Anti-PD-L1 / anti-CD47 natural antibody structure-mimicking heterodimer bispecific antibody and method for producing same

Abstract

The present invention provides an anti-PD-L1 / anti-CD47 bispecific antibody and a method for its preparation, which is a highly stable heterodimer with the characteristics of natural IgG and no heavy-light chain mismatch, and can simultaneously bind to two target molecules and has fewer side effects.

Description

[Technical field]

[0001] The present invention relates to an anti-PD-L1 / anti-CD47 native antibody structure-like heterodimeric bispecific antibody and a method for preparing the same. Specifically, the present invention provides a highly stable heterodimeric anti-PD-L1 / anti-CD47 bispecific antibody that has the characteristics of native IgG and has no heavy-light chain mismatch, and a method for preparing the same. [Background technology]

[0002] Programmed death ligand 1 (PD-L1) is a ligand of the immune checkpoint programmed death-1 (PD-1), and belongs to the B7 family. It is inducibly expressed on the surface of various immune cells, including T cells, B cells, monocytes, macrophages, DC cells, endothelial cells, and epidermal cells. When bound to PD-1, PD-L1 is mainly involved in the negative regulation of T cell activation, and can regulate the strength and duration of immune responses. In addition to being a ligand for PD-1, PD-L1, as a ligand for CD80, can transmit negative regulatory signals to T cells and induce immune tolerance of T cells (Autoimmun Rev, 2013, 12(11): 1091-1100. Front Immunol, 2013, 4: 481. Nat Rev Cancer, 2012, 12(4): 252-264. Trends Mol Med. 2015 Jan; 21(1): 24-33. Clin Cancer Res. 2012 Dec 15; 18(24): 6580-7.). In normal cases, PD-L1 and PD-1 can induce and maintain self-immune tolerance in living tissues, prevent the immune system from being overly activated to damage self-tissues during the inflammatory response, and play a positive role in avoiding the occurrence of autoimmune diseases. In pathological situations, it is involved in the occurrence and development process of tumor immunity and various autoimmune diseases. Many studies have reported that PD-L1 is highly expressed in various tumor tissues, PD-1 is highly expressed in tumor-infiltrating lymphocytes, and excessive expression of PD-L1 and PD-1 is closely associated with poor clinical tumor prognosis (Anticancer Agents Med Chem.2015;15(3):307-13.Hematol Oncol Stem Cell Ther.2014 Mar;7(1):1-17.Trends Mol Med.2015 Jan;21(1):24-33.Immunity.2013 Jul 25;39(1):61-73.J Clin Oncol.2015 Jun 10;33(17):1974-82.).Blocking the interaction of PD-L1 / PD-1 and CD80 / PD-L1 by PD-L1 monoclonal antibodies has shown good antitumor effects in preclinical experimental studies and clinical trials. Currently, PD-L1 monoclonal antibodies have already been approved for the treatment of many tumors, such as non-small cell lung cancer and urothelial carcinoma. However, only some tumor patients can benefit from such monoclonal antibody therapy, while many patients do not respond to such monoclonal antibodies (Expert Opin Ther Targets.2014 Dec;18(12):1407-20.Oncoloy(Williston Park).2014 Nov;28 Suppl 3:15-28.).

[0003] CD47, also known as integrin-associated protein, is a 50Kd transmembrane protein belonging to the immunoglobulin superfamily. It is widely expressed in many types of cells, but its expression is obviously enhanced in many types of tumor cells (Proc Natl Acad Sci USA, 2012, 109(17):6662-6667). The ligand of CD47 is signal-regulatory protein α (SIRPα), which is mainly expressed in macrophages and binds to CD47 to transmit the signal "don't eat me" and inhibit the phagocytosis of macrophages (Curr Opin Immunol, 2009, 21(1):47-52). Anti-CD47 antibodies can be used to block the CD47-SIRPα signal pathway to exert anti-tumor effects. Currently, many anti-CD47 monoclonal antibodies are in the clinical research stage and are used to treat many types of hematological and solid tumors. However, because CD47 is also expressed on the surface of red blood cells, these anti-CD47 treatments may cause side effects such as severe anemia and thrombocytopenia, and may have low bioavailability.

[0004] Therefore, there is a need in the field to investigate novel therapeutic agents that can simultaneously block the PD-L1 and CD47 signaling pathways. Summary of the Invention

[0005] The present invention provides a novel highly stable heterodimeric bispecific antibody that has the structural characteristics of natural IgG and has no heavy chain-light chain mismatch, and is capable of simultaneously blocking PD-L1 and CD47, and a method for preparing the same. This bispecific antibody tends to selectively bind to tumor cells that simultaneously highly express PD-L1 and CD47, thereby exerting an efficient and specific killing effect and having low toxic side effects.

[0006] A first aspect of the present invention relates to a bispecific antibody comprising a first antigen-binding functional region that specifically binds to PD-L1 and a second antigen-binding functional region that specifically binds to CD47, wherein the first antigen-binding functional region that specifically binds to PD-L1 is (A) a heavy chain variable region; (B) a light chain variable region, The heavy chain variable region comprises: (a) HCDR1 comprising the amino acid sequence shown in SEQ ID NO.37; (b) an HCDR2 comprising the amino acid sequence set forth in SEQ ID NO. 38; (c) an HCDR3 comprising the amino acid sequence set forth in SEQ ID NO. 39; The light chain variable region comprises: (a) an LCDR1 comprising the amino acid sequence shown in SEQ ID NO. 40; (b) an LCDR2 comprising the amino acid sequence shown in SEQ ID NO. 41; (c) an LCDR3 comprising the amino acid sequence shown in SEQ ID NO.42.

[0007] In some embodiments, the second antigen-binding functional region that specifically binds to CD47 of the bispecific antibody comprises: (A) a heavy chain variable region; (B) a light chain variable region, The heavy chain variable region comprises: (a) an HCDR1 comprising the amino acid sequence shown in SEQ ID NO. 43; (b) an HCDR2 comprising the amino acid sequence shown in SEQ ID NO. 44; (c) an HCDR3 comprising the amino acid sequence set forth in SEQ ID NO. 45; The light chain variable region comprises: (a) an LCDR1 comprising the amino acid sequence shown in SEQ ID NO. 46; (b) an LCDR2 comprising the amino acid sequence shown in SEQ ID NO. 47; (c) an LCDR3 comprising the amino acid sequence shown in SEQ ID NO.48.

[0008] In some embodiments, the first antigen-binding functional region that specifically binds to PD-L1 of the bispecific antibody comprises: (A) a heavy chain variable region; (B) a light chain variable region, The heavy chain variable region comprises: (a) HCDR1 comprising the amino acid sequence shown in SEQ ID NO.37; (b) an HCDR2 comprising the amino acid sequence set forth in SEQ ID NO. 38; (c) an HCDR3 comprising the amino acid sequence set forth in SEQ ID NO. 39; The light chain variable region comprises: (a) an LCDR1 comprising the amino acid sequence shown in SEQ ID NO. 40; (b) an LCDR2 comprising the amino acid sequence shown in SEQ ID NO. 41; (c) an LCDR3 comprising the amino acid sequence set forth in SEQ ID NO. 42; The second antigen-binding functional region that specifically binds to CD47 is (A) a heavy chain variable region; (B) a light chain variable region, The heavy chain variable region comprises: (a) an HCDR1 comprising the amino acid sequence shown in SEQ ID NO. 43; (b) an HCDR2 comprising the amino acid sequence shown in SEQ ID NO. 44; (c) an HCDR3 comprising the amino acid sequence set forth in SEQ ID NO. 45; The light chain variable region comprises: (a) an LCDR1 comprising the amino acid sequence shown in SEQ ID NO. 46; (b) an LCDR2 comprising the amino acid sequence shown in SEQ ID NO. 47; (c) an LCDR3 comprising the amino acid sequence shown in SEQ ID NO.48.

[0009] The six CDR sequences of the first antigen-binding functional domain that specifically binds to PD-L1 shown in SEQ ID NOs. 37-42 and the six CDR sequences of the second antigen-binding functional domain that specifically binds to CD47 shown in SEQ ID NOs. 43-48 are respectively the CDRs of the heavy and light chain variable domains of monoclonal antibodies obtained by the present inventors using human PD-L1 and CD47 as antigens by hybridoma technology; these monoclonal antibodies differ from PD-L1 antibodies and CD47 antibodies known in the prior art and have higher biological activities, for example higher binding specificity and anti-tumor activity.

[0010] In some embodiments, the six CDR sequences of the first antigen-binding functional region that specifically binds to PD-L1 each have at least 70% identity, such as at least 75%, 80%, 85%, 90%, 95% or more identity, to a sequence as set forth in SEQ ID NO. 37-42, and retain the biological activity of the corresponding parent sequence, or are variant sequences obtained by deleting, substituting and / or adding one or more, such as one, two, three or more amino acid residues, to the sequence as set forth in SEQ ID NO. 37-42, and retaining the biological activity of the corresponding parent sequence.

[0011] In some embodiments, the six CDR sequences of the second antigen-binding functional region that specifically binds to CD47 each have at least 70% identity, for example at least 75%, 80%, 85%, 90%, 95% or more identity, to the sequences shown in SEQ ID NOs. 43-48 and retain the biological activity of the corresponding parent sequence, or are mutant sequences that retain the biological activity of the corresponding parent sequence obtained by deleting, substituting and / or adding one or more, for example one, two, three or more, amino acid residues to the sequences shown in SEQ ID NOs. 43-48.

[0012] In some embodiments, the first antigen-binding functional region that specifically binds to PD-L1 comprises a heavy chain variable region and a light chain variable region, wherein the heavy chain variable region comprises the amino acid sequence set forth in SEQ ID NO:6, or has at least 70% identity, such as at least 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or more identity, to the sequence set forth in SEQ ID NO.6 and retains the biological activity of the corresponding parent sequence, or is a variant sequence obtained by deleting, substituting and / or adding one or more amino acid residues, such as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20 or more amino acid residues, to the sequence set forth in SEQ ID NO.6, and retaining the biological activity of the corresponding parent sequence. The light chain variable region comprises the amino acid sequence set forth in SEQ ID NO:2, or has at least 70% identity, e.g., at least 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or more identity, to the sequence set forth in SEQ ID NO.2 and retains the biological activity of the corresponding parent sequence, or is a variant sequence obtained by deleting, substituting and / or adding one or more amino acid residues, e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20 or more, to the sequence set forth in SEQ ID NO.2, and retaining the biological activity of the corresponding parent sequence.

[0013] In some embodiments, the second antigen-binding functional region that specifically binds to CD47 comprises a heavy chain variable region and a light chain variable region, wherein the heavy chain variable region comprises an amino acid sequence selected from SEQ ID NO. 14, 20, 24, 28, 32, or 36, or has at least 70% identity, such as at least 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or more identity, to a sequence set forth in SEQ ID NO. 14, 20, 24, 28, 32, or 36, and retains the biological activity of the corresponding parent sequence, or is a variant sequence obtained by deleting, substituting and / or adding one or more, such as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20 or more, amino acid residues from the sequence set forth in SEQ ID NO. 6, thereby retaining the biological activity of the corresponding parent sequence. The light chain variable region comprises an amino acid sequence as set forth in SEQ ID NO. 12, 18, 22, 26, 30 or 34, or has at least 70% identity, such as at least 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% or more identity, to the sequence as set forth in SEQ ID NO. 12, 18, 22, 26, 30 or 34, and retains the biological activity of the corresponding parent sequence; or is a variant sequence obtained by deleting, substituting and / or adding one or more amino acid residues, such as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20 or more amino acid residues, to the sequence as set forth in SEQ ID NO. 6, which retains the biological activity of the corresponding parent sequence.

