PD-L1 targeting small molecule polypeptide and application thereof

Abstract

The invention belongs to the technical field of biology, and particularly relates to a PD-L1 targeting micromolecule polypeptide and application thereof. On the basis of machine learning and virtual screening technologies, a series of PD-L1 targeting micromolecule polypeptides are screened out, the in-vivo and in-vitro biological activities of the polypeptides are measured, it is found that the micromolecule polypeptides have the capacity of inducing PD-L1 to form dimerization and good anti-tumor activity, and the anti-tumor activity of the polypeptides is superior to that of PD-L1 monoclonal antibodies and compound inhibitors. Meanwhile, the PD-L1 targeting small-molecule polypeptide has no obvious toxicity, has certain safety, and can be developed into a novel small-molecule polypeptide PD-L1 inhibitor drug.

Description

Technical Field

[0001] The present invention belongs to the field of biotechnology, and specifically relates to a small molecule polypeptide targeting PD-L1 and an application thereof. Background Art

[0002] Cancer, also known as malignant tumors, is a complex disease caused by multiple factors such as genetic or viral pathogens, environmental risk factors, and unhealthy lifestyles. It is common in middle-aged and young people, with poor treatment effects and high mortality. In recent years, the global annual cancer mortality rate has exceeded 10 million, seriously affecting people's health and increasing the social burden. Currently, a variety of treatment methods are used in clinical practice, including radiotherapy, chemotherapy, targeted therapy, and immunotherapy. In recent years, immunotherapy has made significant progress in the field of cancer treatment. Immunotherapy can activate the patient's own immune system, enabling it to more effectively identify and destroy cancer cells. Compared with traditional radiotherapy and chemotherapy, this treatment method has the advantages of strong specificity and few side effects, and has received widespread attention.

[0003] Tumor immunotherapy is a treatment method that targets the body's low or hyperactive immune state, artificially enhances or suppresses the body's immune function to achieve the purpose of treating the disease. Although there are toxic side effects, it is more effective and has fewer side effects than traditional treatments (KENNEDY LK, SALAMAK S. A review of cancer immunotherapy toxicity [J]. CA Cancer J Clin, 2020, 70 (2): 86-104; FARKONA S, DIAMANDIS EP, BLASUTIG I M. Cancer immunotherapy: the beginning of the end of cancer [J]. BMCMed, 2016, 14: 73). Changes in the tumor microenvironment are essential factors in the occurrence and development of tumors. During tumor development, tumor infiltrating lymphocytes (TILs) are ineffective in clearing tumor cells in vivo, but when the immunosuppression in the tumor microenvironment is removed, the cells can proliferate and kill tumors. Therefore, therapeutic methods that target molecules or related influencing factors in the tumor microenvironment and change the tumor microenvironment play an important role in the treatment of cancer (ARNETH B. Tumor microenvironment [J]. Medicine, 2019, 56(1): 15.).

[0004] In tumor immunotherapy, the discovery of tumor immune checkpoint inhibitors is considered to be one of the milestone breakthroughs in tumor treatment. Immune checkpoints are a class of protein molecules expressed on immune cells that play an inhibitory signaling role. By transmitting inhibitory signals into immune cells, they can inhibit the body's over-activated immune response, thereby maintaining immune tolerance and avoiding the occurrence of autoimmune diseases (KITAMURAT, QIAN BZ, POLLARD JW. Immune cell promotion of metastasis[J]. Nat Rev Immunol, 2015, 15(2): 73-86; SCHUMACHER TN, KESMIR C, VANBUUREN MM. Biomarkers in cancer immunotherapy[J]. Cancer Cell, 2015, 27(1): 12-14; KIM N, KIM HS. Targeting checkpoint receptors and molecules for therapeutic modulation of natural killer cells[J]. Front Immunol, 2018, 9: 2041). Tumor cells use this mechanism to express corresponding ligands and bind to immune checkpoint molecules, inhibiting the activation of the immune system and achieving immune escape (MOTZ GT, COUKOS G. Deciphering and reversing tumor immune suppression [J]. Immunity, 2013, 39 (1): 61-73.). Immune checkpoint blockade therapy (ICB) refers to an immunotherapy method that uses immune checkpoint inhibitors (ICIs) to block the binding of immune checkpoint molecules and their ligands, thereby reactivating suppressed immune cells and restoring the tumor-killing effect (Xu Xiangning, Wang Yindi, Lv Zhen, et al. Research progress in anti-tumor treatment based on the PD-1 / PD-L1 immune checkpoint pathway combined with traditional Chinese and Western medicine [J]. Chinese Journal of Basic Medicine of Traditional Chinese Medicine, 2024, 30 (07): 1258-1264.). PD-1 and PD-L1 are two common immune checkpoint proteins. The binding of PD-1 to its ligand PD-L1 can inhibit T cell activation and promote tumor immune escape. Therefore, blocking the PD-1 / PD-L1 interaction is an important means of treating tumors.

