID NOs. 244, 245, and 246.

[0228] CAR structures In certain embodiments, the fusion protein may comprise a chimeric antigen receptor (CAR) or a CAR fragment. The CAR may comprise one or more extracellular ligand-binding domains, optionally a hinge region, a transmembrane region, and an intracellular signaling region. One or more extracellular ligand-binding domains may comprise one or more fusion proteins. Extracellular ligand-binding domains may typically comprise other ligand-binding domains such as single-chain immunoglobulin variable fragments (scFv) or Fab or intrinsic protein ligands. The hinge region may generally comprise a variable-length polypeptide hinge such as one or more amino acids, a CD8 alpha hinge region or an IgG4 region (or others), and combinations thereof. The transmembrane domain may typically comprise a transmembrane region derived from other transmembrane proteins such as CD8 alpha, CD28, or DAP10, DAP12, or NKG2D, and combinations thereof. The intracellular signaling region may comprise one or more intracellular signaling domains such as CD28, 4-1BB, CD3 zeta, OX40, 2B4, or other intracellular signaling domains, and combinations thereof. For example, one or more intracellular signaling domains may include CD28 and CD3 zeta, 4-1BB and CD3 zeta, or CD3 zeta. Lymphocytes such as T cells and NK cells can be modified to produce chimeric antigen receptor cells (e.g., CAR-T). CAR-T cells can recognize one or more specific soluble antigens on target cell surfaces, such as tumor cell surfaces, or on cells in the tumor microenvironment. When the extracellular ligand-binding domain binds to a homologous ligand, the intracellular signaling domain of the CAR can activate the lymphocyte. For example, see Brudno et al., Nature Rev. Clin. Oncol. (2018) 15:31-46; Maude et al., N. Engl. J. Med. (2014) 371:1507-1517; Sadelain et al, Cancer Disc. (2013) 3:388-398 (2018); and U.S. Patent Nos. 7,446,190 and 8,399,645.

[0229] In certain embodiments, a CAR construct comprising a ligand-receptor pair construct described herein is provided. In certain embodiments, the CAR construct comprises an scFv which can be fused to the ligand-receptor pair construct. In certain embodiments, the ligand-receptor pair construct is a single-chain ligand-receptor pair construct which can be fused to the N-terminus of the scFv with or without a linker. In certain embodiments, the single-chain ligand-receptor pair construct comprises a protease-cleavable linker. In certain embodiments, the receptor is fused to the N-terminus of the scFv with or without a first linker, and the ligand is internally fused to a second linker which connects the heavy and light chains of the scFv. In certain embodiments, the linker comprises a protease-cleavage site which can be cleaved by a protease. In certain embodiments, the ligand is fused to the N-terminus of the scFv with or without a first linker, and the receptor is internally fused to a second linker which connects the heavy and light chains of the scFv. In certain embodiments, the first linker is cleavable by a protease, while the second linker is non-cleavable. In certain embodiments, T cells can be modified to express ligand receptor-CARs.

[0230] sequence homology Certain embodiments of this disclosure relate to isolated polynucleotides or sets of polynucleotides encoding the fusion proteins described herein. In this context, polynucleotides may encode all or part of the fusion protein.

[0231] The terms “nucleic acid,” “nucleic acid molecule,” and “polynucleotide” are used interchangeably herein and refer to polymers of nucleotides of any length, which are either deoxyribonucleotides or ribonucleotides, or analogs thereof. Non-exclusive examples of polynucleotides include genes, gene fragments, messenger RNA (mRNA), cDNA, recombinant polynucleotides, plasmids, vectors, isolated DNA of any sequence, isolated RNA of any sequence, nucleic acid probes, and primers.

[0232] A given polynucleotide that "encodes" a polypeptide is a polynucleotide that, under the control of appropriate regulatory sequences, is transcribed in vivo (in the case of DNA) and translated into a polypeptide (in the case of mRNA). The boundaries of the coding sequence are determined by a start codon at the 5' (amino) terminus and a translation termination codon at the 3' (carboxy) terminus. The transcription termination sequence may be located 3' to the coding sequence.

[0233] In certain embodiments, this disclosure relates to polynucleotide and polypeptide sequences that encode at least a portion of a fusion protein described herein, e.g., polynucleotide and polypeptide sequences that are identical or substantially identical to a first or second polypeptide of a biologically functional protein. The terms “identical” or “identical percentage” in the context of two or more polynucleotide or polypeptide sequences mean two or more sequences or subsequences that are the same. Sequences are “substantially identical” if, when compared and aligned to best correspond across a comparison window or a specified region, they have a percentage of identical amino acid residues or nucleotides (e.g., about 80%, about 85%, about 90%, or about 95% identity across a specified region) as measured by one of the commonly used sequence comparison algorithms known to those skilled in the art, or by manual alignment and visual inspection. This definition also refers to complements of test polynucleotide sequences. Identity can exist over a region of at least approximately 50 amino acids or nucleotides, or over a region of 75 to 100 amino acids or nucleotides, or, if not specified, over the entire sequence of the polypeptide or polynucleotide. For sequence comparison, typically a specified reference sequence is compared with the test sequence. When using a sequence comparison algorithm, the test sequence and reference sequence are entered into a computer, and subsequence coordinates are specified if necessary to specify the parameters of the sequence algorithm program. Default program parameters may be used, or alternative parameters may be specified. The sequence comparison algorithm then calculates the percentage sequence identity of the test sequence compared to the reference sequence based on the program parameters.

[0234] When used herein, a “comparison window” refers to a segment of a sequence containing 20 to 1000 consecutive amino acid or nucleotide positions, which may be approximately 50 to 600, 100 to 300, or 150 to 200 consecutive amino acid or nucleotide positions, over which a test sequence can be compared to a reference sequence of the same number of consecutive positions after the two sequences have been optimally aligned. Longer segments up to the maximum full-length sequence may also be used as a comparison window in certain embodiments. Methods for aligning sequences for comparison purposes are known to those skilled in the art. Optimal sequence alignment for comparison can be performed, for example, by the local homology algorithm of Smith & Waterman, 1970, Adv. Appl. Math., 2:482c; by the homology alignment algorithm of Needleman & Wunsch, 1970, J. Mol. Biol., 48:443; by the similarity search of Pearson & Lipman, 1988, Proc. Natl. Acad. Sci. USA, 85:2444; by computer execution of these algorithms (e.g., GAP, BESTFIT, FASTA, or TFASTA (Wisconsin Genetics Software Package, Genetics Computer Group, Madison, WI)); or by manual alignment and visual inspection (see, for example, Ausubel et al., Current Protocols in Molecular Biology, (1995 supplement), Cold Spring Harbor Laboratory Press). Suitable available algorithms for determining percent sequence identity include the BLAST and BLAST2.0 algorithms, described in Altschul et al., 1997, Nuc. Acids Res., 25:3389-3402 and Altschul et al., 1990, J. Mol. Biol., 215:403-410, respectively.Software for performing BLAST analysis is publicly available from the website of the National Center for Biotechnology Information (NCBI).

[0235] Certain embodiments described herein relate to variant sequences containing one or more amino acid substitutions. In some embodiments, the amino acid substitutions are conserved substitutions. Generally, a “conservative substitution” is considered to be the substitution of one amino acid with another amino acid having similar physical, chemical, and / or structural properties. Common conservative substitutions are listed in column 1 of Table 4. Those skilled in the art will understand that the main factors determining what constitutes a conservative substitution are usually the size of the amino acid side chain and its physical / chemical properties, but in certain environments, a given amino acid can be substituted with a broader range of amino acids than those listed in column 1. These additional amino acids tend to have similar properties to the substituted amino acid but differ significantly in size, or are similar in size but differ significantly in physical / chemical properties. This broad range of conservative substitutions is listed in column 2 of Table 4. Those skilled in the art will be able to easily determine the most appropriate set of substituents to choose from, taking into account the specific protein environment in which the amino acid substitution takes place.

[0236] (Table 4) TIFF0007846668000004.tif187164

[0237] Preparation of fusion proteins The fusion proteins described herein can be produced using standard recombination methods known in the art (e.g., U.S. Patent No. 4,816,567 and “Antibodies: A Laboratory Manual,” 2 nd See Edition, Ed. Greenfield, Cold Spring Harbor Laboratory Press, New York, 2014.

[0238] Vector encoding a fusion protein For the recombinant production of the fusion proteins described herein, a polynucleotide or set of polynucleotides encoding the fusion protein is generated, inserted into one or more vectors, and further cloning and / or expression in host cells is performed. The polynucleotide(s) encoding the fusion protein can be generated by standard methods known in the art (e.g., Ausubel et al., Current Protocols in Molecular Biology, John Wiley & Sons, New York, 1994 & update, and “Antibodies: A Laboratory Manual,” 2 nd See Edition, Ed. Greenfield, Cold Spring Harbor Laboratory Press, New York, 2014. As will be understood by those skilled in the art, the number of polynucleotides required for the expression of a fusion protein depends on the format of the fusion protein, for example, whether the fusion protein contains an antibody and the number of polypeptides in the fusion protein. For example, if the fusion protein contains two polypeptide chains, two polynucleotides will be required, each encoding one polypeptide chain. Similarly, in certain embodiments, if the fusion protein contains a biologically functional protein in mAb format, two polynucleotides will be required, each encoding one polypeptide chain. If multiple polynucleotides are required, they can be incorporated into one or more vectors.

[0239] Generally, a polynucleotide or set of polynucleotides is incorporated into an expression vector for expression, along with one or more regulatory elements, such as transcription elements necessary for the efficient transcription of the polynucleotides. Examples of such regulatory elements include, but are not limited to, promoters, enhancers, terminators, and polyadenylation signals. Those skilled in the art will understand that the selection of regulatory elements depends on the host cell selected for the expression of the polypeptide of the fusion protein, and that such regulatory elements can originate from a variety of sources, including bacterial, fungal, viral, mammalian, or insect genes. The expression vector may optionally further contain heterologous nucleic acid sequences that facilitate the expression or purification of the protein to be expressed. Examples include, but are not limited to, signal peptides and affinity tags, such as metal affinity tags, histidine tags, avidin / streptavidin coding sequences, glutathione-S-transferase (GST) coding sequences, and biotin coding sequences. The expression vector may be an extrachromosomal vector or an embedded vector.

[0240] Certain embodiments of this disclosure relate to vectors (such as expression vectors) containing one or more polynucleotides encoding at least a portion of the fusion proteins described herein. The polynucleotide(s) may be contained in one or more vectors. In some embodiments, the polynucleotide is contained in a multicistronic vector.

[0241] Expression vectors used for polynucleotide expression include, but are not limited to, pTT5 and pUC15.

