Methods and uses for attaching therapeutic molecules to the surface of mature red blood cells - Patents.com

Abstract

The present invention discloses a method for in vitro conjugation of mature red blood cells to glycoproteins on the surface of the red blood cell membrane via glycosidic bonds, which preserves the integrity, deformability, oxygen-carrying capacity, and long half-life properties of the mature red blood cell membrane, and the biological activity of the drug for the treatment of tumors, metabolic disorders, and inflammatory diseases.

Description

[Technical Field]

[0001] The present invention discloses a method for preparing modified cells, which belongs to the field of cell engineering. [Background technology]

[0002] CD8 + T cells and NK cells are cytotoxic effector cells of the immune system, capable of recognizing and killing infected or malignant cells. In clinical practice of tumor therapy, CD8 + T cells and NK cells are stimulated by various measures. For example, the use of cytokines such as interleukins (e.g., IL-2, IL-12, IL-15, IL-18, IL-21) stimulates CD8 +They promote the differentiation and activation of T cells and NK cells. However, clinical trials have demonstrated complex effects, including tumor type, administration conditions, short half-life, and potential toxicity due to their effects on non-immune cells, all contributing to insufficient therapeutic efficacy. (2019 - Star Review: The challenges and molecular approaches surrounding interleukin-2-based therapy; 2019 - IL-15 for cancer and autoimmunity; 2021 - Next-generation cytokines for cancer immunotherapy; 2020 - Cancer-killing, decoy-resistant interleukin-18) Furthermore, immune costimulatory molecules, including CD27, CD40, OX40, GITR, and ICOS, are also undergoing clinical trials as second signals for immune cell activation. For example, 4-1BB (CD137, tumor necrosis factor receptor superfamily member 9) belongs to the tumor necrosis factor receptor protein family. It is primarily expressed on the surface of various immune cells, including activated T lymphocytes (CD4+ and CD8+), NK cells, macrophages, and B cells. Preclinical studies have shown that activation of 4-1BB signaling stimulates the activity of cytotoxic T cells and NK cells, resulting in sustained immune memory responses. However, 4-1BB agonists are prone to systemic side effects due to the diffuse expression of 4-1BB. In December 2008, all clinical studies of the 4-1BB agonist urelumab monoclonal antibody were discontinued after two fatal cases of hepatotoxicity occurred. A subsequent detailed analysis of clinical safety data showed that the dose of urelumab monoclonal antibody was the single most important factor associated with severe liver injury (2017 - Results from an Integrated Safety Analysis of Urelumab, an Agonist Anti-CD137 Monoclonal Antibody).Further studies showed that the urelumab monoclonal antibody reaches the liver parenchyma via the blood circulation, activates liver-specific macrophages (Kuffers cells), and causes nonspecific damage to liver cells (2018-Activation of 4-1BB on liver myeloid cells triggers hepatitis).

[0003] Chimeric antigen receptor T-cell immunotherapy (CAR-T) is an adoptive therapy using novel engineered T cells. They recognize tumor cells and activate the engineered T cells, resulting in tumor killing. However, the technical process of CAR-T is complex and costly. The main technical challenge of cell engineering, such as CAR-T, is endowing the engineered cells with new properties without inhibiting their intrinsic functions. Currently, the most common and robust cell engineering methods are limited by technical complexities and safety issues, such as the reproducibility of viral transduction efficiency in primary cells, heterogeneous expression levels of CAR genes, and the potential for endogenous gene disruption.

[0004] Direct modification of cell surfaces using chemical biology tools has become a complementary and generally applicable method in cell therapy. Glycosylation is the covalent attachment of carbohydrates to target molecules (usually proteins or lipids). Protein glycosylation is a template-free enzymatic reaction in which the donor molecule is typically an activated nucleotide sugar that reacts with specific glycoconjugates at acceptor sites (hydroxyl or other functional groups) through the action of glycosyltransferases. Fucose is widely present on the plasma membrane of various cells as a component of the glycans of glycoproteins. Fucosyltransferases are enzymes that transfer L-fucose from the donor substrate GDP-fucose (guanosine diphosphate fucose) to the acceptor substrate. According to current literature reports, the main donor substrate for fucosyltransferases is GDP-fucose, which has a relatively small molecular weight. Fucose is transferred to the N-glycans of glycoproteins in mammalian cells through the action of fucosyltransferases. Wu et al. successfully used Helicobacter pylori-derived fucosyltransferase to transfer large molecules such as antibodies to glycans such as LacNAc and α2,3-sialyl LacNAc on the cell membrane surface. Using this technique, Wu et al. used a natural killer cell line (NK-92MI) and primary murine CD8+ OT-1 T cells to transfer anti-Her2 and anti-PD-L1 antibodies, respectively, via fucosyltransferase. This resulted in the construction of two types of engineered cells that demonstrated specific tumor targeting and suppressive signaling against tumor cell production in a mouse model (see Li J, et al. ACS Cent Sci. 2018 Dec 26;4(12):1633-1641). Therefore, the efficacy of cell therapies such as CAR-T can be significantly improved by using fucosyltransferase to tag target cells bearing acceptor substrates with target molecules bound to donor substrates.Although this approach is a useful complement to CAR-T and adoptive immune cell therapy, it remains a highly customized treatment and is limited by the high cost due to the number of available patient immune cells, the complexity of the in vitro cell expansion process, the cost and labor involved in the operating procedures, and the difficulty of selecting immune cell subtypes. Furthermore, the short half-life of immune cells in vivo makes them susceptible to suppression by the tumor microenvironment, leading to functional exhaustion, limiting their practical application. [Prior art documents] [Non-patent literature]

[0005] [Non-Patent Document 1] Li J, et al. ACS Cent Sci. 2018 Dec 26;4(12):1633-1641 Summary of the Invention [Problem to be solved by the invention]

[0006] In view of the above-mentioned technical deficiencies of the conventional technology, the object of the present invention is to provide a new method applicable to the field of tumor immunotherapy that can efficiently and exclusively activate NK cells and CD8+ T cells present in vivo, either directly or indirectly, in order to distinguish it from the highly customized cell therapy of the conventional technology (which requires increasing the numbers of NK cells and T cells by in vitro expansion before infusion into the patient). [Means for solving the problem]

[0007] Based on the above object of the present invention, first, (1) modifying a chemical molecule or a biological polymer to covalently attach a GDP-fucose derivative to the chemical molecule or the biological polymer; (2) covalently binding the chemical molecule or biopolymer obtained in step (1) to a sugar chain on the surface of the erythrocyte membrane via a glycosidic bond by an enzymatic reaction catalyzed by a fucosyltransferase; The present invention provides a method for attaching a chemical molecule or a biopolymer to the surface of a mature red blood cell, comprising:

[0008] The GDP-fucose derivatives described in the present invention, such as GDP-fucose-(dibenzocyclo)triazole-PEGn (n = 0 to 12) and GDP-fucose-lactic acid-hydroxyacetic acid, can be conjugated to chemical molecules or biological macromolecules. In a specific embodiment of the present invention, the GDP-fucose derivative is guanosine 5'-diphosphate-fucose-triazole-polyethylene glycol-methyltetrazine (GDP-fucose-triazole-PEG4-Tz).

[0009] In one preferred technical means, the chemical molecule is selected from therapeutic molecules or fluorescent moieties.The therapeutic small molecules include but are not limited to glucocorticoids such as dexamethasone, anticoagulants such as warfarin, platelet aggregation inhibitors such as clopidogrel or tegretol, angiotensin inhibitors such as captopril, and antianginal drugs such as trimetazidine; and the fluorescent moieties include but are not limited to Cy5, Cy3.

[0010] In another preferred embodiment, the biopolymer is selected from polynucleotides, polypeptides, and antibodies, where polynucleotides include but are not limited to CpG-containing single-stranded oligonucleotides TCCATGACGTTCCTGACGTT, TCGTCGTTTTGTCGTTTTGTCGTT, and CCTGGATGGGAACTTACCGCTGCA, polypeptides include but are not limited to vascular endothelial growth factor inhibitors, glutathione, somatostatin, glatiramer acetate, and their Fc fusion proteins, and antibodies include but are not limited to anti-VEGF antibodies, anti-4-1BB antibodies, anti-Tie2 antibodies, and anti-CD40 antibodies.

[0011] More preferably, said antibody is selected from a monoclonal antibody, a single chain antibody, a bispecific antibody, a nanobody.

[0012] Particularly preferably, said antibody is an anti-4-1BB activating antibody.

[0013] In another preferred technical means, said biopolymer is selected from interleukin IL-15 isoforms or interleukin IL-12.

[0014] In another preferred technical means, the mature red blood cells are collected from peripheral blood and serum and other blood cells are removed, there are no free protein molecules in the red blood cell suspension, and the red blood cell suspension is freshly prepared or stored under appropriate conditions for no more than 30 days.

[0015] In another preferred technical means, the fucosyltransferase is a human fucosyltransferase or a Helicobacter pylori fucosyltransferase.

[0016] In one more preferred technical means, the GDP-fucose derivative is guanosine 5'-diphosphate-fucose-triazole-polyethylene glycol-methyltetrazine.

[0017] More preferably, the enzymatic reaction of the fucosyltransferase in step (2) is carried out under reaction conditions in which the density of red blood cells is 5×10 7 / mL ~ 5 × 10 9 / mL, the fucosyltransferase concentration is 40 μg / mL to 300 μg / mL, the pharmaceutical polymer is an antibody or Fc fusion protein at a concentration of 60 μg / mL to 120 μg / mL, the reaction temperature is 4°C to 37°C, and the reaction time is 20 minutes to 16 hours.

[0018] In a specific embodiment of the invention, the density of red blood cells is 5×10 7 / m, the optimal binding effect to the Fc fusion protein derivative of human IL-15 (also referred to as N803 in the present invention) is obtained.