[0005] Research on the development of tumor drugs targeting these two targets is not sufficient. Although inhibitors targeting PD-L1 have achieved certain therapeutic effects, most of them are monoclonal antibody compounds. Monoclonal antibody drugs have some disadvantages, such as strong immunogenicity, low tissue / tumor penetration, poor oral bioavailability, difficulty in storage and transportation, and high production costs (Chen Guangfeng, Xie Yanping, Shuai Yongkang, et al. Preliminary study on the performance of new breast cancer targeting small molecule peptides [J]. Anatomical Research, 2024, 46(03): 193-199.). Peptides are amino acid chains composed of several to up to 50 amino acids connected by peptide bonds. Compared with antibodies, they have the advantages of low manufacturing cost, high stability, high affinity, small relative molecular mass, strong tissue permeability, fast blood clearance, low immunogenicity and toxicity, easy preparation and modification (Sun X, Li Y, Liu T, et al. Peptide-based imaging agents for cancer detection [J]. Adv Drug Deliv Rev, 2017, 110-111: 38-51.). Therefore, the development of small molecule peptide inhibitors targeting PD-L1 is of great significance to enrich tumor treatment methods. Summary of the invention

[0006] In view of the problems existing in the prior art, the present invention screened out a small molecule polypeptide targeting PD-L1 based on machine learning and virtual screening technology, and measured its in vivo and in vitro biological activities. It was found that the small molecule polypeptide has the ability to induce PD-L1 to form dimerization and has good anti-tumor activity, and the present invention is completed based on this.

[0007] In a first aspect, the present invention provides a small molecule polypeptide targeting PD-L1, wherein the sequence of the small molecule polypeptide targeting PD-L1 is selected from SEQ ID NO.1-SEQ ID NO.82.

[0008] Furthermore, the small molecule polypeptide targeting PD-L1 is a mutant polypeptide obtained by performing amino acid scanning mutation on the basis of SEQ ID NO.1-SEQ ID NO.30.

[0009] Furthermore, the amino acid scanning mutation includes positively charged amino acid scanning mutation, negatively charged amino acid scanning mutation, neutral amino acid scanning mutation, rare amino acid substitution mutation and D-type amino acid substitution mutation.

[0010] Furthermore, the amino acid sequence of the small molecule polypeptide targeting PD-L1 after amino acid scanning mutation is shown in SEQ ID NO.31-SEQ ID NO.82.

[0011] Furthermore, the small molecule polypeptide targeting PD-L1 is extended or truncated at the N-terminus or C-terminus based on SEQ ID NO.1-SEQ ID NO.30.

[0012] Furthermore, the amino acid sequence of the extended or truncated small molecule polypeptide targeting PD-L1 is shown as SEQ ID NO.31-SEQ ID NO.82.

[0013] Preferably, the sequence of the small molecule polypeptide targeting PD-L1 is SEQ ID NO.45.

[0014] In a second aspect, the present invention provides a method for screening small molecule polypeptides based on machine learning and virtual screening technology:

[0015] S1. Construction of a three-dimensional structure database: Generate tetrapeptide sequences and files, optimize the energy of the peptide system, and build a tetrapeptide database;

[0016] S2. Download ligands from the database as active molecules to prepare for virtual screening;

[0017] S3. Extract three-dimensional structural features for clustering to screen out compounds with similar potential activities as the extracted ligands;

[0018] S4. The peptides obtained by clustering are finally screened by molecular docking and binding free energy calculation;

[0019] S5. Perform solid phase synthesis and biological activity testing on the top-ranked peptides to screen out the peptides with the best activity;

[0020] S6. Perform structural modification on the peptides screened out in S5, and screen out the peptide molecules with the best activity based on binding free energy.

[0021] Furthermore, in step S1, the sequence of the tetrapeptide is generated by Amber's tleap program, covering random combinations of 20 natural amino acids at 4 positions, with a total of 204 combinations.