[0242] Cells containing vectors encoding fusion proteins Suitable host cells for cloning or expressing fusion protein polypeptides include a variety of prokaryotes or eukaryotic cells known in the art. Eukaryotic host cells include, for example, mammalian cells, plant cells, insect cells, and yeast cells (such as cells of the genera Saccharomyces or Pichia). Prokaryote host cells include, for example, Escherichia coli (E. coli) cells, A. salmonicida cells, or Bacillus subtilis (B. subtilis) cells.

[0243] In certain embodiments, the fusion protein is generated in bacteria, for example, when glycosylation and Fc effector function are not required, as described in U.S. Patent Nos. 5,648,237, 5,789,199, and 5,840,523, and Charlton, Methods in Molecular Biology, Vol. 248, pp. 245-254, BKCLo, ed., Humana Press, Totowa, NJ, 2003.

[0244] Eukaryotic microorganisms, such as filamentous fungi or yeasts, in particular fungal and yeast strains whose glycosylation pathways are "humanized" and which result in the production of antibodies with a partial or complete human glycosylation pattern, are suitable expression host cells in certain embodiments (see, for example, Gerngross, 2004, Nat. Biotech. 22:1409-1414, and Li et al., 2006, Nat. Biotech. 24:210-215).

[0245] Suitable host cells for the expression of glycosylated fusion proteins are typically eukaryotic cells. For example, U.S. Patents 5,959,177, 6,040,498, 6,420,548, 7,125,978, and 6,417,429 describe PLANTIBODIES® technology for antibody production in transgenic plants. Mammalian cell lines adapted for growth in suspension are particularly useful for fusion protein expression. Examples include SV40-transformed monkey kidney CV1 cell line (COS-7), human fetal kidney (HEK) cell line 293 or 293 (see, e.g., Graham et al., 1977, J. Gen Virol., 36:59), baby hamster kidney cells (BHK), mouse Sertoli TM4 cells (see, e.g., Mather, 1980, Biol Reprod, 23:243-251); monkey kidney cells (CV1), African green monkey kidney cells (VERO-76), human cervical cancer (HeLa) cells, canine kidney cells (MDCK), buffalo rat liver cells (BRL3A), human lung cells (W138), human liver cells (HepG2), mouse mammary tumor cells (MMT060562), and TRI cells (see, e.g., Mather et al., 1982, Annals NYAcad). Examples of suitable mammalian host cell lines for antibody production include, but are not limited to, MRC5 cells, FS4 cells, Chinese hamster ovary (CHO) cells (including DHFR-CHO cells; see Urlaub et al., 1980, Proc Natl Acad Sci USA, 77:4216), and myeloma cell lines (such as Y0, NS0, and Sp2 / 0). Exemplary mammalian host cell lines suitable for antibody production are outlined in Yazaki & Wu, Methods in Molecular Biology, Vol.248, pp.255-268 (BKCLo, ed., Humana Press, Totowa, NJ, 2003).

[0246] In certain embodiments, the host cell is a transient or stable higher eukaryotic cell line, such as a mammalian cell line. In some embodiments, the host cell is a mammalian HEK293T cell, CHO cell, HeLa cell, NS0 cell, or COS cell. In some embodiments, the host cell is a stable cell line that allows for the maturation glycosylation of the fusion protein.

[0247] Host cells containing an expression vector(s) encoding a fusion protein can be cultured using conventional methods for producing the fusion protein. Alternatively, in some embodiments, host cells containing an expression vector(s) encoding a fusion protein can be used to deliver the fusion protein therapeutically or prophylactically to a target, or the polynucleotide or expression vector can be administered ex vivo to cells derived from the target and then returned to the target's body.

[0248] In some embodiments, the host cell comprises a vector (e.g., transformed) containing polynucleotides encoding the VL and VH of the binding domain as described herein. In some embodiments, the host cell comprises a first vector containing a polynucleotide encoding the VL of the binding domain as described herein and a second vector containing a polynucleotide encoding the corresponding VH of the binding domain. In some embodiments, the host cell is a eukaryote, e.g., Chinese hamster ovary (CHO) cells, human fetal kidney (HEK) cells, or lymphoid cells (e.g., Y0, NS0, Sp20 cells).

[0249] In certain embodiments, the host cell is Expi293® (Thermo Fisher, Waltham, MA). In certain embodiments, the host cell is CHO-S cell (National Research Council Canada) or HEK293 cell.

[0250] Certain embodiments of the present disclosure relate to a method for producing a fusion protein, comprising culturing host cells into which one or more polynucleotides encoding the fusion protein or one or more expression vectors encoding the fusion protein are introduced, under conditions suitable for the expression of the fusion protein, and optionally recovering the fusion protein from the host cells (or from the host cell culture medium).

[0251] Examples of cell culture media that may be used include, but are not limited to, DMEM (Thermo Fisher, Waltham, MA), Opti-MEM (trademark) (Thermo Fisher, Waltham, MA), Opti-MEM (trademark) Reduced Serum Medium (Thermo Fisher, Waltham, MA), RPMI-1640 medium, Expi293 (trademark) Expression Medium (Thermo Fisher, Waltham, MA), and FreeStyle CHO expression medium (Thermo Fisher Scientific, Waltham, MA).

[0252] Cell culture media may contain serum, e.g., fetal bovine serum (FBS), amino acids, e.g., L-glutamine, antibiotics, e.g., penicillin and streptomycin, and / or antifungal agents, e.g., amphotericin, or any other additives commonly used to support cell culture.

[0253] Purification of fusion proteins Typically, fusion proteins are purified after expression. The proteins can be isolated or purified by various methods known to those skilled in the art (e.g., Protein Purification: Principles and Practice, 3). rdSee Ed., Scopes, Springer-Verlag, NY, 1994. Standard purification methods include ion exchange, hydrophobic interaction, affinity, size or gel filtration, and chromatographic techniques, including reversed phase, performed at atmospheric pressure or under high pressure using systems such as FPLC and HPLC. Further purification methods include electrophoresis, immunological, precipitation, dialysis, and chromatofocusing techniques. Ultrafiltration and diafiltration techniques, combined with protein concentration, are also useful. As is well known in the art, various native proteins bind to Fc and antibodies, and these proteins are used for the purification of specific antibodies. For example, bacterial proteins A and G bind to the Fc region. Similarly, bacterial protein L binds to the Fab region of some antibodies. Purification may also be possible with specific fusion partners. For example, antibodies can be fused to glutathione resin when GST fusion is used, or to Ni when His tagging is used. +2 Affinity chromatography, or, if flag tags are used, purification using immobilized anti-flag antibodies, can be employed. The required degree of purification will vary depending on the intended use of the antibody. In some cases, purification may not be necessary.

[0254] In certain embodiments, the fusion protein is substantially pure. When used with respect to the fusion proteins described herein, the term “substantially pure” means that the fusion protein is substantially or essentially free from components that are normally associated with or interact with the protein, such as those found in its naturally occurring environment, for example, in natural cells, or, in the case of recombinant-produced fusion proteins, in host cells. In certain embodiments, a substantially pure fusion protein is a protein preparation having about 30%, about 25%, about 20%, about 15%, about 10%, or about 5% (dry weight) of contaminating proteins.

[0255] Protein purification and / or evaluation of homogeneity can be carried out by any method known in the art, for example, but not limited to, non-reducing / reducing CE-SDS, non-reducing / reducing SDS-PAGE, ultra-high-performance liquid chromatography-size exclusion chromatography (UPLC-SEC), high-performance liquid chromatography (HPLC), mass spectrometry, multi-angle light scattering (MALS), and dynamic light scattering (DLS).

[0256] Post-translation modification In certain embodiments, the fusion proteins described herein include one or more post-translational modifications. Such post-translational modifications may occur in vivo or may be carried out in vitro after the fusion protein has been isolated from host cells.

[0257] Examples of post-translational modifications include various modifications known in the art (see, for example, Proteins-Structure and Molecular Properties, 2nd Ed., TECreighton, WH Freeman and Company, New York, 1993; Post-Translational Covalent Modification of Proteins, BC Johnson, Ed., Academic Press, New York, pgs. 1-12, 1983; Seifter et al., 1990, Meth. Enzymol., 182:626-646; and Rattan et al., 1992, Ann. NYAcad. Sci., 663:48-62). In these embodiments in which the fusion protein contains one or more post-translational modifications, the fusion protein may contain the same type of modification at one or more sites, or it may contain different modifications at different sites.

[0258] Examples of post-translational modifications include glycosylation, acetylation, phosphorylation, amidation, derivatization with known protecting / blocking groups, formylation, oxidation, reduction, proteolytic cleavage, or specific chemical cleavage (by cyanide bromide, trypsin, chymotrypsin, papain, V8 protease, or NaBH4).

[0259] Other examples of post-translational modifications include, for example, the addition or removal of N-linked or O-linked glycans, chemical modification of N-linked or O-linked glycans, treatment of the N-terminus or C-terminus, attachment of chemical moieties to the amino acid backbone, and the addition or deletion of N-terminal methionine residues resulting from expression in prokaryotic host cells. Post-translational modifications may also include modifications by detectable labels, such as enzyme, fluorescent, isotopic, or affinity labels, to enable detection and isolation of the protein. Suitable examples of enzyme labels include, but are not limited to, horseradish peroxidase, alkaline phosphatase, beta-galactosidase, and acetylcholinesterase. Suitable examples of prosthetic group complexes include, but are not limited to, streptavidin / biotin and avidin / biotin. Suitable examples of fluorescent substances include, but are not limited to, umbelliferone, fluorescein, fluorescein isothiocyanate, rhodamine, dichlorotriazinylamine fluorescein, dansilchloride, and phycoerythrin. Examples of luminescent materials include luminol, and examples of bioluminescent materials include luciferase, luciferin, and aequorin. Suitable radioactive materials include iodine, carbon, sulfur, tritium, indium, technetium, thallium, gallium, palladium, molybdenum, xenon, and fluorine.

[0260] Examples of additional post-translational modifications include transfer RNA-mediated amino acid addition to proteins such as acylation, ADP-ribosylation, amidation, covalent bonding of flavins, covalent bonding of heme moieties, covalent bonding of nucleotides or nucleotide derivatives, covalent bonding of lipids or lipid derivatives, covalent bonding of phosphatidylinositol, crosslinking, cyclization, disulfide bond formation, demethylation, covalent crosslinking, cysteine ​​formation, pyroglutamic acid formation, gamma-carboxylation, GPI anchor formation, hydroxylation, iodization, methylation, myristylation, pegylation, prenylation, racemization, selenoylation, sulfated amino acid addition, and ubiquitination.