[0019] Particularly preferably, the enzymatic reaction of the fucosyltransferase is carried out in the presence of Mg 2+ The reaction system is free of fucosyltransferase and contains 150 mM sodium chloride, 2.7 mM potassium chloride, 44 mM glucose, and a pH of 5.9 to 6.3. According to a prior art report, "Jie, L., et al. "A Single-Step Chemoenzymatic Reaction for the Construction of Antibody-Cell Conjugates," ACS Central Science 4.12 (2018)," fucosyltransferase is magnesium (Mg) dependent, and when catalyzing a glycosyltransferase reaction, 20 mM Mg is required in the system. 2+ However, in the pharmaceutical field, additives in the pharmaceutical preparation process can affect the quality and safety of the final drug. In particular, in the latter case, changes in the concentration of metal ions in the blood can lead to dysfunction of various organs. For example, hypermagnesemia initially causes symptoms such as loss of appetite, nausea, vomiting, flushing of the skin, headache, and dizziness, but these symptoms are easily overlooked due to their lack of specificity. When serum magnesium concentrations reach 2-4 mM, obvious changes can appear in the nervous, muscular, and circulatory systems. Furthermore, Mg 2+ The presence of Mg can also affect the integrity and half-life of red blood cells. Commercially available red blood cell storage solutions contain sodium chloride, sodium citrate, citric acid, and glucose as their main components. Red blood cells can be stored at 4°C for 2 weeks without any change in activity or characteristics. 2+ "Ye Hanquan et al., Dynamic Observation of Quality Control of Suspension and Washing of Red Blood Cells Using Red Blood Cell Preservation Solution, Changjiang University Journal (Self-Medical Edition), Late Edition, 2013" is a commonly used solution for collecting, storing, and transporting red blood cells that do not contain magnesium. According to an experimental study, 2+ It has been shown that adding to PBS buffer or red blood cell storage solution still causes red blood cell membrane rupture and hemolysis even at 4 °C (data not shown). Therefore, removing additives and auxiliary ingredients as much as possible in pharmaceutical processes can minimize drug safety risks to patients.

[0020] In the present invention, macromolecular drugs that bind to GDP-fucose derivatives, such as antibodies or fusion proteins that bind to GDP-fucose-triazole-PEG4-Tz (guanosine 5'-diphosphate-fucose-triazole-polyethylene glycol-methyltetrazine) (abbreviated as GDP-fucose-antibodies) are used as substrates for fucosyltransferases. These antibodies or fusion proteins are not naturally occurring low-molecular-weight substrates (e.g., the commercially available derivative GDP-fucose-azide, GDP-fucose-azide, with a molecular weight of only 630 Da), but are macromolecular (antibodies or fusion proteins with a molecular weight of 50 kD or greater). Another substrate for fucosyltransferases is the sugar chain on cell membrane glycoproteins, such as LacNAc. The GDP-fucose-antibody forms a glycosidic bond with LacNAc through the action of fucosyltransferase, releasing GDP, forming (cell membrane) LacNAc-fucose-antibody, thereby binding the antibody molecule to the cell membrane. Analysis of the molecular structure revealed that the steric hindrance between fucosyltransferase and GDP-fucose-antibody is significantly greater than that between fucosyltransferase and low-molecular-weight substrates. 2+ It also affects the action of Mg 2+ It inhibits the enzyme-promoting activity of

[0021] The present invention provides a method for the treatment of red blood cells with Mg as described above. 2+ Taking into consideration the adverse effect of fucosyltransferase substrate damage on the enzyme reaction and the fact that free phosphate molecules inhibit the enzyme reaction as a product, we removed the free phosphate molecules from the reaction system to prevent the formation of Mg 2+ To counteract the adverse effects of the deficiency, the existing PBS buffer solution (137 mM sodium chloride, 10 mM phosphate, 2.7 mM potassium chloride, pH 7.4) was adjusted to an equilibrium solution called phosphate-free NPBS (Non-Phosphate Buffer Solution): 150 mM sodium chloride, 2.7 mM potassium chloride, 44 mM glucose), and the pH was adjusted from the existing 7.2-7.6 to a more acidic environment of 5.9-6.3. The adjusted reaction system is more suited to the in vitro survival conditions of red blood cells. As a result, in this reaction system, Mg2+ It was shown that the removal of β-glucan did not affect the enzymatic reaction.

[0022] Next, the present invention provides modified red blood cells obtained by preparing them according to the above method.

[0023] Finally, the present invention provides the use of the above-described modified red blood cells in the preparation of a medicament for treating a tumor disease.

[0024] In a preferred technical means, said tumor disease is colon cancer or melanoma. [Effects of the Invention]

[0025] The excellent technical effects of the present invention are mainly as follows:

[0026] The present invention relates to a method for activating and / or expanding NK cells and / or CD8+ T cells by binding one or more exogenous stimulatory molecules (e.g., 4-1BBL or a combination thereof) to the surface of mature erythrocytes in peripheral blood. Immune cells (e.g., NK cells and / or CD8+ T cells) circulating in peripheral blood are stimulated with the modified erythrocytes. The present invention has discovered that modified erythroid cells containing IL-12, IL-15 / IL-15RA, 4-1BBL, or a combination thereof, are effective in activating primary CD4+, CD8+, and NK cells and inducing cytotoxicity.

[0027] The modified red blood cells of the present invention have binding sites that do not directly involve amino acid molecules on the cell membrane surface, and the binding reaction is rapid and mild, resulting in good preservation of the red blood cell membrane. It has been shown that the deformability of the red blood cell membrane, intracellular 2,3-DPG, and free hemoglobin are not significantly different from those of natural red blood cells. The immune stimulatory molecules bound to the surface of red blood cells maintain stimulatory signals in the circulatory system and have a long circulatory half-life, providing a safer and more effective method for stimulating immune killer cells.

[0028] The erythroid cells of the present invention are modified into modified erythroid cells. The modified erythroid cells may be nucleated, such as erythroid progenitor cells (e.g., erythroid progenitor cells), or enucleated, such as reticulocytes or red blood cells. The present invention further provides the use of these modified erythroid cells in activating NK cells and / or CD8+ T cells in subjects in need thereof (e.g., cancer subjects or infectious disease subjects). The modified erythroid cells, including IL-12, IL-15 / IL-15RA, 4-1BBL, and combinations thereof, slow tumor growth and effectively reduce tumor burden in the body. Furthermore, pharmaceuticals prepared using the technology of the present invention are in-the-box products, which offer obvious advantages in terms of convenience, availability, and manufacturing costs. [Brief explanation of the drawings]

[0029] [Figure 1] FIG. 1 is a schematic diagram of the attachment of therapeutic molecules to the surface of mature red blood cells and their uses. [Figure 2] FIG. 1 shows a molecular weight analysis diagram by mass spectrometry before and after N803 binding. [Figure 3] This is a histogram of the mass of N803 protein on the erythrocyte membrane detected by flow cytometry. [Figure 4] 1 is a flow cytometry plot of unbound red blood cells and N803 protein-bound red blood cells. [Figure 5] 1 is a flow cytometry plot of the effect of red blood cell density on antibody binding. [Figure 6] 1 is a flow cytometry plot of the effect of fucosyltransferase concentration on antibody binding. [Figure 7] (I) Flow cytometry plot of the effect of N803 protein on antibody binding. [Figure 8] Flow cytometry plot of the effect of N803 protein on antibody binding (II). [Figure 9] Flow cytometry plot of the effect of N803 protein on antibody binding (III). [Figure 10]1 is a flow cytometry plot of the effect of temperature on N803 protein-binding red blood cells. [Figure 11] 10 is a flow cytometry plot of the effect of Mg2+ on 3H3 protein binding to HEK293 cell membranes. [Figure 12] This is a histogram showing the amount of 3H3 protein on the red blood cell membrane as detected by flow cytometry. [Figure 13] This is a histogram of the amount of Avastin protein on the red blood cell membrane detected by flow cytometry. [Figure 14] 1 is an analytical chart of the ATP content of mouse red blood cells after protein binding. [Figure 15] 1 is an analytical chart of the effect of incubation time on the ATP content of mouse erythrocytes. [Figure 16] 1 is an analytical chart showing the effect of protein concentration on the ATP content of mouse red blood cells. [Figure 17] 1 is an analytical chart showing the effect of fucosyltransferase concentration on the ATP content of mouse erythrocytes. [Figure 18] 1 is a chart analyzing the effect of antibody binding on the 2,3-DPG content of red blood cells. [Figure 19] 1 is a chart analyzing the effect of antibody binding on red blood cell membrane integrity. [Figure 20] 1 is a chart analyzing the effect of incubation time on red blood cell membrane integrity. [Figure 21] 1 is a chart analyzing the effect of protein concentration on red blood cell membrane integrity. [Figure 22] 1 is a chart analyzing the effect of fucosyltransferase concentration on red blood cell membrane integrity. [Figure 23] 1 is an analysis chart of the effect of two types of binding methods on the deformability of red blood cells. [Figure 24] 1 is an analysis chart of the effect of two types of binding methods on the uniformity of red blood cells. [Figure 25] This is an evaluation of the half-life of antibody-bound mouse red blood cells in the mouse body. [Figure 26] FIG. 1 is a schematic diagram of a reporter gene system for detecting 4-1BB activation in vitro. [Figure 27] 1 is an analytical chart showing the reporter gene activation ability of mouse erythrocytes bound to the 3H3 antibody. [Figure 28] This is an analysis chart of CD8+ T cells and NK cells produced after costimulation with 3H3-binding red blood cells. [Figure 29] 10 is an analysis chart of interferon-γ secretion by CD8+ T cells induced by anti-4-1BB antibody-bound mouse red blood cells. [Figure 30] FIG. 10 is a graph showing the trend of changes in tumor volume after injection of antibody 3H3-conjugated red blood cells. DETAILED DESCRIPTION OF THE INVENTION

[0030] The present invention will be further described below with reference to specific examples, and the advantages and features of the present invention will become more apparent from the description. However, these examples are merely illustrative and do not limit the scope of protection defined by the claims of the present invention.

[0031] 1 shows a flowchart of the method of the present invention for binding chemical molecules or biopolymers bound to GDP-fucose derivatives by fucosyltransferase to the surface of mature red blood cells, and for perfusing them into a patient's body. Specific embodiments of the technical methods related to the above flow are described below.

[0032] [Example 1] Collection of red blood cells from Balb / C mice and humans 10-12 week-old SPF female Balb / C mice were selected and sterilized under anesthesia. Cardiac blood was collected and transferred to a K2-EDTA-coated centrifuge tube to prevent coagulation. The tube was then stored at 4°C, yielding 0.5-0.8 mL of whole blood per mouse. The blood was centrifuged at 500 × g for 5 minutes at 4°C, and the red blood cell hematocrit (red, lower layer) and plasma (yellow, upper layer) levels were recorded in the test tube. The plasma and intermediate sediment (buffy coat) were slowly and thoroughly aspirated with a micropipette. Bleach was added to the aspirated liquid and discarded in a biohazard container. 0.1 mL of the red blood cell hematocrit was transferred to a 1.5 mL centrifuge tube, 1 mL of pH 7.4 PBS solution was added, the tube was closed, and the tube was inverted several times to mix evenly. The tube was centrifuged at 500 × g for 5 minutes at 4°C, and the supernatant was aspirated and discarded. The PBS washing process was repeated 3-4 times. Add 100 μL to 1000 μL of PBS to the washed red blood cells and store at 4°C until use.