[0022] Further, in step S1, the tetrapeptide file includes a prmtop file and an inpcrd file of the tetrapeptide.

[0023] Furthermore, the tetrapeptide file was imported into the required protein and small molecule force fields using the tleap program, and Na + and Cl - The system charge is kept at 0 and the peptide is solvated.

[0024] Furthermore, during the solvation process, a TIP3PBOX water box was added around the peptide.

[0025] Furthermore, in step S1, the energy optimization of the polypeptide system is performed using the pmemd program of Amber.

[0026] Furthermore, the optimization process is divided into two steps. The first step is to add solute constraints to prevent large changes in solute molecules during solvent optimization; the second step is to optimize and eliminate solute constraints and optimize the entire system; and an rst file is generated at the same time.

[0027] Furthermore, the second step optimizes the entire system using the steepest descent method and the conjugate gradient method.

[0028] Furthermore, in step S1, the construction of the tetrapeptide database includes two steps: extracting the three-dimensional structure of the polypeptide and removing water molecules in the three-dimensional structure, and contains 160,000 tetrapeptides.

[0029] Furthermore, the extracting of the three-dimensional structure of the polypeptide refers to using the ambpdb program of Amber to extract the three-dimensional structure of the polypeptide from the prmtop file and the rst file, and outputting the PDB file of the polypeptide.

[0030] Furthermore, the removal of water molecules in the three-dimensional structure refers to using the pdb4amber program to remove the water molecules in the three-dimensional structure extracted in the previous step, and generating the final structure files in PDB and MOL2 formats.

[0031] Further, in step S2, the ligand refers to the ligand in the three-dimensional structure of the PD-L1 protein.

[0032] Furthermore, in step S3, extracting the features of the three-dimensional structure includes acquiring and converting polypeptide data information.

[0033] Furthermore, the obtaining of polypeptide data information refers to using the parsePDB module of PyUUL to read the PDB file of the three-dimensional structure, obtain the polypeptide data information, and transfer the information to CUDA.

[0034] Furthermore, the polypeptide data information includes data such as coordinates, atom names, channels and atomic radii.

[0035] Furthermore, the converting of the polypeptide data information refers to converting the read polypeptide file information into a surface area point cloud data format using a point cloud conversion module of PyUUL.

[0036] Furthermore, in step S3, considering that the number of atoms that a tetrapeptide can contain is less than 500, the maximum number of points in each batch is set to 500 to ensure a balance between processing efficiency and resource utilization.

[0037] Furthermore, in step S6, the structural modification refers to modification of amino acids in the polypeptide.

[0038] In a third aspect, the present invention provides a fusion protein, wherein the fusion protein comprises the active polypeptide sequence described in the first aspect, and the fusion protein has the function of targeting PD-L1.

[0039] In a fourth aspect, the present invention provides a multimer formed by polymerization of two or more polypeptide monomers, wherein at least one polypeptide monomer is the small molecule polypeptide described in the first aspect; the polypeptide monomers of the multimer are covalently linked.

[0040] Furthermore, the multimer is a homodimer or a heterodimer.

[0041] In a fifth aspect, the present invention provides a nucleic acid molecule encoding the small molecule polypeptide described in the first aspect or the fusion protein or conjugate described in the third aspect.

[0042] In a sixth aspect, the present invention provides a construct comprising the nucleic acid molecule described in the fifth aspect.

[0043] In a seventh aspect, the present invention provides a host cell comprising the nucleic acid molecule described in the fifth aspect and / or the construct described in the sixth aspect, wherein the host cell is transformed or transfected with the nucleic acid molecule described in the third aspect and / or the construct described in the fourth aspect.

[0044] In an eighth aspect, the present invention provides a pharmaceutical composition, comprising the small molecule polypeptide described in the first aspect, the fusion protein described in the third aspect, the multimer described in the fourth aspect, the nucleic acid molecule described in the fifth aspect, the construct described in the sixth aspect and / or the host cell described in the seventh aspect, and optional pharmaceutically acceptable excipients.

[0045] In the ninth aspect, the present invention provides the use of the small molecule polypeptide described in the first aspect, the fusion protein described in the third aspect, the polymer described in the fourth aspect, the nucleic acid molecule described in the fifth aspect, the construct described in the sixth aspect or the host cell described in the seventh aspect in the preparation of drugs for preventing and / or treating tumor-related diseases.

