RU2808138C1 - Cd3-targeting antibody, bispecific antibody and their applications - Google Patents

Abstract

FIELD: biopharmaceuticals.

SUBSTANCE: invention is related to an antibody targeting CD3, a bispecific antibody, their production and use. The CD3-targeting antibody contains the amino acid sequence VH represented by any of SEQ ID NO: 43-55, and the amino acid sequence of VL represented by any of SEQ ID NO: 57-60. The bispecific antibody contains the first functional protein region comprising an antibody targeting CD3 and a second functional protein region. Dedicated nucleic acid, expression vector, genetically modified cell to produce an antibody, pharmaceutical composition and their use in the manufacture of a drug for the treatment of a tumour with CD3 as a target are also proposed.

EFFECT: antibody is able to bind to primate CD3, reduces the toxicity caused by cytokine release syndrome, and the resulting bispecific antibody is stable and has the ability to bind to T cells.

24 cl, 16 dwg, 15 tbl, 8 ex

Description

[1] This application claims priority to Chinese Patent Application No. 2019109413286, filed on September 30, 2019, the contents of which are incorporated herein in their entirety.

Field of technology

[2] The present invention relates to the field of biopharmaceuticals, in particular to a CD3-targeting antibody, a bispecific antibody and their use.

State of the art

[3] T lymphocytes are an important class of cells involved in the adaptive immune response, and T cells recognize antigens through the T cell receptor (TCR). The TCR does not directly recognize antigen surface epitopes, but does specifically recognize antigen-peptide-MHC (pMHC) molecular complexes presented on the surface of antigen-presenting cells (APCs) or target cells. The specificity of the T cell response is mediated by pMHC recognition by the TCR and CD3 molecular complexes. The TCR is a heterodimer consisting of two different transmembrane polypeptide chains, represented by four peptide chains including α, β, γ and δ; Moreover, in accordance with various combinations of peptide chains, TCRs are divided into TCRαβ and TCRγδ. CD3 contains various transmembrane polypeptide chains, i.e. γ, δ, ε and ζ, which interact to form homodimers or heterodimers within TCR-CD3 complexes. For example, TCR-CD3 complexes comprise a TCRαβ dimer, a CD3γε dimer, a CD3δε dimer, and a CD3ζζ dimer. Because the cytoplasmic region of the TCR peptide chain is very short, it is generally assumed that the activation signal generated when the TCR recognizes an antigen is transmitted to the T cell by the CD3 peptide chain.

[4] Due to the important role of CD3 in initiating the immune response, signaling agents targeting the TCR-CD3 signaling pathway, in particular CD3-targeting monoclonal antibodies, are considered effective agents that can modulate the immune process and be used to treat inflammatory or autoimmune diseases. diseases. In fact, the anti-CD3 antibody orthoclone OKT3 was the first approved therapeutic antibody. OKT3 was first approved by the US FDA in 1985 for the treatment of acute rejection after organ transplantation. Although the immunosuppressive capacity resulting from repeated administration of OKT3 provided effective treatment of rejection after kidney transplantation, its use was limited by primary dose response syndrome; which is a syndrome believed to be associated with OKT3-mediated T cell activation and cytokine release. OKT3 was subsequently withdrawn from the market in 2010 due to the induction of a severe cytokine storm and immunogenicity issues associated with murine antibodies, among other factors.

[5] Another problem with CD3 antibodies is that many CD3 antibodies appear to be species specific, for example OKT3 reacts with chimpanzee CD3 but not with other primate CD3 such as macaque CD3 homologues or mouse CD3 homologues. The species specificity of anti-CD3 monoclonal antibodies is a major barrier to their development as antibody therapeutics for the treatment of human diseases. Any new drug candidate must undergo rigorous preclinical testing before it can be used in human clinical trials. The purpose of preclinical testing is to confirm that the drug candidate has the required activity and, most importantly, that the drug candidate is safe. Preclinical safety testing involves administering the drug candidate to members of the species of interest, preferably non-human primates. However, great apes, especially chimpanzees, are considered critically endangered species, and the use of such animals for drug safety testing is strictly limited. Species described in the prior art suitable for testing for safety assessment may include macaques, in particular cynomolgus macaques. However, it is difficult to provide reliable data on the preclinical safety assessment of anti-CD3 antibodies that lack primate species-specific cross-reactivity. Among the known antibodies that bind to human CD3, SP34 is one of the very few that can bind to multiple primate CD3s (eg, human and cynomolgus CD3) (see Salmeron, A. et al, J Immunol 147 (1991) 3047 -3052; Conrad M. L., et al, Cytometry A 71 (2007) 925-933).

[6] Although anti-CD3 monoclonal antibodies have been clinically tested for their effectiveness in certain disease groups, anti-CD3 antibodies have been increasingly used in the development of bispecific antibody drugs in recent years. Globally, bispecific CD3 T cell recruiter (BsTCE) antibodies currently account for more than half of bispecific antibody programs in clinical or preclinical stages. Bispecific antibodies to CD3 BsTCE, on the one hand, demonstrate the same high efficacy as CAR-T cell-based therapies, and on the other hand, they can be produced and commercialized in the same way as traditional monoclonal antibodies. Among the bispecific antibodies currently approved for marketing worldwide, the earliest approved were catumaxomab (approved by the European Medicines Agency (EMA) in 2009 and withdrawn from the United States in 2013) and blinatumomab (approved by the FDA). Food and Drug Administration (FDA) in 2014), both of which are BsTCE. Anti-CD3 antibody is an important component in the construction of BsTCE. The bispecific antibody BsTCE can bind to two targets simultaneously: one end recognizes tumor-associated antigen (TAA) on the surface of tumor cells, while the other end binds to the CD3 molecule on T cells. In the presence of tumor cells, binding of the bispecific antibody BsTCE to the surface of tumor cells can recruit and activate T cells in the vicinity of the tumor cells, which in turn ensures the destruction of tumor cells. In the design and construction of various BsTCE bispecific antibody structures, the selection and optimization of CD3 antibodies is of paramount importance. First, the species specificity of anti-CD3 monoclonal antibodies is critical, especially with regard to cross-reactivity in non-human apes. Second, the affinity of the anti-CD3 antibody for the CD3 complex is also critical; in this case, an antibody to CD3 with high affinity may be characterized by retention of the antibody in the spleen and other areas, which complicates its contact interaction with the tumor; and high affinity may also cause overstimulation of T cells, resulting in high levels of cytokine release. Third, valence bonds involved in the binding of anti-CD3 antibodies also play an important role, and it has previously been found that multivalent forms of bispecific anti-CD3 antibodies can cause side effects by activating T cells without binding tumor-associated antigens, and thus the vast majority of CD3 bispecific antibodies tested are in the form of monovalent CD3 antibodies.

[7] In addition to anti-CD3 antibodies, the structure diagram of BsTCE bispecific antibodies is also extremely important. There are various structures of BsTCE bispecific antibodies, which can be divided into two main categories: IgG-like structures containing Fc, and structures based on antibody fragments without Fc. For example, blinatumomab is a single polypeptide chain structure consisting of two sequentially linked single-chain antibody variable region (scFv) fragments, but this structure has a short half-life, requires continuous intravenous perfusion and is very inconvenient to use. Thus, Fc-containing structures are used in many BsTCE bispecific antibodies to improve molecular stability and pharmacokinetic properties. However, since the CD3-binding domain in BsTCE usually requires a monovalent form, Fc-containing structures are often asymmetric. In the case of these asymmetric Fc-containing structures, there are many technical difficulties to overcome, such as the problem of homodimerization of heavy chains in the asymmetric structure, the problem of light chain mismatch, molecular cross-linking caused by the Fcγ receptor, and functional effects such as ADCC or CDC and etc. To construct bispecific BsTCE antibodies from anti-TAA IgG antibodies and anti-CD3 IgG antibodies, various asymmetric structures can be selected [Figure 16(A)], with one commonly used structure being an IgG-like structure that retains two independent Fab domains, which wherein it contains four distinct polypeptide chains [two distinct heavy chains and two distinct light chains, structure shown in Figure 16(B)], with a molecular weight approximately corresponding to the molecular weight of a conventional monoclonal antibody; however, this structure can lead to the formation of by-products containing several combinations due to the presence of many different polypeptide chains, which creates great difficulty in purifying the expression and carrying out the method for producing the antibody. If an anti-CD3 Fab is modified to form a scFv structure, the “four-stranded” structure can be converted to a “three-stranded” structure [shown in Figure 16(C)], further reducing the number of by-product combinations and hence the complexity of its preparation. In order to construct a bispecific BsTCE antibody, the present inventors attempted to convert the murine IgG antibody SP34 into a scFv, however, no matter which arrangement was chosen (VH/VL) or the extent to which the length of the binding peptide was changed, the scFv was stable could not be obtained; Thus, there is a great need in the art for a stable anti-CD3 monoclonal antibody, especially a stable scFv structure derived therefrom.

[8] Thus, there is a great need in the art for an anti-CD3 antibody that is capable of binding to primate CD3, has suitable CD3 binding ability, and has a stable single-chain scFv structure.