[0261] Masking and programmed activation of fusion proteins According to this disclosure, the fusion protein is masked from engagement with its target(s). The degree to which the fusion protein's binding to its target(s) is reduced can be measured by standard techniques such as enzyme-linked immunosorbent assay (ELISA), biolayer interferometry (BLI), surface plasmon resonance (SPR), fluorescent cell sorting (FACS), flow cytometry, binding equilibrium exclusion (KinExA), mesoscale discovery (MSD), microfluidics, or isothermal titration calorimetry (ITC). In certain embodiments, the fusion protein comprises an antigen-binding domain masked by a ligand-receptor pair, wherein the binding of the antigen-binding domain to its congener antigen is reduced by at least threefold compared to the corresponding unmasked antigen-binding domain, for example, the binding of the antigen-binding domain to its congener antigen is reduced by at least fivefold, at least tenfold, at least twentyfold, at least 25fold or at least 30fold, or at least fortyfold or at least 50fold, or at least 70fold or at least 80fold, or at least 90fold or at least 100fold, or at least 200fold or at least 400fold, or at least 600fold or at least 800fold or at least 1000fold or at least 2000fold or at least 5000fold or at least 10,000fold.

[0262] According to this disclosure, cleavage of at least one protease in the peptide linker between the ligand or receptor of a ligand-receptor pair and a biologically functional protein unmasks (activates) the fusion protein, enabling it to bind to its target(s) of interest. Sensitivity to peptide linker cleavage can be tested in vitro by standard techniques, including those described in the examples herein. The extent to which the binding of the fusion protein to its target(s) is restored after protease cleavage can also be tested by standard techniques such as enzyme-linked immunosorbent assay (ELISA), biolayer interferometry (BLI), surface plasmon resonance (SPR), fluorescent cell sorting (FACS), flow cytometry, binding equilibrium exclusion (KinExA), mesoscale discovery (MSD), microfluidics, or isothermal titration calorimetry (ITC). The restoration of binding of the fusion protein to its target(s) may be partial or complete. Partial recovery of binding is defined as a measurable binding of the relevant domain of the fusion protein (e.g., ligand, receptor, or antigen-binding domain) to its target of interest, and can be, for example, 1 / 100 to 1 / 2 of the binding of the parent domain. Partial recovery may be approximately 1 / 100, 1 / 75, 1 / 50, 1 / 25, 1 / 10, 1 / 5, or 1 / 2 of the binding of the parent domain.

[0263] Treatment method In certain embodiments, the Disclosure includes a method for treating a disease or condition, comprising administering a fusion protein described herein to a subject in need thereof. In certain embodiments, the subject is a mammal. In certain embodiments, the subject is a human.

[0264] In certain embodiments, the methods disclosed herein are for the treatment of cancer. Cancers include, but are not limited to, hematopoietic malignancies (including leukemia, myeloma, and lymphoma), carcinomas (including adenocarcinoma and squamous cell carcinoma), melanoma, and sarcoma. Carcinomas and sarcomas are often referred to as “solid tumors.” In certain embodiments, cancer is a solid tumor. In certain embodiments, cancer is leukemia. In certain embodiments, cancer is lymphoma.

[0265] The fusion protein may exert either cytotoxic or cell proliferation inhibitory effects, potentially resulting in one or more of the following in subjects with tumors: reduction of tumor size, slowing or prevention of tumor size growth, increased disease-free survival from tumor disappearance or removal to recurrence, prevention of initial or subsequent tumor development (e.g., metastasis), increased progression-free survival, reduction of one or more tumor-related adverse symptoms, or increased overall survival.

[0266] In certain embodiments, the methods disclosed herein are for the treatment of immunodeficiency disorders or diseases.

[0267] In certain embodiments, the methods disclosed herein are for the treatment of autoimmune diseases or conditions.

[0268] The methods described herein involve administering the fusion protein described herein to a subject requiring it. The fusion protein may be administered to the subject by an appropriate route of administration. As will be understood by those skilled in the art, the route and / or mode of administration may vary depending on the desired outcome. Typically, immunotherapeutic antibodies are administered by systemic or topical administration. Topical administration may be administered to the tumor site or tumor-inflowing lymph nodes. Generally, fusion proteins are administered by parenteral administration, for example, by intravenous, intramuscular, intradermal, intraperitoneal, subcutaneous, or spinal administration, for example, by injection or infusion.

[0269] Treatment is achieved by administering a "therapeutically effective dose" of the fusion protein. A "therapeutically effective dose" refers to the amount effective in the required dosage and duration to achieve the desired therapeutic outcome. The therapeutically effective dose may vary depending on factors such as the patient's disease state, age, sex, and weight. A therapeutically effective dose is also defined as the amount in which the therapeutically beneficial effects of the fusion protein outweigh any toxic or adverse effects. A "sufficient dose" means an amount sufficient to produce the desired effect, for example, an amount sufficient to modulate the immune response against target cells or tissues by binding the immunomodulatory ligand-receptor to immune cells.

[0270] The appropriate dosage of the fusion protein can be determined by a skilled healthcare professional. The selected dosage level depends on various pharmacokinetic factors, including the activity of the specific fusion protein being employed, the route of administration, the timing of administration, the polypeptide excretion rate, the duration of treatment, other drugs, compounds, and / or substances used in combination with the fusion protein, e.g., anticancer agents, the age, sex, weight, condition, overall health, and medical history of the person being treated, as well as similar factors well known in the medical field.

[0271] Methods for modulating immune cells or immune responses In certain embodiments, the fusion protein described herein is administered to a subject in need of it, for example, a subject with cancer, in order to modulate the immune system of that subject. Thus, in certain embodiments, the fusion protein described herein either downregulates or upregulates the immune response.

[0272] According to this embodiment, administration of a sufficient amount of fusion protein to a subject can affect one of the following in order to activate or upregulate the immune response: modulation of immune checkpoints, modulation of T cell receptor signaling, modulation of T cell activation, modulation of pro-inflammatory cytokines, modulation of interferon-γ production by T cells, modulation of T cell suppression, preparation of survival and / or differentiation of M2 tumor-associated macrophages (TAMs) or myeloid-derived suppressor cells (MDSCs), and / or modulation of cytotoxic or cell proliferation inhibitory effects on cells.

[0273] In certain embodiments, methods for modulating an immune response are provided herein, including inhibition of immune checkpoints, stimulation of immune checkpoints, activation of immune cells, stimulation of T cell receptor signaling, and stimulation of antibody-dependent cell-mediated cytotoxicity (ADCC), T cell-dependent cell-mediated cytotoxicity (TDCC), cell-dependent cell-mediated cytotoxicity (CDC), or antibody-dependent cell-mediated phagocytosis (ADCP).

[0274] In certain embodiments, the fusion protein, upon activation by a protease, can agonize the costimulatory receptors of target leukocytes. Functional effects of costimulatory receptor agonism on leukocytes include activation of T effector cells, differentiation and activation of inflammatory myeloid cells, and / or recruitment of B cells and / or NKT cells. Activation of T effector cells can result in increased production by T cells of one or more cytokines, such as interferon-gamma (IFN-γ), interleukin-2 (IL-2), interleukin-12 (IL-12), interleukin-17 (IL-17), interleukin-21 (IL-21), granulocyte-macrophage colony-stimulating factor (GM-CSF), tumor necrosis factor-α (TNF-α), macrophage inflammatory protein 1β (MIP-1β), and / or CXC motif ligand 13 (CXCL13). Increased production of IL-21 and CXCL13 by T effector cells may, for example, support the differentiation and activation of inflammatory myeloid cells in the TME, mobilize antitumor lymphoid cells such as B cells and NKT cells, and / or support the formation of tertiary lymphoid structures.

[0275] In certain embodiments, the fusion protein activates T effector cells. In some embodiments, the fusion protein increases the production of GM-CSF, TNF-α, MIP-1β, IL-17, IL-12, IL-21, and / or CXC motif ligand 13 (CXCL13) by T effector cells.

[0276] In certain embodiments, the fusion protein reduces monocyte CSF1-dependent survival and activates T effector cells.

[0277] Certain embodiments of this disclosure relate, for example, to a method for modulating the costimulatory receptor agonism of leukocytes in vivo using a fusion protein to treat cancer.

[0278] In certain embodiments, the method relates to the suppression or downregulation of immune cells or immune responses, for example, to treat autoimmune diseases or disorders. Thus, in certain embodiments, the fusion protein is administered in an amount sufficient to modulate immune cells. In certain embodiments, downregulation of the immune response is achieved by modulating immune checkpoints, T cell receptor signaling, T cell activation, pro-inflammatory cytokines, interferon-γ production by T cells, T cell suppression, preparation of survival and / or differentiation of M2 tumor-associated macrophages (TAMs) or myeloid-derived suppressor cells (MDSCs), and / or modification of cytotoxic or cell proliferation-inhibiting effects on cells.

[0279] How to modify the ADCC of target cells In certain embodiments, the fusion protein described herein induces antibody-dependent cell-mediated cytotoxicity (ADCC), resulting in increased lysis of target cells. In certain embodiments, the fusion protein comprises an Fc region with increased binding affinity of Fc to FcRγIIIa (activating receptor), resulting in increased antibody-dependent cell-mediated cytotoxicity (ADCC) and increased lysis of target cells. In certain embodiments, the Fc region comprises a modified CH2 domain with amino acid modifications that result in increased binding affinity of Fc to FcRγIIIa (activating receptor), resulting in increased antibody-dependent cell-mediated cytotoxicity (ADCC).

[0280] In certain embodiments, the fusion protein described herein reduces antibody-dependent cell-mediated cytotoxicity (ADCC). In certain indications, reduction or elimination of ADCC and complement-mediated cytotoxicity (CDC) are desirable. In certain embodiments, the fusion protein may be useful if it includes an Fc region containing a modified CH2 domain, which includes an amino acid modification resulting in increased binding to FcγRIIb or an amino acid modification ("knockout" variant) that reduces or eliminates the binding of the Fc region to all Fcγ receptors. In certain embodiments, the fusion protein includes an Fc region with reduced binding to FcγRIIb (inhibitory receptor).

[0281] Pharmaceutical composition The fusion proteins described herein can be formulated into pharmaceutical compositions. These compositions may include, in addition to one or more fusion proteins, pharmaceutically acceptable excipients, carriers, buffers, stabilizers, or other materials well known to those skilled in the art. Such materials must be non-toxic and must not interfere with the efficacy of the active ingredient. The exact properties of the carrier or other materials may depend on the route of administration, e.g., oral, intravenous, cutaneous or subcutaneous, nasal, intramuscular, or intraperitoneal routes.