[0046] Furthermore, the tumor-related diseases include but are not limited to head and neck squamous cell carcinoma, nasopharyngeal carcinoma, triple-negative breast cancer, non-small cell lung cancer (NSCLC), small cell lung cancer, esophageal cancer, pleural mesothelioma, gastric cancer, hepatocellular carcinoma, cholangiocarcinoma, renal cancer, urothelial carcinoma, cervical cancer, ovarian cancer, melanoma, skin squamous cell carcinoma, certain special types of soft tissue sarcoma, dMMR / MSI-H solid tumors including colorectal cancer, and solid tumors with TMB≥10mut / Mb.

[0047] Beneficial Effects

[0048] 1. The present invention uses machine learning and virtual screening methods to obtain a series of polypeptide molecules of SEQ ID NO.1-SEQ ID NO.82; among them, the polypeptide molecule of SEQ ID NO.45 has the highest activity and can bind to PD-L1 at the molecular level, with a KD of 9.121×10 -7 .

[0049] 2. The small molecule polypeptides screened out in the present invention have the effect of inducing PD-L1 to dimerize at both the molecular and cellular levels.

[0050] 3. The small molecule polypeptides screened in the present invention have great potential in enhancing the killing effect of immune cells on tumor cells; in animal experiments, at the same concentration, the inhibitory activity of the small molecule polypeptide is better than that of the PD-L1 monoclonal antibody control group and the small molecule inhibitor control group (BMS202), and the inhibition rate of tumors increases in a dose-dependent trend; when the concentration is 20 mg / kg, its inhibition rate of tumor weight reaches 83%.

[0051] 4. During the experiment, the weight of mice in the small molecule peptide treatment group hardly changed, indicating that small molecule peptides have no obvious toxicity and have a certain degree of safety.

[0052] 5. PD-L1 is a tumor-related target. The obtained small molecule peptide can inhibit the binding of PD-L1 and PD-1, and is expected to be developed into a new small molecule peptide PD-L1 inhibitor drug. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is a flow chart of the screening method based on machine learning and virtual screening of the present invention.

[0054] Figure 2 It is a schematic diagram of the Alpha Screen assay of polypeptide activity of SEQ ID NO.1-SEQ ID NO.45 provided in the examples of the present invention.

[0055] Figure 3 It is a schematic diagram of the Alpha Screen assay of polypeptide activity of SEQ ID NO.45 provided in an embodiment of the present invention.

[0056] Figure 4 This is the SPR binding curve of SEQ ID NO.45 and PD-L1 provided in the embodiment of the present invention.

[0057] Figure 5 This is a schematic diagram of FRET experiment verifying that SEQ ID NO.45 and BMS202 induce PD-L1 dimerization at the molecular level.

[0058] Figure 6It is a schematic diagram of non-denaturing gel electrophoresis experiment verifying that SEQ ID NO.45 and BMS202 cells induced PD-L1 dimerization.

[0059] Figure 7 This is a schematic diagram of SEQ ID NO.45 enhancing the ability of NK-92 cells to kill tumor cells.

[0060] Figure 8 It is a schematic diagram of the tumor inhibition rate of SEQ ID NO.45 provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0061] The specific embodiments of the present invention are further described below. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention. In addition, the technical features involved in the embodiments described below can be combined with each other as long as they do not conflict with each other.

[0062] The experimental methods in the following examples are conventional methods unless otherwise specified, and the experimental materials used in the following examples are commercially available unless otherwise specified.

[0063] Explanation of terms

[0064] Immune checkpoint inhibitors (ICIs) are a class of monoclonal antibodies. Immune checkpoint molecules and their ligands are negative regulatory molecules expressed in normal tissue cells and immune cells, such as cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) and programmed cell death protein 1 (PD-1). Its basic mechanism of action is to block the binding between the receptors and ligands of immune checkpoint molecules, thereby activating immune effector cells, mainly effector T lymphocytes, and the activated immune effector cells inhibit tumors.

[0065] Programmed cell death receptor ligand 1 (PD-L1): PD-1 is a type I transmembrane protein composed of 288 amino acids, including extracellular IgV and IgC regions, transmembrane regions and intracellular regions, and is mainly expressed in a variety of immune cells such as B cells, T cells, natural killer cells, dendritic cells and tumor-infiltrating lymphocytes. PD-L1, as an endogenous ligand of PD-1, is also a type I transmembrane protein, including extracellular IgV regions, transmembrane regions and intracellular regions that bind to PD-1, and is often expressed in a variety of tumor cells, such as melanoma, colon cancer, lung cancer, pancreatic cancer, ovarian cancer, etc. Under normal physiological conditions, the main function of the PD-1 / PD-L1 pathway is to regulate the intensity of the immune response to an appropriate level, so that the body can successfully complete the immune response while avoiding harmful effects. However, in the tumor microenvironment, PD-L1 on the surface of tumor cells interacts with PD-1 on the surface of T cells, resulting in reduced function of effector T cells, inhibiting anti-tumor immune responses, and allowing tumors to grow. Blocking the binding of PD-1 and PD-L1 can restore T cell activity and enhance the body's ability to kill tumors.