Contents of the present invention

[9] The technical problem that needs to be solved in the field is to overcome the lack of low antigenic, effective and safe anti-CD3 antibodies and bispecific antibodies with asymmetric structures; wherein the present invention provides a CD3-targeting antibody, a bispecific antibody, and their use.

[10] To solve the above technical problem, the technical solution provided by the first aspect of the present invention is a CD3-targeting antibody containing a light chain variable region (VL) and a heavy chain variable region (VH); wherein the amino acid sequence VL is represented by SEQ ID NO: 56 or a mutant variant thereof, VH is a mutant variant of the amino acid sequence set forth by SEQ ID NO: 42 comprising one or more mutations at positions 30, 73, 76, 78, 93 and 94 ( according to Chothia numbering scheme). A mutation may cause the addition, deletion, or substitution of one or more amino acid residues in the original amino acid sequence. The CD3-targeting antibody of the present invention alters T cell binding ability and reduces the level of cytokine release, and thereby is expected to reduce the toxicity associated with cytokine release syndrome.

[11] In a preferred example, VH is characterized by the presence of mutations at positions selected from the following groups:

[12] (a) position 30;

[13] (b) provisions 30, 73 and 76;

[14] (c) provisions 30, 93 and 94;

[15] (d) provisions 30, 73 and 93;

[16] (e) provisions 30 and 93;

[17] (f) provisions 30, 76 and 78;

[18] (g) provisions 73, 76, 93 and 94;

[19] (h) provisions 76, 78 and 93;

[20] (i) provisions 30, 73, 76, 93 and 94;

[21] (j) provisions 30, 76, 78 and 93.

[22] In a preferred example, VH is characterized by the presence of mutations selected from the following groups:

[23] (a) N30S;

[24] (b) N30S, D73N and S76N;

[25] (c) N30S, V93A and R94K;

[26] (d) N30S, D73N and V93A;

[27] (e) N30S and V93T;

[28] (f) N30S, S76N and L78A;

[29] (g) D73N, S76N, V93A and R94K;

[30] (h) S76N, L78A and V93T;

[31] (i) N30S, D73N, S76N, V93A and R94K;

[32] (j) N30S, S76N, L78A and V93T.

[33] Provided that the VH of an antibody is characterized by the presence of mutations as defined above, the antibody of the present invention is further mutated by the VL amino acid sequence set forth in SEQ ID NO: 56 or the VH amino acid sequence set forth in SEQ ID NO: 42, and the resulting amino acid sequence has 80%, 85%, 90%, 95%, 98%, 99% or greater identity with the original amino acid sequence, and amino acid sequences that maintain or improve the function of the antibody are also within the scope of the present invention. .

[34] In a preferred example, the VH amino acid sequence is represented by any of SEQ ID NOs: 43-55, and/or the VL amino acid sequence is presented by any of SEQ ID NOs: 57-60.

[35] In the preferred example

[36] the amino acid sequence of VH is provided under SEQ ID NO: 44, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[37] the amino acid sequence of VH is provided under SEQ ID NO: 51, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[38] the amino acid sequence of VH is provided under SEQ ID NO: 44, and the amino acid sequence of VL is presented under SEQ ID NO: 60; or

[39] the amino acid sequence of VH is provided under SEQ ID NO: 51, and the amino acid sequence of VL is presented under SEQ ID NO: 60; or

[40] the amino acid sequence of VH is provided under SEQ ID NO: 45, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[41] the amino acid sequence of VH is provided under SEQ ID NO: 52, and the amino acid sequence of VL is provided under SEQ ID NO: 58; or

[42] the amino acid sequence of VH is provided under SEQ ID NO: 43, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[43] the amino acid sequence of VH is provided under SEQ ID NO: 43, and the amino acid sequence of VL is presented under SEQ ID NO: 60; or

[44] the amino acid sequence of VH is provided under SEQ ID NO: 50, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[45] the amino acid sequence of VH is provided under SEQ ID NO: 47, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[46] the amino acid sequence of VH is provided under SEQ ID NO: 48, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[47] the amino acid sequence of VH is provided under SEQ ID NO: 49, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[48] the amino acid sequence of VH is provided under SEQ ID NO: 53, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[49] the amino acid sequence of VH is provided under SEQ ID NO: 54, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[50] the amino acid sequence of VH is provided under SEQ ID NO: 43, and the amino acid sequence of VL is presented under SEQ ID NO: 57; or

[51] the amino acid sequence of VH is provided under SEQ ID NO: 44, and the amino acid sequence of VL is presented under SEQ ID NO: 57; or

[52] the amino acid sequence of VH is provided under SEQ ID NO: 43, and the amino acid sequence of VL is presented under SEQ ID NO: 59; or

[53] the amino acid sequence of VH is provided under SEQ ID NO: 44, and the amino acid sequence of VL is presented under SEQ ID NO: 59; or

[54] the amino acid sequence of VH is provided under SEQ ID NO: 51, and the amino acid sequence of VL is presented under SEQ ID NO: 57; or

[55] the amino acid sequence of VH is provided under SEQ ID NO: 55, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or

[56] the amino acid sequence of VH is provided under SEQ ID NO: 46, and the amino acid sequence of VL is provided under SEQ ID NO: 58.

[57] In a preferred example, the antibody comprises a single chain antibody fragment variable (scFv) comprising a VL linker-VH or a VH linker-VL; preferably the linker (ie linker peptide) is (GGGGS) _n [abbreviated as (G ₄ S) _n ] or a variant thereof, where n is a non-zero natural number, preferably from 1 to 20, more preferably the amino acid sequence of the linker presented under SEQ ID NO: 65, SEQ ID NO: 66 or SEQ ID NO: 67; more preferably, the amino acid sequence of the scFv is provided by SEQ ID NO: 73, SEQ ID NO: 74, SEQ ID NO: 75, SEQ ID NO: 78, SEQ ID NO: 79 or SEQ ID NO: 80; even more preferably, the antibody further comprises a crystallizable fragment (Fc), wherein the Fc is linked to the scFv via a hinge.

[58] In a preferred example, the antibody further comprises a constant region, preferably a human constant region; preferably the human constant region comprises a human light chain constant region and a human heavy chain constant region, wherein the human light chain constant region is preferably a human κ light chain constant region given in SEQ ID NO: 61 or a human λ light chain constant region given under SEQ ID NO: 62; more preferably, the human heavy chain constant region is the isotype hIgG1, hIgG2, hIgG3, hIgG4, or a variant thereof, preferably the heavy chain constant region set forth in SEQ ID NO: 63 or SEQ ID NO: 64.

[59] To solve the above technical problem, the technical solution provided by the second aspect of the present invention is a bispecific antibody. The bispecific antibody of the present invention has a three-chain structure, which can reduce the number of by-product combinations, and therefore the complexity of its preparation; however, it is not possible to develop a bispecific antibody through minor modification of the antibody according to existing technology. As described in the prior art, in order to construct the BsTCE bispecific antibody, the present inventors attempted to convert the murine IgG antibody SP34 into scFv, however, no matter which arrangement was chosen (VH/VL) or the extent to which the length of the linker peptide was changed, A stable scFv could not be obtained. After developing several mutation schemes and testing procedures, the present inventors discovered that only some of these mutations could result in scFv maintaining a stable structure. The bispecific antibody of the present invention comprises a first protein functional region and a second protein functional region, wherein the first protein functional region comprises a CD3-targeting antibody according to the first aspect of the present invention; wherein preferably the bispecific antibody contains the following three chains: (1) VL1-linker-VH1-hinge-CH2-CH3 (“overhang”) or VH1-linker-VL1-hinge-CH2-CH3 (“overhang”), related to the first a protein functional region, (2) VH2-CH1-hinge-CH2-CH3 (“hole”) related to a second protein functional region, and (3) VL2-CL related to a second protein functional region; wherein the second functional region of the protein is a non-CD3-targeting antibody, preferably a B7H4-targeting antibody or an ROR1-targeting antibody, and the linker is preferably (G ₄ S) _n , where n is a non-zero natural number , preferably from 1 to 20, and more preferably, the amino acid sequence of the linker is provided by SEQ ID NO: 65, SEQ ID NO: 66 or SEQ ID NO: 67; more preferably, the bispecific antibody comprises VL1-linker-VH1-hinge-CH2-CH3 ("hinge"), set forth in SEQ ID NO: 88, VH2-CH1-hinge-CH2-CH3 ("groove"), set forth in SEQ ID NO: : 86, and VL2-CL, given under SEQ ID NO: 83, or VL1-linker-VH1-hinge-CH2-CH3 (“protrusion”), given under SEQ ID NO: 88, VH2-CH1-hinge-CH2- CH3 ("hole"), shown under SEQ ID NO: 87, and VL2-CL, shown under SEQ ID NO: 85. The bispecific antibody of the present invention overcomes the disadvantage associated with the instability of the arm of a single chain antibody targeting CD3, which when it is stable and has the ability to bind to T cells. The production of a bispecific antibody containing only three chains is easy to achieve while the complexity of its production is reduced.

[60] To solve the above technical problem, the technical solution provided by the third aspect of the present invention is an isolated nucleic acid encoding a CD3-targeting antibody according to the first aspect of the present invention or a bispecific antibody according to the second aspect of the present invention.