[0033] After skin disinfection, 2–3 mL of blood was collected from an antecubital vein by a professional using a disposable vacuum blood collection device from a healthy, anonymous donor. The blood slowly flowed along the wall of the test tube. When it reached the designated marker, the tube was removed and quickly inverted several times to mix evenly. The blood was centrifuged at 500 × g for 5 minutes at 4°C, and the red blood cell hematocrit (red, lower layer) and plasma (yellow, upper layer) levels were displayed on the test tube. The plasma and intermediate sediment (buffy coat) were slowly and thoroughly aspirated with a micropipette. Bleach was added to the aspirated liquid and discarded in a biohazard container. 0.5 mL of the red blood cell hematocrit was transferred to a 15 mL centrifuge tube, 5 mL of pH 7.4 PBS solution was added, the tube was closed, and the tube was inverted several times to mix evenly. The tube was centrifuged at 500 × g for 5 minutes at 4°C, and the supernatant was aspirated and discarded. The PBS washing process was repeated 3–4 times. After washing, 1000 μL of PBS was added to the red blood cells, and the cells were stored at 4° C. until use.

[0034] [Example 2] Binding of GDP-fucose to N803 and detection by mass spectrometry N803 (also known as ALT803) is an Fc fusion protein derivative of human IL-15. For the study "2013 The IL-15-based superagonist ALT 803 promotes the antigen-independent conversion of memory CD8 T cells into innate-like effector cells with antitumor," the corresponding DNA was designed and fused to a human IgG4-Fc sequence according to the published sequence. Souzhou Biotechnology Co., Ltd. synthesized the plasmid DNA and cloned it into the vector pRM293 (pRM293 was obtained by modifying the plasmid pTT5; see Shi C. Purification and characterization of a recombinant G-protein-coupled receptor, Saccharomyces cerevisiae Ste2p, transiently expressed in HEK293 EBNA1 cells. Biochemistry. 2005;44(48):15705-15714). We commissioned Souzhou Biotechnology Co., Ltd. to transiently transfect HEK293 cells (National Research Council, Canada) with the plasmids and perform affinity purification using Mabselect Sure (Protein A, GE Healthcare). For further use, GDP-fucose was conjugated to the purified N803 by crosslinking the amine with N-hydroxylsuccinimide ester (NHS ester). Hereinafter, unless otherwise specified, N803 refers to GDP-fucose-conjugated N803.

[0035] 1. NHS ester-activated crosslinkers and labeling compounds reacted with primary amines on N803 under physiological to slightly alkaline conditions (pH 7.2-9) to form stable amide bonds. Using this principle, the NHS on TCO-PEG4-NHS reacted with primary amines on the N803 protein to obtain TCO-PEG4-N803. The specific method and procedure are as follows:

[0036] Four hundred micrograms of purified N803 protein was reacted with 150 micrograms of TCO-PEG4-NHS (Shanghai Pairui Pharmaceutical Technology Co., Ltd., product number A34125). 20 mM HEPES buffer (pH 7.0-7.5, Thermo Fisher Scientific (China) Co., Ltd., product number 15630) was added, and the mixture was incubated at room temperature for 30 minutes. Five micromolar amounts of Tris buffer were added to terminate the reaction, and the mixture was incubated at room temperature for 5 minutes. The reaction product was then loaded onto a PD SpinTrap G-25 desalting column (Cytiva, product number 28918004) and centrifuged at 800 × g to remove unreacted products and small molecules from the reaction product, yielding pure TCO-PEG4-NHS.

[0037] 2. Based on the reverse-request Diels-Alder cycloaddition reaction between the trans-cyclooctene / tetrazine reaction pair, the bond between TCO (trans-cyclooctene) and Tz (tetrazine) formed a dihydropyridazine bond. Using this principle, TCO-PEG4-N803 was reacted with the GDP-fucose derivative GDP-fucose-triazole-PEG4-Tz to obtain GDP-fucose-triazole-PEG4-PEG4-N803, i.e., GDP-fucose-linked N803. The specifics are as follows:

[0038] The TCO-PEG4-N803 obtained in the previous step was reacted with 30 μg of GDP-Fucose-Triazole-PEG4-Tz (synthesized by Yantang Biotechnology Co., Ltd., product number YT-HJP-3-29). After incubation at room temperature for 30 minutes, the reaction product was applied to a PD SpinTrap G-25 desalting column (Cytiva, product number 28918004) and centrifuged at 800 × g to remove unreacted products and small molecules from the reaction product, yielding pure GDP-fucose-linked N803.

[0039] 3. MALDI-TOF measurement of the molecular weight of N803 protein before and after GDP-fucose derivative binding

[0040] GDP-fucose derivatives were conjugated to lysines in protein molecules by crosslinking amines with N-hydroxylsuccinimide ester (NHS ester). The molecular weight of each lysine increased by approximately 1462.5 Da upon conjugation. By comparing the relative molecular weights of protein molecules before and after conjugation, the number of GDP-fucose derivatives conjugated to the protein could be estimated. Soft ionization mass spectrometry (MSM)—matrix-assisted laser desorption / ionization time-of-flight (MALDI-TOF)—can measure the relative molecular weights of glycoproteins, polypeptides, and amino acids. In this experiment, the molecular weight of N803 protein before and after conjugation with GDP-fucose derivatives was accurately determined by MALDI-TOF. The mass spectrometer used was a G2-XSQ-Tof / Tof (Waters Corporation, US) and was performed by Shanghai Aipudikang Biotechnology Co., Ltd. The test method was as follows: 1) Spotting: 0.5 μL of sample is spotted onto the sample target, and after air drying, 0.5 μL of SA matrix solution is spotted onto the corresponding target and air dried. 2) Calibration: Select the linear method in positive ion mode to test the calibration of the detection range of the sample. 3) Test sample: Select the linear method in positive ion mode to test the molecular weight of the sample. 4) Mass analysis data and mass spectrum processing: The raw data and mass spectra generated by MALDI-TOF are processed by the software provided with the instrument.

[0041] MALDI-TOF analysis revealed that the relative molecular weight of N803 before conjugation was 68,681.5 Da (Figure 2, left, center, and right mass spectra are based on the relative molecular weights of IL15RA-sushi-Fc in N803 before and after conjugation, respectively). The relative molecular weights of N803 after conjugation were 77,621.05 and 83,158.23 Da. Since the molecular weight increased by approximately 1,462.5 Da after conjugation of the GDP-fucose derivative, N803 after conjugation carried approximately 6–10 GDP-fucose derivatives. Hereafter, unless otherwise specified, N803 refers to GDP-fucose-conjugated N803.

[0042] [Example 3] Binding of human peripheral blood erythrocytes to IL-15 derivatives and FACS detection After washing the red blood cells with PBS, serum and other cells were completely removed as needed. 9To 1 mL of red blood cells, 10 μg / mL and 100 μg / mL of fucosyltransferase (prepared as described in CN114369584A and CN114369585A, respectively; in the present invention, the recombinant Helicobacter pylori-derived fucosyltransferase mutant in CN114369585A, whose sequence is represented by SEQ ID NO. 13) and 15 μg / mL and 150 μg / mL of N803 were added, respectively, and the mixture was incubated at room temperature for 30 minutes. The mixture was washed with PBS (pH 7.4), centrifuged at 500 × g, and the supernatant was removed. Erythrocytes were resuspended in PBS (pH 7.4) and incubated with anti-hIgG-Biotin antibody (Beijing Biolab Technology Co., Ltd., product number F030822, diluted 1:200) at 4°C for 30 minutes. After washing once with PBS, they were stained with streptavidin-PE (BioLegend, Inc., product number 405204, diluted 1:200) at 4°C for 30 minutes. After washing once with PBS, they were detected by flow cytometry (Beijing Challen Biotechnology Co., Ltd., product number MateCyte) and analyzed using NovoExpress software.

[0043] Figure 3 shows histograms of the amount of N803 protein on red blood cell membranes detected by flow cytometry under the above-mentioned binding conditions (anti-hIgG-Biotin / Strep-PE). The top, middle, and bottom plots show red blood cells bound with 15 μg / mL N803 and 10 μg / mL fucosyltransferase (top), 150 μg / mL N803 and 100 μg / mL fucosyltransferase (middle), and unbound red blood cells (bottom). N803 detection: First, anti-hIgG-Biotin antibody specifically bound to the human IgG4-Fc sequence of N803 protein. Next, streptavidin-PE (abbreviated as Strep-PE) specifically bound to the biotin on the anti-hIgG-Biotin antibody. Finally, PE fluorescence was detected by flow cytometry. The intensity of PE fluorescence was positively correlated with the amount of N803. As can be seen from the figure, PE (N803) was detected in the bound red blood cells under both conditions and was clearly distinguishable from the unbound red blood cells. It was also found that the amount of N803 bound correlated with the number of antibodies and enzymes, with the greater the number, the greater the amount of binding.

[0044] To further identify the cell types, N803-bound red blood cells were stained with anti-GPA (CD235)-APC (BioLegend, Inc., Product No. 306608, 1:1400 dilution). GPA is a cell membrane protein specific to red blood cells. Figure 4 shows the flow cytometry results of unbound red blood cells (naive hRBC, middle) and N803 protein-bound red blood cells (hRBC-N803, top) for GPA (vertical axis) and N803 (anti-hIgG-PE, horizontal axis). Unstained red blood cells served as a blank control (neg, bottom). Detection Procedure: After washing once with PBS, unbound and N803 protein-bound red blood cells were simultaneously stained with anti-GPA (CD235)-APC and anti-hIgG-PE (Jackson ImmunoResearch Inc., product number 109-116-170, 1:200 dilution) antibodies for 30 minutes at 4°C. After washing once with PBS, the cells were detected by flow cytometry (Beijing Challen Biotechnology Co., Ltd., product number MateCyte) and analyzed using NovoExpress software. As can be seen in Figure 4, over 99% of the red blood cells obtained using the above method were GPA-positive compared to the blank control, indicating that over 99% of the cells were red blood cells. PE (N803) was detected in all bound red blood cells compared to unbound red blood cells, indicating that all of these bound red blood cells were bound to N803.