[0282] Pharmaceutical compositions for oral administration may be in the form of tablets, capsules, powders, or liquids. Tablets may contain a solid carrier or adjuvant such as gelatin. Liquid pharmaceutical compositions generally contain a liquid carrier such as water, petroleum, animal or vegetable oil, mineral oil, or synthetic oil. They may also contain saline solution, dextrose or other sugar solutions, or glycols such as ethylene glycol, propylene glycol, or polyethylene glycol.

[0283] For intravenous, cutaneous, or subcutaneous injection, or injection into the affected area, the active ingredient may be in the form of a parenterally acceptable aqueous solution that is pyrogen-free and has a suitable pH, isotonicity, and stability. Those skilled in the art can readily prepare a suitable solution using an isotonic vehicle such as sodium chloride, Ringer's injection, or Ringer's lactate injection. Preservatives, stabilizers, buffers, antioxidants, and / or other additives may be included as needed.

[0284] In the case of the fusion proteins of this disclosure administered to an individual, the dose is preferably a “therapeutably effective dose” sufficient to provide a benefit to the individual. A “prophylactically effective dose” may also be administered if sufficient to provide a benefit to the individual. The actual dose, as well as the rate and duration of administration, depends on the nature and severity of the protein aggregation disorder being treated. The prescription of treatment, such as the determination of the dose, is the responsibility of the general practitioner or other physician, and typically takes into account the disorder being treated, the individual patient’s condition, the site of delivery, the method of administration, and other factors known to the practitioner. Examples of the techniques and protocols mentioned above can be found in Remington’s Pharmaceutical Sciences, 16th edition, Osol, A. (ed), 1980.

[0285] Depending on the condition being treated, the compositions can be administered alone, in combination with other treatments, simultaneously, or sequentially.

[0286] kit This disclosure also provides kits comprising one or more compositions and instructions for use described herein. Thus, in certain embodiments, a kit comprising a vector for expressing a fusion protein described herein and instructions for use is described herein. In certain embodiments, a kit comprising a host cell containing a vector for expressing a fusion protein and instructions for use is described herein. In certain embodiments, a kit comprising a purified fusion protein and instructions for use is described herein. The purified fusion protein may be lyophilized or provided in a dry form such as powder or granules, and the kit may further contain a solvent suitable for reconstituting the lyophilized or dry component(s).

[0287] A kit typically includes a container and a label and / or accompanying information on or attached to the container. The label or accompanying information includes the descriptions customarily included in the market packaging of a therapeutic product and contains information or instructions regarding the indications, method of use, dosage, administration, contraindications, and / or warnings for the use of the therapeutic product. The label or accompanying information may further include a notice in the format prescribed by a government agency that regulates the manufacture, use, or sale of a pharmaceutical or biological product, indicating that the agency has approved the manufacture, use, or sale for administration to humans or animals. The container holds the composition containing the fusion protein. In some embodiments, the container may have a sterile access port. For example, the container may be a vial with a stopper that can be punctured by an intravenous infusion bag or a subcutaneous needle.

[0288] In addition to the container containing the composition comprising the fusion protein, the kit may include one or more additional containers containing other components of the kit, such as pharmaceutically acceptable buffers (e.g., bacteriostatic water for injection) (BWFI), phosphate-buffered saline, Ringer's solution, or dextrose solution), other buffers, or diluents.

[0289] Suitable containers include, for example, bottles, vials, syringes, and intravenous infusion bags. Containers can be formed from a variety of materials, such as glass or plastic. If necessary, one or more components of the kit may be lyophilized or provided in a dry form such as powder or granules, and the kit may further contain a solvent suitable for reconstituting the lyophilized or dry component(s).

[0290] The kit may further include other materials that are desirable from a commercial or user perspective, such as filters, needles, and syringes. [Examples]

[0291] The following examples are provided for illustrative purposes only and are not intended to limit the scope of the invention in any way. While every effort has been made to ensure accuracy with respect to the numbers used (e.g., quantities, temperatures, etc.), some experimental error and deviation are naturally to be expected.

[0292] Unless otherwise specified, the implementation of this disclosure will involve conventional methods of protein chemistry, biochemistry, recombinant DNA technology, and pharmacology within the scope of the art. Such techniques are adequately described in the literature. For example, TECreighton, Proteins: Structures and Molecular Properties (WH Freeman and Company, 1993); ALLehninger, Biochemistry (Worth Publishers, Inc., current addition); Sambrook, et al., Molecular Cloning: A Laboratory Manual (2nd Edition, 1989); Methods In Enzymology (S. Colowick and N. Kaplan eds., Academic Press, Inc.); Remington's Pharmaceutical Sciences, 18th Edition (Easton, Pennsylvania: Mack Publishing Company, 1990); Carey and Sundberg Advanced Organic Chemistry 3. rd See Ed. (Plenum Press) Vols. A and B (1992).

[0293] Example 1: Design of a masked anti-CD3 × anti-Her2 T cell engager fusion protein A mask was attached to the anti-CD3 Fab × anti-Her2 scFv Fc by ligating one end of the ligand-receptor pair (PD-1-PDL-1) to the N-terminus of the Fab's light chain and the other end to the N-terminus of the heavy chain. The fusion protein construct was designed as follows.

[0294] method The fusion protein was in the modified bispecific Fab×scFv Fc form, containing a semi-antibody with anti-CD3 heavy and light chains, forming a heterodimer with anti-Her2 scFv fused to Fc. The anti-CD3 paratope was described in US20150232557A1 (VL SEQ ID NO: 1; VH SEQ ID NO: 2). The anti-Her2 paratope was in the scFv form based on trastuzumab VL and VH linked by a glycine serine linker, as described in US10000576B1 (SEQ ID NO: 3) (Carter, P. et al. Humanization of an anti-p185HER2 antibody for human cancer therapy. Proc Natl Acad Sci USA 89, 4285-4289, doi:10.1073 / pnas.89.10.4285 (1992)). To enable selective heterodimer pairing, mutations were introduced into the anti-CD3 CH3 and anti-Her2 scFv-Fc CH3 chains as previously described (Von Kreudenstein, T et al. Improving biophysical properties of a bispecific antibody scaffold to aid developability: quality by molecular design. MAbs 5, 646-654, doi:10.4161 / mabs.25632 (2013); (A chain CH3 domain, SEQ ID NO: 4; B chain CH3 domain, SEQ ID NO: 5). Mutations (L234A_L235A_D265S compared to wild-type human IgG1 CH2) were also introduced into both CH2 domains to reduce binding to the Fc gamma receptor (SEQ ID NO: 6).Furthermore, polypeptides based on modified IgV domains of human PD-1 (SEQ ID NO: 7) and / or PD-L1 (SEQ ID NO: 8) (West, SM & Deng, XA Considering B7-CD28 as a family through sequence and structure. Exp Biol Med (Maywood), 1535370219855970, doi:10.1177 / 1535370219855970 (2019)) are linked with linkers ((EAAAK)n, Chen, X., Zaro, JL & Shen, WCF Fusion protein linkers: property, design and functionality. Adv Drug Deliv Rev Using 65,1357-1369,doi:10.1016 / j.addr.2012.09.039(2013)), the anti-CD3 variable domain was fused to the N-terminus of the heavy chain (VH-CH1-hinge-CH2-CH3) and the kappa light chain (VL-CL), respectively. These PD-1 and PD-L1 sites were predicted to dimerize and sterically block epitope binding.In all variants, either the PD-1 or PD-L1 sequence used as one half of the mask contained mutations to increase the affinity of the PD-1:PD-L1 complex, as previously described (Maute, R. Let al. Engineering high-affinity PD-1 variants for optimized immunotherapy and immuno-PET imaging. Proc Natl Acad Sci USA 112, E6506-6514, doi:10.1073 / pnas.1519623112 (2015); SEQ ID NO: 9; Liang, Z. et al. High-affinity human PD-L1 variants attenuate the suppression of T cell activation. Oncotarget 8, 88360-88375, doi:10.18632 / oncotarget.21729 (2017); SEQ ID NO: 10). Furthermore, in all WT PD-1 sites, the unpaired cysteine ​​was mutated to serine to eliminate the disadvantage of the exposed reducing group (SEQ ID NO: 11). Several variants also contained a cleavage sequence for the tumor microenvironment (TME)-associated protease uPa (MSGRSANA SEQ ID NO: 28) to allow partial or complete removal of the mask by exposure of the fusion protein to the protease. A schematic diagram of the construct design for masked Fab and the intended mechanism of action is shown in Figure 1. The final design was a bispecific Fab×scFv Fc molecule containing masked anti-CD3 Fab and anti-Her2 scFv. A schematic diagram is shown in Figure 2, and the sequences used are listed in Table A.

[0295] (Table A) TIFF0007846668000005.tif213165TIFF0007846668000006.tif139165

[0296] Example 2: Production of masked anti-CD3 variants The modified CD3×Her2 Fab×scFv variant sequence designed in Example 1 was introduced into an expression vector, expressed, and purified as follows.

[0297] method A heavy chain expression vector was constructed by ligating a pTT5 vector with a heavy chain vector insert containing a signal peptide (Barash et al., 2002, Biochem and Biophys Res. Comm., 294:835-842, SEQ ID NO: 27) and a heavy chain clone terminated at CH3 G446 (EU numbering). A light chain expression vector was constructed by ligating a pTT5 vector with a light chain vector insert containing the same signal peptide and light chain clone. The obtained heavy chain and light chain expression vectors were sequenced to confirm the correct reading frame and sequence of the coding DNA.

[0298] The heavy and light chains of the modified CD3×Her2 Fab×scFv Fc variant were co-expressed in 25 mL of Expi293F® cell culture medium (Thermo Fisher, Waltham, MA). Expi293® cells were cultured at 37°C in Expi293® expression medium (Thermo Fisher, Waltham, MA) on an orbital shaker rotating at 125 rpm in a humidified atmosphere of 8% CO2. Total cell count: 7.5 × 10⁶ 725 mL of cells were transfected with a total of 25 μg of DNA at an H1:L1:H2 transfection ratio of 40:40:20. Prior to transfection, the DNA was diluted in 1.5 mL of Opti-MEM® I Reduced Serum Medium (Thermo Fisher, Waltham, MA). 80 μL of ExpiFectamine® 293 reagent (Thermo Fisher, Waltham, MA) was diluted in 1.42 mL of Opti-MEM® I Reduced Serum Medium and incubated for 5 minutes. This mixture was then combined with the DNA transfection mix to make a total volume of 3 mL. After 10-20 minutes, the DNA-ExpiFectamine® 293 reagent mixture was added to the cell culture. After incubation at 37°C for 18-22 hours, 150 μL of ExpiFectamine® 293 Enhancer 1 and 1.5 mL of ExpiFectamine® 293 Enhancer 2 (Thermo Fisher, Waltham, MA) were added to each culture. The cells were incubated for 5-7 days, and the supernatant was collected for protein purification.