[0066] BMS202: A potent non-peptide PD-1 / PD-L complex inhibitor that can directly bind to PD-L1 and block the human PD-1 / PD-L interaction, with anti-tumor activity.

[0067] Example 1 Screening of small molecule peptides

[0068] A three-dimensional structure database containing 160,000 peptides was constructed using the xleap program of Amber software as a small molecule compound database for subsequent screening work.

[0069] In order to perform virtual screening, the small molecule ligands in the three-dimensional protein structure of PD-L1 (PDBID: 5N2F) were downloaded from the PDB database as active molecules and mixed into the constructed tetrapeptide database. Subsequently, the three-dimensional features of each peptide and active molecule in the database were extracted using the PyUUL library and used as the input of the downstream K-means algorithm. After that, K-means clustering was performed, the number of clustering categories was set to 2, and then the category in which the active molecule belonged was selected for subsequent molecular docking. Ledock was used in the initial docking stage, and then the top 300 were selected for further docking using Glide. Next, the top 100 were selected to calculate MM / GBSA using Schrödinger's Prime module.

[0070] The top 30 peptides were selected from the top 100 for peptide solid phase synthesis and biological activity determination (Table 1).

[0071] Peptide solid phase synthesis: First, soak the resin in a solvent mixed with N,N-dimethylformamide and dichloromethane in a volume ratio of 1:1 so that the resin can be fully swollen for 4 hours. Then, introduce the first amino acid and let it react with the resin for 1 hour. After completion, use a mixed solvent of methanol, dichloromethane and N,N-diisopropylethylamine (ratio 2:2:1) for end-capping. Then add the correct amino acids in sequence. DMF is used as a solvent for amino acid coupling. HTCU and DIPEA act as condensing agents and activated amino acids, respectively. The reaction time is 1h. Then remove the Fmoc protecting group on the amino acid, couple the amino acid at the next position, and repeat the cycle until the last amino acid. After the peptide sequence is synthesized, use a cleavage solution of trifluoroacetic acid: triisopropylsilane: water = 9:0.5:0.5 to react for 3h to cut the peptide from the resin, then use ether to precipitate the synthesized peptide, and then use a C18 reverse chromatographic column for separation and purification. The molecular weight of the peptide was confirmed by electrospray mass spectrometry, and the purity was >95% by HPLC.

[0072] Bioactivity assay: Alpha Screen technology was used for bioactivity assay, and PD-L1 protein was diluted to 10nM with 1×Assaybuffer. 5μL of PD-L1 protein dilution and 5μL of positive control BMS202 or peptide to be tested were respectively aspirated into 384 white plate wells for incubation (BMS202 and the concentration of the screened peptide were both 500nM). 5μL of 1×Assay buffer was used as a blank control group, i.e. no compound was added to the well. PD-1 protein was diluted to 15nM with 1×Assay buffer. 5μL of PD-1 protein dilution was added to the wells and incubation continued. Acceptor microbeads were diluted 50 times with 1×Assaybuffer, and 5μL of acceptor microbead dilution was aspirated and added to the wells for incubation. Donor microbeads were diluted 50 times with 1×Assaybuffer, and 5μL of donor microbead dilution was aspirated and added to the wells for incubation. Finally, the OD value at 680 nM of the 384-well plate was read using an ELISA reader. The results showed that SEQ ID NO.1 had the best activity, so SEQ ID NO.1 was modified ( Figure 2 ).

[0073] Table 130 Activity and MM / GBSA of 30 peptides

[0074]

[0075] The 20 common amino acids were classified by side chain (Table 2) into 9 categories. An amino acid was randomly selected from these 9 categories as a representative for extension, and AlphaFold2 was used to predict the complex model of the extended peptide and PD-L1. Subsequently, Amber software was used to perform 10ns molecular dynamics simulation and calculate MM / GBSA. The results are shown in Table 3. The three peptides with the best activity are SEQ ID NO.31, SEQ ID NO.32 and SEQ ID NO.33.