[0045] [Example 4] Binding of mouse erythrocytes to IL-15 derivatives and FACS detection to examine the effects of erythrocyte density, enzyme concentration, drug concentration, temperature, and reaction time on the binding effect

[0046] 1. Optimization of red blood cell density under antibody-bound red blood cell reaction conditions GDP-fucose-bound N803 was prepared according to Example 3. Mouse erythrocytes were washed with PBS, and serum and other cells were completely removed as needed. The fucosyltransferase concentration was 100 μg / mL, the N803 concentration was 300 μg / mL, and erythrocytes were added to 5 × 10 9 / mL, 2.5 × 108 / mL, 10 8 / mL, 5 × 10 7 The red blood cells were added to the reaction system at a concentration of 1 / mL. After incubation at room temperature for 30 minutes, they were washed with pH 7.4 PBS, centrifuged at 500 × g, and the supernatant was removed. The red blood cells were resuspended in pH 7.4 PBS and co-incubated with anti-hIgG-Biotin antibody (Beijing Biolab Technology Co., Ltd., product number F030822, 1:200 dilution) at 4°C for 30 minutes. After washing once with PBS, they were stained with Streptavidin-PE (BioLegend, Inc., product number 405204, 1:200 dilution). Detection was performed by flow cytometry (Beijing Challen Biotechnology Co., Ltd., product number MateCyte) and analyzed using NovoExpress software. The results showed that the red blood cell density was 5 × 10 7 / mL, the fluorescence intensity was highest, indicating that the amount of N803 bound was the greatest (Figure 5).

[0047] 2. Optimization of fucosyltransferase concentration under antibody-bound erythrocyte reaction conditions GDP-fucose-bound N803 was prepared according to Example 3. Mouse erythrocytes were washed with PBS and then completely removed from serum and other cells as needed. The concentration of erythrocytes was 5 × 10 8The reaction mixture was incubated at 600 μg / mL with fucosyltransferase at 100 μg / mL, 200 μg / mL, or 300 μg / mL. After incubation at room temperature for 30 minutes, the cells were washed with PBS (pH 7.4) and centrifuged at 500 × g to remove the supernatant. Erythrocytes were resuspended in PBS (pH 7.4) and co-incubated with anti-hIgG-Biotin antibody (Beijing Biolab Technology Co., Ltd., product number F030822, 1:200 dilution) at 4°C for 30 minutes. After washing once with PBS, the cells were stained with Streptavidin-PE (BioLegend, Inc., product number 405204, 1:200 dilution). Analysis was performed by flow cytometry (Beijing Challen Biotechnology Co., Ltd., product number MateCyte) and analyzed using NovoExpress software. The results (Figure 6) showed that N803 could effectively bind to red blood cells under these reaction conditions when the fucosyltransferase concentration was 100 μg / mL to 300 μg / mL.

[0048] 3. Optimization of the concentration of pharmaceutical polymer N803 under antibody-binding red blood cell reaction conditions GDP-fucose-bound N803 was prepared according to Example 1. Mouse erythrocytes were washed with PBS, and serum and other cells were completely removed as needed. The density of the erythrocytes was 10 7The reaction mixture was incubated at room temperature for 30 minutes, then washed with PBS (pH 7.4), centrifuged at 500 × g, and the supernatant was removed. Erythrocytes were resuspended in PBS (pH 7.4) and co-incubated with anti-hIgG-Biotin antibody (Beijing Biolab Technology Co., Ltd., product number F030822, 1:200 dilution) for 30 min at 4 °C. After washing once with PBS, the cells were stained with Streptavidin-PE (BioLegend, Inc., product number 405204, 1:200 dilution). Detection was performed by flow cytometry (Beijing Challen Biotechnology Co., Ltd., product number MateCyte) and analyzed using NovoExpress software. The results (Figures 7-9) showed that N803 could effectively bind to erythrocytes at concentrations between 60 μg / mL and 600 μg / mL.

[0049] 4. Optimization of reaction temperature and reaction time under antibody-bound red blood cell reaction conditions GDP-fucose-bound N803 was prepared according to Example 1. Mouse erythrocytes were washed with PBS, and serum and other cells were completely removed as needed. The concentration of erythrocytes was 10 7The concentrations of fucosyltransferase and N803 were 40 μg / mL and 480 μg / mL, respectively. The reaction conditions were overnight at 4°C (approximately 16 hours), 20 minutes or 1 hour at room temperature, and 20 minutes or 1 hour at 37°C. After incubation, the cells were washed with PBS (pH 7.4), centrifuged at 500 × g, and the supernatant was removed. Erythrocytes were resuspended in PBS (pH 7.4) and co-incubated with anti-hIgG-Biotin antibody (Beijing Biolab Technology Co., Ltd., product number F030822, 1:200 dilution) for 30 minutes at 4°C. After washing once with PBS, the cells were stained with Streptavidin-PE (BioLegend, Inc., product number 405204, 1:200 dilution) and analyzed by flow cytometry (Beijing Challen Biotechnology Co., Ltd., product number MateCyte) using NovoExpress software. The results (Figure 10) showed that N803 bound to erythrocytes at all three temperatures, whereas better separation was achieved at room temperature (10°C–30°C) for 20 minutes.

[0050] [Example 5] Composition of reaction solution, effect of Mg ions on binding reaction According to the literature report "Jie, L., et al. "A Single-Step Chemoenzymatic Reaction for the Construction of Antibody-Cell Conjugates." ACS Central Science 4.12 (2018)," fucosyltransferase is magnesium (Mg) dependent, and when catalyzing the glycosyltransferase reaction, 20 mM Mg is required in the system. 2+However, in the pharmaceutical field, additives used in the drug preparation process can affect the quality and safety of the final drug. In particular, changes in the concentration of metal ions in the blood can lead to dysfunction of various organs. For example, hypermagnesemia initially manifests with symptoms such as loss of appetite, nausea, vomiting, flushed skin, headache, and dizziness, but it is easily overlooked due to its lack of specificity. When serum magnesium levels reach 2-4 mM, obvious changes can appear in the nervous, muscular, and circulatory systems. Therefore, eliminating additives and auxiliary ingredients as much as possible in the pharmaceutical process can minimize the safety risks of drugs to patients.

[0051] Commercially available red blood cell preservative solutions contain sodium chloride, sodium citrate, citric acid, and glucose as their main components. Red blood cells can be stored at 4°C for 2 weeks without any changes in activity or characteristics. They are often used for the collection, storage, and transportation of red blood cells. 2+ "Ye Hanquan et al., Dynamic Observation of Quality Control of Suspension Washing of Red Blood Cells Using Red Blood Cell Preservation Solution, Changjiang University Journal (Self-Science Edition), Late Edition, 2013" (20mM Mg) 2+ It was shown that when added to PBS buffer or red blood cell storage solution, red blood cells still ruptured their red blood cell membranes and hemolysis occurred even at 4°C (data not shown).

[0052] However, Mg 2+ is an important cofactor for enzymes, stabilizing the three-dimensional structure, forming the active center of the enzyme, and acting as a bridge between the substrate molecule and the enzyme protein. In this research field, GDP-fucose is used as a substrate for fucosyltransferase, where the phosphate molecule of GDP is transferred to Mg 2+It is weakly bound to GDP:Mg, which lowers the activation energy of the catalytic reaction and allows the reaction to proceed smoothly (Simonson, T. and Satpati, P. (2013), Simulating GTP:Mg and GDP:Mg with a simple force field: A structural and thermodynamic analysis. J. Comput. Chem., 34: 836-846.).

[0053] In the present study, drugs bound to GDP-fucose derivatives are used as substrates for fucosyltransferases, and the binding moiety is not a natural low-molecular-weight substrate (e.g., the commercially available derivative GDP-Azido-Fucose, GDP-fucose-azide, with a molecular weight of only 630 Da), but a high-molecular-weight substance (an antibody or fusion protein with a molecular weight of 50 kD or more, abbreviated as GDP-Fucose-Antibody). Another substrate for fucosyltransferases is the sugar chain on cell membrane glycoproteins, such as LacNAc. GDP-Fucose-Antibody forms a glycosidic bond with LacNAc through the action of fucosyltransferase, releasing GDP, and forming LacNAc-Fucose-Antibody, thereby binding the antibody molecule to the sugar chain on the cell membrane. Analysis of the molecular structure revealed that the steric hindrance between fucosyltransferases and GDP-Fucose-Antibody is significantly greater than that between fucosyltransferases and low-molecular-weight substrates, and this steric hindrance prevents the formation of Mg. 2+ It also has a significant effect on the involvement of Mg 2+ On the other hand, considering that free phosphate molecules inhibit enzyme reactions as products, removing the phosphate moiety from existing buffer solutions can reduce the inhibitory effect of the products and further reduce the Mg 2+This can compensate for the negative factors caused by the lack of Mg. To confirm this speculation, the existing PBS buffer solution (137 mM sodium chloride, 10 mM phosphate, 2.7 mM potassium chloride, pH 7.4) was adjusted to a phosphate-free equilibrium solution, named NPBS (Non-Phosphate Buffer Solution: 150 mM sodium chloride, 2.7 mM potassium chloride, 44 mM glucose), and the pH was adjusted to 5.9-6.3, which is more compatible with the in vitro survival conditions of red blood cells. As a result, in this reaction system, Mg 2+ It was shown that the removal of β-glucan did not affect the enzymatic reaction.

[0054] Mg 2+ When Mg is added, red blood cells burst, hemolysis occurs, and 2+ Since it was not possible to evaluate the enzymatic reaction of the NPBS reaction system with or without ATP, the reaction system was verified using a human embryonic kidney cell line (HEK293).