[0299] The clarified supernatant sample was applied in batch mode to a 1 mL slurry containing 50% mAb Select SuRe resin (GE Healthcare, Chicago, IL). The column was equilibrated with PBS. After loading, the column was washed with PBS, and the protein was eluted with 100 mM sodium citrate buffer pH 3.5. The pH of the eluted sample was adjusted by adding 10% (v / v) 1 M Tris pH 9 to a final pH of 6-7. After concentration, the entire substance was injected into an AKTA Pure FPLC System (GE Life Sciences) and run on a Superdex 200 Increase 10 / 300 GL (GE Life Sciences) column pre-equilibriumated with PBS pH 7.4. The protein was eluted from the column at a rate of 0.75 mL / min and collected in a 0.5 mL fraction. The peak fraction was pooled and concentrated using a 30 kDa MWCO polyethersulfone concentrator, Vivaspin 20 (MilliporeSigma Burlington MA, USA). After sterile filtration through a 0.2 μm PALL Acrodisc® Syringe Filter equipped with a Supor® Membrane, the protein was quantified based on A280 nm (Nanodrop) and stored at -80°C until further use.

[0300] result The sample contained a significant amount of high molecular weight species, as determined by UPLC-SEC after protein A purification (not shown), and preparative SEC was used to obtain a high-purity sample. The yield after preparative SEC ranged from 1.5 to 5 mg per variant. Sample purity and stability were evaluated in Examples 3 and 4.

[0301] Example 3: Evaluation of the purity and homogeneity of masked anti-CD3 variants The purified variants were evaluated for purity and sample homogeneity using non-reducing / reducing CE-SDS UPLC-SEC as described below.

[0302] method After purification, the purity of the samples was evaluated using a CE-SDS LabChip® GXII (Perkin Elmer, Waltham, MA) by non-reducing and reducing High Throughput Protein Express assays. The procedure was carried out according to the HT Protein Express LabChip® User Guide Version 2, with the following modifications: mAb samples were added in either 2 μL or 5 μL (concentration range 5–2000 ng / µL) to separate wells of a 96-well plate (BioRad, Hercules, CA) along with 7 μL of HT Protein Express Sample Buffer (Perkin Elmer #760328). Reducing buffer was prepared by adding 3.5 μL of DTT (1M) to 100 μL of HT Protein Express Sample Buffer. The mAb samples were then denatured at 90°C for 5 minutes, and 35 μL of water was added to each sample well. The LabChip® instrument was operated using the HT Protein Express Chip (Perkin Elmer #760499) and the HT Protein Express 200 assay setting (14kDa~200kDa).

[0303] UPLC-SEC was performed on an Agilent Technologies 1260 Infinity LC system at 25°C using an Agilent Technologies AdvanceBio SEC 300A column. Before injection, the sample was centrifuged at 10000g for 5 minutes, and 5 μl was injected into the column. The sample was run in PBS, pH 7.4 at a flow rate of 1 mL / min for 7 minutes, and elution was monitored by UV absorbance at 190–400 nm. The chromatogram was extracted at 280 nm. Peak integration was performed using OpenLAB CDS ChemStation software.

[0304] result Representative UPLC-SEC traces of the variant samples after preparative SEC purification in Figures 3A, 3C, 3E, and 3G showed highly homogeneous samples containing 89%–94% of the correct species. The presence of small peaks at short retention times compared to the dominant species indicates the presence of small amounts of high molecular weight species, such as oligomers and aggregates, in all samples.

[0305] Analysis of non-reduced CE-SDS (Figures 3B, 3D, 3E, and 3F) revealed a single dominant species, with only the bands corresponding to the intact chains of all variants being observed in the reduced CE-SDS runs. In particular, the masked heavy and light chains exhibited significantly higher apparent molecular weights than expected (110 kDa vs. 63 kDa for HC, and 54 kDa vs. 37 kDa for LC). This was also reflected in the higher apparent molecular weights of the non-reduced disulfide bond species (215 kDa vs. 152 kDa). Glycosylation of both the PD1 and PD-L1 sites in the design is likely to cause an apparent increase in molecular weight (Tan, S. et al. An unexpected N-terminal loop in PD-1 dominates binding by nivolumab. Nat Commun 8, 14369, doi:10.1038 / ncomms14369 (2017), Li, C. Wet et al. Glycosylation and stabilization of programmed death ligand-1 suppresses T-cell activity. Nat Commun 7, 12632, doi:10.1038 / ncomms12632 (2016)).

[0306] Example 4: Stability evaluation of masked anti-CD3 variants The thermal stability of the purified variant was evaluated by differential scanning calorimetry (DSC) as described below.

[0307] method Representative sets of modified CD3×Her2 Fab×scFv Fc variants were diluted to 0.5–1 mg / ml in PBS. For DSC analysis using NanoDSC (TA Instruments, New Castle, DE, USA), 950 μl of sample and matching buffer (PBS) were added to 96-well plates for both the sample and reference cells. At the start of the DSC run, a PBS blank injection was performed to stabilize the baseline. Then, each sample was injected and scanned from 25°C to 95°C at 1°C / min under a nitrogen pressure of 60 psi. Thermograms were analyzed using NanoAnalyze software. The matching buffer thermogram was subtracted from the sample thermogram, and baseline fitting was performed using a sigmoid curve. The data were then fitted to a two-state scale DSC model.

[0308] result The DSC thermogram (30421, Figure 4) of the unmodified CD3×Her2 Fab×scFv Fc variant showed transition at 68°C and 83°C. m The transpositions in T correspond to individual transpositions that are not degraded for anti-CD3 Fab, anti-Her2 scFv, and CH2 domain unfolding, but T m The transition at 83°C likely corresponds to the unfolding of the CH3 domain in the heavy chain. Thermograms of variants with the PD-1:PD-L1 mask (30430, 30436; Figure 4) also showed two transitions at similar temperatures and similar thermogram traces as the unmasked variant. This suggests that the fused masking domain is responsible for the T of anti-CD3 Fab. m Without affecting Fab, in collaboration with Fab, or non-collaboratively but similar to Fab, scFv, and CH2 m This indicates that it will be unfolded along with [something].

[0309] Example 5: uPa cleavage of anti-CD3 variant To evaluate the partial or complete release of the fusion protein mask from anti-CD3 Fab by cleavage of the introduced protease cleavage site in the linker, samples were treated with uPa in vitro. The reaction was monitored by reduced CE-SDS as follows.

[0310] method For preparative cleavage of the variant, 25–100 ug of purified sample was diluted to a final variant concentration of 0.2 mg / mL in PBS + 0.05% Tween 20, and recombinant human u-plasminogen activator (uPa) / urokinase (R&D Systems #P00749) was added in a protease:substrate molar ratio of 1:50. After incubation at 37°C for 24 hours, the sample fragments were analyzed with reduced CE-SDS as described in Example 2, then frozen and stored at -80°C for further use.

[0311] result Analysis of reduced CE-SDS profiles of uPa-treated and untreated masked variants revealed that, under the conditions investigated, some or all of the mask was effectively removed from the Fab by cleavage at the introduced cleavage site (Figure 5). For successfully cleaved variants (30430, 30436, 31934), bands representing masked heavy and / or light chain fragments completely disappeared upon cleavage, while unmasked heavy and / or light chain fragments remained visible. A broad, low-intensity band corresponding to free PD-1 fragments could be observed for variant 30430, whereas this was not the case for released PD-L1 in variant 30436. The small size and size heterogeneity resulting from glycosylation (Tan, S. et al. An unexpected N-terminal loop in PD-1 dominates binding by nivolumab. Nat Commun 8, 14369, doi:10.1038 / ncomms14369 (2017), Li, C. Wet al. Glycosylation and stabilization of programmed death ligand-1 suppresses T-cell activity. Nat Commun 7, 12632, doi:10.1038 / ncomms12632 (2016)) likely made free PD-1 and PD-L1 fragments almost undetectable, or even undetectable, respectively. No cleavage was observed in variants that did not contain cleavage sequences (30421, 30423).

[0312] Example 6: Masking / Demasking of CD3 Bonding The uncleaved and cleaved samples of the anti-CD3 variant from Example 5 were tested for binding to CD3-binding Jurkat cells by ELISA and for binding to Pan T cells by flow cytometry, as described below.

[0313] method ELISA Human Jurkat cells (Fujisaki Cell Center, Japan) were maintained in RPMI-1640 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS) containing 2 mM L-glutamine and 1X penicillin / streptomycin at 37°C in a humidified +5% CO2 incubator. Samples of the modified CD3×Her2 variant from Example 5 were diluted 2-fold in blocking buffer containing a saturation amount of unrelated human Ig, followed by seven 3-fold serial dilutions in blocking buffer for a total of eight concentration points. Blocking buffer alone was added to the control wells to measure the background signal against cells (negative / blank control).

[0314] All incubations were performed at 4°C. On the day of the assay, exponentially growing cells were centrifuged and seeded in a 1:1 mixture of complete medium and blocking buffer in a 96-well filter plate (MilliporeSigma, Burlington, MA, USA). Equivolutes of 2X variant or control were added to the cells and incubated for 1 hour. The plate was then washed four times using vacuum filtration. HRP-conjugated anti-human IgG Fc gamma-specific secondary antibody (Jackson ImmunoResearch, West Grove, PA, USA) was added to the wells and incubated for another 1 hour. The plate was washed seven times by vacuum filtration, followed by the addition of TMB substrate (Thermo Scientific, Waltham, MA, USA) at room temperature. The reaction was stopped by adding 0.5 volume of 1 M sulfuric acid, and the supernatant was transferred by filtration to a clear 96-well plate (Corning, Corning, NY, USA). The absorbance at 450 nm was read using a Spectramax 340PC plate reader equipped with passcheck correction.

[0315] The binding curves of OD450 versus linear or logarithmic antibody concentrations, with the blank subtracted, were fitted using GraphPad Prism 8 (GraphPad Software, La Jolla, CA, USA). A one-site-specific, four-parameter nonlinear regression curve fitting model using Hillslope was used to determine the Bmax and apparent Kd values ​​for each test sample.