[61] To solve the above technical problem, the technical solution provided by the fourth aspect of the present invention is an expression vector containing an isolated nucleic acid according to the third aspect of the present invention; preferably the expression vector is selected from a retroviral vector, a lentiviral vector, an adenoviral vector and an adeno-associated virus vector.

[62] To solve the above technical problem, the technical solution provided by the fifth aspect of the present invention is a genetically modified cell transfected with an expression vector according to the fourth aspect of the present invention; preferably the genetically modified cell is a eukaryotic cell.

[63] To solve the above technical problem, the technical solution provided by the sixth aspect of the present invention is a pharmaceutical composition comprising a CD3-targeting antibody according to the first aspect of the present invention, a bispecific antibody according to the second aspect of the present invention, a genetically modified cell according to the fifth aspect of the present invention and a pharmaceutically acceptable carrier; preferably the pharmaceutical composition further comprises an antibody directed against an immune checkpoint.

[64] To solve the above technical problem, the technical solution provided by the seventh aspect of the present invention is the use of a CD3-targeting antibody according to the first aspect of the present invention, a bispecific antibody according to the second aspect of the present invention, an isolated nucleic acid according to the third aspect of the present invention, an expression vector according to the fourth aspect of the present invention, a genetically modified cell according to the fifth aspect of the present invention, or a pharmaceutical composition according to the sixth aspect of the present invention in the manufacture of a medicament for treating a tumor.

[65] To solve the above technical problem, the technical solution provided by the eighth aspect of the present invention is a combination of sets comprising set A and set B; wherein kit A comprises a CD3-targeting antibody according to the first aspect of the present invention, a bispecific antibody according to the second aspect of the present invention, a genetically modified cell according to the fifth aspect of the present invention, or a pharmaceutical composition according to the sixth aspect of the present invention; set B contains other antibodies, bispecific antibodies, genetically modified cells or pharmaceutical compositions, other antibodies, bispecific antibodies, genetically modified cells or pharmaceutical compositions targeting CD3, B7H4, ROR1 or other target. Set A and Set B can be used in any order, where Set A can be used before Set B, or Set B can be used before Set A. The drug in Set A is in an injectable form, such as an injection, and the drug in Set B is is present in an injectable form, such as an injection, or in a swallowable form, such as a tablet or pill.

[66] To treat an appropriate tumor, a CD3-targeting antibody according to the first aspect of the present invention, a bispecific antibody according to the second aspect of the present invention, a genetically modified cell according to the fifth aspect of the present invention, a pharmaceutical composition according to the sixth aspect of the present invention, or a combination of kits according to the eighth aspect of the present invention.

[67] Based on the generally known knowledge in the art, the above-mentioned preferred conditions can be optionally combined to obtain preferred examples of the present invention.

[68] All reagents and starting materials used in the present invention are commercially available.

[69] The positive and progressive effects provided by the present invention are as follows:

[70] 1. The monoclonal antibody of the present invention changes the binding ability to T cells and reduces the level of cytokine release, and thereby the toxicity associated with cytokine release syndrome is expected to be reduced.

[71] 2. The bispecific antibody derived therefrom overcomes the disadvantage of the instability of the CD3-targeting single chain antibody arm while being stable and capable of binding to T cells.

[72] 3. The production of a bispecific antibody containing only three chains is easy to perform, and the complexity of its production is reduced.

Brief description of graphic materials

[73] In FIG. Figure 1 shows HPLC-SEC results of single-chain anti-CD3 antibody after one-step purification: (A) PR000275, (B) PR000276, (C) PR000307, and (D) PR000308.

[74] In FIG. Figure 2 shows a sequence alignment of humanized SP34 VH mutant variants.

[75] In FIG. Figure 3 shows a sequence alignment of humanized VL SP34 mutant variants.

[76] In FIG. 4 shows the differences in significant regions of the various VH/VL mutant sequences, wherein (A) shows the VH mutant sequence and (B) shows the VL mutant sequence.

[77] In FIG. 5 shows (A) SDS-PAGE results and (B) HPLC-SEC results of single-chain anti-CD3 antibody PR000510 after one-step purification.

[78] In FIG. 6 shows the binding ability of anti-CD3 antibody PR000260 to (A) recombinant CHOK1 cells overexpressing human CD3 and (B) recombinant CHOK1 cells overexpressing cynomolgus CD3.

[79] In FIG. 7 shows the binding ability of an anti-CD3 antibody to human T cells, including the binding curve and the relative MFI intensity (the MFI fluorescence intensity of the antibody binding to human T cells at certain concentrations and the relative value compared to the parent antibody PR000260 (SP34 )) or maximum MFI, where (A) PR000511, PR000512, PR000513, PR000514 and PR000260 bind to human T cells, (B) PR001848, PR001849 and PR000260 bind to human T cells, (C) PR002467, PR002468, PR0024 69, PR002470, PR002471, PR002472, PR001848 and PR000260 bind to human T cells, (D) PR001848, PR002742, PR002743 and PR000260 bind to human T cells, (E) PR002833, PR002834, PR002835, PR0 02836, PR002837, PR002742, PR001848, PR002469 and PR000260 bind to human T cells, (F) PR003886, PR001848 and PR002742 bind to human T cells, (G) PR001848, PR002469 and PR004616 bind to human T cells.

[80] Figure 8 shows the binding ability of a single chain anti-CD3 antibody to human T cells, including the binding curve and the relative MFI intensity (the MFI fluorescence intensity of an antibody binding to human T cells at a certain concentration and the relative compared to the parent antibody PR000260 (SP34)), where (A) PR000510, PR000624, PR000627 and PR000260 bind to human T cells, (B) PR001850 and PR000260 bind to human T cells.

[81] In FIG. 9 shows the binding ability of an anti-CD3 antibody to cynomolgus T cells.

[82] In FIG. 10 shows the ability of an anti-CD3 antibody to activate human T cells to mediate production of the cytokine IFN-γ, where (A) PR000511, PR000512, PR000513, PR000514 and PR000260 activate T cells, (B) PR001848, PR001849 and PR000260 activate T cells, (C) PR002468, PR002469, PR002471 and PR001848 activate T cells, (D) PR002742, PR001848 and PR000260 activate T cells, (E) PR002833, PR002834, PR002835, PR002836, PR002837 and PR0002 60 activate T cells, (F) PR003886, PR001848 and PR002742 activate T cells, (G) PR001848, PR002469 and PR004616 activate T cells.

[83] In FIG. 11 shows the ability of single chain anti-CD3 antibodies (PR000510, PR000623, PR000624, PR000627 and PR000260) to activate human T cells to produce the cytokine IFN-γ.

[84] In FIG. 12 shows SDS-PAGE results of bispecific antibody samples (A) PR002883 and (B) PR002885 obtained by one-step purification.

[85] In FIG. 13 shows the binding ability of the monoclonal antibody and the bispecific antibody to (A) SK-BR-3 cells and (B) human T cells.

[86] In FIG. 14 shows the ability to kill target cells mediated by the bispecific antibody PR002883 in vitro ; where (A) shows the killing of SK-BR-3 cells and (B) shows the levels of IFN-γ release.

[87] In FIG. 15 shows the binding ability of the monoclonal antibody and the bispecific antibody to (A) Panc-1 cells and (B) human T cells.

[88] In FIG. 16 shows the structure of a monoclonal antibody or a bispecific antibody; (A) IgG structure, (B) asymmetric "four-chain" structure, (C) asymmetric "three-chain" structure, providing a single chain antibody.

Detailed Description of the Preferred Embodiment

[89] The present invention is further illustrated below using examples, however, the present invention is not limited to the scope of the described examples. Experimental methods for which specific conditions are not specified in the following examples were selected in accordance with conventional methods and conditions or in accordance with technical requirements.

[90] As used herein, the term “antibody” generally refers to a protein containing a moiety that binds to an antigen and optionally allows the moiety that binds the antigen to conform the scaffold moiety to a conformation that facilitates binding of the antibody to the antigen. Typically, an antibody may comprise an antibody light chain variable region (VL), an antibody heavy chain variable region (VH), or both. The VH and VL regions can be further divided into hypervariable regions called complementarity determining regions (CDRs), which are dispersed within more conserved regions called framework regions (FRs). Each of VH and VL may consist of three CDR regions and four FR regions, which may be arranged in the following order from amino terminus to carboxy terminus: FR1, CDR1, FR2, CDR2, FR3, CDR3 and FR4. The variable regions of the heavy and light chains contain binding domains that interact with the antigen. Examples of antibodies include, but are not limited to, antibody, antigen binding fragment (Fab, Fab', F(ab)2, Fv fragment, F(ab')2, scFv, di-scFv and/or dAb), immunoconjugate, multispecific antibody (e.g. bispecific antibody), antibody fragment, antibody derivative, antibody analog or fusion protein, etc., provided that they exhibit the required antigen-binding activity.

