Bifunctional fusion protein composed of il-15 and t cell co-stimulatory molecule antibody

Abstract

Provided is a bifunctional fusion protein composed of IL-15 and a T cell co-stimulatory molecule antibody. The T cell co-stimulatory molecule is 4-1BB, ICOS, or OX40, preferably 4-1BB. The fusion protein comprises: (1) a T cell co-stimulatory molecule antibody; (2) a conjugate of IL-15 and an IL-15R Sushi domain, wherein the IL-15 and the IL-15R Sushi domain are linked by means of a second linker fragment; and (3) a first linker fragment, wherein the first linker fragment is used for linking a heavy chain Fc region of the T cell co-stimulatory molecule antibody and the conjugate of IL-15 and the IL-15R Sushi domain, and can be cleaved by matrix metalloproteinases.

Description

Bifunctional fusion protein composed of IL-15 and T cell co-stimulatory molecule antibody Technical field The present invention belongs to the field of biomedical technology. Specifically, it relates to a bifunctional fusion protein composed of IL-15 and T cell co-stimulatory molecule antibody. Background technology Interleukin-15 (IL-15) is a glycoprotein with four α helices, consisting of 114 amino acids and having a molecular weight of 14-15 kDa [1]. IL-15 is mainly secreted by dendritic cells (DC), macrophages and monocytes and acts on other cells. The receptors of IL-15 include IL-15Rα (CD215), IL-2Rβ (CD122) and IL-2Rγ (CD132). CD215 is mainly expressed on the surface of DC, macrophages and monocytes. After being secreted by cells, IL-15 can directly bind to CD215 and be trans-presented to T cells and NK cells expressing IL-2Rβ and IL-2Rγ. After IL-15 binds to the receptor, it can phosphorylate downstream STAT3 and STAT5 and activate the JAK1 and JAK3 signaling pathways [2-4]. IL-15 mainly promotes the proliferation and activation of T cells and NK cells and maintains the homeostasis of lymphocytes. IL-15 can also maintain the survival of memory CD8 + T cells and effector T cells and avoid the activation-induced cell death effect (AICD) of T cells [5]. IL-15 does not bind to IL-2Rα and does not induce the proliferation of regulatory T cells (Treg). In cancer treatment, the use of monomeric IL-15 in clinical practice is severely limited due to its short half-life and weak activity. Since IL-15 needs to bind to IL-15Rα to have strong affinity, the complex form of IL-15 and IL-15Rα has become the main form used in current clinical research. In addition, the mutant N72D of IL-15 can further enhance the activity of IL-15 [6]. By increasing the half-life and activity of IL-15, the anti-tumor activity of IL-15 is improved to a certain extent. However, when used alone, the anti-tumor effect is extremely weak. Therefore, IL-15 is currently mostly used in combination with other treatment methods, such as PD-1 antibodies, CTLA-4 antibodies, CD20 antibodies, and CD52 antibodies. Since NK cells in peripheral tissues express more IL-2Rβ, IL-15 preferentially binds to peripheral NK cells. These forms of IL-15 have no selectivity for the proliferation and activation of NK cells and T cells, causing a large expansion of peripheral T cells, especially NK cells. The increase in their activity and half-life also leads to an increase in side effects. Numerous studies have shown that the toxicity caused by IL-15 is mainly caused by peripheral NK cells [7]. To address the problem of peripheral toxicity caused by IL-15, existing studies mainly fuse IL-15 with other antibodies or polypeptides to change the targeting of the fusion protein, reduce the binding of IL-15 in the periphery, and increase its binding to cells within the tumor, enabling the fusion protein to accumulate in the tumor and exert the activating effect of IL-15 on immune cells. For example, by fusing the RGD polypeptide sequence with IL-15 / IL-15Rα, the RGD polypeptide can specifically bind to integrin αvβ3 highly expressed on tumor cells and endothelial cells, enabling the fusion protein to accumulate within the tumor, increase the drug concentration within the tumor, and exert the activity of IL-15 / IL-15Rα [8]. Considering that the affinity of IL-15 for IL-15Rα is approximately 38 pM, and the affinity of IL-15 for CD122 / CD132 also reaches 1 nM [9, 10], which is relatively close to the affinity of antibodies (10 -8 ~10 -10 M), after fusing IL-15 and an antibody, the targeting of the fusion protein cannot be determined by the antibody. It is necessary to reduce the affinity between IL-15 and the receptor to better target the fusion protein to the tumor. Some studies have mutated IL-15 to reduce its affinity for the receptors IL-15Rα and CD122 / CD132, and at the same time screened for a PD-1 antibody with high affinity. The PD-1 antibody was fused with IL-15 to target the fusion protein to T cells highly expressing PD-1 within the tumor. IL-15 can directly act on T cells within the tumor to exert an anti-tumor effect

[0011] In addition to changing the targeting of the fusion protein by reducing the affinity of IL-15 by mutating it, the extracellular domain of the receptor IL-2Rβ can also be used to block IL-15, and the extracellular domain can be connected to IL-15 through a substrate sequence that can be recognized by matrix metalloproteinase-14 (MMP-14). The MMP-14 recognition sequence in the tumor is cleaved, releasing the activity of IL-15 and activating the anti-tumor immune response.

[0012] . After continuous improvement, IL-15 can effectively amplify CD8+T cells and NK cells in tumors and enhance anti-tumor response. However, tumors often show an immunosuppressive state, and the proliferating immune cells quickly become exhausted or their activity is suppressed by the immunosuppressive cells in the tumor, and they cannot exert anti-tumor effects. The immunosuppressive cells in the tumor microenvironment (TME) mainly include tumor-associated macrophages (TAM), myeloid-derived suppressor cells (MDSC) and Tregs.

[0013] How to effectively inhibit these cells from exerting immunosuppressive effects has also become a key issue that needs to be solved in immunotherapy. Tregs in tumors mainly inhibit anti-tumor immune responses in the following ways: 1) High expression of CTLA-4 competes with DC expression CD80 and CD86; 2) High expression of CD25 to compete with effector T cells for IL-2; 3) Secretion of inhibitory cytokines such as TGF-β and IL-10 to inhibit effector T cell function; 4) Regulate the metabolic levels of tryptophan and adenosine to inhibit effector T cell function

[0014] The increase in the ratio of Tregs / CD8+T cells in tumors is often associated with a poor prognosis. Although some cancer types are more special, in most cancer types, the increase in the number of Tregs is not conducive to the anti-tumor immune response. Therefore, effectively eliminating Tregs in tumors without affecting peripheral Tregs can effectively enhance the anti-tumor immune response. Tregs in the tumor microenvironment often highly express CD25, CTLA-4, ICOS, OX40, CCR4, and CCR8, and antibodies against these molecules are often used to deplete Tregs. Although some of these molecules, such as CD25 and CTLA-4, are highly expressed in intratumoral Tregs, they are also transiently highly expressed on effector T cells. Using the corresponding antibodies to deplete Tregs will also deplete effector T cells, resulting in a significant decrease in the anti-tumor effect. Using antibodies against appropriate marker molecules to selectively deplete Tregs in tumors has become the key to breaking Treg-induced immune tolerance. ICOS, OX40, GITR, and 4-1BB in the Tumor necrosis factor receptor superfamily (TNFRSF) are often constitutively expressed on Tregs and upregulated in tumors. These molecules have received increasing attention. Initially, antibodies against these targets were often used to activate co-stimulatory molecules to activate T cells. Although some studies have shown that antibodies against these molecules can be used to deplete Tregs, there have been no clinical trials for Treg depletion. The 4-1BB molecule is constitutively expressed on Tregs and DCs and inducibly expressed in T cells and NK cells. In T cells, TCR stimulation or CD3 signaling can induce upregulation of 4-1BB expression, which synergizes with TCR signaling to promote the secretion of IL-2 and IFN-γ by T cells and the proliferation of T cells.