[0316] Flow cytometry Antibodies were titrated in v-bottom 96-well plates (Sarstedt AG, Numbrecht, Germany) at a dilution ratio of 1:3 in PBS (Thermo Fisher Scientific, Waltham, MA) containing FACS buffer and 2% FBS, at a total of 20 μL / well, from 300 nM to 1.7 pM. Healthy donor peripheral blood pan T cells (BioIVT, Westbury, NY) were thawed and washed in RPMI1640 medium (A1049101, ATCC modified) (Thermo Fisher Scientific, Waltham, MA) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Waltham, MA). Cells were counted, resuspended in FACS buffer, and added to 96-well plates at a rate of 50,000 cells per well. Cells were incubated with the variant at 4°C for 1 hour, then washed twice with FACS buffer and 1 mg / mL of the secondary antibody AF647 goat anti-human IgG Fc (Jackson ImmunoResearch, West Grove, PA). A 1000-fold dilution of viability dye (Biolegend, San Diego, CA) was also added to the wells. The plates were incubated at room temperature for 30 minutes with shaking (200 rpm). The cells were then washed twice with FACS buffer and resuspended in 100 μL of FACS buffer. For assay readout, the geometric mean of APC fluorescence was measured by flow cytometry using BD LSRFortessa (BD Biosciences, San Jose, CA). Raw data were analyzed using FlowJo, LLC Software (Becton, Dickinson & Company, Ashland, OR). Graphs were generated using GraphPad Prism version 8.1.2 for Mac OS X (GraphPad Software, La Jolla, CA).

[0317] result ELISA As shown in Figure 6, variants containing a complete PD1:PD-L1-based mask attached to the CD3 Fab (30423, 30430, 30436) showed a 40–180-fold reduction in binding compared to the unmasked control (30421). Treatment with uPa partially restored CD3 binding in cleavable variants 30430 and 30436 (within 6–7 times that of the unmasked control). This partial restoration may have been caused by steric hindrance of epitope binding due to the portion of the mask remaining on the mask after cleavage. Concurrently, controls (31929, 31931) that had only PD-1 or PD-L1 attached to either the heavy or light chain, respectively, showed a similar reduction in binding (4–5 times) compared to the unmasked control, similar to the uPa-cleaved samples of the well-masked variants.

[0318] Flow cytometry As shown in Figure 22, variants containing a complete PD1:PD-L1-based mask attached to the CD3 Fab (30423, 30430) showed a >43-fold reduction in binding compared to the unmasked control (30421). Treatment with uPa partially restored CD3 binding in the cleavable variant 30430 (within 29 times that of the unmasked control). This partial restoration may have been caused by steric hindrance of epitope binding due to the portion of the mask remaining on the mask after cleavage. Concurrently, the control (31929) which had only PD-1 attached to the heavy chain showed a similar reduction in binding (14-fold) compared to the unmasked control, similar to the uPa-cleaved sample of the well-masked variant. In another experiment, a variant (32497) containing a non-functional PD-1 domain attached to the heavy chain showed a similar reduction in binding (6-fold) compared to the unmasked control, as seen in the equivalent variant containing functional PD-1 (31929, 5-fold) (Figure 32).

[0319] Example 7: T cell-dependent cell-mediated cytotoxicity of masked and unmasked variants The functional effect of a PD-1:PD-L1-based mask on the ability of the CD3×Her2 Fab×scFv Fc variant to engage with and activate T cells for the killing of Her2-possessing cells was evaluated in a T cell-dependent cell-mediated cytotoxicity (TDCC) assay as follows.

[0320] method Co-culture assay JIMT-1 (Leibniz Institute, Braunschweig, Germany) cultured in a growth medium consisting of DMEM medium (Thermo Fisher Scientific, Waltham, MA) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Waltham, MA), HCC1954 (ATCC, Manassas, VA) and HCC827 (ATCC, Manassas, VA) cultured in a growth medium consisting of ATCC-modified RPMI-1640 (Thermo Fisher Scientific, Waltham, MA) supplemented with 10% fetal bovine serum, and MEM medium (Thermo Fisher Scientific, Waltham, MA) supplemented with 10% fetal bovine serum and 0.01 mg / mL human recombinant insulin (Thermo Fisher MCF-7 (ATCC, Manassas, VA) cultured in growth medium consisting of (Thermo Fisher Scientific, Waltham, MA) was maintained horizontally in an incubator at 37°C with 5% carbon dioxide in a T-175 flask (Corning, Corning, NY). On the assay setup day, variants were titrated in triplicate at a 1:3 dilution ratio directly in a 384-well cell culture treatment optical bottom plate (Thermo Fisher Scientific, Waltham, MA) from 5 nM to 0.08 pM. Tumor cells were rinsed with PBS (Thermo Fisher Scientific, Waltham, MA), harvested using TrypLE Express (Thermo Fisher Scientific, Waltham, MA), diluted in medium, and counted using Vi-Cell (Beckman Coulter, Indianapolis, IN). Primary human pan-T cells (BioIVT, Westbury, NY) vials were thawed in a 37°C water bath, washed in culture medium, and counted using Vi-Cell. The Pan T cell suspension was mixed with tumor cells in a 5:1 effector-to-target ratio, washed, and suspended at 0.55E6 cells / ml. 20 μL of the mixed cell suspension was added to a plate containing titrated variants.The plates were incubated in an incubator at 37°C with 5% carbon dioxide for 48 hours. The samples were then subjected to high-content cell injury assessment, and the supernatant was collected for IFNγ analysis.

[0321] High-Content Cytotoxicity Analysis Cells were stained with Hoechst33342 for nuclear visualization and viability assessment. 10 μL of Hoechst33342 (Thermo Fisher Scientific, Waltham, MA) was diluted 1:1000 in culture medium and added to cells after 48 hours, followed by incubation at 37°C for a further 1 hour. The plates were then subjected to high-content image analysis using CellInsight CX-5 (Thermo Fisher Scientific, Waltham, MA) to distinguish and quantify viable and dead tumor cells as well as effector cells. The plates were scanned using SpotAnalysis.V4 Bioapplication with the following settings: objective lens: 10x, channel 1-386nm: (gain 2, fixed exposure time 0.008 ms).

[0322] Quantitative determination of IFNγ For IFNγ quantification using the MSD U-PLEX 384-well single-spot assay, streptavidin-coated multi-array plates (MA6000 384 SA plates, Meso Scale Diagnostics, Rockville, MD) were blocked with 50 μL of Diluent 100, sealed, and incubated at room temperature for 30 minutes with shaking (800 rpm). After incubation, all wells were aspirated. Biotinylated capture IFNγ antibody was added to Diluent 100 in a ratio of 1:16.5, and 10 μL of the capture antibody solution was added to each well of the blocked plate. The plate was sealed and incubated overnight at 4°C. The following day, the frozen supernatant obtained from the co-culture assay was thawed on wet ice. The plate was washed, and 5 μL of Diluent 43 was added to each well, followed by 5 μL of the thawed supernatant sample or standard. The plate was sealed and incubated at room temperature for 1 hour with shaking (800 rpm). After incubation, the plate was washed, and 10 μL of sulfotag detection antibody, diluted 1:1000 with diluent 3, was added to each well. The plate was sealed and incubated at room temperature for 1 hour with shaking (800 rpm). After incubation, the plate was washed, and 40 μL of MSD GOLD Read Buffer was added to each well. The plate was read using a MESO SECTOR R600 instrument (Meso Scale Diagnostics, Rockville, MD).

[0323] Quantitative analysis of PD-L1 and Her2 receptors Quantitative analysis of Her2 and PD-L1 receptors was performed by flow cytometry using Quantum Simply Cellular anti-human and anti-mouse IgG kits (Bangs Laboratories, Fishers, Indiana), respectively. Tumor cells were rinsed with PBS (Thermo Fisher Scientific, Waltham, MA) and collected using TrypLE Express (Thermo Fisher Scientific, Waltham, MA). Cells were counted using Vi-Cell (Beckman Coulter, Indianapolis, IN), washed, and resuspended at 4 × 10^6 c / mL in PBS (Thermo Fisher Scientific, Waltham, MA) containing FACS buffer-2% FBS. 25 μL of tumor cell suspension was added in triplicate to a 96-well V-bottom plate (Sarstedt AG, Numbrecht, Germany). Anti-Her2-AF647 (trastuzumab, monovalent antibody, Zymeworks, Vancouver, BC), anti-PDL1-APC (clone MIH1, BD Biosciences, San Jose, CA), or unrelated negative control IgG-AF657 (Zymeworks, Vancouver, BC) antibody was added at 15 ug / mL to wells and Eppendorf tubes (Thermo Fisher Scientific, Waltham, MA) containing Quantum Simply Cellular IgG beads (anti-human or anti-mouse) and blank beads. Cells and beads were incubated with the antibody at 4°C for 1 hour in the dark. Cells and beads were washed, resuspended, and analyzed by flow cytometry. For analysis, a standard curve was created for specific lot beads using a spreadsheet provided by Bangs Laboratories (Fishers, Indiana), and surface antigen-binding ability (ABC) was generated by inputting the geometric mean of the cell population using the same spreadsheet. The ABC value represents the number of receptor molecules expressed on the cell surface, assuming a monovalent binding model.The standard curves for determining the range of reliable receptor counts were 3,500 receptors / cell to 330,000 receptors / cell for Her2 and 4,400 receptors / cell to 630,000 receptors / cell for PD-L1.

[0324] result The masking effect observed for the CD3×Her2 Fab×scFv Fc variant in CD3 binding in Example 6 was reproduced when the same sample was examined for function in a TDCC assay using JIMT-1 cells expressing Her2 (Figure 7). The unmasked variant (30421) showed robust tumor cell killing at low variant concentrations, while the efficacy of the masked non-cleaved variant (30423) was reduced 49,000-fold. The well-masked variant (30430) with a cleavable PD-L1 site on the light chain also showed a 5,800-fold reduction in efficacy without uPa treatment. This difference in masking between non-cleaved and cleaved variants was similarly observed for CD3 binding (Example 6). When the mask was cleaved by uPa, the efficacy of 30430 reverted to that of the unmasked (30421) variant. A control variant (31929) bound only to the PD-1 site of the mask showed similar efficacy to 30421 and uPa-treated 30430. An unrelated anti-respiratory syncytial virus (RSV) antibody (22277) did not show T cell activation for tumor cell killing.