[0076] The amino acids in the extended part of the three polypeptides with the best activity were mutated to the same amino acids, and the complex model was predicted and the binding ability based on MM / GBSA was calculated. The results are shown in Table 4, and the binding ability of SEQ ID NO.40 and SEQ ID NO.41 is relatively better.

[0077] SEQ ID NO.40 and SEQ ID NO.41 were further extended according to the above strategy, and the results are shown in Table 5 and Table 6. Among them, the two with the best scores were SEQ ID NO.44 and SEQ ID NO.45, and the peptide solid phase synthesis and biological activity determination were carried out.

[0078] Table 2 Amino acid classification

[0079]

[0080] Table 3 Pentapeptide MM / GB (PB) SA

[0081]

[0082] Table 4 Pentapeptide mutants MM / GB(PB)SA

[0083]

[0084] Table 5 Hexapeptide MM / GB (PB) SA

[0085]

[0086]

[0087] Table 6 Hexapeptide mutants MM / GB(PB)SA

[0088]

[0089] SEQ ID NO.1 was mutated to obtain SEQ ID NO.45 (WMHFNR) and SEQ ID NO.44 (RLHFNR), and the activity was further determined using Alpha Screen technology.

[0090] The results are as follows Figure 3 As shown, SEQ ID NO.45 is more active than SEQ ID NO.44 and SEQ ID NO.1, and its IC 50 It is 0.336μM.

[0091] Example 2 SPR study of the binding dissociation constant of small molecule peptides and PD-L1

[0092] BMS202 was used as a positive control drug, and the SPR method was used to detect the binding effect between PD-L1 and SEQ ID NO.45.

[0093] The results are as follows Figure 4 As shown, SEQ ID NO.45 can bind to PD-L1 at the molecular level, and its KD value is 912.1nM.

[0094] Example 3 FRET study of dimerization of small molecule peptides

[0095] First, the dimerization of SEQ ID NO. 45 was verified at the molecular level. A FRET (fluorescence resonance energy transfer) experiment was performed, in which PD-L1 was coupled with FAM (carboxyfluorescein) and TAMRA (tetramethylrhodamine), and the ratio of PD-L1 of the two markers was 1:1.

[0096] The results are as follows Figure 5 As shown in the figure, when the concentration of SEQ ID NO.45 increased from 250μM to 2mM, the dimerization ratio of PD-L1 showed a gradual increasing trend. At a concentration of 2mM, the dimerization ratio was close to 1.4, indicating that SEQ ID NO.45 has the potential to induce PD-L1 to form dimerization.

[0097] Secondly, non-denaturing gel electrophoresis was used to verify the PD-L1 dimerization induced by different concentration gradients of SEQ ID NO.45 at the cellular level. First, protein samples were collected: MDA-MB-231 cells in the logarithmic growth phase were taken, digested, centrifuged, resuspended and counted. Different concentrations of SEQ ID NO.45 were added to make the final concentrations 0.1, 1, and 10 μM, respectively. At the same time, solvent control wells and positive control wells were set up and placed in an incubator for culture. Collect the cells, discard the supernatant after centrifugation, and then wash with PBS. Add 150 μL of non-denaturing lysis buffer to the cell pellet, blow it evenly, place it on ice for lysis, and mix the sample and store it for later use. Then use Native-Page technology to detect PD-L1: According to the molecular weight of the target protein to be detected, Tris-Gly 10% precast gel was selected for gel electrophoresis. After protein quantification, 20 μg of protein sample was loaded into each well for Native-PAGE gel electrophoresis. After the electrophoresis is completed, carefully remove the gel and place it on the wet transfer membrane in the order of "transfer filter paper-gel-NC membrane transfer filter paper". After the transfer is completed, take out the NC membrane, put it into the Ponceau solution, and place it on a shaker to slowly shake and stain until a clear red protein band appears on the NC membrane. Place the entire membrane in the antibody incubation box. After washing away the residual Ponceau dye solution with TBST solution, add 5% skim milk powder to the box and place it on a shaker at room temperature to slowly shake and block. After the blocking is completed, add TBST and shake quickly to wash away the blocking solution. Take 2μL of PD-L1 primary antibody according to the experimental requirements, dilute it with TBST at a ratio of 1:1000, and place the incubation box on a shaker at low temperature and shake slowly. The next day, use TBST to wash off the primary antibody on the membrane and shake quickly to wash. Add 1-2μL of the corresponding HRP-labeled secondary antibody (diluted with TBST according to the recommended ratio in the manual) and incubate slowly at room temperature. Then wash and remove the secondary antibody in the same way. After washing the membrane, apply an appropriate amount of developer on the NC membrane containing the target protein (be careful to avoid light when preparing the ECL chemiluminescent solution, and mix Regent A and B at a ratio of 1:1 according to the instructions). Then place the strip in a Tanon developer, expose and develop, and analyze the development results.