[91] As used herein, the term “variable” generally refers to the fact that certain portions of the variable domain sequence of an antibody vary substantially, causing different specific antibodies to bind to and be specific for their particular antigen. However, the variability is not distributed evenly throughout the variable region of the antibody. It is concentrated in three segments of the light and heavy chain variable regions, known as the complementarity determining region (CDR) or high variability region (HVR). The portion of the variable domain that is conserved to a higher degree is known as the framework (FR). Each of the natural heavy and light chain variable domains contains four FR regions, most of which adopt a β-sheet conformation and are connected by three CDRs that form a loop linkage and in some cases form part of a β-sheet structure. The CDRs in each chain are in close proximity to each other, being separated by the FR region, and form the antigen-binding region of the antibody together with the CDRs from the other chain, while the constant regions are not directly involved in the binding of the antibody to the antigen, but exhibit various effector functions, such as involvement in antibody-dependent antibody-mediated cytotoxicity. In the current state of the art, CDR antibodies can be detected by various methods, such as Kabat's sequence variability detection scheme (see Kabat et al ., Sequences of proteins of immunological interest, Fifth Edition, National Institutes of Health, Bethesda, Maryland (1991)) and a definition scheme according to Chothia based on the location of structural loop regions (see Al-Lazikani et al ., JMol Biol 273:927-48, 1997). In the present application, a combined detection scheme of Kabat detection and Chothia detection is also used to identify amino acid residues in variable domain sequences and full-length antibody sequences (Table 1).

Table 1. Methods for determining the CDR of antibodies provided in this application (see http://bioinf.org.uk/abs/)

[92] Laa-Lbb may refer to the amino acid sequence from position aa (Chothia numbering scheme) to position bb (Chothia numbering scheme), starting from the N-terminus of the antibody light chain; Haa-Hbb may refer to the amino acid sequence from position aa (Chothia numbering scheme) to position bb (Chothia numbering scheme), starting at the N-terminus of the antibody heavy chain. For example, L24-L34 may refer to the amino acid sequence from position 24 to position 34, starting from the N-terminus of the antibody light chain according to the Chothia numbering scheme; H26-H32 may refer to the amino acid sequence from position 26 to position 32, starting from the N-terminus of the antibody heavy chain according to the Chothia numbering scheme.

[93] Effector functions mediated by the antibody Fc domain, such as ADCC and CDC, are also highly important biological functions, with different IgG isoforms having different functions in relation to ADCC or CDC, for example, IgG1 and IgG3 have strong ADCC and CDC effects, in while IgG2 and IgG4 are characterized by relatively weak effects. In addition, the original Fc effector function can also be modulated by amino acid mutations or modifications to alter the ability of the Fc to bind to the Fc receptor. For example, the double mutation “LALA” (L234A/L235A) in IgG1 significantly reduces the affinity for FcγRIIIA (CD16A) and thus reduces the effect of ADCC. In addition, the P329G mutation significantly reduces IgG1 binding of several Fcγ receptors (see Schlothauer T, Herter S, Koller CF, et al. Protein Eng Des Sel. 2016 Oct; 29(10): 457-466). In the present application, in order to reduce the binding of anti-CD3 antibodies to the Fcγ receptor, a double mutation "LALA" (L234A/L235A) or a triple mutation "LALAPG" (L234A/L235A/P329G) was introduced into the Fc of these anti-CD3 antibodies.

Example 1. Preparation and analysis of a recombinant antibody with determination of its characteristics

1.1. Preparation of recombinant IgG antibodies

[94] Once the light and heavy chain variable domain sequences encoding the antibody molecule have been obtained, the recombinant antibody molecule can be produced by expressing the light and heavy chain variable domain sequences fused to the corresponding human antibody light and heavy chain constant domain sequences using conventional recombinant techniques. DNA. In this example, an antibody heavy chain variable domain (VH) sequence was genetically synthesized and inserted into a plasmid vector for expression in mammalian cells encoding the heavy chain constant domain sequence of a human IgG1 isotype antibody to produce an encoded full-length IgG1 antibody heavy chain, and thus a double mutation "LALA" (L234A/L234A) (SEQ ID NO: 63) or triple mutation "LALAPG" (L234A/L235A/P329G) (SEQ ID NO: 64) was introduced into the constant region of the heavy chain of IgG1 to reduce the level of binding of the antibody to the receptor Fcγ. The antibody light chain variable domain (VL) sequence was genetically synthesized and inserted into a plasmid vector for expression in mammalian cells encoding the human κ light chain constant domain sequence (SEQ ID NO: 61) to produce an encoded full-length κ antibody light chain; alternatively, VL was genetically synthesized and inserted into a plasmid vector for expression in mammalian cells encoding the human antibody λ light chain constant domain sequence (SEQ ID NO: 62) to produce the full-length λ light chain, yielding the encoded antibody.

[95] By cotransfecting a mammalian host cell (eg, human embryonic kidney cell HEK293) with plasmids encoding the antibody heavy chain and plasmids encoding the light chain of the antibody, it is possible to obtain purified recombinant antibody as a complex of correctly paired light and heavy chains using conventional techniques for the expression and purification of recombinant proteins. Specifically, HEK293 cells were propagated in FreeStyle™ F17 Expression Medium (Thermo, no. A1383504). Before transient transfection, the cell concentration was adjusted to 6-8×10 ⁵ cells/ml and they were incubated in a shaker at 37°C, 8% CO ₂ for 24 hours at a cell concentration of 1.2×10 ⁶ cells/ml. 30 ml of cultured cells were obtained. The heavy chain encoding plasmid and the light chain encoding plasmid were mixed in a 2:3 ratio, and a total of 30 μg of plasmids was dissolved in 1.5 ml of Opti-MEM reduced serum medium (Thermo, no. 31985088) and filtered through membrane with 0.22 µm pores. Then, 1.5 ml of Opti-MEM was dissolved in 120 μl of PEI (Polysciences, no. 23966-2) at a concentration of 1 mg/ml and incubated for 5 minutes. PEI was slowly added to the plasmid, followed by incubation for 10 minutes at room temperature. The mixed solution of plasmid and PEI was slowly added dropwise into the culture flask while the culture flask was shaken. Transfected cells were incubated at 37°C, 8% CO ₂ in a shaker for 5 days. Cell viability was determined after 5 days. Cultures were then harvested by centrifugation at 3300 g for 10 minutes to ensure collection of the supernatant. Impurities in the supernatant were removed by high-speed centrifugation. A gravity elution column (Bio-Rad, #7311550) containing MabSelect™ (GE Healthcare Life Science, #71-5020-91 AE) was equilibrated with PBS (pH 7.4) and washed with 2-5 PBS - multiple of the column volume. A sample of the supernatant was loaded into the column and washed with an amount of PBS equal to 5-10 times the column volume. The target protein was then eluted with 0.1 M glycine at pH 3.5, then adjusted to neutral pH with Tris-HCl at pH 8.0, finally concentrated using an ultrafiltration tube (Millipore, # UFC901024), and then replacement with PBS buffer and a purified solution of recombinant antibodies was obtained. Finally, the concentration was measured using NanoDrop (Thermo Scientific™ NanoDrop™ One), and the purified recombinant antibody solution was dosed and stored for use as a reserve.

1.2. Production of monovalent recombinant scFv-his antibody

[96] The VH and VL sequences of the antibody were linked with a flexible peptide (linker) to produce a single polypeptide chain encoding both VH and VL, i.e. fragment represented by the variable region of a single chain antibody (scFv). If a linker peptide of suitable length is selected, such as (G ₄ S) ₃ (SEQ ID NO: 65) or (G ₄ S) ₄ (SEQ ID NO: 66), then VH and VL can be suitably folded and assembled into functional antibodies. Different scFv structures (VH-linker-VL or VL-linker-VH) can be designed depending on the different arrangements of VH and VL and differences in the linker peptide. A single scFv contains an antigen-binding region consisting of a pair of VH and VL, which typically binds only one antigen molecule and is therefore called a monovalent binding molecule.

[97] To facilitate purification in this example, the C-terminus of the scFv was fused to a His tag consisting of 6-histidine. A plasmid encoding scFv and a His tag was genetically synthesized and inserted into a plasmid vector for expression in mammalian cells to produce a plasmid encoding scFv-his, which was transfected into a mammalian host cell (eg, human embryonic kidney cell HEK293), and then recombinant the protein can be purified using conventional recombinant protein expression and purification techniques. Specifically, HEK293 cells were propagated in FreeStyle™ F17 Expression Medium (Thermo, no. A1383504). Before transient transfection, the cell concentration was adjusted to 6-8×10 ⁵ cells/ml and they were incubated in a shaker at 37°C, 8% CO ₂ for 24 hours at a cell concentration of 1.2×10 ⁶ cells/ml. 30 ml of cultured cells were obtained. 30 μg of plasmid was dissolved in 1.5 ml of Opti-MEM reduced serum medium (Thermo, no. 31985088) and filtered through a 0.22 μm pore membrane. Then, 1.5 ml of Opti-MEM was dissolved in 120 μl of PEI (Polysciences, no. 23966-2) at a concentration of 1 mg/ml and incubated for 5 minutes. PEI was slowly added to the plasmid, followed by incubation for 10 minutes at room temperature. The mixed solution of plasmid and PEI was slowly added dropwise into the culture flask while the culture flask was shaken. Transfected cells were incubated at 37°C, 8% CO ₂ in a shaker for 5 days. Cell viability was determined after 5 days. Cultures were then harvested by centrifugation at 3300 g for 10 minutes to ensure collection of the supernatant. Impurities in the supernatant were removed by high-speed centrifugation. Gravity elution column (Bio-Rad, no. 7311550) containing Ni Sepharose excel sepharose gel (GE Healthcare Life Science, no. 17-3712-01), Ni Sepharose excel (GE Healthcare Life Science, no. 17-3712-01), equilibrated with PBS (pH 7.4) and washed with 2-5 times the column volume of PBS. A sample of the supernatant was loaded into the column and washed with an amount of PBS equal to 5-10 times the column volume. First, nonspecifically adsorbed heteroproteins were eluted using buffer A (containing 20 mM imidazole, 150 mM phosphate, pH 8.0), and then the target protein using buffer B (containing 500 mM imidazole, 150 mM phosphate, pH 8.0), and finally , used ultrafiltration tubes (Millipore, no. UFC901024) to concentrate and exchange the solution with PBS buffer and obtain a solution of purified recombinant antibodies. Finally, the concentration was measured using NanoDrop (Thermo Scientific™ NanoDrop™ One), and the purified recombinant antibody solution was dosed and stored for use as a reserve.