[0055] GDP-fucose-bound 3H3 was prepared according to Example 3. Human embryonic kidney cell line (HEK293) was washed with PBS, and the medium was completely removed as needed. For HEK293 cell-bound 3H3, the same conditions as for erythrocyte-bound N803 were selected: fucosyltransferase concentration 100 μg / mL, 3H3 concentration 120 μg / mL, and HEK293 cell density 5 × 10 7 / mL, and the reaction buffer was NPBS (Mg 2+ without) and NPBS (20 mM Mg 2+After incubation at room temperature for 30 minutes, the cells were washed with NPBS, centrifuged at 200 × g, and the supernatant was removed. 3H3-bound HEK293 cells (HEK293-3H3) were resuspended in PBS and co-incubated with anti-mouse IgG-Biotin antibody (Beijing Biolab Technology Co., Ltd., product number F030805, 1:200 dilution) at 4°C for 30 minutes. After washing once with PBS, the cells were stained with Streptavidin-PE (BioLegend, Inc., product number 405204, 1:200 dilution). Detection was performed by flow cytometry (Agilent Technologies (China), product number NovoCyte) and analyzed using NovoExpress software.

[0056] Figure 11 shows histograms of the amount of 3H3 protein (anti-mIgG-Biotin / Strep-PE) on the HEK293 cell membrane detected by flow cytometry under the above binding conditions. The HEK293 cells shown in the upper, middle, and lower figures were incubated in NPBS (Mg 2+ HEK293 cells (top) bound in the reaction buffer (NPBS (20 mM Mg)). 2+ The figures show HEK293 cells bound under conditions (middle) and unbound HEK293 cells (bottom, negative). 3H3 detection: First, an anti-mIgG-Biotin antibody was used to specifically bind to the mouse IgG2aa sequence of the 3H3 protein. Next, streptavidin-PE (abbreviated as Strep-PE) was used to specifically bind to the biotin on the anti-mIgG-Biotin antibody. Finally, PE fluorescence was detected by flow cytometry. The intensity of PE fluorescence was positively correlated with the amount of 3H3. As can be seen from the figure, PE (3H3) was detected in bound red blood cells under both conditions and was clearly distinguishable from unbound red blood cells. 2+ HEK293 cells bound in the absence of 3H3 showed a fluorescence intensity of 20 mM Mg. 2+ Approximately 20% higher than in buffer conditions, more antibody molecules are bound to the cell membrane, i.e., NPBS (no Mg 2+ ) buffer conditions were found to have better binding efficiency.

[0057] Unless otherwise specified, the erythrocyte binding reaction system contains Mg 2+ NPBS buffer without ATP was used, and PBS buffer was used for other reaction conditions (e.g., flow cytometry detection, ATP concentration detection, 2,3-DPG concentration detection, and washing, storage, functional detection of red blood cell-bound products, as well as various other non-red blood cell detections).

[0058] [Example 6] Mouse erythrocyte-binding anti-mouse 4-1BB antibody and FACS detection 3H3 is an anti-mouse 4-1BB antibody (Rickert, K., Wet et al. Combining phage display with de novo protein sequencing for reverse engineering of monoclonal antibodies. MAbs 8, 501-512 (2016)). The corresponding DNA was designed and fused to the constant region sequence of mouse IgG2aa according to the published sequence. Souzhou Biotechnology Co., Ltd. synthesized the plasmid DNA and cloned it into vector pRM293 (pRM293 was obtained by modifying the plasmid pTT5; see Shi, C. Purification and characterization of a recombinant G-protein-coupled receptor, Saccharomyces cerevisiae Ste2p, transiently expressed in HEK293 EBNA1 cells. Biochemistry. 2005;44(48):15705-15714). We commissioned Souzhou Biotechnology Co., Ltd. to transiently transfect HEK293 cells (National Research Council, Canada) with the plasmids and perform affinity purification using Mabselect Sure (Protein A, GE Healthcare). A GDP-fucose derivative (GDP-Fucose-Triazole-PEG4-Tz; unless otherwise noted, the GDP-fucose derivative used in this study was GDP-Fucose-Triazole-PEG4-Tz) was conjugated to purified 3H3 via amine reaction (the conjugation method was the same as for N803 conjugation described above). Hereafter, unless otherwise noted, 3H3 refers to GDP-fucose-conjugated 3H3.

[0059] The fucosyltransferase concentration was 100 μg / mL, the 3H3 concentration was 200 μg / mL, and the red blood cells were diluted to a density of 2 × 10 93H3-labeled red blood cells (mRBC-3H3) were added to the reaction system at 1 / mL and incubated at room temperature for 30 minutes. The resulting cells were then centrifuged at 500 × g and the supernatant removed. The mRBC-3H3 was then co-incubated with anti-mouse IgG-Biotin antibody (Beijing Biolab Technology Co., Ltd., product number F030805, 1:200 dilution) resuspended in PBS, pH 7.4, for subsequent FACS detection at 4°C for 30 minutes. After washing once with PBS, the cells were stained with Streptavidin-PE (BioLegend, Inc., product number 405204, 1:200 dilution) and analyzed by flow cytometry (Beijing Challen Biotechnology Co., Ltd., product number MateCyte) using NovoExpress software. The results (Figure 12) demonstrated that 3H3 can effectively bind to red blood cells under these reaction conditions. Figure 12 shows a histogram of the amount of 3H3 protein (anti-mIgG-Biotin / Strep-PE) on the erythrocyte membrane detected by flow cytometry under the above-mentioned binding conditions. The red blood cells shown at the top and bottom are 3H3-bound (top) and unbound (bottom, negative) red blood cells, respectively. 3H3 detection: First, anti-mIgG-Biotin antibody was used to specifically bind to the mouse IgG2aa sequence of 3H3 protein. Next, streptavidin-PE (abbreviated as Strep-PE) was used to specifically bind to the biotin on the anti-mIgG-Biotin antibody. Finally, PE fluorescence was detected by flow cytometry. The intensity of PE fluorescence was positively correlated with the amount of 3H3. As can be seen from the figure, PE (3H3) was detected on bound red blood cells compared to unbound red blood cells, and was clearly distinguishable from unbound red blood cells, indicating that 3H3 can effectively bind to red blood cells under these reaction conditions.

[0060] [Example 7] Mouse erythrocyte-bound anti-human VEGF (avastin) antibody and FACS detection Avastin (Bevacizumab) is a humanized monoclonal antibody IgG1 developed by Roche and purchased as a vascular endothelial growth factor (VEGF) inhibitor from TargetMol Chemicals Inc., "2020 - Bevacizumab (Avastin®) in cancer treatment: A review of 15 years of clinical use." GDP-fucose was conjugated to purified Avastin via amine reaction (the conjugation method is the same as the N803 conjugation method described above) for necessary use. Hereafter, unless otherwise specified, Avastin refers to GDP-fucose-conjugated Avastin. The fucosyltransferase concentration was 100 μg / mL, and the Avastin concentrations were 200 μg / mL and 1.5 mg / mL. Red blood cells were collected in a 100 mL tube. 8 After incubation at room temperature for 30 minutes, the mRBC-Avastin-labeled red blood cells (mRBC-Avastin) were collected and centrifuged at 500 × g. The supernatant was removed. For subsequent FACS detection, the mRBC-Avastin was resuspended in PBS (pH 7.4) and co-incubated with anti-hIgG-Biotin antibody (Beijing Biolab Technology Co., Ltd., product number F030822, 1:200 dilution) at 4°C for 30 minutes. After washing once with PBS, the cells were stained with Streptavidin-PE (BioLegend, Inc., product number 405204, 1:200 dilution). Detection was performed by flow cytometry (Beijing Challen Biotechnology Co., Ltd., product number MateCyte) and analyzed using NovoExpress software.

[0061] Figure 13 shows histograms of the amount of Avastin protein (anti-hIgG-Biotin / Strep-PE) on red blood cell membranes detected by flow cytometry under the above-mentioned binding conditions. The red blood cells shown at the top, middle, and bottom are red blood cells bound with 1500 μg / mL Avastin (top), 200 μg / mL Avastin (middle), and unbound red blood cells (bottom, negative), respectively. Avastin detection: First, anti-hIgG-Biotin antibody was used to specifically bind to Avastin protein (human IgG1), and then streptavidin-PE (abbreviated as Strep-PE) was used to specifically bind to the biotin on the anti-hIgG-Biotin antibody. Finally, PE fluorescence was detected by flow cytometry. The intensity of PE fluorescence was positively correlated with the amount of Avastin. As can be seen from the figure, PE (Avastin) was detected in the bound red blood cells under both conditions compared to the unbound red blood cells, and there was almost no difference in the PE fluorescence intensity between 200 μg / mL Avastin-bound red blood cells and 1500 μg / mL Avastin-bound red blood cells, indicating that Avastin can effectively bind to red blood cells under both reaction conditions.

[0062] [Example 8] Evaluation of the function of red blood cells themselves when modified red blood cells are used as therapeutic drugs This example demonstrates that the method described in the present invention does not have a destructive effect on red blood cells and does not affect the function of the red blood cells themselves.

[0063] The lifespan of red blood cells ranges from 100 to 130 days, with an average of approximately 125 days. Mouse red blood cells have a lifespan of approximately 30 to 45 days. Normal red blood cells lack a nucleus or organelles, are biconcave, disc-shaped, and capable of plastic deformation and oxygen transport. Blood transfusion was the first clinically employed cell therapy, primarily for the treatment of anemia. Maintaining normal red blood cell function outside the body is a hot topic in transfusion research, with the goal of ensuring that transfused red blood cells maintain their normal lifespan and oxygen-carrying function in vivo. The most common methods for assessing red blood cell function include measuring red blood cell ATP content, red blood cell 2,3-DPG content, free hemoglobin content, and red blood cell deformability.

[0064] 1. ATP content in red blood cells Mature red blood cells do not have any organelles, including mitochondria, and cannot produce ATP for energy supply through mitochondria like most cells. Red blood cells mainly produce ATP through glucolysis, i.e., anaerobic respiration, and this ATP is used to maintain the ion balance of the red blood cell and to maintain the ion pumps (sodium pump and calcium pump) on the red blood cell membrane to maintain the plasticity of the cell membrane and its special biconcave disc shape. A lack of ATP causes a disruption of the ion balance inside and outside the red blood cell membrane, resulting in the inability of red blood cells to produce Na. + The inflow of K + The excretion of Ca 2+ As a result of increased influx of erythrocytes, red blood cells absorb excess water, causing them to swell and burst. Therefore, ATP levels in red blood cells are essential for maintaining normal erythrocyte morphology, deformability, and oxygen-carrying function.

