CN117563001A - Application of galectin-1 as immune checkpoint in preparation of tumor immunotherapy medicament and medicament - Google Patents

Abstract

The invention discloses application of galectin-1 (Gal-1) serving as an immune check point in preparation of tumor immunotherapy medicaments and the medicaments, relates to the fields of medicaments and tumor therapies, and discloses a novel tumor immunotherapy strategy for targeting the novel immune check point. The invention discovers that lactose and derivatives thereof can inhibit immune checkpoints by competing Gal-1 on the surfaces of cancer cells and immune cells, and have no obvious toxic and side effects; knockout of B4GATL1 to reduce protein galactosylation can significantly reduce Gal-1 levels of tumor metastasis to T cells, thereby enhancing T cell activation and function; in addition, the lactose synthase component LALBA is over-expressed in the tumor of the mice, so that the de novo synthesis of lactose in the tumor microenvironment can be realized, and the CD8 is further enhanced ⁺ T cell mediated immune responses. Combines the anti-tumor efficacy and the economic cost of lactose, and provides a widely feasible idea for cancer treatment.

Description

Galectin-1 as an immune checkpoint in the preparation of tumor immunotherapy drugs applications and the drug

This application claims priority from Chinese invention patent application number 2022114354538, filed on November 16, 2022, and Chinese invention patent application number 2022114593007, filed on November 16, 2022, each of which is hereby incorporated by reference in its entirety. incorporated herein for all purposes.

Technical field

The invention belongs to the fields of medicine and tumor treatment, and in particular provides the application of galectin-1 as an immune checkpoint in the preparation of tumor immunotherapy drugs and the drug.

Background technique

Tumor immunity is by far the most effective cancer treatment method; in particular, the use of biological macromolecules or small molecules to target tumor immune checkpoints (such as PD1, PDL1, CTLA-4, etc.) has been used in the treatment of various tumors. showed good results; the application of tumor immune checkpoint inhibitor therapy has also exposed some key problems that need to be solved; for example, it only has good efficacy on some tumor types, but is ineffective on other tumor types; even on the same type of tumor There are also significant differences in the efficacy of different patients with tumors; in-depth research on the mechanism of tumor immunity will provide new ideas to solve these problems; on the other hand, the high cost of tumor immune checkpoint inhibitor treatment is the main limitation limiting its practical application Obstacles; for example, due to the combined effect of the Chinese government’s centralized medical procurement policy and fierce competition among manufacturers, the annual cost of anti-PD1 antibody treatment has dropped from hundreds of thousands of US dollars to tens of thousands of RMB, but it is still affordable for most families of cancer patients. A heavy burden; looking around the world, how to provide affordable tumor immunotherapy for tumor patients is an urgent task.

Antibodies are the main biological macromolecules of tumor immune checkpoint inhibitors, which have high requirements for production and storage, resulting in high prices. There are already some small molecule inhibitors of tumor immune checkpoints under development, or the production is complex or difficult. There are many side effects due to poor targeting.

Galectins, as a family of galactoside-binding proteins, have powerful immunomodulatory activities (reviewed by Rabinovich et al., J Leuk Biol., 2002, 71:741-752). In particular, galectin-1 is a negative regulator of T cell responses and can induce T cell apoptosis (Perillo, N L et al., Nature, 1995, 378(6558):736-739). Galectin-1 is also secreted by activated T cells, thereby acting as an autoregulator of T cell activation and inhibiting antigen-induced T cell proliferation (Blaser C. et al., Euro. JImmunal, 1998, 28:2311-2319).

However, the current understanding of galectin inhibitors is still very broad, and it is believed that they can be used for pathogenic infections, autoimmune diseases, transplant rejection, graft-versus-host disease, allergies, inflammatory diseases, and cancers and tumors. , and its special significance in tumor immunotherapy has not been found.

In view of this, this invention is proposed.

Contents of the invention

The present invention surprisingly found that cancer cells in tumor tissue can transfer abnormally high levels of galectin-1 to the surface of adjacent immune cells, especially tumor-targeting cytotoxic T cells. Tumor-derived galectin-1 binds to galactosylated proteins on the surface of cytotoxic T cells to regulate T cell activation and function. Lactose and N-acetyllactosamine can competitively bind to galectin from tumor cells on these T cells, allowing the T cells to restore their killing ability. This discovery makes galectin and its inhibitors particularly significant in the treatment of tumors compared with general inflammation treatments. Lactose, in particular, is not degraded outside the intestines and stomach and can competitively bind Gal-1 to the Gal-1 binding site on the surface of T lymphocytes, blocking the abnormally high levels of Gal-1 secreted by tumor cells from adjacent T lymphocytes. The combination of lymphocytes rescues the function of T lymphocytes. Therefore, galectin, especially Gal-1, can be used as a new immune checkpoint in the treatment of tumors.