1.3. Preparation of divalent recombinant antibody scFv-Fc

[98] In this example, a recombinant scFv-Fc molecule was constructed by fusion of the Fc sequence of the human IgG1 constant region (Glu216-Lys447, containing the hinge region, CH2 domain and CH3 domain) with scFv at its C-terminus, and by homodimerization of the Fc was formed a divalent scFv-Fc dimer molecule that was capable of binding two antigen molecules simultaneously. In this case, a double mutation “LALA” (L234A/L235A) or a triple mutation “LALAPG” (L234A/L235A/P329G) was also introduced into Fc to reduce the binding of the antibody to the Fcγ receptor. The polypeptide sequence encoding scFv-Fc was genetically synthesized and inserted into a plasmid vector for expression in mammalian cells to produce a plasmid encoding scFv-Fc, which was then transfected into mammalian host cells (eg, human embryonic kidney cell HEK293) and then used the method of protein expression and purification described in example 1.1 to obtain a purified recombinant protein.

1.4. Protein purity analysis using HPLC-SEC

[99] The purity and aggregate form of protein samples were analyzed using size exclusion chromatography (SEC) to separate molecules by size. A TSKgel G3000SWxl analytical column (Tosoh Bioscience, no. 08541, 5 μm, 7.8 mm × 30 cm) was connected to a high pressure liquid chromatograph (HPLC) (Agilent Technologies, Agilent 1260 Infinity II) and equilibrated with PBS buffer at room temperature for at least 1 hour. The system was injected with an appropriate amount of protein sample (at least 10 μg), filtered through a 0.22 μm pore membrane, and the following HPLC program was set: the column was loaded with sample buffer PBS at a flow rate of 1.0 ml/min for a maximum of 20 minutes. HPLC generated analytical HPLC reports that, when generated, reported the retention times of components of different molecular sizes in the sample.

Example 2 Recombinant expression of a mouse-human chimeric anti-CD3 antibody representing the SP34 antibody.

[100] SP34 is a mouse-derived anti-human CD3e antibody that binds a number of different primate CD3s and has T cell activating function. The VH and VL variable region sequences of SP34 are disclosed in WO 2016071004A1. In the present application, the amino acid sequence of VH SP34 is provided under SEQ ID NO: 42, and its corresponding mouse germline V gene is IGHV10-1; the amino acid sequence of VL SP34 is provided under SEQ ID NO: 56, and its corresponding mouse germline V gene is IGLV1. In this example, the VH sequence of SP34 was fused to the human IgG1 heavy chain constant domain sequence (SEQ ID NO: 63) containing the double "LALA" mutation (L234A/L235A) to produce a full-length heavy chain of a mouse-human chimeric SP34 isotype antibody. IgG1; the amino acid sequence of SP34 VL was fused to the human antibody λ light chain constant domain sequence (SEQ ID NO: 62) to generate the full length λ light chain of the mouse-human chimeric antibody SP34.

[101] Mouse-human chimeric recombinant antibody SP34 PR000260 was prepared according to the method of Example 1.1. The following Table 2 shows the recombinant expression data for PR000260.

Table 2. Expression and purification of recombinant antibody PR000260

Example 3: Conversion of anti-CD3 antibody, murine SP34 antibody, into recombinant scFv antibody

[102] The VH sequence (SEQ ID NO: 42) and the VL sequence (SEQ ID NO: 56) of SP34 were linked by a flexible peptide (linker) to produce a single polypeptide chain encoding both VH and VL, i.e. fragment represented by the variable region of a single chain antibody (scFv). Depending on the different positions of VH and VL and different lengths of linker peptides (SEQ ID NO: 65, SEQ ID NO: 66), it was possible to construct different scFv structures, and a His tag consisting of 6-histidine was fused to C- end of the scFv to ensure purification. The linker peptide shown under SEQ ID NO: 67 could also be used to construct the scFv according to the present application.

[103] In this example, four recombinant scFv antibody molecules (PR000275, PR000276, PR000307, PR000308) were prepared according to the method of Example 1.2. The sequence numbers of these four recombinant scFv antibody molecules are listed in Table 3 below; the following Table 4 shows the expression data of these four recombinant molecules; and the HPLC-SEC results for these four molecules after one-step purification are shown in Figure 1, where (A) shows the result for PR000275, (B) shows the result for PR000276, (C) shows the result for PR000307, and (D) shows result for PR000308. It can be seen that when using the VH and VL sequences of SP34 to construct scFv, no matter what alignment model was used (VH/VL) or to what extent the length of the linker peptide was changed, a stable scFv could not be obtained.

Table 3. Structure and sequence number for four recombinant antibody molecules

Table 4. Expression and purification of scFv recombinant antibody molecules

Example 4: SP34 Sequence Optimization

4.1. Humanization of variable region sequences and mutation of framework region

[104] In this example, the “CDR transplantation” method was used to humanize the sequence, i.e. transplanting a mouse antibody VH CDR into a human antibody VH framework; and transplanting a mouse antibody VL CDR into a human antibody VL framework. The VH or VL framework region sequence of a human antibody can be derived from human germline gene sequences or antibody sequences that have been rearranged by V(D)J, or consensus sequences of a particular human antibody VH or VL gene family. In this example, framework sequences provided by human germline gene sequences were used as humanized template sequences, i.e. the human V germline fragment provided the framework sequences FR1, FR2 and FR3, and the human J germline fragment provided the framework sequence FR4. Finally, the humanized variable region (VH or VL) sequences were designed as (human)FR1-(mouse)CDR1-(human)FR2-(mouse)CDR2-(human)FR3-(mouse)CDR3-(human)FR4.

[105] In this example, the sequence of the human V germline gene fragment IGHV3-73*01 or the human V germline fragment IGHV3-23*01 was used as a humanized template in combination with the sequence of the human J germline fragment IGHJ1*01 to obtain framework sequences. Amino acid mutations were introduced at one or more sites at position 30, position 73, position 76, position 78, position 93, or position 94 (according to the Chothia numbering scheme) to produce several different VH mutant sequences.

[106] In this example, the sequence of the human V germline fragment IGLV7-46*02 in combination with the sequence of the human J germline fragment IGLJ2*01 or the sequence of the human V germline fragment IGKV1-39*01 in combination with the sequence of the human V germline fragment germline J gene IGKJ4*01 was used as a humanized template to obtain the framework sequence. Amino acid mutations were introduced at zero or more sites at position 2, position 36, position 46, position 49, position 66, position 69, position 71, or position 87 (according to the numbering scheme according to Chothia) to produce several different VL mutant sequences .

[107] The following Table 5 lists the sequence numbers of the antibody variable region, its optimized mutant sequences (FV) and the sequences of the CDR and FR regions determined according to Chothia.

Table 5. SP34 antibody variable region and its optimized mutant sequences (FV), as well as a list of CDR and FR region sequences determined according to Chothia

[108] Figure 2 provides a list of VH mutant sequences compared. Figure 3 provides a list of VL mutant sequences compared. The sequence differences between VH mutants and VL mutants at significant sites are listed in (A) and (B) in Figure 4, respectively. As seen in Figures 2-4, mutations of the antibody of the present invention in VH occurred at one or more sites selected from position 30, position 73, position 76, position 78, position 93 or position 94 of the amino acid sequence shown under SEQ ID NO : 42. Mutations in VL occurred at position 2, position 36, position 46, position 49, position 66, position 69, position 71 and/or position 87 of the sequence shown under SEQ ID NO: 56. More detailed information about the mutations can be found in Table 5 for sequence characteristics VH3730, VH3731, VH3732, VH3733, VH3734, VH3735, VH3230, VH3231, VH3232, VH3233, VH3234, VH3235, VH3236, VL7460, VL7461, VK1392 and VK 1393.

4.2. Recombinant antibody molecules containing mutant variants with optimized sequences

[109] The VH mutant and VL mutant sequences obtained in Example 4.1 were paired and combined, and a recombinant IgG antibody was constructed according to the method described in Example 1.1, where a double mutation "LALA" or a triple mutation "LALAPG" was introduced into IgG1 heavy chain constant region to reduce Fc effector function. Table 6 lists the sequences of recombinant antibody molecules that have been sequence optimized. Table 7 shows data regarding the expression of recombinant antibodies. With the exception of three IgG molecules constructed with the VH mutant VH3230, which had very low expression yields, all other IgG molecules had acceptable expression yields.