[0015] Further studies have shown that 4-1BB co-stimulation not only induces effector cells to produce effector factors but also facilitates the differentiation of memory T cells and effector cells and protects T cells from apoptosis.

[0016] In tumors, the 4-1BB molecule is upregulated in both T cells and NK cells, especially highly expressed in Tregs. Among them, the function of the 4-1BB molecule in Th1 CD4 + T cells and Tregs is not yet clear. Activation of the 4-1BB signal helps activate NK cells and enhance their ADCC and cytotoxic effects.

[0017] In addition, the high expression of the 4-1BB molecule in Tregs makes it an important target for effectively reducing and eliminating intratumoral Tregs, providing an auxiliary means for immunotherapy. In a mouse model, using a 4-1BB depletion antibody can effectively inhibit tumor growth and clear intratumoral Tregs. The effect of clearing Tregs is equivalent to that of CTLA-4 and OX40 antibodies. At the same time, after treatment with the 4-1BB antibody, the remaining Tregs in the tumor show a weaker inhibitory phenotype.

[0018] Although some CD8 in tumors +T cells also express 4-1BB. Using 4-1BB antibodies may affect the number of these T cells while deleting Tregs. However, multiple studies have shown that the anti-tumor effect of using 4-1BB antibodies of IgG1 subtype in mouse tumor models is far less than that of IgG2a subtype antibodies, indicating that the clearance of Tregs is superior to activating T cells with 4-1BB antibodies for the generation of anti-tumor immune responses [19, 20]. Based on this, the present invention is proposed. [References] [1] FEHNIGER, T.A. Interleukin 15: biology and relevance to human disease[J]. 2001, 97(1): 14-32. [2] WALDMANN, THOMAS A. The biology of interleukin-2 and interleukin-15: implications for cancer therapy and vaccine design[J]. Nature Reviews Immunology, 2006, 6(8): 595-601. [3] WALDMANN T A, MILJKOVIC M, CONLON K C. Interleukin-15(dys)regulation of lymphoid homeostasis: Implications for therapy of autoimmunity and cancer[J]. Journal of Experimental Medicine, 2020, 217(1). [4] FEHNIGER T A, CALIGIURI M A. Fehniger TA,Caligiuri MAInterleukin 15:biology and relevance to human disease.Blood 97:14-32[J]. Blood, 2001, 97(1): 14-32. [5] MARKS-KONCZALIK J, DUBOIS S, LOSI J M, et al. IL-2-induced activation-induced cell death is inhibited in IL-15 transgenic mice[J]. Proc Natl Acad Sci U S A, 2000, 97(21): 11445-50. [6] ZHU X, MARCUS W D, XU W, et al. Novel Human Interleukin-15 Agonists[J]. Journal of Immunology, 2009, 183(6): 3598-607. [7] GUO Y, LUAN L, RABACAL W, et al. IL-15 Superagonist-Mediated Immunotoxicity: Role of NK Cells and IFN-γ[J]. The Journal of Immunology, 2015: jimmunol.1500300. [8] CHEN S, HUANG Q, LIU J, et al. A targeted IL-15 fusion protein with potent anti-tumor activity[J]. Cancer Biology&Therapy, 2015, 16(9): 1415-21. [9] Soluble interleukin-15 receptor alpha(IL-15R alpha)-sushi as a selective and potent agonist of IL-15 action through IL-15R beta / gamma. Hyperagonist IL-15 x IL-15R alpha fusion proteins[J]. Journal of Biological Chemistry, 2006, 281.

[0010] GIRI J G, AHDIEH M, EISENMAN J, et al. Utilization of the beta and gamma chains of the IL receptor by the novel cytokine IL[J]. The EMBO Journal, 1994, 13(12).

[0011] An Engineered IL15 Cytokine Mutein Fused to an Anti-PD1 Improves Intratumoral T-cell Function and Antitumor Immunity[J]. Cancer Immunology Research, 2021, 9(10):1141 - 57.

[0012] Tumor-conditional IL-15 pro-cytokine reactivates anti-tumor immunity with limited toxicity[J]. Cell Research.

[0013] VONDERHEIDE, ROBERT H, WEIPING, et al. Understanding the tumor immune microenvironment(TIME)for effective therapy[J].

[0014] OHUE Y, NISHIKAWA H. Regulatory T(Treg)cells in cancer: Can Treg cells be a new therapeutic target?[J]. Cancer Science, 2019, 110(7).

[0015] MELERO I, BACH N, HELLSTR M K E, et al. Amplification of tumor immunity by gene transfer of the co-stimulatory 4-1BB ligand: synergy with the CD28 co-stimulatory pathway[J]. European Journal of Immunology, 1998.

[0016] SHUFORD W W, KLUSSMAN K, TRITCHLER D D, et al. 4-1BB Costimulatory Signals Preferentially Induce CD8+ T Cell Proliferation and Lead to the Amplification In Vivo of Cytotoxic T Cell Responses[J]. Journal of Experimental Medicine, 1997, 186(1): 47-55.

[0017] KOHRT H E, COLEVAS A D, HOUOT R, et al. Targeting CD137 enhances the efficacy of cetuximab[J]. Journal of Clinical Investigation, 2019, 129(6): 2595-.

[0018] MAJ T, WANG W, CRESPO J, et al. Oxidative stress controls regulatory T cell apoptosis and suppressor activity and PD-L1-blockade resistance in tumor[J]. Nature Immunology, 2017.

[0019] ASHLEY V, MENK, NICOLE E, et al. 4-1BB costimulation induces T cell mitochondrial function and biogenesis enabling cancer immunotherapeutic responses[J]. The Journal of experimental medicine, 2018.