The present invention starts from the molecular mechanism, innovatively discovers new mechanisms and targets targeting tumor immune checkpoints, and screens natural products for regulating tumor immunity, ultimately providing new methods for tumor immunotherapy.

The present invention surprisingly found that the activation level and target cell killing activity of CD8 ⁺ T cells with B4GALT1 knocked out were significantly enhanced; in in vivo and in vitro tumor infiltration experiments, B4GALT1 gene knocked out OT-1 T cells in tumors had faster proliferation rate, resulting in significant tumor growth inhibition. β-1,4-galactosyltransferase 1 (B4GALT1) is a key enzyme in the N-glycan biosynthesis pathway. Its inactivation can lead to CD8 ⁺ T both in vitro and in vivo TCR activation and function enhancement of cells. It is worth mentioning that the present invention found that in tumor cells, inactivation of B4GALT1 can promote CD8 ⁺ T cell-mediated immune killing by regulating the interferon gamma signaling pathway and antigen presentation.

The present invention further found that the activity of B4GALT1 was inhibited in the mouse colon cancer cell line MC38 tumor, and its immune response ability was explored in vitro and in vivo (data not shown in this article). The results showed:

(1) In MC38 with B4GALT1 gene knockout, IFNγ, TNFα, and IFNα signaling pathways were all significantly up-regulated. CRISPR/Cas9 screening studies in tumor cells in recent years have proven that these signaling pathways play an important role in tumor immunotherapy. At the same time, after IFNγ treatment, the B2m expression level and OVA antigen presentation of B4GALT1-knocked out MC38 cells were significantly enhanced.

(2) In vitro killing experiments show that knocking out the B4GALT1 gene in MC38 cells can enhance the specific in vitro killing function of OT-1T cells. During this process, the present inventors noticed that the IFNγ expression level in OT-1T cells increased significantly when co-cultured with B4GALT1 knockout MC38 cells compared with co-culture with control MC38 cells.

(3) In the in vivo tumor growth experiment, the present invention found that in terms of subcutaneous tumors in wild-type mice, the growth rate of MC38 tumors with B4GALT1 knockout was significantly slower than that in the control group, while in immunodeficient NPG (NOD -PrkdcscidIL2rgnull) mice but this phenomenon was not observed. In order to further understand which type of immune cells mediates this tumor immune response, the present invention used specific antibodies to eliminate different immune cells in wild-type mice and then implanted them subcutaneously. Finally, it was found that CD8 ⁺ and CD4 ⁺ were eliminated T cells can significantly restore the growth rate of MC38 tumors with B4GALT1 knockout, especially after deletion of CD8 ⁺ T cells; however, deletion of natural killer cells cannot restore the growth rate.

This shows that B4GALT1 knockout MC38 tumor cells can enhance immune response activity both in vitro and in vivo, and CD8 ⁺ T cells play a major role in this process.

The present invention found that exogenous expression of LALBA can secrete lactose in tumor cells, and this process is dependent on the activity of B4GALT1. α-Lactalbumin Alpha (LALBA) can form a heterodimer with B4GALT to participate in lactose synthesis. Overexpression of LALBA in tumor cells can promote the response of tumor cells to immune cells, thereby inhibiting tumor growth. Even if only a part of tumor cells overexpress LALBA, it can still exert a tumor growth inhibitory effect. In addition, orthotopic injection of tumors with lentivirus or adeno-associated virus encoding LALBA can also achieve the effect of inhibiting tumors, and mice with complete tumor regression can establish an immune memory of the tumor and will not be re-implanted with the tumor. Visible tumors form. However, tumor cells overexpressing mutant LALBA do not produce lactose and have no significant inhibitory effect on tumor growth.

The present invention creatively discovered that there is a phenomenon in tumor tissue, in which cancer cells can transfer abnormally high levels of galectin-1 to the surface of adjacent immune cells, especially tumor-targeted cytotoxic T cells. Subsequently, tumor-derived galectin-1 interacts with galactosylated proteins on the surface of cytotoxic T cells to regulate T cell activation and function. Through anti-Gal-1 antibody staining, the present invention confirms that lactose treatment can remove Gal-1 from the surface of OT-ⅠT cells, and at the same time, it also reduces the level of Gal-1 on the surface of MC38 cells. And it was found that lactose could significantly increase the expression of Ifnγ and Tnfα in exhausted CD8 ⁺ PD-1 + T cells isolated from tumors.

The present invention further found that by competitively binding to tumor cell-derived galectin on these T cells, the T cells can be restored to their killing ability.

The present invention also found that lactose, as a β-galactoside, can interfere with the function of the N-glycosyl group in both CD8 ⁺ T cells and tumor cells. Lactose was injected through the tail vein and it was found that the lactose solution could significantly inhibit the growth of "hot" tumors, and CD8 ⁺ T cells played an important role in this inhibitory process. Transcriptome sequencing analysis showed that lactose treatment can promote the proliferation and enhance the function of tumor-specific CD8 ⁺ T cells. By constructing humanized mice with the human immune system, the activating role of lactose in human tumor immunity was confirmed under three-dimensional culture conditions of human tumors.