[0014] In some embodiments, the first antigen-binding functional region that specifically binds to PD-L1 comprises a heavy chain variable region and a light chain variable region, wherein the heavy chain variable region comprises the amino acid sequence set forth in SEQ ID NO.6 and the light chain variable region comprises the amino acid sequence set forth in SEQ ID NO.2; and, The second antigen-binding functional region that specifically binds to CD47 comprises a heavy chain variable region and a light chain variable region selected from the following: a) a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 14, and a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 12; b) a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 20, a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 18; c) a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 24, a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 22; d) a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 28, a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 26; e) a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 32, and a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 30; or f) a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO.36; a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO.34.

[0015] In some embodiments, the first antigen-binding functional region and the second antigen-binding functional region of the bispecific antibody are selected from a Fab fragment, an scFv fragment, and a variable domain fragment Fv.

[0016] In a further embodiment, the first antigen-binding functional domain and the second antigen-binding functional domain of the bispecific antibody are both Fab fragments.

[0017] In some embodiments, the Fab fragment of the bispecific antibody comprises a first heavy chain variable region and a second heavy chain variable region that are different, and a first light chain variable region and a second light chain variable region that are different.

[0018] In some embodiments, one of the first and second antigen-binding functional regions of the bispecific antibody is a Fab fragment and the other is an scFv.

[0019] In some embodiments, the bispecific antibody comprises a first Fc chain and a second Fc chain.

[0020] Here, the first Fc chain and the second Fc chain are both Fc fragments of immunoglobulin G containing amino acid substitutions, and the first Fc chain and the second Fc chain together form a heterodimer capable of binding to an Fc receptor.

[0021] Here, the first Fc chain and the second Fc chain are linked to the first antigen-binding functional region and the second antigen-binding functional region, respectively, via a covalent bond or a linker.

[0022] and one of the first Fc chain and the second Fc chain comprises amino acid substitutions of T366L and D399R at positions 366 and 399, and the other comprises amino acid substitutions of L351E, Y407L and K409V at positions 351, 407 and 409, wherein the amino acid positions are numbered according to the Kabat EU index numbering system.

[0023] Unless otherwise specified, all amino acid positions in antibodies herein are numbered according to the Kabat EU index numbering system.

[0024] In some embodiments, the weight percentage of the first Fc chain and the first antigen-binding functional region linked thereto via a covalent bond, and the second Fc chain and the second antigen-binding functional region linked thereto via a covalent bond of the bispecific antibody based on all polypeptide chains in a formed homodimer is less than 50%, for example less than 45%, less than 40%, less than 35%, less than 30%, less than 25%, less than 20%, or less, in a solution in the presence of a reducing agent and when the solution does not contain any polypeptides other than the first Fc chain and the first antigen-binding functional region linked thereto via a covalent bond, and the second Fc chain and the second antigen-binding functional region linked thereto via a covalent bond.

[0025] In some embodiments, the first antigen-binding functional region of the bispecific antibody comprises the amino acid sequence of SEQ ID NOs: 2 and 6.

[0026] In some embodiments, the second antigen-binding functional region of the bispecific antibody comprises the amino acid sequence of SEQ ID NO: 12, 14, 18, 20, 22, 24, 26, 28, 30, 32, 34, 36. It is clear to those skilled in the art that the combination of the amino acid sequences should follow biological rules, i.e., the variable regions and antigen-binding fragments of the light and heavy chains or the light and heavy chains should match each other, for example, SEQ ID NO: 12 and SEQ ID NO: 14 match each other, and SEQ ID NO: 18 and SEQ ID NO: 14 match each other.

[0027] In some embodiments, the first light chain and the second light chain of the bispecific antibody both comprise the amino acid sequence of SEQ ID NO:4.

[0028] In some embodiments, one of the first and second heavy chains of the bispecific antibody comprises the amino acid sequence of SEQ ID NO:8 or 10, and the other comprises the amino acid sequence of SEQ ID NO:16.

[0029] In some embodiments, the bispecific antibody comprises a pair of a first heavy chain and a first light chain that specifically binds to PD-L1, where the first heavy chain has a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO.6 and a heavy chain constant region comprising the amino acid sequence set forth in SEQ ID NO.8 or 10. The first light chain has a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO.2 and a light chain constant region comprising the amino acid sequence set forth in SEQ ID NO.4.

[0030] In some embodiments, the bispecific antibody comprises a pair of a second heavy chain and a second light chain that specifically binds to CD47, wherein the second heavy chain has a heavy chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 14, 20, 24, 28, 32 or 36 and a heavy chain constant region comprising the amino acid sequence set forth in SEQ ID NO. 16. The second light chain has a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 12, 18, 22, 26, 30 or 34 and a light chain constant region comprising the amino acid sequence set forth in SEQ ID NO.4.

[0031] A second aspect of the invention relates to an isolated polynucleotide encoding a bispecific antibody as described above.

[0032] In some embodiments, the nucleotide sequence encoding the amino acids of the first antigen-binding functional region comprises the nucleotide sequence of SEQ ID NOs: 1 and 5.

[0033] In some embodiments, the nucleotide sequence encoding the amino acid of the second antigen-binding functional region comprises the nucleotide sequence of SEQ ID NO: 11, 13, 17, 19, 21, 23, 25, 27, 29, 31, 33, 35. As described above, the amino acid sequence combination method should follow biological rules, that is, the nucleotide sequences encoding the light chain and heavy chain or variable regions thereof and the antigen-binding fragments should match each other, for example, SEQ ID NO: 11 and SEQ ID NO: 13 match each other, and SEQ ID NO: 17 and SEQ ID NO: 13 match each other.

[0034] In some embodiments, the nucleotide sequences encoding the amino acids of the first light chain and the second light chain both comprise the nucleotide sequence of SEQ ID NO:3.

[0035] In some embodiments, the nucleotide sequence encoding the amino acids of one of the first heavy chain and the second heavy chain comprises the nucleotide sequence of SEQ ID NO:7 or 9, and the nucleotide sequence encoding the other comprises the nucleotide sequence of SEQ ID NO:15.

[0036] A third aspect of the invention relates to a recombinant expression vector comprising an isolated polynucleotide as described above.

[0037] In some embodiments, the expression vector is the plasmid vector X0GC obtained by pCDNA-based modification.

[0038] A fourth aspect of the invention relates to a host cell comprising an isolated polynucleotide as described above or a recombinant expression vector as described above.

[0039] In some embodiments, the host cell is selected from human embryonic kidney cells HEK293 or HEK293T, HEK293E, HEK293F obtained by modification based on HEK293 cells, hamster ovary cells CHO or CHO-S, CHO-dhfr-, CHO / DG44, ExpiCHO obtained by modification based on CHO cells, Escherichia coli or Escherichia coli BL21, BL21(DE3), Rosetta, Origami obtained by modification based on E. coli, yeast or Pichia pastoris obtained by modification based on yeast, Saccharomyces cerevisiae, Kluyveromyces lactis, Hansenula polymorpha, insect cells or cells High5, SF9 obtained by modification based on insect cells, plant cells, mammalian mammary gland cells, somatic cells.

[0040] A fifth aspect of the invention relates to a composition comprising a bispecific antibody as described above or an isolated polynucleotide as described above or a recombinant expression vector as described above or a host cell as described above and a pharma- ceutically acceptable carrier.

[0041] A sixth aspect of the invention relates to a method for producing a bispecific antibody as described above comprising the steps of:

[0042] 1) expressing in a host cell an isolated polynucleotide as described above or a recombinant expression vector as described above, respectively; 2) reducing each expressed protein in the host cell; 3) mixing the reduced protein and then oxidizing the mixture.

[0043] In some embodiments, the reduction step comprises 1) performing the reduction reaction in the presence of a reducing agent selected from 2-mercaptoethylamine, dithiothreitol, tris(2-carboxyethyl)phosphine or other chemical derivatives, and 2) removing the reducing agent. In some embodiments, the reducing agent is dithiothreitol at a concentration of 0.1 mM or more, e.g., 0.1 mM, 0.2 mM, 0.3 mM, 0.4 mM, 0.5 mM or more, and the reaction conditions are at 4° C. for at least 3 hours, e.g., 3.5 hours, 4 hours, 4.5 hours, 5 hours, 5.5 hours, 6 hours or more.

[0044] In some embodiments, the oxidation step is oxidation in air, but further comprises carrying out the oxidation reaction in the presence of an oxidizing agent selected from L-dehydroascorbic acid or a chemical derivative thereof, in some embodiments, the oxidizing agent is L-dehydroascorbic acid at a concentration of 0.5 mM or more, e.g., 0.6 mM, 0.7 mM, 0.8 mM, 0.9 mM, 1.0 mM, 1.2 mM, 1.5 mM or more, and the reaction conditions are at 4° C. for at least 5 hours, e.g., 5 hours, 6 hours, 7 hours, 8 hours, 9 hours or more.

[0045] In some embodiments, the method further comprises the steps of isolation and purification, hi some embodiments, the isolation and purification comprises by cation exchange resin, anion exchange resin, reverse phase chromatography, affinity chromatography, size exclusion chromatography and combinations thereof.

[0046] A seventh aspect of the invention relates to the use of a bispecific antibody as described above, and / or an isolated polynucleotide as described above, and / or a recombinant expression vector as described above, and / or a host cell as described above, and / or a composition as described above, in the manufacture of a medicament for preventing and / or treating a disease in a subject.

[0047] An eighth aspect of the present invention relates to a bispecific antibody as described above, and / or an isolated polynucleotide as described above, and / or a recombinant expression vector as described above, and / or a host cell as described above, and / or a composition as described above for use as a medicament for preventing and / or treating a disease in a subject.

[0048] A ninth aspect of the invention relates to a method for preventing and / or treating a disease comprising the step of administering to a subject in need thereof a bispecific antibody as previously described, and / or an isolated polynucleotide as previously described, and / or a recombinant expression vector as previously described, and / or a host cell as previously described, and / or a composition as previously described.

[0049] In some embodiments, the subject is a mammal, preferably a human.

[0050] In some embodiments, the disease is selected from leukemia, lymphoma, myeloma, brain cancer, squamous cell carcinoma of the head and neck, non-small cell lung cancer, nasopharyngeal cancer, esophageal cancer, gastric cancer, pancreatic adenocarcinoma, gallbladder cancer, liver cancer, colorectal cancer, breast cancer, ovarian cancer, cervical cancer, endometrial cancer, uterine sarcoma, prostate cancer, bladder cancer, renal cell carcinoma, melanoma, small cell lung cancer, and bone cancer, and preferably, the disease is selected from leukemia, lymphoma, and myeloma.