[0020] BUCHAN S L,DOU L,REMER M,et al.Antibodies to Costimulatory Receptor 4-1BB Enhance Anti-tumor Immunity via T Regulatory Cell Depletion and Promotion of CD8 T Cell Effector Function[J].Immunity,2018. Summary of the Invention The present invention first relates to a bifunctional fusion protein composed of IL-15 and an antibody against a T cell costimulatory molecule; The T cell costimulatory molecules are: 4-1BB, ICOS, OX40; preferably 4-1BB; The fusion protein includes: (1) An antibody against a T cell costimulatory molecule; (2) A conjugate of IL-15 and the Sushi domain of IL-15R, wherein IL-15 and the Sushi domain of IL-15R are connected by a second linker segment; and (3) A first linker segment; the first linker segment is used to connect the heavy chain Fc region of the antibody against the T cell costimulatory molecule and the conjugate of IL-15 and the Sushi domain of IL-15R, and can be cleaved by a matrix metalloproteinase; Preferably, the antibody against the T cell costimulatory molecule is an IgG1 type antibody. Preferably, the first linker segment includes: 2-4 G4S linker units and a short peptide contained therein that can be recognized and cleaved by a matrix metalloproteinase; More preferably, the amino acid sequence of the short peptide that can be recognized and cleaved by a matrix metalloproteinase is as shown in SEQ ID NO.6; Most preferably, the amino acid sequence of the first linker segment is as shown in SEQ ID NO.12 or SEQ ID NO.11; Preferably, the second linker segment includes: 2-5 GnS linker units, wherein n is an integer from 1 to 4; more preferably, the amino acid sequence of the second linker segment is as shown in SEQ ID NO.5. Furthermore, the bifunctional fusion protein includes: (1) The first structural unit, from the N-terminus to the C-terminus in sequence: 4-1BB antibody heavy chain, the first linker segment, the conjugate of IL-15 and IL-15R Sushi domain; (2) The second structural unit: the 4-1BB antibody light chain paired with the 4-1BB antibody heavy chain; Furthermore, the bifunctional fusion protein dimerizes through the Fc region of the 4-1BB antibody heavy chain to form a homodimer. Preferably, The 4-1BB antibody is an antibody formed by fusing the Fab region of a human or murine antibody with the Fc region of IgG1, The amino acid sequence of the heavy chain (VH+CH1) of the murine 4-1BB antibody is as shown in SEQ ID NO.2; The amino acid sequence of the heavy chain (VH+CH1) of the human 4-1BB antibody is as shown in SEQ ID NO.10; The amino acid sequence of the light chain (VL-CL) of the murine 4-1BB antibody is as shown in SEQ ID NO.1; The amino acid sequence of the light chain (VL-CL) of the human 4-1BB antibody is as shown in SEQ ID NO.9; The Fc region of the IgG1 is the Fc region of human IgG1, or a variant of the Fc region of human IgG1 with the ADCC effect knocked out; The amino acid sequence of the Fc region of human IgG1 is as shown in SEQ ID NO.3, and the amino acid sequence of the variant of the Fc region of human IgG1 with the ADCC effect knocked out is as shown in SEQ ID NO.4. The IL-15 and the IL-15R Sushi domain are human or murine proteins, The amino acid sequence of the murine IL-15 is as shown in SEQ ID NO.7, and the amino acid sequence of the murine IL-15R Sushi domain is as shown in SEQ ID NO.8; The amino acid sequence of the human IL-15 is as shown in SEQ ID NO.13, and the amino acid sequence of the human IL-15R Sushi domain is as shown in SEQ ID NO.14. The present invention also relates to a nucleotide fragment encoding the bifunctional fusion protein, a vector containing the nucleotide fragment, and a host. The present invention also relates to the following applications of the bifunctional fusion protein, or the nucleotide fragment encoding the bifunctional fusion protein, or the vector containing the nucleotide fragment, or the host: (1) Preparation of anti-tumor drugs; (2) Preparation of combined anti-tumor drugs. Preferably, the anti-tumor drug is: a drug that reduces and clears intratumoral Treg cells and induces the proliferation of CD8 + T cells; Preferably, the tumor is: a tumor in an immunosuppressive state; Preferably, the combined anti-tumor drug is: a combined anti-tumor drug in combination with an immune checkpoint inhibitor; the immune checkpoint inhibitors include but are not limited to: PD-1 / PD-L1 antibodies, CTLA4 antibodies, TIGIT antibodies, LAG-3 antibodies, TIM-3 antibodies. The beneficial effects of the present invention are as follows. (1) IL-15 has a strong ability to expand CD8+ T cells and NK cells, but the expanded T cells do not show high anti-tumor ability. By changing the immunosuppressive state in the tumor, the therapeutic effect of IL-15 can be improved; (2) Based on the ability of the 4-1BB antibody to clear Tregs in the tumor and relieve the inhibitory state of Tregs on effector T cells, the bifunctional antibody of the fusion protein composed of the 4-1BB antibody and IL-15 can simultaneously exert the functions of clearing Tregs and expanding CD8+ T cells; (3) By designing a short substrate sequence that can be recognized and cleaved by MMP highly expressed in the tumor between the antibody and IL-15, the short linker peptide can maintain the activity of IL-15 and avoid peripheral toxicity. At the same time, after the fusion protein enters the tumor, the linker peptide can be cleaved by MMP to release IL-15 and restore its activity, which has great development potential. (4) This fusion protein can overcome the tolerance of the PD-L1 antibody and has a synergistic effect when used in combination with the PD-L1 antibody. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1. The Anti-4-1BB-hIgG1 antibody can control the growth of small tumors, but has poor therapeutic effects in large tumors. 1A. Inhibitory effect of the Anti-4-1BB antibody on the growth of small tumors; 1B. Inhibitory effect of the Anti-4-1BB antibody on the growth of large tumors; 1C. Clearance effect of the Anti-4-1BB antibody on Tregs in the tumor; 1D. Flow cytometry diagram of the clearance effect of the Anti-4-1BB antibody on Tregs in the tumor. Figure 1. The Anti-4-1BB-hIgG1 antibody can control the growth of small tumors, but has poor therapeutic effects in large tumors. 1A. Inhibitory effect of the Anti-4-1BB antibody on the growth of small tumors; 1B. Inhibitory effect of the Anti-4-1BB antibody on the growth of large tumors; 1C. Clearance effect of the Anti-4-1BB antibody on Tregs in the tumor; 1D. Flow cytometry diagram of the clearance effect of the Anti-4-1BB antibody on Tregs in the tumor. Figure 2. Antitumor effects of different doses of IL-15 and its effect on lymphocyte expansion. 2A. High-dose IL-15 inhibits tumor growth; 2B. IL-15 significantly expands immune cells in the periphery and within tumors; 2C. IL-15 treatment causes an increase in ALT levels in the blood. Figure 3. IL-15 and Anti-4-1BB combination therapy has a synergistic effect. 3A. IL-15 cannot overcome the inhibitory effect of Tregs on CD8+ T cells; 3B. Synergistic antitumor effect of Anti-4-1BB antibody and IL-15. Figure 4. Construction and expression of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein. 4A. Schematic diagram of the structure of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein; 4B. SDS-PAGE analysis of the constructed Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein, R is the Reducing band, and NR is the Non-reducing band. Figure 5. In vitro and in vivo functional detection of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein. 5A. MMP-mediated in vitro cleavage of the fusion protein. Buffer represents the band of the fusion protein incubated in a cleavage solution without MMP14 for 12 hours (h). The bands of MMP 6h or 12h are the protein bands of the fusion protein after cleavage by MMP14 for 6h or 12h; 5B. Activation of the STAT5 signaling pathway in the CTLL-2 cell line by the fusion protein and the cleaved protein; 5C. Expansion effect of the fusion protein on immune cells in peripheral blood. Figure 6. Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein has a synergistic therapeutic effect. 6A. Tumor inhibitory effect of different doses of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein (MC38 tumor model); 6B. Effect of different doses of the fusion protein on the body weight of mice after treatment; 6C. Comparison of the therapeutic effects of the fusion protein with Mix and the fusion protein without MMP (the MMP sequence is replaced by 2G4S); 6D. Results of the re-challenge test in the Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein treatment group (MC38 tumor model). Figure 7. The therapeutic effect of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein is independent of innate immune cells, and the therapeutic effect of the fusion protein on Rag-bearing tumor mice. Figure 8. The function of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein depends on CD8 T cells. 