[0325] TDCC was repeated using JIMT-1 as a Her2 and PD-L1 positive cell line, and expanded to three other cell lines with different receptor levels, using different T cell donors than in the previous experiment. Cytotoxicity data from two repeats are shown in Figure 23. For repeat n=1, the level of the cytokine IFNγ was also monitored instead of T cell immune activation (Figure 24). The number of receptors was determined for all cell lines used and is shown in Figure 25. The potency of the unmasked control (30421) was determined to be 0.03 pM (HCC1954: high Her2, high PD-L1) to 3 pM (MCF-7: medium Her2, low PD-L1) for cytotoxicity in different cell lines. The potency of this unmasked control, as determined by IFNγ release, was 8.4 pM (HCC1954: high Her2, high PD-L1) to 50 pM (HCC829: low Her2, medium PD-L1). Masking of non-cleaved (30423) and cleaved (30430) masked variants, as measured by increased EC50, was observed in all cell lines, ranging from 72 to >450-fold in cytotoxicity readout and 8.2 to >350-fold in IFNγ readout. The variant with only the PD-1 site bound to the heavy chain (31929), the cleaved masked variant after uPa treatment (30430 + uPa), and the combination of unmasked control and saturated dose of anti-PD-L1 antibody (30421 + 120 nM atezolizumab) showed higher potency compared to the unmasked control (30421) in cell lines with marked PD-L1 expression (HCC1954, JIMT-1, HCC827) because they also have the ability to engage PD-L1 (EC50 in cytotoxicity was 0.019 to 0.84 times lower). In cell lines with extremely low PD-L1 expression (MCF-7), there was no significant difference in cytotoxicity readout between the unmasked control (30421) and these variants capable of engaging PD-L1 (31929, 30421 + 120 nM atezolizumab, 30430 + uPa). However, these variants containing anti-PD-L1 sites showed higher IFNγ release efficacy in all test cell lines compared to the unmasked control (30421).The unrelated anti-RSV antibody (22277) did not show activity in TDCC in any of the cell lines.

[0326] Example 8: PD1 and PD-L1 binding analysis of masked anti-CD3 variants Instead of the biological activity of the PD-1 and PD-L1 sites used as masking domains, the binding of modified variants to CHO cells expressing PD-L1 and PD-1 was determined as follows.

[0327] method CHO cell transfection CHO-S cells (National Research Council Canada) were cultured in FreeStyle CHO expression medium (Thermo Fisher Scientific, Waltham, MA) containing 1% fetal bovine serum (Thermo Fisher Scientific, Waltham, MA). Transfection was performed using the Neon Transfection System (Thermo Fisher Scientific, Waltham, MA). CHO-S cells were counted, washed twice with PBS and once with resuspension buffer R (Thermo Fisher Scientific, Waltham, MA), and then resuspended at 100 E6 cells / mL. PD-1, PDL-1, or GFP plasmid DNA (GenScript, Piscataway, NJ) was added at a rate of 1 ug / 1 E6 cells. 3 mL of electrolytic buffer E2 (Thermo Fisher Scientific, Waltham, MA) was packed into neon tubes. Transfection of each plasmid was performed using a 100 μL neon chip (Thermo Fisher Scientific, Waltham, MA) with the following settings: voltage -1620, width -10, pulse -3. The transfected cells were transferred to pre-warmed flasks at a concentration of 1 E6 cells / mL for each condition.

[0328] PD1 / PDL1 by flow cytometry The variants purified in Example 2 and the variants treated with uPa in Example 5 were directly titrated at a 1:3 dilution ratio in 200 nM v-bottom 96-well plates (VWR, Radnor, PA, USA). CHO-PD1, CHO-PDL-1, and CHO-GFP cells were thawed, washed with RPMI1640 medium (A1049101, ATCC modified) (Thermo Fisher Scientific, Waltham, MA, USA) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Waltham, MA, USA), and resuspended in FACS buffer (PBS + 2% FBS). CHO-PD1 cells and CHO-PDL-1 cells were each combined with CHO-GFP cells in a 2:1 ratio, and 20 μL of the cell suspension was added to plates containing the titrated variants. The cells were incubated with the variants at 4°C for 1 hour. After incubation, cells were washed twice with FACS buffer, and 1 ug / mL of the secondary antibody AF647 goat anti-human IgG Fc (Jackson Immuno Research, West Grove, PA, USA) was added to the wells along with a 1000-fold dilution of viability indicator dye (Biolegend, San Diego, CA, USA). The plates were incubated at room temperature for 30 minutes. Cells were washed twice with FACS buffer and resuspended in 50 uL of FACS buffer.

[0329] For assay interpretation, the geometric mean of APC fluorescence was measured by flow cytometry using BD LSRFortessa (BD Life Sciences, Gurugram, India). Nonspecific binding was determined by measuring the geometric mean of APC fluorescence in GFP-positive cells. Graphs were generated using GraphPad Prism version 8.1.2 for Mac OS X (GraphPad Software, La Jolla, CA, USA).

[0330] result As shown in Figure 8, binding to PD-L1(A) or PD-1(B) was not observed in masked variants (30423, 30426, 30430, 30436) without uPa treatment (-uPa). Variants containing only affinity-mature PD-1 or PD-L1 sites bound to either the heavy or light chain showed binding with IC50s of 0.3 nM and 6 nM, respectively. Non-cleavable variants (30423, 30426) did not bind to PD-L1 or PD-1 when treated with protease (+uPa), but partial binding was restored in samples treated with uPa containing a uPa cleavage sequence between Fab and the PD-1:PD-L1 mask. Specifically, binding to PD-L1 was partially restored in 30430, within 53 times that of the associated one-sided mask control 31929(A). Binding to PD-1 was partially restored in 30436, within 12 times the one-sided mask control 31931(B). This is consistent with the properties of immunomodulators (PD-1 in 30430, PD-L1 in 30436) designed to remain on these variants if cleaved by proteases. Variants without the PD-1:PD-L1-based mask (30421) and the unrelated control (22277) did not show binding to PD-L1 or PD-1, as expected. In another experiment using JIMT-1 as the target cell, a variant containing a non-functional PD-1 domain attached to the heavy chain (32497) showed reduced TDCC potency compared to the unmasked control (v30421) (55-fold EC). 50 (Figure 33), equivalent variants including functional PD-1 (31929) showed increased TDCC potency (0.2x EC) compared to the previously observed unmasked control (v30421). 50 ).

[0331] Example 9 Investigation of functional enhancement of PD-1 mask in hybrid PD-1 / PD-L1 reporter gene assay In addition to the variant's T cell engagement function, a custom hybrid PD-1 / PD-L1 reporter gene assay (RGA) was performed as follows to investigate the blocking of PD-1:PD-L1 checkpoint engagement by the PD-1 site of the mask.

[0332] method JIMT-1 (Leibniz Institute, Braunschweig, Germany) cultured in a growth medium consisting of DMEM medium (Thermo Fisher Scientific, Waltham, MA) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Waltham, MA); HCC1954 (ATCC, Manassas, VA) and HCC827 (ATCC, Manassas, VA) cultured in a growth medium consisting of ATCC-modified RPMI-1640 (Thermo Fisher Scientific, Waltham, MA) supplemented with 10% fetal bovine serum; MEM medium (Thermo Fisher Scientific, Waltham, MA) supplemented with 10% fetal bovine serum and 0.01 mg / mL human recombinant insulin (Thermo Fisher MCF-7 cells (ATCC, Manassas, VA) cultured in growth medium consisting of Scientific, Waltham, MA, and Jurkat T cells (PD-1 / PD-L1 Blockade Bioassay Promega Cat# J1250, Madison, WI) stably expressing human PD-1 and NFAT-induced luciferase cultured in ATCC-modified RPMI-1640 medium supplemented with 10% fetal bovine serum, were maintained in an incubator at 37°C with 5% carbon dioxide in T-75 or T-175 flasks (Corning, Corning, NY) before assay setup. On the experimental day, variants were titrated in triplicate at a 1:3 dilution ratio of 20 μL total per well in 384-well low-flange white flat-bottom polystyrene TC-treated microplates (Corning Cat# 3570, Corning, NY) from 150 nM to 0.85 pM. Tumor cells were dissociated using cell dissociation buffer and mixed with Jurkat cells in a 1:1 ratio in RPMI1640 supplemented with 1% fetal bovine serum. 20 μL of the mixed cell suspension was added to a plate containing the titrated variant. The plate was incubated at 37°C in 5% carbon dioxide for 16 hours.After incubation, 40 μL of Bio-Glo® Luciferase Assay reagent (Promega Cat# G7940, Madison, WI) was added to all wells, ensuring that no bubbles formed. The plate was then read after 10 minutes using a microplate reader (Biotek Synergy H1, Winooski, VT) in luminescence mode with a gain of 150. A schematic diagram of the assay setup is shown in Figure 9A.

[0333] result Figure 9B shows the analysis of custom RGAs to investigate the functional enhancement by masking. Treatment of cells with an unmasked variant (30421) capable of crosslinking T cells and tumor cells, combined with a saturated dose (150 nM) of anti-PD-L1 antibody, resulted in a high RGA response. While the unmasked bispecific CD3×Her2 antibody productively crosslinked T cells and tumor cells, high concentrations of anti-PD-L1 antibody robustly blocked PD-1:PD-L1 checkpoint engagement, resulting in a high signal across all test variant concentrations. Conversely, treatment with the unmasked variant (30421) alone significantly reduced the signal due to PD-1 and PD-L1 engagement between modified T cells and JIMT-1 cells. The uncleaved (30423) and cleaved (30430) masked variants, untreated with uPa (-uPa), showed a significant decrease in activity at variant concentrations below 10 nM compared to the unmasked 30421, indicating productive inhibition of T cell engager function due to steric hindrance of the CD3 paratope. The untreated uPa-30430 sample was more potent in inducing an RGA response than 30423. When treated with uPa (+uPa), the cleaved masked variant (30430) showed higher RGA activity than the unmasked control (30421) at variant concentrations above 100 pM, indicating PD-1:PD-L1 checkpoint engagement blocking by the functional PD-1 site of the mask remaining on the variant after CD3 paratope demasking and cleavage. Consistent with these findings, a control with only the PD-1 domain bound to the CD3 Fab heavy chain showed a similar profile and, compared to the unmasked control (30421), exhibited increased RGA activity at variant concentrations above 100 pM. An unrelated anti-RSV antibody (22277) did not show RGA activity.

[0334] As determined in Example 7, the RGA was repeated using JIMT-1 as a Her2 and PD-L1 positive cell line and expanded to three other cell lines with different receptor levels. The data from the RGA performed here are shown in Figure 26. Masking of the masked variants, both non-cleaved (30423) and cleaved (30430), as measured by EC50 increase, was observed in all cell lines and ranged from 4 to 530 times compared to the unmasked control (30421). The potency of the unmasked control (30421) was comparable across cell lines (EC50 = 20-50 pM), but variants tested in cell lines with fewer Her2 and / or PD-L1 receptors (HCC827 and MCF-7) showed stronger masking than those in cell lines with higher receptor expression (HCC1954 and JIMT-1). Treatment with uPa (+uPa) restored the cleavable masked variant (30430) to 1.7–3.6 times the potency of the unmasked control. The variant with only the PD-1 site bound to the heavy chain (31929), the cleavable masked variant after uPa treatment (30430+uPa), and the combination of the unmasked control and a saturated dose of anti-PD-L1 antibody (30421+120nM atezolizumab) showed higher efficacy (1.6–3.3 times higher maximum RLU) in cell lines with marked PD-L1 expression (HCC1954, JIMT-1, HCC827) because they also have the ability to engage PD-L1. These variants also showed high potency in cell lines with high TAA and PD-L1 expression (HCC1954, JIMT-1) (0.2–0.4 times EC50). In cell lines with extremely low PD-L1 expression (MCF-7), n...