[0051] In other words, the present invention has designed a novel anti-PD-L1 / anti-CD47 natural antibody structure-like heterodimer bispecific antibody that has the characteristics of natural IgG and has no heavy-light chain mismatch, and is highly stable. The bispecific antibody prepared in the present invention can simultaneously bind to two target molecules, PD-L1 and CD47, and when applied to the treatment of complex diseases, it can exert better effects than a single therapeutic agent and has fewer side effects. In addition, compared to the combination therapy of multiple drugs, the bispecific antibody is a single therapeutic molecule, which not only makes it easier for patients and medical professionals to use, but also simplifies the complicated process of drug development. [Brief description of the drawings]

[0052] [Figure 1] Chromatograms of the elution peaks of anti-PD-L1 expression products are shown. [Diagram 2] A chromatogram of the elution peak of anti-CD47 expression product is shown. [Diagram 3] 1 shows the structure of the anti-PD-L1 / anti-CD47 heterodimer antibody molecule. [Figure 4] 1 shows the structure of a half antibody molecule containing one heavy and one light chain. [Diagram 5] Figure 5 shows the results of SEC-HPLC analysis of half antibody molecules containing one heavy chain and one light chain, where Figure 5A and B show the results for an anti-PD-L1 half antibody molecule and an anti-CD47 half antibody molecule, respectively. [Figure 6] Figure 2 shows the results of SEC-HPLC analysis of anti-PD-L1 / anti-CD47 heterodimer antibody molecules. [Figure 7] Figure 2 shows the results of CE analysis of anti-PD-L1 / anti-CD47 heterodimer antibody molecules. [Figure 8] Figure 8A shows the affinity of several anti-PD-L1 / anti-CD47 heterodimer antibodies to PD-L1, and Figure 8B shows the affinity of several anti-PD-L1 / anti-CD47 heterodimer antibodies to CD47. [Figure 9]Figure 9A shows the blocking activity of multiple anti-PD-L1 / anti-CD47 heterodimer antibodies against PD-L1 / PD-1 binding, Figure 9B shows the blocking activity of multiple anti-PD-L1 / anti-CD47 heterodimer antibodies against PD-L1 / CD80 binding, and Figure 9C shows the blocking activity of multiple anti-PD-L1 / anti-CD47 heterodimer antibodies against CD47 / SIRPα binding. [Figure 10] Figure 10A shows the binding activity of anti-CD47 monoclonal antibodies to HCC827 and RBCs, and Figures 10B and C show the binding activity of anti-PD-L1 / anti-CD47 heterodimer antibodies to HCC827 and RBCs. [Figure 11] 1 shows the regulatory activity of anti-PD-L1 / anti-CD47 heterodimer antibodies on T cells. [Figure 12] This shows the phagocytic activity of tumor cells by macrophages mediated by anti-PD-L1 / anti-CD47 heterodimer antibody. [Figure 13] 1 shows the anti-tumor effect of anti-PD-L1 / anti-CD47 heterodimer antibodies in a xenograft tumor model. [Figure 14] Figure 2 shows the anti-tumor effect of anti-PD-L1 / anti-CD47 heterodimer antibodies in a gene knock-in mouse allogeneic tumor model. [Figure 15] Figure 2 shows that the anti-PD-L1 / anti-CD47 heterodimer antibody BH3012h has a stronger anti-tumor effect than the corresponding anti-PD-L1 bivalent monoclonal antibody and anti-CD47 bivalent monoclonal antibody. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0053] Definition: As used herein, the term "antibody" is used in the broadest sense to refer to a protein that contains an antigen-binding site and includes natural and artificial antibodies of various structures, including, but not limited to, complete antibodies and antigen-binding fragments of antibodies.

[0054] As used herein, the term "bispecific antibody" comprises antigen-binding domains that specifically bind to epitopes on two different biological molecules. Unless otherwise specified, the order of the antigens to which the bispecific antibody binds in the name of the described bispecific antibody is arbitrary. That is, in some embodiments, the terms "anti-PD-L1 / CD47 bispecific antibody" and "anti-CD47 / PD-L1 bispecific antibody" may be used interchangeably. In some embodiments, a bispecific antibody comprises two half antibodies, where each half antibody comprises a single heavy chain variable region and at least a portion of any heavy chain constant region, and a single light chain variable region and at least a portion of any light chain constant region. In some embodiments, a bispecific antibody comprises two half antibodies, where each half antibody comprises a single heavy chain variable region and a single light chain variable region, and does not comprise more than one single heavy chain variable region and does not comprise more than one single light chain variable region. In some embodiments, a bispecific antibody comprises two half antibodies, where each half antibody comprises a single heavy chain variable region and a single light chain variable region, and where the first half antibody binds a first antigen but not a second antigen, and the second half antibody binds the second antigen but not the first antigen.

[0055] The term "complementarity determining region" or "CDR region" or "CDR" (which may be used interchangeably herein with hypervariable region "HVR") refers to the regions of an antibody variable domain that form sequence-wise highly variable and structurally determined loops ("hypervariable loops") and / or contain antigen contact residues ("antigen contact points"). The CDRs are primarily responsible for binding to the antigen epitope. Herein, the three CDRs of the heavy chain are referred to as HCDR1, HCDR2 and HCDR3, and the three CDRs of the light chain are referred to as LCDR1, LCDR2 and LCDR3.

[0056] It should be noted that the boundaries of the CDRs of the variable regions of the same antibody obtained based on different numbering schemes may be different. That is, the CDR sequences of the variable regions of the same antibody defined by different numbering schemes are different. Therefore, when a specific CDR sequence defined in the present invention is used to define an antibody, the scope of the antibody further includes antibodies whose variable region sequences include the specific CDR sequence, but whose so-called CDR boundaries differ from the boundaries of the specific CDRs defined in the present invention by applying a different scheme (e.g., a rule or combination of different numbering schemes). "Covalent bond" refers to the formation of a single molecule by covalently linking two Fc chains or any one of the Fc chains and the antigen-binding functional region linked thereto in a heterodimeric bispecific antibody. Here, the Fc chain comprises a first antigen-binding functional region and a second antigen-binding functional region linked by one or more covalent bonds (e.g., disulfide bonds), and the first Fc chain and the second Fc chain are each linked to the antigen-binding functional region via a covalent bond (e.g., an imine bond or an amide bond).

[0057] An antigen-binding functional region refers to a region that can specifically interact with a target molecule such as an antigen. Its action is highly selective. Usually, a sequence that recognizes one target molecule cannot recognize the sequence of another molecule. Representative antigen-binding functional regions include antibody variable regions, structural variants of antibody variable regions, receptor-binding domains, ligand-binding domains, or enzyme-binding domains.

[0058] Interchain binding by one or more disulfide bonds means that the first Fc chain and the second Fc chain are linked by one or more disulfide bonds to form a heterodimeric fragment. In the present invention, the one or more disulfide bonds can be formed when the first Fc chain and the second Fc chain, or the first Fc chain and the second Fc chain and the antigen-binding functional region linked thereto, are synthesized in the same cell, or can be formed by an in vitro reduction-oxidation process after the first Fc chain and the second Fc chain, or the first Fc chain and the second Fc chain and the antigen-binding functional region linked thereto are separately synthesized in different cells.

[0059] The first Fc chain and the second Fc chain refer to a binding fragment formed by a covalent bond including a disulfide bond, and each chain includes at least a part of the heavy chain constant region of an immunoglobulin. The first Fc chain and the second Fc chain have different amino acid sequences, and differ by at least one amino acid. In the first Fc chain and the second Fc chain of the present invention, since there is a strong repulsive force between the same chains and an attractive force between different chains, the first Fc chain and the second Fc chain, or the first Fc chain and the second Fc chain and the antigen-binding functional region linked thereto, tend to form a heterodimer when both are expressed in a cell. When the first Fc chain and the second Fc chain, or the first Fc chain and the second Fc chain and the antigen-binding functional region linked thereto, are expressed separately in two host cells, the first Fc chain or the first Fc chain and the antigen-binding functional region linked thereto do not tend to form homodimers, and the second Fc chain or the second Fc chain and the antigen-binding functional region linked thereto do not tend to form homodimers. In the present invention, when the first Fc chain and the second Fc chain, or the first Fc chain and the second Fc chain and the antigen-binding functional region linked thereto are expressed separately in two host cells in the presence of a reducing agent, the proportion of homodimers is less than 50%, i.e., the proportion of monomers (one Fc chain or one Fc chain and the antigen-binding functional region linked thereto) is more than 50%.

[0060] Immunoglobulins have a symmetrical structure with four polypeptide chains, including two identical heavy chains that are relatively long, have a relatively large relative molecular weight, contain 450-550 amino acid residues, and have a relative molecular weight between 55,000 Da and 70,000 Da, and two identical light chains (L chains) that are relatively short, have a relatively small relative molecular weight, contain about 210 amino acid residues, and have a relative molecular weight of about 24,000 Da. The sequence of about 110 amino acids near the N-terminus varies greatly in the heavy and light chains of different immunoglobulins and is called the variable region (V region). In contrast, the remaining amino acid sequence near the C-terminus is relatively stable and is called the constant region (C region). Each heavy chain of an antibody is composed of a heavy chain variable region (abbreviated as VH herein) and a heavy chain constant region (abbreviated as CH herein), and each light chain is composed of a light chain variable region (abbreviated as VL herein) and a light chain constant region (abbreviated as CL herein). In the heavy chain, the variable region occupies about 1 / 4 of the length of the heavy chain, while the constant region occupies about 3 / 4 of the length of the heavy chain. Regarding the five known classes of Ig, namely IgG (γ), IgA (α), IgD (δ), IgM (μ) and IgE (ε), the H chains of the first three Igs have three constant regions, namely CH1, CH2 and CH3. The H chains of the last two types (IgM and IgE) have one VH region and four constant regions, namely CH1 to CH4. The constant region is also the framework of the immunoglobulin molecule and one of the regions that activates the immune response. The examples of the present invention relate to IgG, but it is clear to those skilled in the art that the class of the antibody of the present invention can be converted by known methods. For example, the first IgM antibody of the present invention can be converted in class to the IgG antibody of the present invention. It should be noted that the IgG subclass can be converted to another subclass by class conversion technology, for example, IgG1 can be converted to IgG2. Thus, the effector functions of the antibodies of the invention can be converted by isotype switching to, for example, IgG1, IgG2, IgG3, IgG4, IgD, IgA, IgE or IgM antibodies for various therapeutic applications.In one example, the antibody of the invention is an IgG1 antibody, such as IgG1,κ.

[0061] In the present invention, the portion of the constant region includes at least the region where the first Fc chain and the second Fc chain interact with each other. For IgG, the region is a portion of amino acids located in the CH3 region, including at least GLN347, TYR349, THR350, LEU351, SER354, ARG355, ASP356, GLU357, LYS360, SER364, THR366, LEU368, LYS370, ASN390, LYS392, THR394, PR0395, VAL397, ASP399, SER400, PHE405, TYR407, LYS409, and LYS439.

[0062] "The first Fc chain and the second Fc chain are each linked to one antigen-binding functional region via a covalent bond or a linker" refers to the first Fc chain and the second Fc chain being each linked to an antigen-binding fragment of an antibody, or a single-chain antibody capable of recognizing an antigen, or a structural variant of another antibody fragment capable of recognizing an antigen, or a receptor capable of recognizing a ligand, or a ligand capable of recognizing a receptor, via a covalent bond or a linker. Here, the covalent bond is a type of chemical bond, in which two or more atoms share their outer electrons and ideally reach an electronically saturated state, thereby forming a relatively stable chemical structure called a covalent bond. In other words, a covalent bond is an interaction formed between atoms by sharing an electron pair. Atoms of the same element or different elements may be bonded together via a covalent bond. The covalent bond between the first Fc chain and the second Fc chain of the present invention includes, but is not limited to, an amide bond formed by a dehydration reaction between the amino group of one amino acid molecule and the carboxyl group of another amino acid molecule, an amide bond or an imine bond formed from an aldehyde group of ethylene glycol or polyethylene glycol, or other compounds or polymers thereof, and an amino group of one amino acid molecule. Here, the linker is an amino acid sequence or a compound or a compound polymer that can link two polypeptide chains by a covalent bond, where the amino acid sequence includes, but is not limited to, a small peptide segment such as GGGSGGGGSGGGGGS. The first Fc chain or the second Fc chain can be linked to a single chain antibody capable of recognizing an antigen, or a structural variant of another antibody fragment capable of recognizing an antigen, via an amide bond.