8A. One day after injection of 200 μg of depleting antibody, the depletion efficiency of CD4 T cells and CD8 T cells in peripheral blood was detected by flow cytometry; 8B. Tumor growth curves of different treatment groups. Figure 9. Intratumoral T cells play a key role in the treatment with Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein, and the therapeutic effect of the fusion protein on tumor-bearing mice injected with FTY720. Figure 10. The immune response induced by local treatment of in situ tumors can control the growth of distant tumors. 10A. Therapeutic effects of different treatment groups on the left and right sides (drug administration side) of tumors in the MC38 model; 10B. Therapeutic effect of in situ tumors in the B16F10 tumor model; 10C. Therapeutic effect on lung metastatic tumors in the B16F10 tumor model Figure 11. Treatment with Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein deletes intratumoral Tregs and increases the ratio of CD8 T cells to Tregs. 11A. Flow cytometry plots of intratumoral Treg cells in different treatment groups; 11B. Proportion of intratumoral Tregs in CD4+T in different treatment groups; 11C. Flow cytometry plots of intratumoral CD8+T cells in different treatment groups; 11D. Proportion of intratumoral CD8+T in CD45+ cells in different treatment groups. Figure 12. The function of the antibody partly depends on the binding of Fc to FcγR. Figure 13. aPDL1 antibody can cooperate with Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein to enhance the anti-tumor effect. Figure 14. Anti-OX40-MMP-IL-15-IL15Ra fusion protein has a good anti-tumor effect. Figure 15. In vitro functional detection of humanized Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein. 15A. SDS-PAGE detection of humanized fusion protein; 15B. In vitro cleavage of the fusion protein by MMP. The band of Buffer-expressed fusion protein incubated in the cleavage solution without MMP14 for a certain time, and the bands of MMP 6h or 12h are the protein bands of the fusion protein after cleavage by MMP14 for 6h or 12h; 15C. Activation of STAT5 signal and EC50 of the fusion protein and the cleaved protein on the HEK-IL2 cell line. Figure 16. Detection of the immune cell expansion activity of humanized Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein in hu-PBMC mice in vivo. Figure 17. Therapeutic efficacy and safety of humanized Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein in hu-PBMC tumor-bearing mice. 17A. Antitumor effects of the fusion protein and combination treatment groups; 17B. Body weight changes in the fusion protein and combination treatment groups. Detailed implementation manners Experimental materials 1. Strains and plasmids Strains: Top10 E. coli and DH5α E. coli competent cells (Beijing TransGen Biotech Co., Ltd.) Plasmids: pEE12.4-IgGκ, containing the signal peptide of mouse IgGκ, for the expression of cytokines and antibodies. All primers used in the experiments were designed by DNAMAN software and synthesized by Genewiz. 2. Experimental animals Wild-type C57BL / 6, BALB / c mice and BALB / c-nude mice were purchased from Vital River Laboratories (Beijing, China). Unless otherwise specified, all experiments used female mice aged 8 - 10 weeks. All mice were housed in a specific pathogen-free (SPF) barrier environment. The animal husbandry and experimental operations complied with the relevant regulations of the Animal Management Committee of the Institute of Biophysics, Chinese Academy of Sciences. 3. Cell lines MC38 is a mouse colorectal cancer cell line with a C57 background. CT26 is a mouse colorectal cancer cell line with a BALB / C background. The above cell lines were all cultured in DMEM complete medium (containing 10% inactivated fetal bovine serum, 2 mmol / l L-glutamine, 0.1 mmol / l non-essential amino acids, 100 U penicillin and 100 μg / ml streptomycin). TIB-210TM hybridoma cell line (ATCC TIB-210), used for expressing the deletion antibody of CD8+ T cells (clone: 2.43). TIB-207TM hybridoma cell line (ATCC-TIB-207), used for expressing the deletion antibody of CD4+ T cells (clone: GK1.5). HB-197TM hybridoma cell line (ATCC-HB-197), used for expressing the antibody that blocks mouse FcγRII / III (clone: 2.4G2). The FreeStyleTM 293F cell line (Invitrogen) is a suspension cell line derived from the HEK293 cell strain and is cultured in SMM293-TII or CD OptiCHOTM medium. It is mainly used for transient transfection to express fusion proteins. The CTLL-2 cell line is a murine T cell line used to detect the biological activity of IL-2. The above cell lines are cultured in RPMI1640 complete medium (containing 10% heat-inactivated fetal bovine serum, 2 mmol / L L-glutamine, 0.1 mmol / L non-essential amino acids, 100 U penicillin, 100 μg / ml streptomycin, and 100 IU / ml recombinant IL-2). IL15&IL-15R Sushi The amino acid sequence of murine IL-15 is shown in SEQ ID NO.7, and the amino acid sequence of the murine IL-15R Sushi region is shown in SEQ ID NO.8. The gene sequence of human IL-15 is shown in SEQ ID NO.13, and the gene sequence of the human IL-15R Sushi region is shown in SEQ ID NO.14. The sIL15 mentioned in the examples corresponds to IL15 RA Fc in CN201810420739.6, and its amino acid sequence is shown in SEQ ID NO.15: The Fab region of the Anti-4-1BB antibody The light chain variable region + light chain constant region (light chain) of a human or murine 4-1BB antibody; The heavy chain variable region + CH1 region (heavy chain) of a human or murine 4-1BB antibody. The Fc region of the Anti-4-1BB antibody The Fc region of human IgG1 (CH2 + CH3) or a variant of the Fc region of human IgG1 with ADCC effect knocked out. Mouse tumor inoculation and treatment (1) Tumor inoculation and measurement: Establishment of a tumor model, 5×10 5 MC38 single cells are suspended in 100 μL of PBS and subcutaneously inoculated into the back of C57BL / 6 mice; When performing a Re-challenge experiment with the same tumor cells on mice with tumor regression, the inoculation number of tumor cells is 5 times that of the initial tumor modeling, and the inoculation site is subcutaneously on the other side of the back of the mice. The tumor size is monitored twice a week, and the long diameter (a), short diameter (b), and height (c) of the tumor are measured using vernier calipers. The volume of the mouse tumor = a × b × c / 2. (2) Treatment: The antibody or antibody fusion protein is administered by intraperitoneal injection. In some experiments, intratumoral administration is also used. The specific dosage will be described in the specific experiments. Preparation of monoclonal antibody (mouse ascites method) The CD4 + T cell-depleting antibody GK1.5, CD8 + T cell-depleting antibody TIB210, and the FcRII / III blocking antibody are all from the corresponding hybridoma cells (TIB-210TM, TIB-207TM, HB-197TM), and are produced and purified by our laboratory. Cell depletion in mice CD4 + T cells, CD8 + Depletion of CD4 On the day before fusion protein treatment, 200 μg of GK1.5 or TIB210 antibody was intraperitoneally injected respectively to deplete CD4 + T cells, CD8 + T cells, and then injected once every 3 days, and the injection times were adjusted according to the treatment cycle. The depletion efficiency was detected by flow cytometry. Blockade of T cell egress FTY720 (purchased from Sigma) is an immunosuppressant that can reduce the egress of T cells from lymphoid organs to the peripheral blood circulation. In the present invention, FTY720 blockade was performed at different stages of mouse tumor inoculation to alter the tumor microenvironment. Blockade was performed during the treatment of mouse tumors: 20 μg of FTY720 was intraperitoneally injected 1 day before tumor treatment, and then 10 μg was intraperitoneally injected every other day. The blockade time was determined according to the treatment cycle, which would result in no newly migrated T cells in the tumor tissue during the tumor treatment process. With the FTY720 blockade protocol, the importance of infiltrating lymphocytes in tumor tissue can be studied. Unless otherwise specified, the molecular structure of the Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein molecule in Examples 1-5 below is the molecular structure shown in Figure 4A. Example 1. The Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein significantly improves the treatment effect compared with single treatment 1. Anti-4-1BB antibody is effective against small tumors but ineffective against large tumors We used an Anti-4-1BB antibody of the human IgG1 subtype and verified the function of the antibody in the MC38 tumor model. Compared with the untreated control group, the antibody treatment group could effectively control the growth of small tumors (Figure 1A), but the antibody treatment failed to completely eliminate the tumors when the tumors were large (Figure 1B). Different doses of the 4-1BB antibody could eliminate Tregs in the tumors to varying degrees (Figure 1C). C57BL / 6 mice were subcutaneously inoculated with 5×10 5 MC38 tumor cells. Tumor-bearing mice began to be grouped for treatment on the 7th or 12th day (n = 5 / group): intraperitoneal injection (ip) of 200 μg Anti-4-1BB antibody, once every three days for a total of three treatments. The results showed that antibody treatment alone had a good anti-tumor effect only on small tumors. 