[0065] The ATP measurement kit (Beyotime Biotechnology Co., Ltd., product number S0026) was developed based on the fact that ATP is required to supply energy when firefly luciferase catalyzes luciferin to produce fluorescence. When firefly luciferase and luciferin are in excess, the amount of fluorescence produced is proportional to the concentration of ATP within a certain concentration range. The operating procedure is outlined below. 10 6 Erythrocytes were centrifuged to pellet the supernatant, discarded, and gently flicked to disperse the cells. 20 μL of erythrocyte lysis solution was added. Cells were thoroughly lysed using an appropriate vortex. After lysis, the cells were centrifuged at 12,000 g for 5 minutes at 4°C. The supernatant was removed and placed on ice for subsequent measurements. The ATP measurement reagent, containing luciferin enzyme and substrate, was diluted appropriately and added to the lysed erythrocyte sample. The mixture was quickly mixed uniformly with a micropipette and allowed to stand for at least 2 seconds. The RLU was measured using a luminometer (cartridge: SpectraMax Paradigm multimode microplate reader, Meigu Molecular Instruments (Shanghai) Co., Ltd.), and the RLU value of each cell was calculated (Figure 14). Figure 14 shows mouse erythrocytes bound to anti-mouse 4-1BB antibody 3H3 and anti-mouse 4-1BB antibody LOB, respectively, washed, and analyzed by flow cytometry. The ATP content in the erythrocytes was determined according to the procedure described above. The results showed that antibody-bound red blood cells showed no significant changes in ATP content compared to non-antibody-bound native red blood cells. 3H3 and LOB12.3 (hereafter abbreviated as LOB) are anti-mouse 4-1BB antibodies (Rickert, K., Wet et al. Combining phage display with de novo protein sequencing for reverse engineering of monoclonal antibodies. MAbs 8, 501-512 (2016)).

[0066] The red blood cell binding conditions were a cell density of 5 x 10 9As shown in Figure 15, when the incubation time was 30 minutes, 60 minutes, or 120 minutes, the ATP content in the antibody-bound red blood cells was detected, and extending the incubation time did not show any significant effect on the ATP content in the red blood cells.

[0067] The red blood cell binding conditions were a cell density of 5 x 10 9 As shown in Figure 16, when the antibody or Fc fusion protein concentrations were 100 μg / mL, 200 μg / mL, and 400 μg / mL, the ATP content in the antibody-bound red blood cells was detected, and when the antibody or Fc fusion protein concentrations were 100 μg / mL to 400 μg / mL, no significant effect on the ATP content in the red blood cells was observed.

[0068] The red blood cell binding conditions were a cell density of 5 x 10 9 As shown in Figure 17, when the antibody or Fc fusion protein concentration was 200 μg / mL, the incubation time was 30 minutes, and the fucosyltransferase concentrations were 50 μg / mL, 100 μg / mL, and 200 μg / mL, the ATP content in the antibody-bound red blood cells was detected, and when the fucosyltransferase concentration was 50 μg / mL to 200 μg / mL, no significant effect on the ATP content in the red blood cells was observed.

[0069] 2. 2,3-DPG content in red blood cells 2,3-Diphosphoglycerate (2,3-DPG) in red blood cells is an intermediate product of glycolytic metabolism and is commonly used as an indicator of red blood cell oxygen-carrying capacity. When bound to hemoglobin (Hb), it reduces the oxygen affinity of Hb, thereby promoting the release of oxygen from oxidized Hb. 2,3-DPG almost completely disappears after approximately one week of blood storage at 4°C. It has been reported that its decrease shifts the oxygen dissociation curve to the left, affecting oxygen utilization by tissues. Therefore, 2,3-DPG levels reflect the oxygen-carrying capacity of red blood cells and determine the ease of oxygen release by hemoglobin in tissues.

[0070] The Mouse 2,3-DPG Enzyme-Linked Immunosorbent Assay Kit (Jiangsu Jingmei Biotechnology Co., Ltd.) was used to measure the content of 2,3-DPG in samples using a double-antibody sandwich immunoassay. A microplate was coated with purified mouse 2,3-DPG antibody, prepared as a solid-phase antibody. Test samples and HRP-labeled 2,3-DPG antibody were added sequentially to the wells to form antibody-antigen-enzyme-labeled antibody complexes. The final color-developing substrate, catalyzed by HRP, showed a non-uniform color that correlated positively with the content of 2,3-DPG. The operating procedure is outlined below. 10 7 The red blood cells were centrifuged to pellet the supernatant, discarded, and 50 μL of red blood cell lysis solution was added. Cells were thoroughly lysed using an appropriate vortex. After lysis, the cells were centrifuged at 12,000 g for 5 minutes at 4°C. The supernatant was removed and added to the coated wells. After incubation at 37°C for 30 minutes and thorough washing, enzyme-labeled antibody was added. After incubation at 37°C for 30 minutes and thorough washing, a color development solution was added. After 10 minutes of color development at 37°C, a stop solution was added, and the absorbance (OD value) was read at a wavelength of 450 nm within 15 minutes (Figure 18). Figure 18-A shows that the 2,3-DPG content of antibody-bound red blood cells was not significantly different from that of non-antibody-bound native red blood cells.

[0071] The human 2,3-DPG enzyme-linked immunosorbent assay kit was purchased from Wenzhou Kemiao Biotechnology Co., Ltd. The principle and results were similar to those of the mouse red blood cell test. Figure 18-B shows that the 2,3-DPG content of protein-bound red blood cells did not change significantly compared with that of non-antibody-bound native red blood cells.

[0072] 3. Free hemoglobin content Hemoglobin (Hb) is present in red blood cells. When red blood cells are destroyed, hemoglobin is released into the blood, increasing the amount of free hemoglobin in the plasma. Therefore, the level of free hemoglobin reflects the integrity of red blood cells. The measurement principle of the trace free hemoglobin measurement kit (Beijing Biolab Technology Co., Ltd.) is based on the fact that the heme in the hemoglobin molecule has peroxidase-like activity, catalyzing the release of nascent oxygen from H2O2 and oxidizing phenol and 4-AAP to red substances, with the color intensity proportional to the Hb content. The operating procedure is outlined below. The color reagent was prepared immediately before use according to the ratio specified in the instruction manual. The color reagent and sample were mixed uniformly at a ratio of 50:3, and then placed in a 37°C water bath for 20 minutes. The absorbance (OD value) was read at a wavelength of 510 nm. As shown in Figure 19, a 5 × 10 6 The mouse erythrocyte-bound anti-mouse 4-1BB antibody LOB was resuspended in 50 μL of PBS solution and centrifuged at 12,000 g for 5 minutes. The supernatant was collected and the free hemoglobin content in the supernatant was determined according to the procedure described above. The results showed that antibody-bound erythrocytes did not cause significant hemolysis and did not release significant free hemoglobin compared with native erythrocytes.

[0073] The red blood cell binding conditions were a cell density of 5 x 10 9 / mL, fucosyltransferase concentration 100 μg / mL, antibody or Fc fusion protein concentration 200 μg / mL, and incubation times of 30, 60, and 120 minutes, respectively, and the detected concentrations of free hemoglobin showed no significant effect on the integrity of the red blood cell membrane even with prolonged incubation times, as shown in Figure 20 .

[0074] The red blood cell binding conditions were a cell density of 5 x 10 9As shown in Figure 21, the concentration of free hemoglobin was detected at antibody or Fc fusion protein concentrations of 100 μg / mL, 200 μg / mL, and 400 μg / mL, respectively, with a fucosyltransferase concentration of 100 μg / mL, an incubation time of 30 minutes, and antibody or Fc fusion protein concentrations of 100 μg / mL, 200 μg / mL, and 400 μg / mL, respectively, and no significant effect on the integrity of the erythrocyte membrane was observed when the antibody or Fc fusion protein concentrations were between 100 μg / mL and 400 μg / mL.

[0075] The red blood cell binding conditions were a cell density of 5 x 10 9 As shown in Figure 22, the concentration of free hemoglobin was detected at 50 μg / mL, antibody or Fc fusion protein concentration of 200 μg / mL, incubation time of 30 minutes, and fucosyltransferase concentrations of 50 μg / mL, 100 μg / mL, and 200 μg / mL, respectively, and no significant effect on the integrity of the red blood cell membrane was observed when the fucosyltransferase concentration was between 50 μg / mL and 200 μg / mL.

[0076] 4. Red blood cell deformability Erythrocyte deformability is one of the most important factors affecting blood apparent viscosity and effective perfusion in the in vivo microcirculation. It is also an important determinant of erythrocyte lifespan. Currently, there are numerous methods for measuring erythrocyte deformability, which can be divided into two major categories. The first category involves comparing the average deformability of erythrocyte populations using erythrocyte suspensions, such as laser diffraction. The second category involves measuring the deformability and mechanical properties of cell membranes using individual erythrocytes, such as differential bottom adhesion, micropipette, and electron spin resonance spectroscopy. In this study, we used a laser diffraction method (erythrocyte deformometer LBY-BX, Beijing Pulisheng Instrument Co.) to evaluate the average deformability of erythrocyte populations. The proportion of elongated erythrocytes in a given suspension medium, i.e., the deformation index (DI), can be measured at different shear rates using a laser diffractometer to reflect erythrocyte deformability. As shown in Figure 23, two different erythrocyte labeling methods were compared. The experimental procedure is outlined below.

[0077] The fucosyltransferase-mediated mouse erythrocyte binding conditions created by the present invention were set at a cell density of 5×10 9 The washed red blood cells were designated mRBC-3H3, with a protein concentration of 200 μg / mL, a fucosyltransferase concentration of 100 μg / mL, and an incubation time of 30 minutes at room temperature. Alternatively, Sulfo-NHS-Biotin (Beijing Biolab Technology Co., Ltd., product number GS4320) was purchased and used to link amines with N-hydroxylsuccinimide ester (NHS ester) to biotinylate lysines on the surface of proteins in red blood cell membranes. This method is also commonly used for labeling most cell membranes, such as the red blood cell labeling method described in a 2021 paper by Rubius Therapeutics (2021 - Anti-tumor effects of RTX-240 an engineered red blood cell) and the 1987 method of labeling mouse red blood cells with biotin to measure their lifespan in mice (1987 - Biotinylated Erythrocytes In Vivo Survival and In Vitro Recovery). The same cell density was used, and 30 μg / mL of Sulfo-NHS-Biotin was used for 30 minutes of incubation at room temperature, followed by washing. The resulting red blood cells bound by this method were designated mRBC-Biotin. The labeling efficiency and uniformity of the red blood cells obtained by both binding methods were examined by flow cytometry (Figure 24). Both methods yielded ideally uniform modified red blood cells. Mouse red blood cells containing unbound pharmaceuticals were collected at a density of 5 × 10 9 / mL, and a 30-minute incubation in the PBS solution and a washing step were also performed at room temperature, but no antibody, no fucosyltransferase, and no sulfo-NHS-biotin were added. These cells were designated naive mRBCs.