[0063] The first Fc chain and the second Fc chain tend to form heterodimers rather than homodimers, since in the first Fc chain and the second Fc chain, there is a repulsive force between the same polypeptide chains, while there is an attractive force between different polypeptide chains, when co-expressed in a cell, the first Fc chain and the second Fc chain, or the first Fc chain and the second Fc chain and the antigen-binding functional region linked thereto tend to form a heterodimer. When the first Fc chain and the second Fc chain, or the first Fc chain and the second Fc chain and the antigen-binding functional region linked thereto are separately expressed in two host cells, the first Fc chain or the first Fc chain and the antigen-binding functional region linked thereto do not tend to form homodimers, and the second Fc chain or the second Fc chain and the antigen-binding functional region linked thereto do not tend to form homodimers.

[0064] The Kabat EU index numbering system refers to the assignment of numbers to each amino acid in an antibody sequence based on the method proposed by Kabat, which has become a standard method in the art for assigning numbers to each residue. Kabat's scheme can be extended beyond his work to other antibodies. Based on conserved amino acids, the antibody of interest is matched to one of the consensus sequences identified by Kabat.

[0065] Fc domain refers to the fragment crystallizable region (Fc) corresponding to the CH2 and CH3 domains of Ig, which is the portion of Ig that interacts with effector molecules or cells.

[0066] IgG, short for Immunoglobulin G (IgG), is the major antibody component in serum. Human IgG is classified into four subclasses, IgG1, IgG2, IgG3, and IgG4, based on antigenic differences in the r chain of the IgG molecule.

[0067] A half antibody molecule is a structure formed by one heavy chain and one light chain of an antibody, which may or may not be linked via a covalent bond, and is a monovalent antibody structure that recognizes an antigen.

[0068] Fab fragment is an antigen binding fragment (Fab) and is a molecular recognition sequence corresponding to the two arms of an antibody molecule, which is composed of an intact light chain and the VH and CHI domains of the heavy chain. scFv is a molecular recognition sequence and is a structural variant of an antibody fragment obtained by genetically modifying the light chain variable region and the heavy chain variable region of an antibody. The extracellular region of a membrane receptor is a molecular recognition sequence. A membrane receptor usually includes an extracellular region located outside the cell that can recognize and bind to the corresponding antigen or ligand, a transmembrane region that anchors the receptor to the surface of the cell, and an intracellular region located inside the cell that has kinase activity or can transmit a signal. The ligand of a cell membrane receptor refers to a protein, small peptide, or compound that is recognized and bound by the extracellular region of the membrane receptor. Cytokines are low molecular weight soluble proteins produced by various cells induced by immunogens, mitogens, or other stimuli, and have various functions such as innate immunity, adaptive immunity, hematopoiesis, cell growth and regulation of adult pluripotent stem cells (APSCs), and repair of damaged tissues. Cytokines are classified into interleukins, interferons, tumor necrosis factor superfamily, colony stimulating factors, chemokines, growth factors, etc. Protein expression tags refer to amino acid sequences, either small peptides or long amino acids, that are added to the N- or C-terminus of a protein of interest. Addition of tags can be advantageous for correct folding of proteins, isolation and purification of proteins, and reduction of intracellular degradation of proteins. Commonly used tags include, but are not limited to, HA, SUMO, His, GST, GFP, and Flag.

[0069] The heterodimeric bispecific antibodies of the present invention may have one or more substitutions, deletions, additions, and / or insertions. For example, some amino acids can be substituted with other amino acids in the protein structure without significant loss of binding ability to other polypeptides (e.g., antigens) or cells. Since the biological functional activity of a protein is determined by the binding ability and the characteristics of the protein, some amino acid sequence substitutions can be made to the protein sequence without significant loss of their biological effect or activity.

[0070] Often, a polypeptide variant will contain one or more conservative substitutions, which refers to the replacement of an amino acid with another amino acid having similar characteristics, such that one skilled in the art of peptide chemistry would expect the secondary structure and hydrophilicity of the polypeptide to remain substantially unchanged.

[0071] Amino acid substitutions are typically based on the relative similarity of the amino acid side chain substituents, e.g., their hydrophobicity, hydrophilicity, charge, size, etc. Exemplary substitutions taking into account various such characteristics are known to those of skill in the art and include arginine and lysine, glutamic acid and aspartic acid, serine and threonine, glutamine and asparagine, as well as valine, leucine, and isoleucine.

[0072] The term "identity" as used in the present invention has the meaning commonly known in the art, and the rules and standards for measuring identity between different sequences are also known to those skilled in the art, and refers to the percentage of identical residues between a polynucleotide or polypeptide sequence variant and a non-variant sequence after aligning the sequences and introducing gaps (if necessary to achieve the maximum homology percentage). In the present invention, if the identity limitation is satisfied, the resulting variant sequence is also required to have the biological activity possessed by the parent sequence. Methods and means for screening variant sequences by the above-mentioned activities are known to those skilled in the art. Those skilled in the art can easily obtain such variant sequences from the teachings disclosed herein. In specific embodiments, polynucleotide and polypeptide variants have at least about 70%, at least about 75%, at least about 80%, at least about 90%, at least about 95%, at least about 98%, or at least about 99%, or at least about 99.1%, 99.2%, 99.3%, 99.4%, 99.5%, 99.6%, 99.7%, 99.8% or 99.9% polynucleotide or polypeptide identity to the polynucleotides or polypeptides described herein. Due to redundancy in the genetic code, variants exist that encode the same amino acid sequences as these sequences.

[0073] In another embodiment of the present invention, a polynucleotide composition is provided that can hybridize to the polynucleotide sequence provided by the present invention, or a fragment thereof, or a complementary sequence thereof, under moderate to high stringency conditions. Hybridization techniques are known in the field of molecular biology. For illustrative purposes, moderately stringent conditions suitable for testing hybridization of the polynucleotide of the present invention with other polynucleotides include pre-washing with a solution of 5xSSC, 0.5% SDS, 1.0 mM EDTA (pH 8.0), hybridizing overnight under 5xSSC conditions at 50°C to 60°C, and washing twice for 20 minutes with 2x, 0.5x, and 0.2xSSC containing 0.1% SDS at 65°C. Those skilled in the art will appreciate that the stringency of hybridization can be easily manipulated, for example, by changing the salt content of the hybridization solution and / or the hybridization temperature. For example, in another embodiment, suitable highly stringent hybridization conditions include those described above, except that the hybridization temperature is increased, for example, to 60-65°C or 65-70°C.

[0074] The host cell of the present invention may be any cell for expressing a foreign gene, including, but not limited to, E. coli, yeast, insect cells, plant cells, and mammalian cells.

[0075] Vectors of the invention include vectors capable of replicating in any type of cell or organism, including, for example, plasmids, phages, cosmids, and minichromosomes. In some embodiments, a vector comprising a polynucleotide of the invention is a vector suitable for propagation or replication of the polynucleotide, or a vector suitable for expression of a polypeptide of the invention. Such vectors are known in the art and commercially available.

[0076] "Vector" includes shuttle vectors and expression vectors. Typically, a plasmid construct further comprises an origin of replication (e.g., CoEl origin of replication) for plasmid replication and selection in bacteria, and a selection marker (e.g., ampicillin or tetracycline resistance). "Expression vector" refers to a vector that contains the control sequences or regulatory elements necessary to express the antibodies of the invention, including antibody fragments, in bacteria or eukaryotic cells.

[0077] The vector of the present invention may be any vector for expressing a foreign gene, including, but not limited to, a plasmid vector, where the plasmid vector includes at least a replication origin, a promoter, a gene of interest, a multicloning site, and a selection marker gene, and preferably, the vector of the present invention includes, but is not limited to, a plasmid vector obtained by modification based on pCDNA, such as, for example, the X0GC vector.

[0078] Subjects of the present invention include birds, reptiles, mammals, etc. Preferably, mammals include rodents, primates, and more preferably, primates include humans.

[0079] The range of diseases according to the present invention includes, but is not limited to, tumors.Preferably, the tumors include leukemia, lymphoma, myeloma, brain tumor, head and neck squamous cell carcinoma, non-small cell lung cancer, nasopharyngeal carcinoma, esophageal cancer, gastric cancer, pancreatic adenocarcinoma, gallbladder cancer, liver cancer, colorectal cancer, breast cancer, ovarian cancer, cervical cancer, endometrial cancer, uterine sarcoma, prostate cancer, bladder cancer, renal cell carcinoma, melanoma, small cell lung cancer, and bone cancer.

[0080] The pharmaceutically acceptable carrier refers to the common pharmaceutical carrier in the pharmaceutical field, for example, diluent, excipient, water, etc., filler such as starch, sucrose, lactose, microcrystalline cellulose, binder such as cellulose derivative, alginate, gelatin and polyvinylpyrrolidone, wetting agent such as glycerin, disintegrating agent such as sodium carboxymethyl starch, hydroxypropyl cellulose, crosslinked carboxymethyl cellulose, agar, calcium carbonate and sodium bicarbonate, absorption promoter such as quaternary ammonium compound, surfactant such as hexadecanol, sodium dodecyl sulfate, adsorption carrier such as kaolinite, bentonite, lubricant such as talc powder, calcium stearate and magnesium stearate, finely powdered silica gel and polyethylene glycol. In addition, other adjuvants such as flavorings and sweeteners may be added to the composition.

[0081] The present invention will be further described below with reference to the following non-limiting examples. Those skilled in the art will recognize that various modifications can be made to the present invention without departing from the gist of the present invention. Such modifications are also included within the scope of the present invention.

[0082] The following experimental methods are all conventional methods unless otherwise stated. The experimental materials used are readily available from commercial companies unless otherwise stated. The antibodies used in the following examples of the present invention are all commercially available standard antibodies.