2. Anti-tumor effects of different doses of IL-15 and its effect on lymphocyte expansion Since the 4-1BB antibody of the Human IgG1 subtype can effectively delete Tregs in the tumor, and Tregs are a type of immunosuppressive cell with a large inhibitory effect on T cell function in the tumor, but the Anti-4-1BB antibody cannot treat advanced tumors well, we need to further exert the function of T cells. Considering that the deletion of Tregs relieves the inhibition of T cells, but the number of T cells does not change significantly. To effectively expand CD8 + T cells, we chose to use IL-15 as an adjuvant to expand T cells. To verify the expansion effect of IL-15, we treated MC38 tumor-bearing mice with two different doses of sIL-15, namely the fusion protein in the form of IL-15-IL15Ra-Fc, at 10 μg and 50 μg, once every 3 days for 3 consecutive times. Two days after the second treatment, the spleen and tumor tissues of the mice were taken to detect the number and proportion of immune cells. Seven days after the last treatment, the ALT level in the peripheral blood of the mice was detected. The results showed that there was a significant difference in the tumor volume between the mice in the 50 μg dose treatment group and the untreated group, while there was no significant difference in the tumor volume between the mice in the 10 μg dose treatment group and the untreated group (Figure 2A). Both the low-dose and high-dose groups could significantly increase the expansion of CD8 + T cells in the spleen and tumors (Figure 2B). After IL-15 treatment, the ALT level in the peripheral blood of the mice increased significantly (Figure 2C). 3. IL-15 and Anti-4-1BB combination therapy has a synergistic effect IL-15 has a good proliferative effect on T cells. However, Tregs cells have a strong inhibitory effect on the proliferation of T cells. To detect whether the amplification of T cells by IL-15 can overcome the inhibitory effect of Tregs in vitro, anti-CD3 antibody was used to stimulate CFSE-labeled CD8 + T cells in vitro, 1 μg / mL of sIL-15 was added, and different proportions of Tregs were added simultaneously. After incubation for 3 days, the change in the intensity of CFSE in CD8 + T cells was detected to measure the proliferation ratio of T cells. The results showed that the addition of IL-15 could not overcome the inhibitory effect of Tregs on the amplification of CD8 + T cells. Compared with the group without adding Tregs, the proliferation ratio of T cells decreased after the addition of Tregs (Figure 3A) (the light-colored peaks in the figure represent the division generations of CD8 + T cells. As the number of added Treg cells increased, the division generations of CD8 + T cells decreased significantly). Given the limited therapeutic effects of 4-1BB antibody and IL-15 itself, and their complementary functions, we combined the two to detect whether they have a synergistic effect on anti-tumor efficacy. MC38 tumor-bearing mice were used. Combination treatment was started on the 13th day after tumor inoculation, with 200 μg of Anti-4-1BB antibody and 15 μg of sIL-15, administered by intraperitoneal injection once every 3 days for 3 consecutive times, and the tumor volume was measured. The results showed that the combined use of Anti-4-1BB antibody and sIL-15 had a better effect than their respective single-agent treatments, with a synergistic anti-tumor effect (Figure 3B). 4. Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein has a synergistic anti-tumor effect 4.1. Construction and expression of the fusion protein The combined treatment of Anti-4-1BB antibody and sIL-15 has a synergistic anti-tumor effect. However, considering that the treatment with sIL-15 can cause peripheral toxicity, how to design a fusion protein that can both delete tumor-infiltrating Tregs by the antibody and expand tumor-infiltrating T cells by sIL-15, while maximizing the activity of IL-15 within the tumor and not expanding immune cells peripherally to avoid peripheral toxicity. We designed a fusion protein by connecting IL-15-IL-15Ra to the carboxyl terminus of the antibody. The antibody and IL-15 were connected by a short Linker to shield the activity of IL-15. After replacing the short Linker with a substrate sequence recognized by MMP highly expressed in tumors, the fusion protein could be cleaved in tumors to release the activity of IL-15. Each molecule contains a complete Anti-4-1BB antibody heavy chain and an IL-15-IL-15Ra molecule, and there is an MMP substrate sequence between the antibody and the cytokine (Figure 4A). The light chain of the antibody and the plasmid of this fusion protein were co-transfected into 293F cells. After seven days, the cell supernatant was collected. After centrifuging the supernatant, the protein was purified using a protein A column, and the purified antibody was identified by protein gel. SDS-PAGE analysis of the protein molecular composition proved that the molecule could express the light chain of the Anti-4-1BB antibody and the fused heavy chain (Figure 4B). 4.2. Verification of the function of the fusion protein (1) In vitro activity detection of the fusion protein The MMP substrate sequence in the fusion protein can be cleaved by MMP2, MMP9 or MMP14. We co-incubated the fusion protein with the activated MMP14 enzyme in vitro and then performed SDS-PAGE analysis to verify the in vitro cleavage effect of the fusion protein. The results showed that the fusion protein could be effectively cleaved in vitro (Figure 5A). IL-15 often induces the activation of the downstream STAT5 signaling pathway after binding to its receptor. We used the CTLL2-STAT5-Luc cell line, blocked the 4-1BB molecules expressed in CTLL-2 cells with the Anti-4-1BB antibody of the same clone, and then added different concentrations of the fusion protein to stimulate the cells. After 4 hours, the luciferase activity in the cells was detected. The results showed that the ability of the fusion protein to activate the STAT5 signaling pathway decreased. After in vitro enzymatic cleavage, the activation ability was completely restored (Figure 5B), indicating that the activity of IL-15 in the fusion protein was well blocked and its activity was fully restored after MMP cleavage. (2) Detection of the activity of the fusion protein in expanding immune cells in mice in vivo The activity of IL-15 in the fusion protein was well blocked. To further verify whether the fusion protein could expand immune cells in the peripheral tissues of tumor-bearing mice, we intraperitoneally injected 100 μg of the fusion protein or an equimolar mixture of Anti-4-1BB antibody and sIL-15, and treated once every three days. Two days after the second treatment, the number of CD4 + T cells, CD8 + T cells and NK cells in the peripheral blood was detected. The results showed that, compared with the combined treatment group, the fusion protein did not cause a large expansion of T cells and NK cells in the peripheral blood, indicating good safety (Figure 5C). 4.3. The fusion protein has good safety and antitumor effects MC38 tumor cells were subcutaneously inoculated into C57BL6 mice, and treatment began on D13 after tumor inoculation; 10, 50, and 200 μg of the fusion protein were respectively injected intraperitoneally. Treatment was carried out three times on D13, D16, and D19, and the tumor size was measured twice a week. 5 The results showed that as the treatment dose of the fusion protein increased, the treatment effect of the fusion protein also gradually improved (Figure 6A). Meanwhile, we measured the body weight of the mice every 1 - 2 days before and after treatment to verify the effect of the fusion protein on the body weight of the mice. The results showed that as the dosage increased, the body weight of the mice did not change significantly (Figure 6B). In addition, we compared the antitumor effects of 50 μg of the fusion protein with a mixture of equimolar amounts of Anti - 4 - 1BB antibody and sIL - 15, and a fusion protein in which the MMP sequence in the fusion protein was replaced with a 2×G4S linker. The results showed that the treatment effect of the fusion protein was better than that of the mixed treatment and the fusion protein without the MMP sequence (Figure 6C). Furthermore, in the fusion protein treatment group, a tumor Re - challenge experiment was carried out two months after tumor clearance. In the fusion protein treatment group, mice with complete tumor clearance were subjected to a tumor Re - challenge test two months after tumor regression. Five - fold the dose (1.5×10 cells) of MC38 tumor cells were subcutaneously inoculated. For Re - challenge, only tumor cells were inoculated, and no drugs were administered to each group after inoculation. 