[0078] The percentage of elongated red blood cells in a PBS suspension medium was measured using a laser diffractometer under the same conditions. As shown in Figure 23, the deformability index of unbound mouse red blood cells (naive mRBC) was between 0.15 and 0.2, which is considered a relatively normal deformability index in this experiment. Fucosyltransferase-mediated mouse red blood cell-bound cells (mRBC-3H3) also had a deformability index between 0.15 and 0.2, similar to normal red blood cells. Red blood cells treated with classical amine crosslinking (mRBC-Biotin) had a deformability index of approximately 0.05, indicating a significant decrease in the deformable percentage. Therefore, the fucosyltransferase-mediated antibody-bound red blood cells prepared by this invention showed no significant effect on deformability compared to non-antibody-bound native red blood cells.

[0079] [Example 9] Evaluation of the in vivo half-life of antibody-bound mouse erythrocytes in mice The life span of human peripheral blood red blood cells is approximately 120 days, while that of mouse peripheral blood red blood cells is approximately 20 to 40 days (1958 - The Life Span of Red Cells in the Rat and the Mouse as Determined by Labeling with DFP in Vivo; 2015 - Determination of RBC Survival in C57BL6 and C57BL6-Tg(UBC-GFP) Mice). A long half-life is one of the advantages of red blood cells as drug vectors. A series of studies in Example 8 demonstrated that the red blood cell-binding drug prepared according to the present invention does not significantly impair the function of the red blood cells themselves. This characteristic was also reflected in in vivo tests using mice, where no significant shortening of the half-life of drug-bound red blood cells was observed. Mouse erythrocytes were bound to the anti-mouse 4-1BB antibodies 3H3 and LOB, and the control antibody mATNP (anti-trinitrophenyl antibody, P. Barber, M.B. Rittenberg, Anti-trinitrophenyl (TNP) antibody detection by neutralization of TNP-coliphage T4, Immunochemistry, Volume 6, Issue 2, 1969, Pages 163-174), respectively, according to the methods described in Examples 3 and 4. 5 × 10 8 The transfusion was then perfused into the mice via a vein. Starting one day after transfusion, venous blood was collected from the medial canthus of the mice and detected by flow cytometry according to the staining method described in Examples 2 and 3. The results showed that antibody-bound red blood cells were clearly detected in the peripheral blood until the end of the study, 28 days after transfusion (Figure 25). The half-life of antibody-bound red blood cells was calculated using a nonlinear regression method, and was found to be over 20 days, close to the lifespan of red blood cells in normal mice. This indicates that the conjugation method described in this invention does not affect the lifespan of mouse red blood cells. Rubius Therapeutics conjugated protein molecules to the surface of cell membranes using an NHS ester and amine crosslinking method, and 1 × 10 8 and 1 × 10 9The red blood cells were perfused into mice twice weekly to determine the half-life and pharmacokinetics over a 14-day period (Anti-tumor effects of RTX-240 an engineered red blood cell, Cancer Immunol Immunother 2021 Sep;70(9):2701-2719). The number of red blood cells perfused in mice in this study was 1 × 10 compared to 1 × 10 8 As with the group, only 1 × 10 9 However, the results showed that Rubius Therapeutics' 1 × 10 8 After 14 days, only about 1% of the drug-bound red blood cells remained in the group, and the number of drug-bound red blood cells was 1 × 10 9 It was also shown that only about 15% of the group remained. Although specific data were not disclosed, the half-life was significantly shorter than 14 days. No significant effect was observed on the half-life of the fucosyltransferase-mediated antibody-bound red blood cells produced by the present invention. The half-lives of mATP-mIgG2aa1.1 (abbreviated as mATNP), LOB-mIgG2aa.1 (abbreviated as LOB), and 3H3-mIgG2aa.1 (abbreviated as 3H3) were analyzed by nonlinear biphasic decay using Graphpad Prism software. The half-lives of all of these were close to 30 days, which is similar to the lifespan of normal mouse peripheral blood red blood cells. mIgG2aa1.1 reduces the Fc-binding complement and FcgR functions by adding amino acid mutations to the Fc portion of mouse IgG2aa (Kovarik J. Highly reduced binding to high and low affinity mouse Fc gamma receptors by L234A / L235A and N297A Fc mutations engineered into mouse IgG2a. Mol Immunol. 2015 Feb;63(2):456-63.).

[0080] [Example 10] Functional evaluation of anti-mouse 4-1BB antibody-bound mouse erythrocytes as an immunomodulator 3H3 is an anti-mouse 4-1BB activating antibody. 4-1BB, also known as CD137, is a member of the tumor necrosis factor (TNF) receptor family and is expressed on the membrane surface of immune cells such as CD4+ and CD8+ T cells, NKT cells, NK cells, DCs, and macrophages. Upon binding to ligands or activating antibodies, 4-1BB phosphorylates the cytoplasmic IκB / p65 / p50 trimer, releasing p65 / p50 (NF-κB), which then translocates from the cytoplasm to the nucleus and binds to specific sequences on nuclear DNA, promoting the transcription of related genes and contributing to cytokine production and secretion. Taking advantage of the characteristics of the 4-1BB signaling pathway, we constructed a reporter gene system to detect 4-1BB activation in vitro. Briefly, as shown in Figure 26, mouse 4-1BB protein was expressed on the cell membrane surface of a human embryonic kidney cell line (HEK293). A DNA sequence specifically recognized by NF-κB was introduced into the cells as a promoter, downstream of which was a luciferase reporter gene (Luc). When 4-1BB on the cell membrane received an activation signal, the downstream luciferase reporter gene was expressed, and the luciferase expression level was proportional to the activity of the transcription factor. Upon addition of a specific luciferase substrate, luciferase reacted with the substrate, emitting fluorescence. The intensity of this fluorescence was measured to measure luciferase activity and determine whether 4-1BB was activated. The luciferase reporter gene pGL4.32 (CAT# E8491) was purchased from Promega Corporation, an affiliate of Promega (Beijing) Biotech Co., Ltd., and the mouse 4-1BB expression plasmid was synthesized by Souzhou Biotechnology Co., Ltd.HEK293 cells carrying a reporter gene were transiently transfected with the pGL4.32 plasmid and the mouse 4-1BB expression plasmid 24 hours after which native mouse erythrocytes and 3H3 antibody-conjugated mouse erythrocytes were added and incubated for 6 hours. Luminescence was then measured using a Bio-Glo (SpectraMax Paradigm multimode microplate reader, Meigu Molecular Instruments (Shanghai) Co., Ltd.) cartridge. TM Quantitation was performed using a luciferase assay system (Promega Corporation, CAT# G7940), and the average relative light units (RLU) were calculated. Figure 27 shows that mouse erythrocytes containing the 3H3 antibody can effectively activate mouse 4-1BB. 1 x 10 cells transfected with mouse 4-1BB plasmid DNA were used. 5 1 × 10 HEK293 cells were incubated with 1 × 10 native mouse erythrocytes and 3H3 antibody-conjugated mouse erythrocytes. 6 These were added and labeled as Naive mRBC and mRBC-3H3, respectively. As a negative control, mouse erythrocytes were not added and labeled as Negative CTRL. 5 HEK293 cells were transfected with pGL4.32 plasmid DNA carrying a CMV promoter, allowing the cells to express a downstream luciferase reporter gene without external stimuli, and were labeled as Positive CTRL as a positive control. The results showed that the RLUs of the negative control, positive control, naive mRBC, and mRBC-3H3 groups were 7.1 × 10, respectively. 4 ±4.1×10 4 , 1.2 × 10 6 ±2.7×10 4 , 1.0×10 5 ±9.3×10 4 , 8.6×10 5 ±1.9×10 5It was shown that 3H3 antibody-conjugated mouse erythrocytes increased reporter gene activation 8.6-fold compared to naive RBCs. In an in vitro detection model, 3H3 antibody-conjugated mouse erythrocytes were able to activate mouse 4-1BB ligand, suggesting the potential for these erythrocytes to be used as immunomodulators. This potential was further confirmed using mouse primary lymphocytes in Example 11 below, and subsequently demonstrated antitumor properties in mouse in vivo experiments in Example 12.

[0081] [Example 11] CD8 induced by anti-mouse 4-1BB antibody-conjugated mouse erythrocytes as an immunomodulator + Functional assessment of cell in vitro stimulation Cytotoxic T lymphocytes (CTLs) are generally CD8 + T cells are a key component of the adaptive immune system and play a crucial role in immune defense against intracellular pathogens such as viruses, bacteria, and tumors. Anti-mouse 4-1BB antibody-conjugated mouse red blood cells express CD8 T cells, which are involved in the immune system, particularly tumor killing. + To confirm its immunomodulatory properties, we isolated and prepared mouse splenic lymphocytes and examined their response to anti-mouse 4-1BB antibody-conjugated mouse red blood cells in vitro. The procedure was as follows: Mice were sacrificed by vertebral corpuscle removal. The lower abdomen was disinfected with an alcohol swab. The abdomen was then aseptically opened. The spleen was removed from the abdominal cavity and placed in a dish containing PBS buffer. Any excess tissue was washed and removed. The spleen was then cut into three pieces and placed on a 200-mesh nylon mesh with sterile forceps. The mesh was then placed in a small dish containing 5 mL of fresh PBS buffer. The spleen was then crushed by rotating it with a 5 mL syringe plunger. The nylon mesh was then removed, and the cell suspension was collected in a 15 mL centrifuge tube and centrifuged at 2000 rpm for 5 minutes at 4°C. The supernatant was discarded. The splenocytes were resuspended in 5 mL of red blood cell lysis solution, thoroughly blown off the cells with a pipette, and incubated at room temperature (on ice) for 5 minutes. The cells were then centrifuged at 2000 rpm at 4°C for 5 minutes, and the supernatant was discarded. The cells were resuspended in 1000 μL of PBS buffer and transferred to a 1.5 mL centrifuge tube for future use. Mouse lymphocytes were collected at a concentration of 1.5 × 10 6Splenocytes were seeded in a 96-well microplate at a density of 1 / mL and treated with 1 μg / mL anti-mouse CD3 antibody (Thermo Fisher Scientific (China) Co., Ltd., product number 16-0031-85). Simultaneously, pre-prepared 3H3 antibody-conjugated mouse red blood cells were added to costimulate the splenocytes, or anti-mouse 4-1BB antibody (10 nM 3H3 protein, 100 nM LOB protein) was added as a control. After 48 h, cells were collected and stained with anti-CD8 (Elabscience Biotechnology Co., Ltd., product number E-AB-F1104J, 1:50 dilution), anti-CD4 (Elabscience Biotechnology Co., Ltd., product number E-AB-F1097C, 1:200 dilution), and NK1.1 (Elabscience Biotechnology Co., Ltd., product number E-AB-F0987H, 1:200 dilution), respectively, and analyzed by flow cytometry.