Table 6. List of sequences of chimeric antibody SP34 or sequence optimized antibody

Table 7. Expression Yield and Purity of Recombinant Antibody

4.3. Recombinant scFv molecules containing sequence-optimized mutant variants

[110] The sequences of the VH mutant variants and the VL mutant variants obtained in Example 4.1 were paired and combined, and a plurality of recombinant scFv divalent antibody molecules were produced in accordance with the method described in Example 1.3. The following tables 8, 9 respectively list information regarding the sequences and protein expression of the scFv. As can be seen from Table 9, PR000510 and PR000627 are molecules characterized by particularly improved expression and stability. Figure 5 shows the results of (A) SDS-PAGE and (B) HPLC-SEC PR000510, which show that they have good monomeric purity without the presence of obvious aggregates.

Table 8. Information regarding the structure and sequence of scFv molecules constructed from sequence optimized mutant variants

Table 9. ScFv Antibody Expression Data After Sequence Optimization

Example 5: Use of FACS to Determine the Binding Ability of Anti-CD3 Antibody and CD3 Expressing Cells

[111] The FACS flow analysis method was used to analyze the binding of anti-CD3 antibody to CD3-expressing cells, wherein the CD3-expressing cells could be: CHOK1 cells overexpressing human CD3 or HEK293 cells (plasmids encoding ORF chains γ, δ, ε and ζ CD3 of human origin, and plasmids encoding ORFs of the human TCR α and β chains, were cotransfected into CHOK1 (ATCC, CCL-61) or HEK293 (ATCC, CRL-1573) host cells to construct stable cell lines expressing the structure represented by human TCR/CD3 complex); CHOK1 or HEK293 cells overexpressing cynomolgus CD3; Human T cells of all types (isolated using a kit for isolating human T cells of all types (Miltenyi, No. 130-096-535) from PBMC); Cynomolgus T cells of all types. Specifically, the collected cells were washed twice with PBS containing 2% FBS (FACS buffer) and resuspended in FACS buffer, distributed into 96-well plates at 1 x 10 ⁵ cells per well, centrifuged at 500 g for 5 minutes . The supernatant was removed and 100 μl of pregradient anti-CD3 antibody was added, then incubated for 1 hour at room temperature and washed twice with FACS buffer. Cells were resuspended in FACS buffer in which Alexa Fluor 488 AffiniPure secondary antibody, a goat anti-human IgG antibody that is specific for the Fcγ fragment, was diluted (Jackson ImmunoResearch, no. 109-545-098), then incubated at room temperature in darkness for 30 minutes. Cells were washed twice with FACS buffer and resuspended in 200 μl FACS buffer. The fluorescence signal value was read by flow cytometry (BD FACS CANTOII or ACEA NovoCyte) and the resulting data was processed and analyzed using FlowJo software version 10 (FlowJo, LLC). GraphPad Prism 8 software was used for data processing and graphical analysis, and parameters such as binding curves and EC50 values could be obtained using four-parameter nonlinear fitting.

[112] Figure 6 shows the binding ability of the anti-CD3 antibody prepared in Example 2 to recombinant CHOK1 cells overexpressing human CD3 (Figure 6(A)) and recombinant CHOK1 cells overexpressing cynomolgus CD3 (Figure 6(B)) . The results indicated that the chimeric antibody SP34 PR000260 had high binding capacity to both human and cynomolgus CD3.

[113] Figure 7 (A) - (G) shows the binding ability of the anti-CD3 antibodies produced in Example 4.2 (including PR000260 and its mutant variants) to human T cells of all types, respectively. MFI fluorescence intensity for anti-CD3 antibody binding to all types of human T cells and relative to parent antibody PR000260 was calculated at antibody concentrations of 7.4 or 10 μg/mL. Specifically, after sequence optimization, SP34 IgG isotype antibodies PR000512, PR000513, PR001849, and PR002837 exhibited binding abilities comparable to PR000260 (i.e., the chimeric SP34 antibody); while PR000514 had slightly higher binding capacity than PR000260; PR000511, PR001848, PR002469, PR002472, PR002742, PR002833, PR002834, PR002835, PR002836, PR003886 and PR004616 had lower T cell binding capacity; PR002467, PR002468, PR002470, PR002471 and PR002743, on the other hand, barely bound T cells (or the signal could not be detected at the current antibody concentration). The above results indicate that the present invention, by optimizing the sequence of the anti-CD3 antibody, has produced several new antibodies that have varying abilities to bind to human T cells and can be used in various application scenarios.

[114] Figure 8 (A) and (B) shows the binding ability of the single chain anti-CD3 antibody in scFv-Fc format produced in Example 4.3 to human T cells of all types. MFI fluorescence intensity for anti-CD3 antibody binding to all types of human T cells and relative to parent antibody PR000260 was calculated at antibody concentrations of 7.4 or 10 μg/mL. In particular, after humanization optimization of the scFv format antibody SP34, PR000624 had comparable or slightly higher binding capacity when compared to PR000260; PR000510 and PR000627 had comparable or slightly lower binding capacity when compared to PR000260; and PR001850 had significantly lower T cell binding capacity when compared with PR000260. The above results indicate that the present invention has also produced several stable single chain antibodies in scFv form through sequence optimization of anti-CD3 antibodies that are capable of binding to human T cells and are suitable for use in scenarios such as the construction of bispecific antibodies.

[115] Figure 9 shows the ability of some of the anti-CD3 antibodies produced in Example 4.2 to bind to all types of cynomolgus T cells. It can be seen that the different molecules had different binding abilities to all types of cynomolgus T cells and were positively correlated with their binding abilities to all types of human T cells; those. molecules that bind strongly to all types of human T cells also bind strongly to all types of cynomolgus T cells, and vice versa.

Example 6 Determination of Anti-CD3 Antibody-Induced Activation on Human T Cells

[116] Gradient dilutions of anti-CD3 antibodies (eg, 50, 10, 5, 1, 0.5, 0.05 μg/ml) were coated into 96-well cell culture plates, with triplicate wells for each concentration and 50 μl of solution per well, incubation was carried out overnight at 4°C. The density of human PBMCs (MiaoTong Biology) or human T cells of all types (isolated using the Human T Cells of All Types Isolation Kit (Miltenyi, No. 130-096-535) from PBMCs) was adjusted to 7.5× ¹⁰⁵ / ml and human anti-CD28 antibody was added at a concentration of 1 μg/ml, after which 200 μl per well of the resulting mixture was added to the cell culture plate and incubated in a CO ₂ incubator. After 72 hours of incubation, the supernatant was collected and its IFN-γ content was determined using an IFN-γ ELISA kit (Thermo, no. 88-7316-77). GraphPad Prism 8 software was used for data processing and graphical analysis.

[117] Figure 10 (A) - (G) shows the ability of each anti-CD3 antibody (including the chimeric SP34 antibody) produced in Example 4.2 to activate human T cells, respectively. When the antibody concentration was 1 μg/ml, PR000511, PR000512, PR000513, and PR000514 produced significantly lower levels of IFN-γ after T cell activation than PR000260; when the antibody concentration was 10 μg/ml, PR000512, PR000513 and PR000514 activated cells to provide slightly lower levels of IFN-γ than PR000260 (Figure 10(A)). When the antibody concentration was 0.5 μg/ml and 5 μg/ml, the level of IFN-γ provided upon cell activation with PR001848 was significantly lower than that for PR000260 (Figure 10(B)). In addition, the activation effect of T cells with antibodies PR002468, PR002469, PR002471, PR002742, PR002833, PR002834, PR002835, PR002836, PR002837, PR001848, PR003886 and PR004616 was also revealed ((C) - (G) in Figure 10 ) in concentrations 0, 5 μg/ml, 5 μg/ml and 50 μg/ml, and the results showed that the level of IFN-γ produced by T cells activated by these antibodies was much lower than the levels provided by activation with PR000260, at this did not detect IFN-γ release by T cells activated by PR002468 and PR002471, and only weak levels of IFN-γ were detected when using 50 μg/ml PR002469 and PR002835; the level of IFN-γ produced by T cells activated with PR002742 and PR003886 was comparable and slightly lower than with PR001848; the level of IFN-γ produced by T cells activated with PR002469 and PR004616 was comparable and significantly lower than with PR001848. The above results indicate that the present invention, by optimizing the sequence of the anti-CD3 antibody, has produced several new antibodies that have varying activation abilities in human T cells, can control the release of different levels of cytokines, and can be used in various application scenarios .

[118] Figure 11 shows the ability of the anti-CD3 antibody in scFv-Fc format obtained in Example 4.3 to activate human T cells. PR000510, PR000623, PR000624, and PR000627 at concentrations of 1 μg/ml and 10 μg/ml showed lower IFN-γ levels than PR000260, but higher than the isotype control antibody, indicating that these four molecules limited cytokine release. regulating T cell activation levels. The above results indicate that the present invention has also produced several stable single chain antibodies in the form of scFv through sequence optimization of anti-CD3 antibodies that exhibit weaker activation of human T cells, exhibiting lower levels of cytokine release, and are suitable for use in scenarios such as the construction of bispecific antibodies.