6 The results showed that compared with the control group, re - inoculation of three - fold the dose of MC38 cells did not cause tumor growth, indicating that the fusion protein treatment induced the generation of memory immune cells in mice (Figure 6D). Example 2. Anti - 4 - 1BB - MMP - IL - 15 - IL15Ra fusion protein can activate CD8 T cells 1. The treatment effect of Anti - 4 - 1BB - MMP - IL - 15 - IL15Ra fusion protein is independent of innate immune cells To explore which group of cells mediated the antitumor function of the fusion protein. We first used Rag1 gene - knockout mice. Since these mice lack the Rag gene, TCR and BCR cannot rearrange, so T and B cells cannot develop maturely and lack T and B lymphocytes. Rag1 - / - mice were subcutaneously inoculated with 5×10 ... 5MC38 tumor cells were intraperitoneally injected with 50 μg of the fusion protein on days 12, 15, and 18 to verify whether the therapeutic effect of the drug depends on innate immune cells or adaptive immune cells. After treatment, it was found that the fusion protein had no anti-tumor effect in Rag1- / - mice (Figure 7). This indicates that the effect of the drug depends on adaptive immune cells. 2. The therapeutic effect of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein depends on CD8 + T cells We further verified the role of T cells in the treatment process of the fusion protein. On day 12 after subcutaneous inoculation of 5×10 5 MC38 tumor cells on the back of C57BL / 6 mice, treatment was started. Anti-4-1BB-MMP-IL-15-IL15Ra (50 μg) was intraperitoneally injected (administered on days 12, 15, and 18 after tumor inoculation). One day before treatment, 200 μg of CD4 T cell depletion antibody (clone number: GK1.5, prepared in our laboratory) and 200 μg of CD8 T cell depletion antibody (clone number: TIB210, prepared in our laboratory) were intraperitoneally injected, for a total of four injections. We depleted CD8 + T cells and CD4 + T cells using anti-CD4 and anti-CD8 monoclonal antibodies and detected the depletion efficiency. The results showed that the antibodies could effectively deplete T cells in peripheral blood (Figure 8A). Depleting CD4 + T cells during treatment with the fusion protein did not affect the anti-tumor effect of the drug. However, after depleting CD8 + T cells, the anti-tumor effect completely disappeared (Figure 8B). 3. The therapeutic effect of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein depends on CD8 + T cells in the tumor To further explore whether the CD8+ T cells relied on by the drug are the group of T cells existing in the tumor itself that play a role, or whether they activate T cells outside the tumor through the drug and then migrate into the tumor to play a role. We used the inhibitor FTY720 that inhibits the migration of T cells from lymph nodes to the periphery to block the migration of T cells from lymph nodes to the tumor, thereby maintaining the stability of T cells in the tumor. The experimental protocol is as follows: On the back of C57BL / 6 mice, 5×10 5MC38 tumor cells were intraperitoneally injected with 50 μg of the fusion protein on days 12, 15, and 18 after tumor inoculation. In the FTY720 treatment group, 15 μg of FTY720 was intraperitoneally injected on day 11 after tumor inoculation, and 10 μg was continuously injected on days 13, 15, 17, 19, and 21 respectively. After FTY720 blockade, the treatment with the fusion protein still had an effect (Figure 9). The above results indicate that the anti-tumor effect of the fusion protein depends on the pre-existing CD8 + T cells in the tumor. 4. Protective effect of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein treatment on distal metastatic tumors To verify whether the immune protection generated by in-situ tumor treatment also has an inhibitory effect on tumors at other metastatic sites, we conducted a mouse dual-tumor model experiment. We simultaneously inoculated tumor cells on the left and right sides of the subcutaneous area of mice and only performed in-situ treatment on one side to observe whether the distal tumors could also be controlled. The experimental protocol was as follows: C57BL6 mice were inoculated with 5×10 5 MC38 tumor cells, and intratumoral local treatment with 30 μg of the fusion protein was performed on days 13, 16, and 19 after tumor inoculation. On day 13, 1×10 6 MC38 tumor cells were inoculated on the left side of the mice, and the tumor volumes on both the left and right sides were measured twice a week. The results showed that in-situ treatment with intratumoral injection only on the right tumor could not only eliminate the in-situ tumor but also control the untreated distal tumor on the contralateral side (Figure 10A). In addition, we further verified the therapeutic effect of the fusion protein treatment on lung metastasis during melanoma metastasis using the B16F10 mouse tumor model. The experimental protocol was as follows: C57BL6 mice were inoculated with 5×10 5 B16F10 tumor cells, and intratumoral local treatment with 30 μg of the fusion protein was performed on days 9, 12, and 15 after tumor inoculation. Starting from day 8 after tumor inoculation, 15 μg of FTY720 was intraperitoneally injected every 2 days until the end of the experiment. On day 8 after tumor inoculation, 5×10 5 B16F10 tumor cells were injected into the tail vein of each mouse. Ten days after the third treatment, the mice were perfused, the lungs of the mice were taken, and the melanoma metastases in the lungs were counted to verify the therapeutic effect on the metastases after the fusion protein treatment. The results showed that the fusion protein had significant anti-tumor effects on subcutaneous tumors in situ after intratumoral treatment, and was able to partially eliminate the in situ tumors. Blocking T cell migration with FTY720 did not affect the control of in situ tumors (Figure 10B). After FTY720 blockade, the number of lung metastases in the fusion protein treatment group was similar to that of the untreated control group, while the number of lung metastases in the fusion protein treatment group was significantly reduced (Figure 10C). Example 3: Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein can delete intratumoral Tregs and increase the CD8 / Treg cell ratio 1. Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein deletes intratumoral Tregs and increases the CD8 / Treg cell ratio The Anti-4-1BB antibody of the Human IgG1 subtype can effectively delete Tregs in tumors, and IL-15 can expand CD8+ T cells. To verify the function of the fusion protein in vivo, MC38 tumor-bearing mice were treated on the 13th and 16th days after tumor inoculation by intraperitoneal injection of 50 μg of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein. On the 18th day, tumor tissues were taken and the proportions of intratumoral Treg cells and CD8+ T cells were analyzed by flow cytometry. Through flow cytometry analysis, we found that the Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein could effectively delete intratumoral Treg cells and increase the proportion of CD8T, showing the best therapeutic effect (Figure 11A-D). 2. The therapeutic effect of Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein depends on the binding of Fc to FcγR Through the analysis of intratumoral lymphocytes in MC38 tumor-bearing mice, the Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein has the effect of deleting Tregs. To verify whether Treg deletion is the main mechanism for the Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein to exert its anti-tumor effect, we constructed Anti-4-1BB-MMP-IL-15-IL15Ra-no ADCC antibody by point mutation of the Fc region. The mutated Fc does not bind to the FcγR receptor and loses the deletion function mediated by ADCC and ADCP, but still retains the amplification effect of IL-15 on immune cells. Using the MC38 tumor model, C57BL6 mice were inoculated with MC38 tumor cells and treatment started on the 13th day after inoculation. Intraperitoneal injection of 50 μg of wt Fc or no ADCC Fc antibody was given once every three days for a total of three treatments. The results showed that after inoculating tumors, when comparing the therapeutic effects of wt Fc and no ADCC Fc, it was found that the mutated Fc reduced the therapeutic effect of the fusion protein (Figure 12). In the Anti-4-1BB-MMP-IL-15-IL15Ra-no ADCC treatment group, the fusion protein still retained partial anti-tumor effects, indicating that the function of the Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein mostly relied on the deletion of Treg cells. Example 4. Treatment with Anti-4-1BB-MMP-IL-15-IL15Ra combined with PDL1 antibody 1. Upregulation of PDL1 in intratumoral immune cells after treatment with the fusion protein When treating MC38 tumor-bearing mice, although the tumor could be effectively controlled, the tumor could not be eliminated in some mice. When the tumor was relatively large, the treatment with the single fusion protein could only control tumor growth; moreover, the treatment with the fusion protein often activated the immune cells in the tumor, and the increased activation degree often led to the negative feedback of the immune system, resulting in the upregulation of the expression of immune checkpoint molecules, thereby inhibiting the activation of immune cells. To better improve the therapeutic effect, after treatment with the