[0098] The results are as follows Figure 6 As shown in the figure, in the presence of SEQ ID NO.45 and BMS202 (1 μM), in addition to the electrophoretic band of monomeric PD-L1, an additional electrophoretic band at the dimer position appeared. This result clearly shows that SEQ ID NO.45 also has the effect of inducing PD-L1 to dimerize at the cellular level.

[0099] Example 4: Exploring the in vitro immune anti-tumor effects of small molecule peptides

[0100] It has been reported that NK cells in tumor patients can express PD-1 protein. The application of PD-L1 inhibitors can block the inhibitory effect of tumor cells on NK cells and restore the anti-tumor activity of NK cells. In order to evaluate the in vitro immune anti-tumor effect of SEQ ID NO.45, this study conducted an experiment on the ability of SEQ ID NO.45 to enhance the killing ability of NK-92 cells. First, tumor cells were treated with different concentrations of SEQ ID NO.45, and then the killing effect of NK cells on MDA-MB-231 and HCC827 was detected using the LDH method.

[0101] The results are as follows Figure 7 As shown, SEQ ID NO.45 is more active than BMS202 (1 μM), and SEQ ID NO.45 significantly enhances the killing ability of NK cells against MDA-MB-231 (human breast cancer cells) and HCC827 (human lung cancer cells) in a dose-dependent manner. 10 μM of SEQ ID NO.45 can increase the killing rate of NK-92 against MDA-MB-231 and HCC827 from about 35% to about 60%. These results illustrate the great potential of SEQ ID NO.45 in enhancing the killing effect of immune cells on tumor cells.

[0102] Example 5 Antitumor effect of small molecule peptides in tumor-bearing mice

[0103] By constructing an MC-38 tumor-bearing mouse model, the anti-tumor effect of SEQ ID NO.45 in tumor-bearing mice was studied. After successful modeling, we randomly divided the tumor-bearing mice into six groups: SEQ ID NO.45 low-dose group (5 mg / kg), SEQ ID NO.45 medium-dose group (10 mg / kg), SEQ ID NO.45 high-dose group (20 mg / kg), PD-L1 monoclonal antibody control group (α-PD-L1, 5 mg / kg), small molecule inhibitor control group (BMS202, 20 mg / kg) and normal saline group. Subsequently, the corresponding drug treatment was given every two days by intraperitoneal injection. During the administration process, the tumor volume and body weight changes of tumor-bearing mice were monitored every two days. When the tumor volume reached 1000mm 3 We killed the mice by cervical dislocation, dissected the mice, took pictures of the tumors, and calculated the tumor inhibition rate.

[0104] The results are as follows Figure 8As shown, SEQ ID NO.45 can show good anti-tumor effect in the low-dose group. Whether in terms of tumor volume increase rate or tumor weight inhibition, it is superior to the PD-L1 monoclonal antibody control group (α-PD-L1), the small molecule inhibitor control group (BMS202) and the saline group. SEQ ID NO.45 can increase the inhibition rate of tumor weight (TGI) in a dose-dependent manner. In the high-dose group, the inhibition rate of tumor weight by SEQ ID NO.45 reached 83%. During the experiment, the weight of mice in the SEQ ID NO.45 treatment group hardly changed. These results show that SEQ ID NO.45 has a significant inhibitory effect on tumor proliferation in vivo and has no obvious toxicity.