[0083] Working Example Example 1 Vector construction of anti-PD-L1 / anti-CD47 heterodimer antibody molecule The X0GC expression vectors containing the light and heavy chains of anti-human PD-L1 antibodies, respectively, were constructed (derived from patent WO2008 / 048037). The nucleotide sequence of the light chain variable region is shown in SEQ ID NO:1, and the amino acid sequence is shown in SEQ ID NO.2. The nucleotide sequence of the light chain constant region is shown in SEQ ID NO:3, and the amino acid sequence is shown in SEQ ID NO.4. The nucleotide sequence of the heavy chain variable region is shown in SEQ ID NO:5, and the amino acid sequence is shown in SEQ ID NO.6. The nucleotide sequence of the heavy chain constant region is shown in SEQ ID NO.7 or SEQ ID.9, and the amino acid sequence is shown in SEQ ID NO.8 or SEQ ID NO.10. A fragment containing an enzyme cleavage site, a signal peptide sequence, and an antibody PD-L1 light chain variable region sequence was synthesized at GENEWIZ Co., Ltd., and a vector containing the synthesized target fragment and an X0GC vector ligated with a light chain constant region sequence were subjected to double enzyme digestion (Double Digests) with BamHI (Thermo, Catalog No. FD0054) and Pfl23II (Thermo, Catalog No. FD0854). The enzyme cleavage reaction system consisted of 6μL of 10x buffer, 2μL each of BamHI and Pfl23II, 40μL of vector, and 10μL of H2O, and the enzyme cleavage reaction system was reacted at 37℃ for 1 hour. The enzyme cleavage product was ligated with T4 DNA ligase (Thermo, Catalog No. EL0011) in a reaction system containing 1 μL of 10× ligase buffer, 0.5 μL of ligase, 7.5 μL of the light chain variable region fragment recovered from the gel, and 1 μL of the X0GC vector ligated with the light chain constant region sequence recovered from the gel, and the ligation was carried out for 30 minutes at 22° C. The ligation product was transformed into E. coli DH5α competent cells (Tiangen, CB104, the same applies below).A fragment containing an enzyme cleavage site, a signal peptide sequence, and an antibody PD-L1 heavy chain variable region sequence was synthesized by Kin Wei Ji Co., Ltd., and a vector containing the synthesized target fragment and an X0GC vector ligated with a heavy chain constant region sequence were subjected to double enzyme cleavage with HindIII (Thermo, Catalog No. FD0505) and NheI (Thermo, Catalog No. FD0973). The enzyme cleavage reaction system consisted of 6 μL of 10x buffer, 2 μL each of HindIII and NheI, 40 μL of vector, and 10 μL of H2O, and the enzyme cleavage reaction system was reacted at 37°C for 1 hour. The enzyme cleavage products were ligated with T4 DNA ligase (Thermo, Cat. No. EL0011) in a reaction system containing 1μL of 10x ligase buffer, 0.5μL of ligase, 7.5μL of heavy chain variable region fragment recovered from the gel, and 1μL of X0GC vector ligated with the heavy chain constant region recovered from the gel, and the ligation reaction was carried out at 22℃ for 30 minutes. The ligation products were transformed into E.coli DH5α competent cells (Tiangen, CB104). X0GC expression vectors of the heavy and light chains of anti-human PD-L1 antibody were obtained for expressing the heavy and light chains of anti-human PD-L1 antibody in eukaryotic cells, respectively.

[0084] The present invention simultaneously constructed X0GC expression vectors containing the heavy and light chain variable region sequences of anti-human CD47 mouse-derived antibody CD47-18-5, respectively. The nucleotide sequence of the light chain variable region is shown in SEQ ID NO.11, and the amino acid sequence is shown in SEQ ID NO.12. The nucleotide sequence of the light chain constant region is shown in SEQ ID NO.3, and the amino acid sequence is shown in SEQ ID NO.4. The nucleotide sequence of the heavy chain variable region is shown in SEQ ID NO.13, and the amino acid sequence is shown in SEQ ID NO:14. The nucleotide sequence of the heavy chain constant region is shown in SEQ ID NO.15, and the amino acid sequence is shown in SEQ ID NO:16. The X0GC expression vectors of the heavy and light chains of anti-human CD47 antibody were obtained to express the heavy and light chains of anti-human CD47 antibody CD47-18-5 in eukaryotic cells, respectively.

[0085] The present invention also simultaneously constructed X0GC expression vectors containing the heavy and light chain variable region sequences of anti-human CD47 mouse-derived antibody CD47-18-10, respectively. The nucleotide sequence of the light chain variable region is shown in SEQ ID NO.17, and the amino acid sequence is shown in SEQ ID NO.18. The nucleotide sequence of the light chain constant region is shown in SEQ ID NO:3, and the amino acid sequence is shown in SEQ ID NO.4. The nucleotide sequence of the heavy chain variable region is shown in SEQ ID NO.19, and the amino acid sequence is shown in SEQ ID NO.20. The nucleotide sequence of the heavy chain constant region is shown in SEQ ID NO.15, and the amino acid sequence is shown in SEQ ID NO.16. The X0GC expression vectors of the heavy and light chains of anti-human CD47 antibody were obtained to express the heavy and light chains of anti-human CD47 antibody CD47-18-10 in eukaryotic cells, respectively.

[0086] The present invention simultaneously constructed X0GC expression vectors containing the heavy and light chain variable regions of the humanized sequence z3 of the anti-human CD47 antibody CD47-18-5, respectively. The nucleotide sequence of the light chain variable region is shown in SEQ ID NO.21, and the amino acid sequence is shown in SEQ ID NO.22. The nucleotide sequence of the light chain constant region is shown in SEQ ID NO:3, and the amino acid sequence is shown in SEQ ID NO.4, the nucleotide sequence of the heavy chain variable region is shown in SEQ ID NO.23, and the amino acid sequence is shown in SEQ ID NO.24. The nucleotide sequence of the heavy chain constant region is shown in SEQ ID NO.15, and the amino acid sequence is shown in SEQ ID NO.16. The X0GC expression vectors of the heavy and light chains of the anti-human CD47 antibody were obtained to express the heavy and light chains of the humanized sequence z3 of the anti-human CD47 antibody CD47-18-5 in eukaryotic cells, respectively.

[0087] The present invention simultaneously constructed X0GC expression vectors containing the heavy and light chain variable regions of humanized sequence z4 of anti-human CD47 antibody CD47-18-5, respectively. The nucleotide sequence of the light chain variable region is shown in SEQ ID NO.25, and the amino acid sequence is shown in SEQ ID NO.26. The nucleotide sequence of the light chain constant region is shown in SEQ ID NO:3, and the amino acid sequence is shown in SEQ ID NO.4. The nucleotide sequence of the heavy chain variable region is shown in SEQ ID NO.27, and the amino acid sequence is shown in SEQ ID NO.28. The nucleotide sequence of the heavy chain constant region is shown in SEQ ID NO.15, and the amino acid sequence is shown in SEQ ID NO.16. The X0GC expression vectors of the heavy and light chains of anti-human CD47 antibody were obtained to express the heavy and light chains of humanized sequence z4 of anti-human CD47 antibody CD47-18-5 in eukaryotic cells, respectively.

[0088] The present invention simultaneously constructed X0GC expression vectors containing the heavy and light chain variable regions of the humanized sequence z7 of the anti-human CD47 antibody CD47-18-5, respectively. The nucleotide sequence of the light chain variable region is shown in SEQ ID NO.29, and the amino acid sequence is shown in SEQ ID NO.30. The nucleotide sequence of the light chain constant region is shown in SEQ ID NO:3, and the amino acid sequence is shown in SEQ ID NO.4. The nucleotide sequence of the heavy chain variable region is shown in SEQ ID NO.31, and the amino acid sequence is shown in SEQ ID NO.32. The nucleotide sequence of the heavy chain constant region is shown in SEQ ID NO.15, and the amino acid sequence is shown in SEQ ID NO.16. The X0GC expression vectors of the heavy and light chains of the anti-human CD47 antibody were obtained to express the heavy and light chains of the humanized sequence z7 of the anti-human CD47 antibody CD47-18-5 in eukaryotic cells, respectively.

[0089] The present invention simultaneously constructed X0GC expression vectors containing the heavy and light chain variable regions of the humanized sequence z11 of the anti-human CD47 antibody CD47-18-5, respectively. The nucleotide sequence of the light chain variable region is shown in SEQ ID NO.33, and the amino acid sequence is shown in SEQ ID NO.34. The nucleotide sequence of the light chain constant region is shown in SEQ ID NO:3, and the amino acid sequence is shown in SEQ ID NO.4. The nucleotide sequence of the heavy chain variable region is shown in SEQ ID NO.35, and the amino acid sequence is shown in SEQ ID NO.36. The nucleotide sequence of the heavy chain constant region is shown in SEQ ID NO.15, and the amino acid sequence is shown in SEQ ID NO.16. The X0GC expression vectors of the heavy and light chains of the anti-human CD47 antibody were obtained to express the heavy and light chains of the humanized sequence z11 of the anti-human CD47 antibody CD47-18-5 in eukaryotic cells, respectively.

[0090] Example 2 Expression of anti-PD-L1 / anti-CD47 heterodimeric antibody molecules Expression vectors containing the heavy and light chains of anti-human PD-L1 antibodies were co-transfected into ExpiCHO cells (ExpiCHOTM cells, Cat. No. A29127, Invitrogen), respectively. Various expression vectors containing the heavy and light chains of anti-human CD47 antibodies were also co-transfected into ExpiCHO cells, respectively.

[0091] The day before transfection, cells were diluted to 3.5 × 10 6 On the day of transfection, cells were seeded at an inoculation density of 6 × 10 cells / mL in fresh ExpiCHO expression medium (ExpiCHOTM Expression Medium, Cat. No. A29100-01, Invitrogen). 6The cells were diluted at a dilution density of 1000 cells / mL. The plasmid was taken out according to the transfection volume, and the final concentration of the plasmid was 0.5 μg / mL. The plasmid was diluted with OptiPROTMSFM medium (OptiPROTMSFM, Catalog No. 12309-019, Invitrogen) to 4% of the transfection volume, and mixed by inversion. A 6.4-fold amount of the plasmid ExpiFectamineTM transfection reagent (ExpiFectamineTM CHO Transfection Kit, Catalog No. A29129, Invitrogen) was taken out, and the transfection reagent was diluted with OptiPROTMSFM medium to 4% of the transfection volume, and mixed by inversion. The diluted transfection reagent was added to the diluted plasmid, mixed gently, left at room temperature for 1 to 5 min, and gradually added dropwise to the cells. Then, it was placed in a cell incubator (CO2 concentration was 8%) and incubated on a shaker at 120 rpm at 37 °C for 20 hours. 0.006x transfection volume of ExpiCHOTM Enhancer (ExpiFectamineTM CHO Transfection Kit, Catalog No. A29129, Invitrogen) and 0.24x transfection volume of ExpiCHOTM Feed (ExpiCHOTM Feed, Catalog No. A29101-02, Invitrogen) were slowly added dropwise to the cells. They were incubated at 32°C in a shaker at 120 rpm. The 10-day transfected cell culture supernatant was collected by centrifugation.

[0092] The expression level was measured by ELISA. Before purification by chromatography column, the precipitate was removed by filtration through a 0.2 μm filter membrane. This step was performed at 4°C.

[0093] Example 3 Purification of anti-PD-L1 / anti-CD47 heterodimer antibody molecule expression product Purification was performed at 4°C using an AKTA explorer 100 protein purification system (GE Healthcare) and an affinity chromatography column MabSelect SuRe (16mmI.D., 22mL, GE Healthcare). First, the chromatography column was equilibrated with mobile phase A (20mM sodium phosphate buffer, pH 7.4), and after stabilizing the baseline, the supernatant of the cells treated as described above was loaded, and after loading, mobile phase A was used for equilibration. The samples were anti-PD-L1 expression products and anti-CD47 expression products, respectively. The column was then washed with 5 column volumes of mobile phase B (mobile phase A containing 1M sodium chloride), then washed with 2 column volumes of mobile phase A, and then eluted with 5 column volumes of mobile phase C (100mM glycine, pH 3.3), and the elution peak, i.e., the peak of the target protein, was collected. The flow rate in the above elution steps was 5mL / min in all cases. The chromatogram of the elution peak of the anti-PD-L1 expression product is shown in Figure 1, and the elution peak of the anti-CD47 expression product (consisting of the heavy and light chain sequences of humanized sequence z11 of antibody CD47-18-5) is shown in Figure 2. The elution peaks of the other anti-CD47 expression products are similar. The indicated elution peak (grey area shown in the figure) was collected and the pH was adjusted to 7.2 by adding 1 M Tris alkaline solution dropwise, and then sterile filtered and stored at 2-8°C.

[0094] Example 4. Preparation and purification of anti-PD-L1 / anti-CD47 heterodimeric antibody molecules The structure of the anti-PD-L1 / anti-CD47 heterodimer antibody molecule is shown in Figure 3.