fusion protein, we detected the expression of PD-1 and PD-L1 in the immune cells in the tumors of mice, hoping to find a suitable immune checkpoint inhibitory antibody and use it as a combined treatment plan to observe whether the therapeutic effect could be improved. Protocol: Subcutaneously inoculate 5×10 5 MC38 tumor cells on the back of C57BL / 6 mice. Treat with 50 μg of the fusion protein on the 12th and 15th days, and take the tumor tissues of the mice on the 18th day to analyze the expression of PD-1 and PD-L1 in the intratumoral immune cells. The results showed that after treatment, the expression level of PD-L1 in CD45 + immune cells in the tumors of mice increased significantly, while the expression of PD-1 in CD8 + T cells decreased significantly (Figure 13). This suggested that we could use the combination of the fusion protein and the PD-L1 antibody to further relieve the degree of immunosuppression in the tumor. To further improve the therapeutic effect of the fusion protein, the fusion protein and the PD-L1 antibody were used to treat late-stage tumor-bearing mice (tumor volume about 150 - 200 mm 3 ). Protocol: Subcutaneously inoculate 1×10 6 MC38 tumor cells on the back of C57BL / 6 mice. Intraperitoneally inject 200 μg of aPDL1 antibody on the 13th and 16th days, and intraperitoneally inject 50 μg of the Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein on the 13th, 16th, and 19th days. The results showed that for advanced tumors, the PD-L1 antibody could further improve the therapeutic effect of the fusion protein and overcome immune tolerance. Finally, it should be noted that the above embodiments are only used to help those skilled in the art understand the essence of the present invention and are not used to limit the protection scope of the present invention. Example 5: Anti-OX40-MMP-IL-15-IL15Ra fusion protein has good anti-tumor effects 1. Construction and expression of the fusion protein The above research has proved that the combined treatment of Anti-4-1BB antibody and sIL-15 has a synergistic anti-tumor effect. Considering that OX40 and 4-1BB are both members of the TNF superfamily and are also T cell activating factors, it is speculated that the fusion of Anti-4-1BB antibody and sIL-15 may also produce a synergistic anti-tumor effect. Here, we verified again that IL-15-IL-15Ra was connected to the carboxyl terminus of immunoglobulin Fc, and a short substrate sequence that could be recognized by MMP highly expressed in tumors was fused between Fc and IL-15 for expression, so that the fusion protein could be cleaved in tumors to release the activity of IL-15. Each molecule contains a complete antibody Fc and an IL-15-IL-15Ra molecule, and there is an MMP substrate sequence between the antibody and the cytokine. We co-incubated the fusion protein with activated MMP14 enzyme in vitro, and then performed SDS-PAGE analysis to verify the in vitro cleavage effect of the fusion protein. The results showed that the fusion protein could be effectively cleaved in vitro (Figure 14A). 2. In vitro activity detection of the fusion protein (lymphocyte proliferation promotion experiment (CCK8 assay)) CTLL2 cells were cultured in 1640 complete medium containing 100 U / ml commercial recombinant IL2 cytokine for 24 hours; then washed 2-3 times with complete medium without IL2, and the cells were diluted to 2×10 4 / ml; the diluted samples sIL15-Fc, Fc-MMP14-sIL15 (without MMP14 cleavage), and Fc-MMP14-sIL15 (with MMP14 cleavage) were cultured with complete medium without IL2. The initial concentration was 5 μg / ml, and 5-fold dilutions were made for 10 dilution gradients; 100 μl of cell suspension and 100 μl of the sample were added to a 96-well cell culture plate, and mixed well with a pipette tip; after culturing for 72 hours, 20 μl of CCK8 was added, and after continuing to culture for 3 hours, the OD values at two wavelengths of 450 nM and 630 nM were detected with an enzyme-linked immunosorbent assay reader. The detection results showed that: (1) The biological activities of Fc-MMP14-sIL-15(-) and sIL15Fc differed by about 300-fold, indicating that the biological activity of Fc-MMP14-sIL15 was very low without MMP14 cleavage; (2) The biological activities of Fc-MMP14-sIL-15(+) and sIL15Fc were comparable, indicating that the biological activity of Fc-MMP14-sIL15 was restored to the level of sIL15-Fc upon MMP14 cleavage (Figure 14B). 3. The fusion protein has good anti-tumor effects 5×10 5 MC38 tumor cells were subcutaneously inoculated into C57BL6 mice, and when the tumors grew to 100 mm 3 ; αOX40 15 μg, sIL15-Fc 15 μg or αOX40-MMP14-sIL15 30 μg were administered for treatment, and the administration was carried out 3 times at an interval of 2 days. The control group was given PBS in the same manner. The volume of the tumors was measured (volume = length × width × height / 2). The results showed that αOX40-MMP14-sIL15 administered intraperitoneally provided good control of the tumors. This indicated that the fusion protein had better anti-tumor effects compared to the combined administration of equimolar amounts of αOX40 and sIL15-Fc (Figure 14C). Example 6. The humanized Anti-4-1BB-MMP-IL-15-IL15Ra fusion protein has good anti-tumor effects and safety 6.1 Construction and expression of the humanized fusion protein To verify that the humanized Anti-4-1BB-MMP-IL-15-IL15R fusion protein had similar anti-tumor effects, we constructed fusion proteins by linking the heavy chain of the antibody that could recognize the human 4-1BB molecule with human IL-15-IL-15Ra in the same way. Fusion proteins with two linker lengths, one being G4S-MMP-G4S and the other being GS-MMP-GGS, were constructed and analyzed by SDS-PAGE (Figure 15A) (R: reducing gel, NR: non-reducing gel). 6.2 Verification of the function of the fusion protein (1) In vitro activity detection of the fusion protein The MMP substrate sequence in the humanized fusion protein can be cleaved by MMP2, MMP9 or MMP14. We co-incubated the fusion protein with activated MMP14 enzyme in vitro and then performed SDS-PAGE analysis to verify the in vitro cleavage effect of the fusion protein. The results showed that the fusion proteins with different Linker lengths could be effectively cleaved in vitro (Figure 15A). We used the HEK-IL2 cell line to detect the activation of STAT5 signal in this cell line by human IL-15 in the fusion protein. The results showed that the ability of the un-cleaved fusion protein to activate the downstream signal of the HEK-IL2 cell line decreased. After in vitro enzymatic cleavage, the activation ability was basically restored (Figure 15B), indicating that the activity of IL-15 in the humanized fusion protein was well blocked and its activity was also very fully restored after MMP cleavage. (2) Expansion of immune cells in hu-PBMC mice in vivo by the fusion protein The activity of IL-15 in the humanized fusion protein was well blocked. To further verify whether the fusion protein would expand human immune cells in the peripheral tissues of tumor-bearing PBMC mice, we intraperitoneally injected 100 μg of the fusion protein or an equimolar mixture of Anti-4-1BB antibody and sIL-15, and treated once every three days. Two days after the second treatment, we detected the number of CD4 + T cells, CD8 + T cells and CD3 + T cells in peripheral blood. The fusion protein with Linker GS-MMP-GGS was designated as Human GS, the fusion protein with Linker G4S-MMP-G4S was designated as Human G4S, and the combined treatment group was designated as Abs+sIL-15. The results showed that compared with the combined treatment group, the fusion protein did not cause a large expansion of T cells in peripheral blood and had good safety (Figure 16). 6.3. The fusion protein has good safety and anti-tumor effect NSG mice are immunodeficient mice caused by knockout mutations of Prkdc gene and Il2rg gene, lacking mature T, B and NK cells, not generating immunoglobulins, and having abnormal dendritic cell (DC) function, which is very suitable for human cell or tissue transplantation. We used 6-8-week-old NSG mice and subcutaneously inoculated 1×10 6 CFPAC1 cells on the right dorsal side on day D0. Seven days later, 5×10 6 PBMC cells were inoculated by tail vein injection. Seven days after the inoculation of PBMC cells, the mice were grouped according to tumor size to establish a tumor-bearing hu-PBMC mouse model. Treat with 100 μg of the fusion protein and an equimolar amount of a mixture of Anti-4-1BB antibody and sIL-15 by intraperitoneal injection respectively. Treat three times on days D14, D17, and D20, and measure tumor size and body weight twice a week. The results show that the therapeutic effect of the fusion protein is better than that of the control group (Figure 17A). The fusion protein does not cause weight loss. In the mixed treatment group, weight loss exceeded 10% after the second treatment, leading to death, indicating that the safety of the fusion protein is better than that of the mixed treatment group (Figure 17B). The "x" in Figures 17A and B indicates that all the mice in the mixed treatment group died.