Example 7 Bispecific Antibody Targeting B7H4 Containing Anti-CD3 Antibody Presented in scFv Format

[119] B7H4, a member of the B7 family of transmembrane proteins, is highly expressed in various solid tumor tissues, such as breast, ovarian and endometrial cancer, while it has no or very weak expression in normal tissues, which makes B7H4 a very specific tumor-associated antigen target. A bispecific antibody molecule targeting both B7H4 and CD3 was designed that is capable of selectively activating T cells in the vicinity of tumor cells by targeting and binding to B7H4 on the surface of tumor cells, thereby allowing specific killing of tumor cells.

7.1. Obtaining Antibody to B7H4

[120] The variable region sequence of the anti-B7H4 antibody can be obtained from WO 2016040724, and the recombinant IgG isotype antibody PR000014 binding the B7H4 target was constructed according to the methods of Example 1.1. The following Table 10 provides information regarding the sequence of the anti-B7H4 antibody PR000014.

Table 10. Anti-B7H4 Antibody Light and Heavy Chain Sequence Information PR000014

7.2. Generation of a bispecific antibody targeting B7H4 containing an anti-CD3 antibody presented in scFv format

[121] The sequence of the anti-B7H4 antibody PR000014 obtained in Example 7.1 and the sequence of the single-chain anti-CD3 antibody PR000627 obtained in Example 4.3 were used to design the bispecific antibody molecule PR002883 targeting B7H4 × CD3, which contained three polypeptide chains: a heavy chain, comprising a scFv of a single chain anti-CD3 antibody (SEQ ID NO: 88), a heavy chain containing a VH of an anti-B7H4 antibody (SEQ ID NO: 86), and a light chain containing a VL of an anti-B7H4 antibody (SEQ ID NO: 83). The structure is shown in Figure 16 (C). Since the molecule had a special asymmetric structure, various amino acid mutations were introduced into the constant regions of the two heavy chains to reduce the formation of homologous heavy chain dimers. At the same time, a triple mutation “LALAPG” (L234A/L235A/P329G) was introduced into the heavy chain constant region to prevent cross-linking and reduce effector function due to Fcγ receptor binding.

[122] Recombinant PR002883 bispecific antibody protein was prepared using the method described in Example 1.1 in combination with plasmids in some ratio (eg, 1:1:1 or other ratios) and subsequent one-step affinity purification. The sequence of the bispecific antibody PR002883 is presented in Table 11; bispecific antibody expression is shown in Table 12.

Table 11. Bispecific Antibody Chains and Related Sequence Information

Table 12. Bispecific antibody expression

[123] Figure 12 (A) shows the results for bispecific antibody PR002883 after one-step purification by SDS-PAGE analysis. It showed that its main by-products are incompletely assembled molecules with a small number of highly polymeric components, the amount of which can be reduced by optimizing the purification step or optimizing the transfection rate of plasmids.

7.3. Binding to tumor cells expressing B7H4

[124] In this example, the ability of a bispecific antibody to bind SK-BR-3 (ATCC, HTB-30) tumor cells expressing human B7H4 was tested. Specifically, a suspension of SK-BR-3 cells was collected, the cell density was adjusted to 1×10 ⁶ /ml and inoculated at 100 μl/well into a 96-well V-bottom plate (Corning, no. 3894); the antibody to be tested was then added at 2-fold the final concentration obtained by the 3-fold gradient dilution in a volume of 100 μl/well. Cells were incubated at 4°C in the dark for 2 hours. After this, the cells were washed twice with 100 μl/well of pre-chilled PBS and centrifuged at 500 g, 4°C, for 5 minutes. The supernatant was removed. Next, 100 μl/well of secondary antibody Alexa Fluor 488 AffiniPure, a goat anti-human IgG antibody specific for the Fcγ fragment (Jackson ImmunoResearch, no. 109-545-098), was added and the cells were incubated in the absence of light at 4°C. C in the dark for 1 hour. The cells were then washed twice with 100 μl/well of pre-chilled PBS and centrifuged at 500 g for 5 minutes. The supernatant was removed. Finally, the cells were resuspended with 200 μl/well of pre-chilled PBS. The fluorescence signal value was read by flow cytometry (BD FACS CANTOII or ACEA NovoCyte) and the resulting data was processed and analyzed using FlowJo software version 10 (FlowJo, LLC). GraphPad Prism 8 software was used for data processing and graphical analysis, and parameters such as binding curves and EC50 values could be obtained using four-parameter nonlinear fitting.

[125] Figure 13(A) shows the binding ability of the monoclonal antibody obtained in Example 7.1 and the bispecific antibody obtained in Example 7.2 to SK-BR-3 cell. It can be seen that the bispecific antibody PR002883 had comparable or even better binding ability compared to the monoclonal antibody PR000014.

7.4. Human T cell binding

[126] The ability of bispecific antibody PR002883 to bind to all types of human T cells was determined by the method described in Example 5. As shown in Figure 13 (B), PR002883 is able to bind to all types of human T cells.

7.5. In vitro killing of SK-BR-3 cells characterized by high levels of B7H4 expression by a bispecific antibody and release of cytokines

[127] To study the ability to kill target cells in vitro mediated by the bispecific antibody B7H4 × CD3, human PBMCs were used as effector cells, and SK-BR-3 cells characterized by a high level of B7H4 expression (ATCC, HTB-30) used as target cells for in vitro killing assays and detection of cytokine release. Specifically, 50 μl of RPMI1640/10% FBS medium was added to each well of an E-plate (ACEA Biosciences Inc., no. 05232368001), followed by equilibration in an incubator containing 5% CO ₂ at 37°C for 30 minutes. , and then the E-plate was placed in the xCELLigence RTCA instrument (ACEA Biosciences) to test for normality. SK-BR-3 density was adjusted to 0.4 x 10 ⁶ cells/ml using RPMI1640/10% FBS media, then inoculated into an E-plate with 50 µl cells/well, and then the E-plate was placed on xCELLigence RTCA overnight to identify the cellular index. PBMC density was adjusted to 4 x 10 ⁶ cells/ml using RPMI1640/10% FBS medium and inoculated into an E-plate with 50 µl cells/well, then the antibody to be tested, presented at 4 times the final concentration obtained by 5 -fold gradient dilution was added at 50 μl/well, with the highest final antibody concentration being 0.2 nM, and there were 7 concentrations for each antibody, the final ratio of effector cells to target cells was 10:1, and provided two repetitions. At the same time, a blank control was added to the plate: SKBR3 + PBMC + RPMI1640/10% FBS medium; The E-plate was incubated at 37°C in an incubator containing 5% CO ₂ for 24 hours. After incubation, the E-plate was placed on the xCELLigence RTCA instrument to detect the cellular index.

[128] The specific killing effect of the antibody was calculated using the cell index determined during the assay in the following formula:

% cell kill = (1 - test sample/blank) * 100%.

[129] Cell culture supernatant was collected to detect the release of the cytokine IFN-γ. Refer to the instructions for use of the IFN-γ Human Uncoated ELISA Kit, Thermo, no. 88-7316-77) for ELISA detection.

[130] As shown in (A) and (B) of Figure 14, bispecific antibody PR002883 could activate T cells to release cytokines such as IFN-γ and effectively kill SK-BR-3 tumor cells. Almost 100% of tumor cells were killed when the bispecific antibody concentration was 0.01 μg/ml (Figure 14(A)).

Example 8 Bispecific Antibody Targeting ROR1 Containing Anti-CD3 Antibody Presented in scFv Format

[131] ROR1 was an inactive transmembrane protein tyrosine kinase protein that is overexpressed in many tumors but is virtually not expressed in normal tissues. ROR1 promotes cell proliferation and migration in chronic lymphocytic leukemia through signaling through the Wnt signaling pathway after interacting with Wnt5a, serving as a receptor, and promotes epithelial-mesenchymal transformation (EMT) in solid tumors. The tumor-specific expression of ROR1 makes it a suitable tumor-associated antigen target for therapeutic development. Design of a bispecific antibody molecule that targets both ROR1 and CD3 can selectively activate T cells in the vicinity of tumor cells by targeting and binding to ROR1 on the surface of tumor cells, thereby allowing specific killing of tumor cells.

8.1. Obtaining Antibody to ROR1

[132] The variable region sequence of the anti-ROR1 antibody could be obtained from WO 2016094873, and the design of the recombinant IgG antibody PR000374 targeting ROR1 according to the methods of Example 1.1. The following Table 13 provides sequence information for the anti-ROR1 antibody PR000374.

Table 13. Sequence listing of anti-ROR1 antibody representing PR000374

8.2. Generation of a bispecific antibody targeting ROR1 containing an anti-CD3 antibody presented in scFv format

[133] Construction of the bispecific antibody molecule PR002885 targeting ROR1 x CD3 was carried out using the sequence of the anti-ROR1 antibody PR000374 obtained in Example 8.1 and the sequence of the single chain anti-CD3 antibody PR000627 obtained in Example 4.3, which contained three polypeptide chains: a heavy chain containing the single chain anti-CD3 antibody scFv (SEQ ID NO: 88), a heavy chain containing the VH anti-ROR1 antibody (SEQ ID NO: 87), and a light chain containing the VL anti-ROR1 antibody (SEQ ID NO: 85). The structure is shown in Figure 16 (C). Since the molecule had a special asymmetric structure, various amino acid mutations were introduced into the constant regions of the two heavy chains to reduce the formation of homologous heavy chain dimers. At the same time, a triple mutation “LALAPG” (L234A/L235A/P329G) was introduced into the heavy chain constant region to prevent cross-linking and reduce effector function due to Fcγ receptor binding.