[0082] The left side of Figure 28 shows CD8 counts obtained by flow cytometry after stimulation with 1 μg / mL anti-mouse CD3 antibody and costimulation with five different anti-mouse 4-1BB antibodies. + The percentage of T cells is 1 × 10 5 CD8 produced by costimulation with 3H3-binding erythrocytes (1E5mRBC-3H3) + The proportion of T cells was 9.1%, 1 × 10 6 CD8 produced by costimulation with 3H3-binding erythrocytes (1E6mRBC-3H3) + The proportion of T cells was 12.6%, 1 × 10 7 CD8 produced by costimulation with 3H3-binding erythrocytes (1E7mRBC-3H3) + The percentage of T cells was 17.5%, CD8 generated by costimulation with 10 nM 3H3 protein. + The percentage of T cells was 7.2%, CD8 generated by costimulation with 100 nM LOB protein. + The percentage of T cells was 8.1%. The results showed that the CD8 T cells obtained by stimulation with anti-mouse 4-1BB antibody (10 nM 3H3 protein or 100 nM LOB protein) +The percentage of T cells was only 7% to 8%, and CD8 T cells were obtained by stimulation with anti-mouse 4-1BB antibody-conjugated mouse erythrocytes. + The proportion of T cells was high (9%-17.5%) and showed an increase with increasing red blood cell count.

[0083] The right side of Figure 28 shows the percentage of NK cells obtained by flow cytometry after stimulation with 1 μg / mL anti-mouse CD3 antibody and costimulation with five different conditions of anti-mouse 4-1BB antibody. 5 The percentage of NK cells generated by costimulation with 3H3-binding erythrocytes (1E5mRBC-3H3) was 11.5%, 1 × 10 6 The percentage of NK cells generated by costimulation with 3H3-binding erythrocytes (1E6mRBC-3H3) was 13.5%, 1 × 10 7 The percentage of NK cells generated by costimulation with 3H3-conjugated red blood cells (1E7mRBC-3H3) was 15.3%, the percentage of NK cells generated by costimulation with 10 nM 3H3 protein was 9.5%, and the percentage of NK cells generated by costimulation with 100 nM LOB protein was 10.3%. The results showed that the percentage of NK cells generated by stimulation with anti-mouse 4-1BB antibody (10 nM 3H3 protein or 100 nM LOB protein) was only about 10%, while the percentage of NK cells generated by stimulation with anti-mouse 4-1BB antibody-conjugated mouse red blood cells was high (11.5%-15.3%) and increased as the number of red blood cells increased.

[0084] As described above, anti-mouse 4-1BB antibody-conjugated mouse erythrocytes were identified as CD8 + It is possible to effectively stimulate the proliferation of T and NK cells.

[0085] CD8 + To further investigate the effect on T lymphocytes, CD8 + T lymphocytes were isolated and purified. Single cells from mouse spleens were sorted by MojoSort. TM After resuspending in Buffer (BioLegend, CAT# 480007) and counting, 7The cells were separated, and 10 μL of Biotin-Antibody Cocktail was added and incubated on ice for 15 minutes. 10 μL of Streptavidin Nanobeads were then added and mixed thoroughly, and the mixture was incubated on ice for 15 minutes. The mixture was then placed in a magnetic rack and the supernatant was collected. + The percentage of T exceeded 95%. + 10 ml of T in a 96-well microplate 6 The cells were seeded at a density of 1000 / mL, anti-mouse CD3 antibody was added at 2 μg / mL, and the prepared 3H3 antibody-bound mouse red blood cells were added. After 48 hours of co-incubation, the incubation supernatant was collected and the interferon-γ concentration in the supernatant was detected using a mouse interferon-γ ELISA kit (Beijing Solarbio Science & Technology Co., Ltd., product number SEKM-0031).

[0086] Figure 29 shows the results of 2×10 4 Unbound red blood cells (hereafter abbreviated as 2E4mRBC-CTRL, left) or 2 × 10 4 The concentrations of interferon-γ secreted by CD8+ T cells in the three stimulated groups were measured by ELISA using 3H3-conjugated red blood cells (hereafter abbreviated as 2E4mRBC-3H3, center) and a blank control without added red blood cells (hereafter abbreviated as only anti-CD3, right). The concentrations of interferon-γ produced by the 2E4mRBC-CTRL group were 7 pg / mL, the 2E4mRBC-3H3 group was 37 pg / mL, and the only anti-CD3 group was 1 pg / mL. The results showed that anti-mouse 4-1BB antibody-conjugated mouse red blood cells significantly increased interferon-γ secretion by CD8+ T cells (Figure 29). Interferon-γ is primarily produced by activated T cells and NK cells and has antiviral, immunoregulatory, and antitumor properties.

[0087] In vitro studies have shown that anti-mouse 4-1BB antibody-conjugated mouse erythrocytes act as an immune activator, inhibiting CD8 + It has been shown that it can effectively activate T and NK cells.

[0088] [Example 12] Treatment of CT26 colon cancer tumor model with anti-mouse 4-1BB antibody-conjugated mouse erythrocytes as an immunomodulatory agent Balb / C mice were purchased from Chuangmo Biotechnology Beijing Co., Ltd. and bred at Chuangmo Biotechnology Beijing Co., Ltd. Approximately 2*10 5 CT-26 cells (a mouse colon cancer cell line, provided by Chuangmo Biotechnology Beijing Co., Ltd.) were subcutaneously injected into the right scapular region of 5- to 12-week-old mice. Tumor volume was measured along three orthogonal axes (a, b, c) and calculated as tumor volume = abc / 2. Mouse erythrocytes were conjugated with the anti-mouse 4-1BB antibody 3H3 according to the method described in Example 3 or Example 4. After tumor formation (9 to 12 days), the mice received 5 × 10 antibody-conjugated or non-antibody-conjugated erythrocytes via the tail vein once a week. 8 After the red blood cell injection, tumor volume and mouse body weight were measured three times a week, and the general behavior of the mice was recorded.

[0089] Figure 30 shows the trend of tumor volume changes in mice after the first injection of antibody 3H3-conjugated red blood cells or non-antibody-conjugated red blood cells, which were continuously monitored for 14 days. On day 14, the mean tumor volume in mice injected with non-antibody-conjugated red blood cells was 1720.8 mm 3 The mean tumor volume of mice injected with antibody 3H3-conjugated red blood cells was 890.3 mm 3 The tumor volume in mice injected with 3H3 antibody-conjugated red blood cells was significantly reduced compared to mice injected with non-antibody-conjugated red blood cells. [Industrial Applicability]

[0090] In vivo efficacy studies demonstrated that anti-mouse 4-1BB antibody-conjugated mouse erythrocytes could be used as an immunomodulatory agent for treating CT26 colon cancer.

Claims

1. (1) binding a GDP-fucose derivative to a chemical molecule or a biopolymer; (2) covalently binding the chemical molecule or biopolymer obtained in step (1) to a sugar chain on the surface of the erythrocyte membrane via a glycosidic bond by an enzymatic reaction catalyzed by a fucosyltransferase; 1. A method for binding a chemical molecule or a biological macromolecule to the surface of a mature red blood cell, comprising:

2. 10. The method of claim 1, wherein the chemical molecule is selected from a therapeutic molecule or a fluorescent moiety.

3. The method of claim 1, wherein the biopolymer is selected from the group consisting of a polynucleotide, a polypeptide, and an antibody.

4. 4. The method of claim 3, wherein the antibody is selected from a monoclonal antibody, a single-chain antibody, a bispecific antibody, and a nanobody.

5. The method according to claim 4, wherein the antibody is an anti-4-1BB activating antibody.

6. The method according to claim 3, wherein the biopolymer is selected from interleukin IL-15 isoforms and interleukin IL-12.

7. 2. The method of claim 1, wherein the mature red blood cells are collected from peripheral blood and removed of serum and other blood cells, there are no free protein molecules in the red blood cell suspension, and the red blood cell suspension is freshly prepared or stored under appropriate conditions for no more than 30 days.

8. The method according to claim 1, wherein the fucosyltransferase is a human fucosyltransferase or a Helicobacter pylori fucosyltransferase.

9. 9. The method according to claim 8, wherein the GDP-fucose derivative is guanosine 5'-diphosphate-fucose-triazole-polyethylene glycol-methyltetrazine.

10. The enzymatic reaction of the fucosyltransferase in step (2) is carried out under the reaction conditions of a red blood cell density of 5×10 ７ / mL to 5 x 10 ９ 10. The method according to claim 9, wherein the reaction temperature is 4°C to 37°C, the reaction time is 20 minutes to 16 hours, the fucosyltransferase concentration is 40 μg / mL to 300 μg / mL, the pharmaceutical polymer is an antibody or an Fc fusion protein at a concentration of 60 μg / mL to 120 μg / mL, the reaction temperature is 4°C to 37°C, and the reaction time is 20 minutes to 16 hours.

11. The enzymatic reaction of the fucosyltransferase is carried out in the presence of Mg ２＋ The method according to claim 11, characterized in that the reaction system is free of: 150 mM sodium chloride, 2.7 mM potassium chloride, 44 mM glucose, pH 5.9 to 6.

3.

12. Modified red blood cells, characterized in that they are obtained by preparing them using the method according to any one of claims 1 to 10.

13. 13. Use of the modified red blood cells of claim 12 in the preparation of a medicament for treating a tumor disease.

14. 14. The use according to claim 13, characterized in that the tumor disease is colon cancer or melanoma.

Images