Claims

1. A small molecule polypeptide targeting PD-L1, wherein the sequence of the small molecule polypeptide targeting PD-L1 is selected from SEQ ID NO.1-SEQ ID NO.82; the small molecule polypeptide targeting PD-L1 is a mutant polypeptide obtained by performing amino acid scanning mutation on the basis of SEQ ID NO.1-SEQ ID NO.30, wherein the amino acid scanning mutation includes positively charged amino acid scanning mutation, negatively charged amino acid scanning mutation, neutral amino acid scanning mutation, rare amino acid substitution mutation and D-type amino acid substitution mutation, and the amino acid sequence of the small molecule polypeptide targeting PD-L1 after the amino acid scanning mutation is shown in SEQ ID NO.31-SEQID NO.82; the small molecule polypeptide targeting PD-L1 is a polypeptide with N-terminal or C-terminal extension or truncation on the basis of SEQ ID NO.1-SEQ ID NO.30, and the amino acid sequence of the small molecule polypeptide targeting PD-L1 after the extension or truncation is shown in SEQ ID NO.31-SEQ ID NO.82; preferably, the small molecule polypeptide targeting PD-L1 has a sequence of SEQ ID NO.

45.

2. A method for screening small molecule peptides based on machine learning and virtual screening technology: S1. Construction of a three-dimensional structure database: Generate tetrapeptide sequences and files, optimize the energy of the peptide system, and build a tetrapeptide database; S2. Download ligands from the database as active molecules to prepare for virtual screening; S3. Extract three-dimensional structural features for clustering to screen out compounds with similar potential activities as the extracted ligands; S4. The peptides obtained by clustering are finally screened by molecular docking and binding free energy calculation; S5. Perform solid phase synthesis and biological activity testing on the top-ranked peptides to screen out the peptides with the best activity; S6. Structural modification of the polypeptide amino acids screened out in S5 is performed, and the polypeptide molecules with the best activity are screened out based on the binding free energy.

3. The screening method according to claim 2, in step S1, the sequence of the tetrapeptide is generated by Amber's tleap program, covering random combinations of 20 natural amino acids at 4 positions, with a total of 204 combinations; the tetrapeptide file is imported into the required protein and small molecule force field by using the tleap program, and Na + and Cl- to keep the system charge at 0, and obtain it after the polypeptide is solvated; the energy optimization of the polypeptide system is performed using Amber's pmemd program, including two steps of adding solute constraints and optimizing and eliminating solute constraints; the construction of the tetrapeptide database includes two steps of extracting the three-dimensional structure of the polypeptide and removing the water molecules in the three-dimensional structure, and contains 160,000 tetrapeptides; In step S3, the feature extraction of the three-dimensional structure is specifically as follows: using the parsePDB module of PyUUL to read the PDB file of the three-dimensional structure, obtain data such as coordinates, atom names, channels and atomic radii, and transfer this information to CUDA; then, using the point cloud conversion module of PyUUL to convert the read polypeptide file information into the form of surface area point cloud data.

4. A fusion protein, comprising the small molecule polypeptide sequence according to claim 1, wherein the fusion protein has the function of targeting PD-L1.

5. A multimer, formed by the polymerization of two or more polypeptide monomers, wherein at least one polypeptide monomer is the small molecule polypeptide according to claim 1; the polypeptide monomers of the multimer are covalently linked; the multimer is a homodimer or a heterodimer.

6. A nucleic acid molecule encoding the small molecule polypeptide according to claim 1 or the fusion protein or conjugate according to claim 4.

7. A construct comprising the nucleic acid molecule of claim 6.

8. A host cell comprising the nucleic acid molecule of claim 6 and / or the construct of claim 7, wherein the host cell is transformed or transfected by the nucleic acid molecule of claim 6 and / or the construct of claim 7.

9. A pharmaceutical composition, comprising the small molecule polypeptide according to claim 1, the fusion protein according to claim 4, the multimer according to claim 5, the nucleic acid molecule according to claim 6, the construct according to claim 7 and / or the host cell according to claim 8, and optional pharmaceutically acceptable excipients.

10. Use of the small molecule polypeptide according to claim 1, the fusion protein according to claim 4, the multimer according to claim 5, the nucleic acid molecule according to claim 6, the construct according to claim 7 or the host cell according to claim 8 in the preparation of a medicament for preventing and / or treating tumor-related diseases; the tumor-related diseases include but are not limited to head and neck squamous cell carcinoma, nasopharyngeal carcinoma, triple-negative breast cancer, non-small cell lung cancer (NSCLC), small cell lung cancer, esophageal cancer, pleural mesothelioma, gastric cancer, hepatocellular carcinoma, cholangiocarcinoma, renal cancer, urothelial carcinoma, cervical cancer, ovarian cancer, melanoma, skin squamous cell carcinoma, certain special types of soft tissue sarcomas, dMMR / MSI-H solid tumors including colorectal cancer, and solid tumors with TMB≥10mut / Mb, etc.

Images