Claims

1. A fusion protein comprising a biologically functional protein, a ligand-receptor pair, a first peptide linker, and a second peptide linker, The biologically functional protein is an antibody or an antigen-binding antibody fragment, comprising at least a first polypeptide and a second polypeptide, and The ligand-receptor pair comprises the extracellular component of an immunoglobulin superfamily receptor and its homologous ligand or receptor-binding fragment, The first polypeptide comprises a first VH polypeptide, the second polypeptide comprises a first VL polypeptide, the first VH and VL polypeptides form a first antigen-binding domain, the ligand is fused to one of the first VH or VL polypeptides via a first peptide linker, and the receptor is fused to the other of the first VH or VL polypeptide via a second peptide linker, and the ligand-receptor pair sterically inhibits the binding of the first antigen-binding domain to its homologous antigen. The first and second peptide linkers are of sufficient length to enable the pairing of the ligand and the receptor, and at least one of the first and second peptide linkers includes a protease cleavage site. The aforementioned fusion protein.

2. The fusion protein according to claim 1, wherein the binding of the first antigen-binding domain to its congener antigen is reduced by 10 times or more compared to the parent antigen-binding domain that is not fused to the ligand-receptor pair.

3. The fusion protein according to claim 1 or 2, wherein the first antigen-binding domain is Fab.

4. The fusion protein according to any one of claims 1 to 3, wherein the first antigen-binding domain binds to an antigen expressed on cancer cells or immune cells.

5. The fusion protein according to any one of claims 1 to 3, wherein the first antigen-binding domain binds to an antigen expressed on a T cell.

6. The fusion protein according to any one of claims 1 to 3, wherein the first antigen-binding domain binds to a tumor-associated antigen (TAA).

7. The fusion protein according to any one of claims 1 to 3, wherein the first antigen-binding domain binds to an antigen selected from the group consisting of differentiation antigen group 3 (CD3), human epidermal growth factor receptor 2 (HER2), epidermal growth factor receptor (EGFR), mesothelin (MSLN), tissue factor (TF), differentiation antigen group 19 (CD19), tyrosine protein kinase Met (c-Met), and cadherin 3 (CDH3).

8. The fusion protein according to any one of claims 1 to 7, wherein the biologically functional protein comprises a second antigen-binding domain comprising a second VH polypeptide and a second VL polypeptide.

9. The fusion protein according to claim 8, wherein the fusion protein comprises a second ligand-receptor pair, the ligand of the second ligand-receptor pair is fused to one of the second VH or VL polypeptide via a third peptide linker, and the receptor of the second ligand-receptor pair is fused to the other of the second VH or VL polypeptide via a fourth peptide linker, at least one of the third and fourth peptide linkers comprises a protease cleavage site, and the ligand-receptor pair sterically inhibits the binding of the second antigen-binding domain to its homologous antigen.

10. The fusion protein according to claim 8 or 9, wherein it binds to two different antigens, one of which is an antigen expressed by T cells and the other antigen is an antigen expressed by cancer cells.

11. The fusion protein according to claim 5 or 10, wherein the antigen expressed by the T cell is CD3.

12. An anti-CD3 paratope comprising VH containing HCDR1, HCDR2, and HCDR3, and VL containing LCDR1, LCDR2, and LCDR3, (a) HCDR1, HCDR2, and HCDR3 each contain the amino acid sequences shown in SEQ ID NOs. 207, 208, and 209, and LCDR1, LCDR2, and LCDR3 each contain the amino acid sequences shown in SEQ ID NOs. 211, 212, and 214, (b) HCDR1, HCDR2, and HCDR3 each contain the amino acid sequences shown in SEQ ID NOs. 224, 225, and 226, and LCDR1, LCDR2, and LCDR3 each contain the amino acid sequences shown in SEQ ID NOs. 228, 229, and 230, (c) HCDR1, HCDR2, and HCDR3 each contain the amino acid sequences shown in SEQ ID NOs. 232, 233, and 234, and LCDR1, LCDR2, and LCDR3 each contain the amino acid sequences shown in SEQ ID NOs. 236, 237, and 238, or (d) HCDR1, HCDR2, and HCDR3 each contain the amino acid sequences shown in SEQ ID NOs: 240, 241, and 242, respectively, and LCDR1, LCDR2, and LCDR3 each contain the amino acid sequences shown in SEQ ID NOs: 244, 245, and 246, The aforementioned anti-CD3 paratope The fusion protein according to claim 11, comprising:

13. The fusion protein according to any one of claims 1 to 12, wherein the biologically functional protein includes a dimeric Fc region.

14. The fusion protein according to claim 13, wherein the dimeric Fc region is a heterodimer Fc.

15. The fusion protein according to any one of claims 1 to 14, wherein the ligand-receptor pair is selected from the group consisting of PD1-PDL1, PD1-PDL2, CTLA4-CD80, CD28-CD80, CD28-CD86, CTLA4-CD86, PDL1-CD80, ICOS-ICOSL, NCRSRLG1-NKp30, and CD47-SIRPa.

16. The fusion protein according to any one of claims 1 to 14, wherein the ligand-receptor pair is PD1-PDL1, CTLA4-CD80, or CD47-SIRPa.

17. The ligand-receptor pair is PD1-PDL1, and (a) The ligand PDL1 comprises the amino acid sequence shown in SEQ ID NO: 8, and / or (b) The receptor PD1 comprises the amino acid sequence shown in SEQ ID NO: 9 The fusion protein according to claim 16.

18. The ligand-receptor pair is CTLA4-CD80, and (a) The ligand CD80 comprises the amino acid sequence shown in SEQ ID NO: 25, SEQ ID NO: 185, SEQ ID NO: 187 or SEQ ID NO: 189, and / or (b) The receptor CTLA4 comprises the amino acid sequence shown in SEQ ID NO: 26 The fusion protein according to claim 16.

19. The fusion protein according to claim 15, wherein the ligand-receptor pair is selected from the group consisting of CTLA4-CD80, PDL1-CD80, and CD28-CD80, and the ligand CD80 comprises the amino acid sequence shown in SEQ ID NO: 25, having a mutation selected from the group consisting of (a) H18Y, A26E, E35D, M47S, I61S, and D90G; (b) E35D, M47S, N48K, I61S, K89N; (c) E35D, D46V, M47S, I61S, D90G, K93E; (d) H18Y, A26E, E35D, M47S, I61S, V68M, A71G, D90G; (e) I58S, V68S, L70S; (f) M47S, I61S, or (g) V22S.

20. The fusion protein according to any one of claims 1 to 19, wherein the receptor and the ligand are fused to the N-terminus of the first and second polypeptides, respectively.

21. The aforementioned proteases are serine protease, MMP1, MMP2, MMP3, MMP7, MMP8, MMP9, MMP10, MMP11, MMP12, MMP13, MMP14, MMP15, MMP16, MMP17, MMP18 (collagenase 4), MMP19, MMP20, MMP21, adamaricin, ceralisin, astacin, caspase 1, caspase 2, caspase 3, caspase 4, caspase 5, caspase 6, caspase 7, caspase 8, caspase 9, caspase 10, caspase 11, caspase 12, A fusion protein according to any one of claims 1 to 20, selected from the group consisting of caspase 13, caspase 14, cathepsin A, cathepsin B, cathepsin D, cathepsin E, cathepsin K, cathepsin S, granzyme B, guanidinobenzoatase (GB), hepsin, elastase, regmine, matryptase, matryptase 2, meprin, neurosin, MT-SP1, neprilysin, plasmin, PSA, PSMA, TACE, TMPRSS3, TMPRSS4, uPA, calpain, FAP, and KLK.

22. The fusion protein according to any one of claims 1 to 21, wherein the protease is uPA or matryptase.

23. The fusion protein according to any one of claims 1 to 20, wherein the protease cleavage site includes the amino acid sequence MSGRSANA (SEQ ID NO: 28).

24. The fusion protein according to any one of claims 1 to 23, wherein each of the peptide linkers is independently 3 to 50 or 5 to 20 amino acid lengths.

25. The fusion protein according to any one of claims 1 to 24, wherein one of the first or second peptide linkers does not contain a protease cleavage site.

26. Each of the peptide linkers can independently (a) (Gly ｎ Ser) linker, where the (Gly ｎ Ser) linker is (Gly ３ Ser) ｎ (Gly ４ Ser) １ (Gly ３ Ser) １ (Gly ４ Ser) ｎ (Gly ３ Ser) ｎ (Gly ４ Ser) ｎ and (Gly ４ Ser) ｎ (where n is an integer from 1 to 5), or contains an amino acid sequence selected from the group consisting of (b) (EAAAK) ｎ It includes a linker (where n is an integer from 1 to 5), or (c) containing a polyproline linker, optionally PPP or PPPP, or a glycine-proline linker, optionally GPPPG, GGPPPGG, GPPPPPG or GGPPPPPGG, or (d) an immunoglobulin hinge region sequence containing an amino acid sequence having up to 30 percent difference in amino acid sequence identity compared to the amino acid sequence of the wild-type immunoglobulin hinge region, A fusion protein according to any one of claims 1 to 25.

27. The fusion protein according to any one of claims 1 to 25, wherein at least one of the peptide linkers comprises the amino acid sequence EAAAK EAAAK (SEQ ID NO: 38).

28. below: The Fab region and the Fc region, The Fab region comprises a VH polypeptide and a VL polypeptide that form an antigen-binding domain. The aforementioned Fab region and Fc region; and A ligand-receptor pair comprising the extracellular component of an immunoglobulin superfamily receptor and its related ligand or receptor-binding fragment, The ligand is fused to the N-terminus of one of the VH or VL polypeptides via a first peptide linker, and the receptor is fused to the N-terminus of the other VH or VL polypeptide via a second peptide linker. The first and second peptide linkers are of sufficient length to enable the pairing of the ligand and the receptor. At least one of the first and second peptide linkers includes a protease cleavage site, and The ligand-receptor pair sterically inhibits the binding of the antigen-binding domain to its homologous antigen. The ligand-receptor pair A fusion protein containing [the specified ingredient].

29. The fusion protein according to claim 28, further comprising an additional Fab region or scFv.

30. A pharmaceutical composition comprising a fusion protein according to any one of claims 1 to 29 for the treatment of cancer.

31. A pharmaceutical composition comprising a fusion protein according to any one of claims 1 to 29 for modulating an immune response.

32. One or more vectors encoding the fusion protein according to any one of claims 1 to 29.

33. A cell comprising one or more vectors according to claim 32.

Images