Claims

1. A bifunctional fusion protein composed of IL-15 and a T cell costimulatory molecule antibody, characterized in that, The fusion protein described above includes: (1) A T cell co-stimulatory molecule antibody; (2) An IL-15 and IL-15R Sushi domain conjugate, where the IL-15 and IL-15R Sushi domains are connected by a second linker fragment; and (3) A first linker fragment; the first linker fragment is used to connect the heavy chain Fc region of the T cell co-stimulatory molecule antibody and the IL-15 and IL-15R Sushi domain conjugate, and can be cleaved by a matrix metalloproteinase; The T cell co-stimulatory molecule is: 4-1BB, ICOS, OX40; preferably 4-1BB; preferably, the T cell co-stimulatory molecule antibody is an IgG1 type antibody; Preferably, the first linker fragment is: 2-4 GnS linker units and a short peptide contained therein that can be recognized and cleaved by a matrix metalloproteinase, where n is an integer from 1 to 4; 2. The bifunctional fusion protein according to claim 1, wherein The amino acid sequence of the short peptide that can be recognized and cleaved by a matrix metalloproteinase is as shown in SEQ ID NO.6; The amino acid sequence of the first linker fragment is as shown in SEQ ID NO.12; 3. The bifunctional fusion protein according to claim 1, wherein The second linker fragment is: 2-5 G4S linker units; preferably, the amino acid sequence of the second linker fragment is as shown in SEQ ID NO.

5.

4. The bifunctional fusion protein according to any one of claims 1-4, characterized in that, The bifunctional fusion protein includes: (1) A first structural unit, which is, from the N-terminus to the C-terminus in sequence: a 4-1BB antibody heavy chain, a first linker fragment, an IL-15 and IL-15R Sushi domain conjugate; (2) A second structural unit: a 4-1BB antibody light chain paired with the 4-1BB antibody heavy chain; And, the bifunctional fusion protein dimerizes through the Fc region of the 4-1BB antibody heavy chain to form a homodimer.

5. The bifunctional fusion protein according to claim 4, wherein The 4-1BB antibody is an antibody formed by fusing the Fab region of a human or murine antibody with the Fc region of IgG1, The amino acid sequence of the heavy chain (VH+CH1) of the murine 4-1BB antibody is as shown in SEQ ID NO.2; The amino acid sequence of the heavy chain (VH+CH1) of the human 4-1BB antibody is as shown in SEQ ID NO.10; The amino acid sequence of the light chain (VL+CL) of the murine 4-1BB antibody is as shown in SEQ ID NO.1; The amino acid sequence of the light chain (VL+CL) of the human 4-1BB antibody is as shown in SEQ ID NO.9; The Fc region of IgG1 is the Fc region of human IgG1, or a variant of the Fc region of human IgG1 with the ADCC effect knocked out.

6. The bifunctional fusion protein according to claim 5, wherein The amino acid sequence of the Fc region of human IgG1 is as shown in SEQ ID NO.3, and the amino acid sequence of the variant of the Fc region of human IgG1 with the ADCC effect knocked out is as shown in SEQ ID NO.

4.

7. The bifunctional fusion protein according to claim 4, wherein IL-15 and the IL-15R Sushi domain are human or murine proteins; preferably, The amino acid sequence of murine IL-15 is shown in SEQ ID NO.7, and the amino acid sequence of the Sushi domain of murine IL-15R is shown in SEQ ID NO.8; The amino acid sequence of human IL-15 is shown in SEQ ID NO.13, and the amino acid sequence of the Sushi domain of human IL-15R is shown in SEQ ID NO.

14.

8. A nucleotide fragment encoding the bifunctional fusion protein according to any one of claims 1-7, a vector containing the nucleotide fragment, and a host.

9. The following applications of the bifunctional fusion protein according to any one of claims 1-7, or the nucleotide fragment encoding the bifunctional fusion protein according to claim 8, or the vector containing the nucleotide fragment, or the host: (1) Preparation of an anti-tumor drug; (2) Preparation of a combined anti-tumor drug.

10. According to the application described in claim 9, wherein The anti-tumor drug described is: a drug that reduces and eliminates intratumoral Treg cells and induces the proliferation of CD8 + T cells; preferably, the tumor is: a tumor in an immunosuppressed state; the combined anti-tumor drug is a combined anti-tumor drug used in combination with an immune checkpoint inhibitor, and the combined anti-tumor drug can overcome immune checkpoint inhibitor tolerance; the immune checkpoint inhibitors include but are not limited to: PD-1 / PD-L1 antibodies, CTLA4 antibodies, TIGIT antibodies, LAG-3 antibodies, TIM-3 antibodies.

Images