[134] Recombinant PR002885 bispecific antibody protein was prepared using the method described in Example 1.1 in combination with plasmids in some ratio (eg, 1:1:1 or other ratios) and subsequent one-step affinity purification. The sequence of the bispecific antibody PR002885 is presented in Table 14; bispecific antibody expression is shown in Table 15.

Table 14. Bispecific antibody chains and corresponding sequence number

Table 15. Bispecific antibody expression

[135] Figure 12 (B) shows the results for bispecific antibody PR002885 after one-step purification by SDS-PAGE analysis. It showed that its main by-products are incompletely assembled molecules with a small number of highly polymeric components, the amount of which can be reduced by optimizing the purification step or optimizing the transfection rate of plasmids.

8.3. Binding to tumor cells expressing ROR1

[136] In this example, the ability of a bispecific antibody to bind Panc-1 tumor cells (ATCC, CRL-1469) expressing human ROR1 was tested. Specifically, a suspension of Panc-1 cells was collected, the cell density was adjusted to 1×10 ⁶ /ml, and they were inoculated at 100 μl/well into a 96-well V-bottom plate (Corning, no. 3894); the antibody to be tested was then added at 2-fold the final concentration obtained by the 3-fold gradient dilution in a volume of 100 μl/well. Cells were incubated at 4°C in the dark for 2 hours. After this, the cells were washed twice with 100 μl/well of pre-chilled PBS and centrifuged at 500 g, 4°C for 5 minutes. The supernatant was removed. Next, 100 μl/well of Alexa Fluor 488 AffiniPure goat anti-human IgG Fcγ fragment-specific fluorescent secondary antibody (Jackson ImmunoResearch, no. 109-545-098) was added and the cells were incubated in the absence of light at 4 °C in the dark for 1 hour. The cells were then washed twice with 100 μl/well of pre-chilled PBS and centrifuged at 500 g for 5 minutes. The supernatant was removed. Finally, the cells were resuspended with 200 μl/well of pre-chilled PBS. The fluorescence signal value was read by flow cytometry (BD FACS CANTOII or ACEA NovoCyte) and the resulting data was processed and analyzed using FlowJo software version 10 (FlowJo, LLC). GraphPad Prism 8 software was used for data processing and graphical analysis, and parameters such as binding curves and EC50 values could be obtained using four-parameter nonlinear fitting.

[137] Figure 15(A) shows the binding ability of the monoclonal antibody obtained in Example 8.1 and the bispecific antibody obtained in Example 8.2 to a Panc-1 cell. It can be seen that both the bispecific antibody PR002885 and the monoclonal antibody PR000374 can bind Panc-1.

8.4. Human T cell binding

[138] The ability of the bispecific antibody PR002885 to bind to all types of human T cells was determined by the method described in Example 5. As shown in Figure 15 (B), PR002885 is able to bind to all types of human T cells.

Claims (43)

1. An antibody targeting CD3 containing a light chain variable region (VL) and a heavy chain variable region (VH); Where the amino acid sequence of VH is provided under SEQ ID NO: 44, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 51, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 44, and the amino acid sequence of VL is presented under SEQ ID NO: 60; or the amino acid sequence of VH is provided under SEQ ID NO: 51, and the amino acid sequence of VL is presented under SEQ ID NO: 60; or the amino acid sequence of VH is provided under SEQ ID NO: 45, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 52, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 43, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 43, and the amino acid sequence of VL is presented under SEQ ID NO: 60; or the amino acid sequence of VH is provided under SEQ ID NO: 50, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 47, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 48, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 49, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 53, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 54, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 44, and the amino acid sequence of VL is presented under SEQ ID NO: 57; or the amino acid sequence of VH is provided under SEQ ID NO: 44, and the amino acid sequence of VL is presented under SEQ ID NO: 59; or the amino acid sequence of VH is provided under SEQ ID NO: 51, and the amino acid sequence of VL is presented under SEQ ID NO: 57; or the amino acid sequence of VH is provided under SEQ ID NO: 55, and the amino acid sequence of VL is presented under SEQ ID NO: 58; or the amino acid sequence of VH is provided under SEQ ID NO: 46, and the amino acid sequence of VL is provided under SEQ ID NO: 58. 2. The antibody of claim 1, wherein the antibody comprises a single chain antibody (scFv) comprising a VL-linker-VH or a VH-linker-VL. 3. The antibody of claim 2, wherein the linker is (G ₄ S) _n or a variant thereof, where n is a non-zero natural number. 4. The antibody of claim 3, wherein the amino acid sequence of the linker is SEQ ID NO: 65, SEQ ID NO: 66 or SEQ ID NO: 67. 5. The antibody of claim 4, wherein the amino acid sequence of the scFv is provided by SEQ ID NO: 73, SEQ ID NO: 74, SEQ ID NO: 75, SEQ ID NO: 78, SEQ ID NO: 79 or SEQ ID NO: 80. 6. The antibody of claim 5, wherein the antibody further comprises an Fc, wherein the Fc is linked to the scFv via a hinge. 7. The antibody of claim 1, wherein the antibody further comprises a constant region. 8. The antibody of claim 7, wherein the constant region is a human constant region, wherein the human constant region includes a human light chain constant region and a human heavy chain constant region. 9. The antibody of claim 8, wherein the human light chain constant region is the human κ light chain constant region set forth in SEQ ID NO: 61 or the human λ light chain constant region set forth in SEQ ID NO: 62. 10. The antibody of claim 9, wherein the human heavy chain constant region is represented by the isotypes hIgG1, hIgG2, hIgG3, hIgG4, or a variant thereof. 11. The antibody of claim 10, wherein the human heavy chain constant region is the heavy chain constant region set forth in SEQ ID NO: 63 or SEQ ID NO: 64. 12. A bispecific antibody for treating a CD3-targeted tumor, comprising a first protein functional region and a second protein functional region, wherein the first protein functional region comprises an antibody targeting CD3 according to any one of claims. 1-11. 13. The bispecific antibody of claim 12, wherein the bispecific antibody comprises the following three chains: (1) VL1-linker-VH1-hinge-CH2-CH3 (“overhang”) or VH1-linker-VL1-hinge-CH2-CH3 (“overhang”) "ridge"), related to the first functional region of the protein, (2) VH2-CH1-hinge-CH2-CH3 ("groove"), related to the second functional region of the protein, and (3) VL2-CL, related to the second functional region of the protein ; wherein the second functional region of the protein is an antibody that does not target CD3. 14. The bispecific antibody of claim 13, wherein the second functional region of the protein is an antibody targeting B7H4 or an antibody targeting ROR1. 15. The bispecific antibody of claim 13 or 14, wherein the linker is (G ₄ S) _n , where n is a non-zero natural number. 16. The bispecific antibody according to claim 15, wherein the amino acid sequence of the linker is represented by SEQ ID NO: 65, SEQ ID NO: 66 or SEQ ID NO: 67. 17. Bispecific antibody according to any one of paragraphs. 12-16, wherein the bispecific antibody comprises VL1-linker-VH1-hinge-CH2-CH3 (“bump”), shown under SEQ ID NO: 88, VH2-CH1-hinge-CH2-CH3 (“groove”), shown under SEQ ID NO: 86, and VL2-CL, given under SEQ ID NO: 83, or VL1-linker-VH1-hinge-CH2-CH3 (“protrusion”), shown under SEQ ID NO: 88, VH2-CH1-hinge -CH2-CH3 ("hole"), shown under SEQ ID NO: 87, and VL2-CL, shown under SEQ ID NO: 85. 18. An isolated nucleic acid for producing an antibody, encoding an antibody targeting CD3, according to any one of claims. 1-11. 19. Isolated nucleic acid for producing an antibody, encoding a bispecific antibody according to any one of paragraphs. 12-17. 20. An expression vector for producing an antibody containing the isolated nucleic acid according to claim 18 or 19. 21. A genetically modified cell to produce an antibody, transfected with the expression vector according to claim 20. 22. A pharmaceutical composition for treating a tumor with CD3 as a target, containing an antibody targeting CD3 according to any one of claims. 1-11, bispecific antibody according to any one of paragraphs. 12-17, a genetically modified cell according to claim 21 and a pharmaceutically acceptable carrier. 23. The pharmaceutical composition according to claim 22, wherein the pharmaceutical composition further comprises an antibody directed against an immune checkpoint. 24. Use of an antibody targeting CD3 according to any one of claims. 1-11, a bispecific antibody according to any one of paragraphs. 12-17, an isolated nucleic acid according to claim 18 or 19, an expression vector according to claim 20, a genetically modified cell according to claim 21, or a pharmaceutical composition according to claim 22 or 23 in the manufacture of a medicinal product for the treatment of a tumor with CD3 as a target.