[0095] The products purified by the above MabSelect SuRe (16mmI.D., 22mL, GE Healthcare) method were subjected to in vitro recombination to obtain heterodimers. First, in the process of reducing the protein solution purified and collected as described above with DTT, disulfide bonds were cleaved, and disulfide bonds in the hinge regions of the homodimeric antibody molecules contained in the anti-PD-L1 and anti-CD47 products were also cleaved to form half antibody molecules containing one heavy chain and one light chain, the structures of which are shown in Figure 4. The reduced samples were analyzed by SEC-HPLC (TOSOH, TSKgel superSW3000) with 1mM DTT reducing agent in the mobile phase buffer, and the results are shown in Figure 5A and B, respectively, where the proportions of homodimeric molecules of anti-PD-L1 and anti-CD47 were both less than 20%, and the proportions of half antibody molecules were both more than 80%.

[0096] The reduced anti-PD-L1 and anti-CD47 half antibody molecules were then mixed in an equal molar ratio and the recombination reaction was carried out at 4℃ for 24 hours; during the recombination process, a heterodimeric bispecific antibody containing both anti-PD-L1 and anti-CD47 half antibody molecules was formed from the anti-PD-L1 and anti-CD47 half antibody molecules via the non-covalent interactions between CH2-CH3; the protein solution was then concentrated by ultrafiltration through an ultrafiltration concentration tube (nominal molecular weight cut-off of 10 kDa), the solution was replaced with phosphate solution (PBS, pH=7.4) to stop the reduction, and an oxidation reaction was carried out with air or an oxidizing agent to reform the disulfide bonds of the heterodimeric bispecific antibody.

[0097] The anti-PD-L1 / anti-CD47 heterodimer antibody molecules obtained by reducing and oxidizing the above anti-PD-L1 and anti-CD47 expression products were concentrated by ultrafiltration through an ultrafiltration concentration tube (nominal cut-off molecular weight of 10 kDa), and the solution was replaced with 10 mM sodium phosphate buffer (pH 6.0). Purification was performed at room temperature using an AKTA explorer 100 protein purification system (GE Healthcare) and an ion chromatography column Source 15S (16 mm I.D., 22 mL, GE Healthcare). First, the chromatography column was equilibrated with mobile phase A (10 mM sodium phosphate, pH 7.0) and the baseline was stabilized, and then the protein solution treated as described above was loaded at a flow rate of 5 mL / min, and after loading, mobile phase A was used for equilibration. The column was then washed with a 20 column volume gradient of A (10 mM sodium phosphate, pH 5.8) to B (10 mM sodium phosphate, pH 5.8) (0% B to 100% B, 170 min, flow rate 2 mL / min). The main elution peak was collected and the collected protein solution was concentrated by ultrafiltration through an ultrafiltration concentration tube (nominal molecular weight cut-off of 10 kDa), the solution was replaced with phosphate solution (PBS, pH=7.4), sterilized by filtration, and stored at 4°C. The purified anti-PD-L1 / anti-CD47 heterodimer antibody molecule BH3012h (in which the anti-CD47 moiety contains the heavy and light chain variable region sequences of the humanized sequence z11 of the antibody CD47-18-5) was subjected to purity analysis by SEC-HPLC, the result of which is shown in Figure 6, the purity was 97.3%. CE analysis was performed, and the results are shown in FIG. 7. The purity was 95.3%.Other anti-PD-L1 / anti-CD47 heterodimer antibody molecules, such as BH3014 (in which the anti-CD47 portion comprises the heavy and light chain variable region sequences of humanized sequence z3 of antibody CD47-18-5), BH3015 (in which the anti-CD47 portion comprises the heavy and light chain variable region sequences of humanized sequence z4 of antibody CD47-18-5), BH3016 (in which the anti-CD47 portion comprises the heavy and light chain variable region sequences of humanized sequence z7 of antibody CD47-18-5), The results for the anti-PD-L1 / anti-CD47 heterodimer bibody 18-5 (wherein the anti-CD47 moiety contains the heavy and light chain variable region sequences of murine antibody CD47-18-5) and the anti-PD-L1 / anti-CD47 heterodimer bibody 18-10 (wherein the anti-CD47 moiety contains the heavy and light chain variable region sequences of murine antibody CD47-18-10) were all similar.

[0098] Example 5 Target binding activity of anti-PD-L1 / anti-CD47 heterodimer antibodies The binding ability of anti-PD-L1 / anti-CD47 heterodimer antibodies to single antigens was measured by enzyme-linked immunosorbent assay (ELISA). The specific procedure was as follows: Recombinant human PD-L1 (Beijing Sino Biological Co., Ltd., Catalog No. 10084-H08H) or human CD47 (Acro Biosystems Co., Ltd. (Beijing), Catalog No. CD7-H5227) was coated onto a 96-well high-binding ELISA plate (Costar, Catalog No. 42592) using carbonate buffer solution at pH=9.6 at a coating concentration of 1μg / mL at 100μL / well. Coating was performed overnight at 4℃. Washed five times with PBST. Blocked with PBST containing 1% BSA at 300μL / well and incubated at 25℃ for 1 hour. Washed five times with PBST. Heterodimer antibody samples and controls serially diluted in PBST containing 1% BSA were added at 100 μL / well and incubated for 1 hour at 25°C. The plates were washed five times with PBST. Horseradish peroxidase-labeled anti-human IgG antibody (Chemicon, Cat. No. AP309P) diluted 1:10000 in PBST containing 1% BSA was then added at 100 μL / well and incubated for 1 hour at 25°C. The plates were washed five times with PBST. The color substrate TMB was added at 100 μL / well and allowed to develop for 10 minutes at room temperature. Color development was stopped by adding 100 μL / well of 1M H2SO4. The absorbance at 450 nm was read on a microplate reader.

[0099] The results are shown in Figure 8A, which shows that the anti-PD-L1 / anti-CD47 heterodimer antibodies BH3012h, BH3014, BH3015, and BH3016 all have high affinity for PD-L1 and comparable activity, which is slightly weaker than the antigen binding activity of PD-L1 bivalent monoclonal antibodies. As shown in Figure 8B, the anti-PD-L1 / anti-CD47 heterodimer antibodies BH3012h, BH3014, BH3015, and BH3016 have different affinities for CD47, with BH3012h having the strongest binding activity, followed by BH3016, then BH3015, and BH3014 having the weakest binding activity.

[0100] Example 6 Blocking activity of anti-PD-L1 / anti-CD47 heterodimer antibodies against ligand-receptor binding The detection process for blocking activity of PD-L1 binding to PD-1 and CD80 was as follows: Recombinant human PD-L1-Fc (Acro Biosystems Co., Ltd., Beijing, Cat. No. PD1-H5258) was coated onto a 96-well high-binding ELISA plate using carbonate buffer solution, pH=9.6, at a coating concentration of 1μg / mL with a coating volume of 100μL / well. Coating was performed overnight at 4℃. Washed five times with PBST. Blocked with PBST containing 1% BSA at 300μL / well and incubated at 25℃ for 1 hour. Washed five times with PBST. 50μL / well of the heterodimer antibody samples and controls serially diluted in PBST containing 1% BSA were added, and 50μL of biotin-labeled PD-1-Fc (Beijing Hanmi Pharm) was added to a concentration of 40nM (final concentration of 20nM) or 50μL of biotin-labeled CD80-Fc (Acro Biosystems Co., Ltd., Beijing, Catalog No. B71-H82F2) was added to a concentration of 100nM (final concentration of 50nM) and incubated at 25℃ for 90 minutes. Washed 5 times with PBST. 100μL / well of Streptavidin-HRP (BD Pharmingen, Catalog No. 554066) diluted 1:1000 in PBST containing 1% BSA was then added, and incubated at 25℃ for 1 hour. Washed 5 times with PBST. The colorimetric substrate TMB was added at 100 μL / well and allowed to develop for 10 minutes at room temperature. Color development was stopped by adding 100 μL / well of 1M H2SO4. The absorbance at 450 nm was read on a microplate reader.

[0101] The detection process for blocking activity of CD47 binding to SIRPα was as follows: Recombinant human CD47-Fc (CD7-H52A5, Acro Biosystems Co., Ltd., Beijing) was coated on a 96-well high-adsorption ELISA plate at a coating concentration of 1 μg / mL with a coating volume of 100 μL / well using carbonate buffer solution at pH=9.6. Coating was performed overnight at 4°C. Washed five times with PBST. Blocking was performed with PBST containing 1% BSA at 300 μL / well and incubated at 25°C for 1 hour. Washed five times with PBST. Heterodimeric antibody samples and controls serially diluted in PBST containing 1% BSA were added at 50 μL / well, and biotin-labeled SIRPα (Acro Biosystems Co., Ltd., Beijing, CDA-H82F2) was added at 50 μL / well to a concentration of 4 nM (final concentration 2 nM) and incubated at 25°C for 90 min. Washed five times with PBST. Then, Streptavidin-HRP (BD Pharmingen, Cat. No. 554066) diluted 1:1000 in PBST containing 1% BSA was added at 100 μL / well and incubated at 25°C for 1 h. Washed five times with PBST. Chromogenic substrate TMB was added at 100 μL / well and allowed to develop at room temperature for 10 min. Color development was stopped by adding 100 μL / well of 1 M H2SO4. The absorbance at 450 nm was read on a microplate reader.

[0102] The results are shown in Figure 9A, anti-PD-L1 / anti-CD47 heterodimer antibodies BH3012h, BH3014, BH3015, BH3016 can all block the binding of PD-L1 to PD-1 with comparable activity and slightly weaker than PD-L1 bivalent monoclonal antibodies. As shown in Figure 9B, anti-PD-L1 / anti-CD47 heterodimer antibodies BH3012h, BH3014, BH3015, BH3016 can all block the binding of PD-L1 to CD80 with comparable activity. As shown in Figure 9C, the anti-PD-L1 / anti-CD47 heterodimer antibodies BH3012h, BH3014, BH3015, and BH3016 have different blocking activities against the binding of CD47 to SIRPα, with BH3012h being the strongest, followed by BH3016, then BH3015, and BH3014 being the weakest.

[0103] Example 7 Binding of anti-PD-L1 / anti-CD47 heterodimer antibodies to tumor cells / red blood cells HCC827 tumor cells express PD-L1 and CD47, and red blood cells (RBCs) express only CD47. HCC827 and RBCs were mixed, and flow cytometry (FCM) was used to detect whether there was selectivity in the binding of the heterodimer to the two types of cells in the mixed cells.

[0104] The specific steps of this method are as follows: HCC827 cells (purchased from ATCC) and RBC cells (obtained from healthy individuals) were collected. They were washed once with cold PBS (GIBCO, Catalog No. 14190-235) containing 2% FBS (Hyclone, Catalog No. SH30084.03). 1 × 10 6 HCC827 cells and 10 x 10 per tube 6The cells were mixed with RBC cells, resuspended in 200 μL of cold DPBS (GIBCO, Cat. No. 14190-136) containing 2% FBS, and serially diluted heterodimer antibody samples and controls were added. The flow cytometry tubes were incubated on ice for 30 minutes. The tubes were washed twice with PBS containing 2% FBS. The cells were resuspended in 200 μL of cold PBS containing 2% FBS and FITC-labeled anti-human IgG antibody (Beijing Zhongsugi Jinqiao, Cat. No. ZF0306) diluted 1:1000. The cells were incubated on ice for 30 minutes in the dark. The tubes were washed twice with PBS containing 2% FBS. The cells were resuspended in 500 μL of cold PBS, and the cell suspension was detected and analyzed by flow cytometry (BD, FACS Calibur) to read the fluorescence intensity of each of the two types of cells in the mixed cells.

[0105] The results are shown in Figure 10, which shows that there was little difference in binding between HCC827 and RBCs mixed with a CD47 monoclonal antibody (containing the heavy and light chain variable region sequences of humanized sequence z11 of antibody CD47-18-5 and a human IgG4 constant region sequence) (Figure A). Meanwhile, the anti-PD-L1 / anti-CD47 heterodimer bound to HCC827 cells co-expressing PD-L1 and CD47, but showed weak or no binding to RBCs expressing only CD47, demonstrating binding selectivity (Figures B and C).

[0106] Example 8. Regulatory activity of anti-PD-L1 / anti-CD47 heterodimer antibodies on T cells The modulatory activity of anti-PD-L1 / anti-CD47 heterodimer antibodies on T cell immune responses was measured by mixed lymphocyte reaction (MLR).

[0107] Obtaining human dendritic cells (DCs): Human PBMC cells (Lonza, Cat. No. CC-2702) were resuscitated and collected. 5 × 10 human PBMC cells were 6The cells were resuspended in serum-free RPMI 1640 medium (GIBCO, Catalog No. 22400-089) at a cell density of 100 / mL, inoculated into cell culture flasks, and incubated at 37° C. in a carbon dioxide incubator for 90 minutes. The culture supernatant and suspended cells were discarded, and the adherent cells were incubated in complete medium (RPMI 1640 containing 10% FBS) and supplemented with 100 ng / mL GM-CSF (Beijing Sino Biological Co., Ltd., Catalog No. 10015-HNAH) and 100 ng / mL IL-4 (Beijing Sino Biological Co., Ltd., Catalog No. 11846-HNAE). After incubation for 3 days, the culture medium was replaced and incubated for another 3 days. The medium was then replaced with complete medium (RPMI 1640 containing 10% FBS) containing 100 ng / mL GM-CSF, 100 ng / mL IL-4, and 20 ng / mL TNF-α, and incubated for 1 day to obtain DC cells.

[0108] Obtaining human T cells: Human PBMC cells were resuscitated and collected to ensure that the PBMCs and the PBMCs from which DC cells were derived were derived from different individuals. Human T cells were isolated by following the instructions of the Pan T cell isolation kit (Miltenyi Biotech, Cat. No. 5150414820). Briefly, first, PBMCs were washed once with PBS, and then PBMCs were diluted with 10 7 Resuspend the cells in 40 μL of isolation buffer (PBS containing 2 mM EDTA and 0.5% BSA, pH = 7.2) (the amounts used below are all 10 7 10 μL of Pan T cell Biotin Antibody Cocktail (calculated in cells) and 10 μL of Pan T cell Biotin Antibody Cocktail were added and incubated at 4° C. for 5 minutes. 30 μL of isolation buffer and 20 μL of Pan T cell MicroBead Cocktail were added and incubated at 4° C. for 10 minutes. T cells were obtained by passing through a MACS isolation column.

[0109] The collected human DC cells and human T cells were resuspended in complete medium (RPMI 1640 containing 10% FBS) and seeded into a 96-well plate. The seeded DC cells and T cells were each 1 × 10 4 / well, 1 x 10 5 The plates were incubated at 100x for 24 hours at 37°C in a carbon dioxide incubator for 5 days. After incubation, the supernatants in the wells were removed and the cytokine IFN-γ (RayBiotech, catalog number ELH-IFNg) was detected according to the kit manual.

[0110] As shown in Figure 11, human T cells were activated and secreted IFN-γ by stimulation with allogeneic DC cells. The addition of PD-L1 antibody enhanced T cell activation and promoted cytokine secretion. Anti-PD-L1 / anti-CD47 heterodimer antibody also showed strong T cell regulatory activity and significantly promoted the secretion of cytokine IFN-γ.

[0111] Example 9: Anti-PD-L1 / anti-CD47 heterodimer antibody-mediated phagocytosis of tumor cells by macrophages Preparation of mature human macrophages: Human PBMC cells (Lonza, Cat. No. CC-2702) were resuscitated and collected. 5 × 10 human PBMC cells were 6The cells were resuspended in serum-free RPMI 1640 medium at a cell density of 10,000 cells / mL, inoculated into cell culture bottles, and incubated at 37°C in a carbon dioxide incubator for 90 minutes. The culture supernatant and suspended cells were discarded, and then the adherent cells were incubated in complete medium (RPMI 1640 containing 10% FBS) and 25 ng / mL M-CSF (Beijing Sino Biological Co., Ltd., Catalog No. 11792-HNAN) was added and incubated for 7 days. Then, the macrophages were collected and resuspended in complete medium (RPMI 1640 containing 10% FBS) containing 25 ng / mL M-CSF and 50 ng / mL IFN-γ (Beijing Sino Biological Co., Ltd., Catalog No. 11725-HNAS). The cell suspension was inoculated into 48-well cell culture plates at 50,000 cells / well, and incubated for 1 day to prepare the macrophages for maturation.

[0112] Raji tumor cells were stained according to the instruction manual of the CFSE kit (Life technologies, Catalog No. C34554). Briefly, CFSE was diluted with PBS to a working concentration of 5 μM, preheated at 37°C, centrifuged at 1000 rpm for 5 minutes, Raji cells were collected, resuspended in preheated CFSE working solution, and incubated at 37°C for 15 minutes. The cells were washed once with complete medium, resuspended in complete medium, incubated for 30 minutes, washed twice with complete medium, and resuspended in complete medium.

[0113] The 48-well plate was washed three times with complete medium. The Raji cells stained with CFSE, the test heterodimer samples, and the control were pre-incubated for 15 minutes, then added to the 48-well culture plate and incubated for 2 hours at 37°C in a carbon dioxide incubator. After the incubation, the 48-well plate was washed three times with complete medium, and 10 μg / mL wheat germ agglutinin and Alexa Fluor 555 (Life Technologies, Catalog No. W32464) diluted in complete medium were added, and incubated for 15 minutes in the dark. The 48-well plate was further washed three times with complete medium, and fixed with 4% paraformaldehyde for 15 minutes. The 48-well plate was further washed three times with complete medium, and complete medium was added. Photographs were taken under a fluorescent microscope, and the cells were counted. The phagocytosis index (%) was calculated as follows: number of phagocytosed green-labeled Raji cells / number of red-labeled macrophages present x 100.

[0114] As shown in Figure 12, a CD47 monoclonal antibody (comprising the heavy and light chain variable region sequences of the humanized sequence z11 of the antibody CD47-18-5 and a human IgG4 constant region sequence) can mediate the phagocytosis of Raji tumor cells by macrophages, and the anti-PD-L1 / anti-CD47 heterodimer antibodies BH3012h, BH3015, and BH3016 can also mediate the phagocytosis of Raji tumor cells by macrophages.

[0115] Example 10 Pharmacodynamic study of anti-PD-L1 / anti-CD47 heterodimer antibodies in humanized mice against A375 tumors Six to eight week old female immunodeficient NCG mice (Nanjing Institute of Biological Medicine) were used as experimental materials. After the mice were adapted to the environment for one week, human PBMCs were first administered to partially humanize the mouse immune system, and 5 × 10 6 Each mouse was inoculated with 5 × 10 cells. 6 A375 human melanoma cells were inoculated subcutaneously on the right dorsal side. Tumor volumes were approximately 100 mm 3When the tumor volume reached 100%, the tumor-bearing mice were divided into groups of 6 mice each according to the tumor volume. The vehicle (PBS), 35 nmol / kg PD-L1 monoclonal antibody, 35 nmol / kg CD47 monoclonal antibody, and 70 nmol / kg anti-PD-L1 / anti-CD47 heterodimer antibody (considering that the monoclonal antibody is a bivalent antibody, both the anti-PD-L1 and anti-CD47 bispecific antibodies are monovalent) were administered intraperitoneally twice a week for 2 consecutive weeks. The tumor volume was measured 3 times a week from the day of administration, and its long diameter a and short diameter b were measured. The formula for calculating tumor volume was tumor volume (mm 3 )=(a×b 2 ) / 2. The duration of tumor volume measurement was 3 weeks, i.e., one more week was observed after the administration was stopped.

[0116] The results are shown in Figure 13. In xenograft tumor models, anti-PD-L1 / anti-CD47 heterodimer antibodies have stronger anti-tumor effects than PD-L1 monoclonal antibodies and CD47 monoclonal antibodies.

[0117] Example 11 Pharmacodynamic study of anti-MC38 tumors in gene knock-in mice with anti-PD-L1 / anti-CD47 heterodimer antibodies Six to eight week old female B-hPD-L1 / hSIRPα / hCD47 triple humanized mice (Jiangsu Biocytogen Co., Ltd.) were used as experimental materials. After the mice were adapted to the environment for one week, 5 × 10 5 B-hPD-L1 / hCD47 MC38 mouse colon tumor cells (mouse-derived PD-L1 and CD47 knockout, human-derived PD-L1 and CD47 overexpression, MC38 cells from Shanghai Shunran biotech Co., Ltd.) were inoculated subcutaneously into the right dorsal region. Tumor volumes were approximately 100 mm 3When the tumor volume reached 100%, the tumor-bearing mice were divided into groups of 6 mice each according to the tumor volume. Vehicle (PBS), 1 mg / kg PD-L1 monoclonal antibody, 1 mg / kg CD47 monoclonal antibody, and 2 mg / kg anti-PD-L1 / anti-CD47 heterodimer antibody (considering that monoclonal antibodies are bivalent antibodies, both anti-PD-L1 and anti-CD47 of the bispecific antibodies are monovalent) were administered intraperitoneally twice a week for 3 consecutive weeks. The tumor volume was measured twice a week from the day of administration, and its long diameter a and short diameter b were measured. The formula for calculating tumor volume was tumor volume (mm 3 )=(a×b 2 ) / 2. The duration of tumor volume measurement was 3 weeks, i.e., one more week was observed after the administration was stopped.

[0118] The results are shown in Figure 14. In the allogeneic tumor model, the anti-PD-L1 / anti-CD47 heterodimer antibody has a stronger anti-tumor effect than the PD-L1 monoclonal antibody and the CD47 monoclonal antibody.

[0119] Example 12 Pharmacodynamic study of anti-PD-L1 / anti-CD47 heterodimer antibodies against anti-CT26 tumors in wild type mice Six to eight week-old female wild-type Balb / c mice (purchased from Beijing Weitong Lihua) were used as experimental materials. After adapting the mice to the environment for one week, 5 × 10 5 CT26 / hPD-L1 / hCD47 mouse colon tumor cells (purchased from Jiangsu Cancer Center) were inoculated subcutaneously into the right dorsal region. The tumor volume was approximately 100 mm 3When the tumor volume reached 100%, the tumor-bearing mice were divided into groups of 8 mice each according to the tumor volume, and were intraperitoneally administered with vehicle (DPBS, GIBCO, Cat. No. 14190-136), 17.5 nmol / kg anti-PD-L1 monoclonal antibody (monoclonal antibody corresponding to the anti-PD-L1 moiety in the anti-PD-L1 / anti-CD47 heterodimer antibody), 17.5 nmol / kg anti-CD47 monoclonal antibody (monoclonal antibody corresponding to the anti-CD47 moiety in the anti-PD-L1 / anti-CD47 heterodimer antibody), or 35 nmol / kg anti-PD-L1 / anti-CD47 heterodimer antibody (considering that the monoclonal antibody is a bivalent antibody, both anti-PD-L1 and anti-CD47 in the bispecific antibody are monovalent) twice a week for 3 consecutive weeks. From the day of administration to the end of administration, the tumor volume was measured twice a week. After the end of administration, the tumor volume was measured once, and the long diameter a and short diameter b were measured each time. The tumor volume was calculated as follows: tumor volume (mm 3 )=(a×b 2 ) / 2. The duration of tumor volume measurement was 4 weeks, i.e., one additional week was observed after the administration was stopped.

[0120] The results are shown in Figure 15, which shows that in an allogeneic tumor model, the anti-PD-L1 / anti-CD47 heterodimer antibody BH3012h has a stronger anti-tumor effect than the corresponding anti-PD-L1 bivalent monoclonal antibody and anti-CD47 bivalent monoclonal antibody.

Claims

1. A bispecific antibody comprising a first antigen-binding functional region that specifically binds to PD-L1 and a second antigen-binding functional region that specifically binds to CD47, The first antigen-binding functional region that specifically binds to PD-L1 is (A) a heavy chain variable region; (B) a light chain variable region, The heavy chain variable region comprises: (a) an HCDR1 comprising the amino acid sequence set forth in SEQ ID NO. 37; (b) an HCDR2 comprising the amino acid sequence set forth in SEQ ID NO. 38; (c) an HCDR3 comprising the amino acid sequence set forth in SEQ ID NO. 39; The light chain variable region comprises: (a) an LCDR1 comprising the amino acid sequence set forth in SEQ ID NO. 40; (b) an LCDR2 comprising the amino acid sequence set forth in SEQ ID NO. 41; (c) an LCDR3 comprising the amino acid sequence set forth in SEQ ID NO. 42; The second antigen-binding functional region that specifically binds to CD47 is (A) a heavy chain variable region; (B) a light chain variable region, The heavy chain variable region comprises: (a) an HCDR1 comprising the amino acid sequence set forth in SEQ ID NO. 43; (b) an HCDR2 comprising the amino acid sequence set forth in SEQ ID NO. 44; (c) an HCDR3 comprising the amino acid sequence set forth in SEQ ID NO. 45; The light chain variable region comprises: (a) an LCDR1 comprising the amino acid sequence set forth in SEQ ID NO. 46; (b) an LCDR2 comprising the amino acid sequence set forth in SEQ ID NO. 47; (c) an LCDR3 comprising the amino acid sequence set forth in SEQ ID NO. 48; Bispecific antibodies.

2. The bispecific antibody of claim 1, wherein the first antigen-binding functional region that specifically binds to PD-L1 comprises a heavy chain variable region and a light chain variable region, the heavy chain variable region comprises the amino acid sequence shown in SEQ ID NO. 6, and the light chain variable region comprises the amino acid sequence shown in SEQ ID NO.

2.

3. 3. The bispecific antibody according to claim 1 or 2, wherein the second antigen-binding functional region that specifically binds to CD47 comprises a heavy chain variable region and a light chain variable region, the heavy chain variable region comprises an amino acid sequence selected from SEQ ID NO. 14, 20, 24, 28, 32, or 36, and the light chain variable region comprises an amino acid sequence shown in SEQ ID NO. 12, 18, 22, 26, 30, or 34.

4. The first antigen-binding functional region that specifically binds to PD-L1 comprises a heavy chain variable region and a light chain variable region, the heavy chain variable region comprises the amino acid sequence shown in SEQ ID NO. 6, and the light chain variable region comprises the amino acid sequence shown in SEQ ID NO. 2; and, The second antigen-binding functional region that specifically binds to CD47 includes a heavy chain variable region and a light chain variable region selected from the following: A bispecific antibody according to any one of claims 1 to 3. a) a heavy chain variable region comprising the amino acid sequence shown in SEQ ID NO. 14, and a light chain variable region comprising the amino acid sequence shown in SEQ ID NO.

12. b) a heavy chain variable region comprising the amino acid sequence shown in SEQ ID NO. 20, and a light chain variable region comprising the amino acid sequence shown in SEQ ID NO. 18; c) a heavy chain variable region comprising the amino acid sequence shown in SEQ ID NO. 24, and a light chain variable region comprising the amino acid sequence shown in SEQ ID NO.

22. d) a heavy chain variable region comprising the amino acid sequence shown in SEQ ID NO. 28, and a light chain variable region comprising the amino acid sequence shown in SEQ ID NO.

26. e) a heavy chain variable region comprising the amino acid sequence shown in SEQ ID NO. 32, and a light chain variable region comprising the amino acid sequence shown in SEQ ID NO.

30. f) a heavy chain variable region comprising the amino acid sequence shown in SEQ ID NO. 36, and a light chain variable region comprising the amino acid sequence shown in SEQ ID NO. 34;

5. The bispecific antibody according to any one of claims 1 to 4, wherein the first antigen-binding functional region and the second antigen-binding functional region are selected from a Fab fragment, an scFv fragment and a variable domain fragment Fv.

6. The bispecific antibody according to claim 1 , wherein the first antigen-binding functional region and the second antigen-binding functional region are both Fab fragments.

7. 7. The bispecific antibody of claim 6, wherein the Fab fragment comprises a first heavy chain variable region and a second heavy chain variable region that are distinct and a first light chain variable region and a second light chain variable region that are distinct.

8. The bispecific antibody according to any one of claims 1 to 5, wherein one of the first antigen-binding functional region and the second antigen-binding functional region is a Fab fragment and the other is an scFv.

9. 1. A bispecific antibody comprising a first Fc chain and a second Fc chain, the first Fc chain and the second Fc chain are both Fc fragments of immunoglobulin G containing amino acid substitutions, and the first Fc chain and the second Fc chain together constitute a heterodimer capable of binding to an Fc receptor; the first Fc chain and the second Fc chain are linked to the first antigen-binding functional region and the second antigen-binding functional region, respectively, via a covalent bond or a linker; and one of the first Fc chain and the second Fc chain comprises amino acid substitutions of T366L and D399R at positions 366 and 399, and the other comprises amino acid substitutions of L351E, Y407L and K409V at positions 351, 407 and 409, wherein the amino acid positions are numbered according to the Kabat EU index numbering system. A bispecific antibody according to any one of claims 1 to 5.

10. The bispecific antibody according to claim 9, wherein the weight percentage of the first Fc chain and the first antigen-binding functional region linked thereto via a covalent bond, and the second Fc chain and the second antigen-binding functional region linked thereto via a covalent bond, based on all polypeptide chains of the homodimer formed, is less than 50%, in a solution in the presence of a reducing agent, when the solution does not contain any polypeptides other than the first Fc chain and the first antigen-binding functional region linked thereto via a covalent bond, and the second Fc chain and the second antigen-binding functional region linked thereto via a covalent bond.

11. A bispecific antibody comprising a pair of a first heavy chain and a first light chain that specifically binds to PD-L1, The first heavy chain has a heavy chain variable region comprising the amino acid sequence shown in SEQ ID NO. 6 and a heavy chain constant region comprising the amino acid sequence shown in SEQ ID NO. 8 or 10; 11. The bispecific antibody of claim 1, wherein the first light chain has a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 2 and a light chain constant region comprising the amino acid sequence set forth in SEQ ID NO.

4.

12. A bispecific antibody comprising a second heavy chain and a second light chain pair that specifically binds to CD47, The second heavy chain has a heavy chain variable region comprising the amino acid sequence shown in SEQ ID NO. 14, 20, 24, 28, 32 or 36, and a heavy chain constant region comprising the amino acid sequence shown in SEQ ID NO. 16; 12. The bispecific antibody of claim 1, wherein the second light chain has a light chain variable region comprising the amino acid sequence set forth in SEQ ID NO. 12, 18, 22, 26, 30 or 34 and a light chain constant region comprising the amino acid sequence set forth in SEQ ID NO.

4.

13. An isolated polynucleotide encoding a bispecific antibody according to any one of claims 1 to 12.

14. A recombinant expression vector comprising the isolated polynucleotide of claim 13.

15. 15. A host cell comprising the isolated polynucleotide of claim 13 or the recombinant expression vector of claim 14.

16. Human embryonic kidney cells HEK293 or HEK293 cell-based modification HEK293T, HEK293E, HEK293F, hamster ovary cells CHO or CHO cell-based modification CHO-S, CHO-dhfr-, CHO / DG44, ExpiCHO, E. coli or E. coli-based modification E. coli BL21, BL21 (DE3), Rosetta, Origami, yeast or yeast-based modification Pichia pastoris, Saccharomyces cerevisiae, Kluyveromyces lactis, Hansenula polymorpha, insect cells or insect cell-based modification High5, SF9, plant cells, mammalian mammary gland cells, somatic cells, the host cell according to claim 15 selected from the group consisting of.

17. 17. A composition comprising a bispecific antibody according to any one of claims 1 to 12, or an isolated polynucleotide according to claim 13, or a recombinant expression vector according to claim 14, or a host cell according to claim 15 or 16, and a pharma- ceutically acceptable carrier.

18. 1) expressing the isolated polynucleotide of claim 13 or the recombinant expression vector of claim 14 in a host cell; 2) reducing each expressed protein in the host cell; 3) mixing the reduced protein and then oxidizing the mixture; A method for producing a bispecific antibody according to any one of claims 1 to 12.

19. 19. The method according to claim 18, wherein the reduction step comprises: 1) performing a reduction reaction in the presence of a reducing agent selected from 2-mercaptoethylamine, dithiothreitol, or tris(2-carboxyethyl)phosphine; and 2) removing the reducing agent.

20. The method according to claim 18 or 19, wherein the oxidation step is oxidation in air, and further comprises a step of carrying out the oxidation reaction in the presence of an oxidizing agent selected from L-dehydroascorbic acid or a chemical derivative thereof.

21. 21. The method of any one of claims 18 to 20, further comprising an isolation and purification step.

22. 18. Use of a bispecific antibody according to any one of claims 1 to 12, and / or an isolated polynucleotide according to claim 13, and / or a recombinant expression vector according to claim 14, and / or a host cell according to claim 15 or 16, and / or a composition according to claim 17 in the manufacture of a medicament for preventing and / or treating a disease in a subject.

23. A bispecific antibody according to any one of claims 1 to 12, and / or an isolated polynucleotide according to claim 13, and / or a recombinant expression vector according to claim 14, and / or a host cell according to claim 15 or 16, and / or a composition according to claim 17 for use as a medicament for preventing and / or treating a disease in a subject.

24. The subject is a mammal.

24. The use according to claim 22, or the heterodimeric bispecific antibody, isolated polynucleotide, recombinant expression vector, host cell or composition according to claim 23.

25. The subject is a human.

25. The use, heterodimeric bispecific antibody, isolated polynucleotide, recombinant expression vector, host cell or composition according to claim 24.

26. 24. The use of claim 22 or the heterodimeric bispecific antibody, isolated polynucleotide, recombinant expression vector, host cell or composition of claim 23, wherein the disease is selected from leukemia, lymphoma, myeloma, brain tumor, head and neck squamous cell carcinoma, non-small cell lung cancer, nasopharyngeal carcinoma, esophageal cancer, gastric cancer, pancreatic adenocarcinoma, gallbladder cancer, liver cancer, colorectal cancer, breast cancer, ovarian cancer, cervical cancer, endometrial cancer, uterine sarcoma, prostate cancer, bladder cancer, renal cell carcinoma, melanoma, small cell lung cancer, bone cancer.

Images