The present invention also successively confirmed the role of lactose in human tumor immunity in humanized mice with reconstructed human immune systems and under in vitro culture conditions of human tumors. These all provide insights into the clinical application of lactose. possibility. More importantly, since lactose is cheap and easy to obtain, with extremely high production worldwide, if it can be used directly or combined with other treatments for clinical treatment of tumors, it will provide a new way for many cancer patients and will Greatly reduce the economic cost of tumor treatment.

However, in the present invention, it was found that only intravenous injection of lactose can exert a tumor growth inhibitory effect, while no significant effect was observed via oral or intraperitoneal injection. This may be because oral or intraperitoneal injection is difficult to achieve a sufficiently high effectiveness within the tumor. concentration; on the other hand, a stable inhibitory effect can only be observed when the injection is started early after tumor transplantation. In the late stage of tumor transplantation, lactose injection has no significant effect, which suggests that the single use of lactose in clinical applications may not be able to inhibit the treatment of intermediate and advanced tumors. be suppressed. In addition, lactose injected intravenously will be rapidly consumed in the blood through the renal system and excreted from the body through urine. Therefore, the present invention requires lactose injection every two days. This shortcoming may be solved by combining lactose with a protein carrier. Injection is performed after coupling to compensate. The present invention is testing this method, hoping to use carrier proteins to maintain the lactose concentration in the blood for a relatively long time, thereby reducing the frequency of injections.

Galectins are a family of conserved galactosylation-binding proteins that regulate immune cell functions during normal physiological processes as well as various pathological processes, including tumors. The present findings demonstrate that cancer cells can transfer abnormally high levels of galectin-1 to the surface of neighboring immune cells, particularly tumor-targeting cytotoxic T cells. Subsequently, tumor-derived galectin-1 interacts with galactosylated proteins on the surface of cytotoxic T cells to regulate T cell activation and function. The present invention conducted a series of Gal-1 transfer experiments, and the results show that cell-to-cell proximity, cell surface galactosylation level and cell surface Gal-1 level are the main driving forces for intercellular Gal-1 transfer. This suggests an unattended and undefined method of intercellular communication, namely proximity-dependent intercellular protein spreading (PDICPS). The low-affinity binding of Gal-1 to its galactosylation target is an intrinsic feature of this proximity-dependent intercellular protein diffusion.

Targeting this new immune checkpoint will be a new strategy for tumor immunotherapy. On the one hand, knocking out B4GATL1 to reduce protein galactosylation can significantly reduce the level of Gal-1 transferred from tumors to T cells, thereby enhancing T cell activation and function. On the other hand, lactose and its derivatives can inhibit this immune checkpoint by competing for Gal-1 on the surface of cancer cells and immune cells.

Targeting this new immune checkpoint with small molecules will be a feasible strategy for tumor immunotherapy. Lactose and related derivatives containing galactosides can inhibit this immune checkpoint by competing for Gal-1 on the surface of cancer cells and immune cells. In particular, lactose, as a structurally simulated competitive galectin inhibitor, can effectively inhibit tumor growth through systemic intravenous injection into the circulation system of mice with different tumor burdens. However, the present invention surprisingly found that lactose only plays a role in activating anti-tumor immunity in "hot" tumors with strong immunogenicity, but has no obvious effect in "cold" tumors with weak immunogenicity. This confirms that the main mechanism of lactose inhibiting tumor growth depends on the immune system, and also suggests that the treatment of "cold" tumors with lactose may require combined measures to enhance tumor immunogenicity. In order to further explore the potential of lactose as an immune checkpoint inhibitor, the present invention implanted human gastric cancer SGC7901 tumors into humanized mice and observed that systemic intravenous injection of lactose can significantly inhibit the growth of SGC7901 tumors. Subsequently, in an in vitro culture model of surgically removed patient tumors, lactose treatment demonstrated immune responses similar to anti-PD-1 pembrolizumab treatment. In addition, in the artificial membrane permeability experiment, the hypotonicity of lactose also shows that it is lactose itself, not its degradation products, that exerts the tumor inhibitory effect in the body. Comprehensive experimental analysis found that lactose activates CD8 ⁺ T cells infiltrating in tumors by competing for Gal-1 on CD8 ⁺ T cells and relieves Gal-1's inhibition of T cell activation.

Those in the art know that Galon and Bruni expanded the classification of tumors into four categories (see Galon, J. and Bruni, D., Approaches to treat immune hot, altered and cold tumors with combination immunotherapies", Nature Reviews Drug Discovery (18), March 2019,197-218): hot, altered-repellent, altered-immunosuppressive, and cold to facilitate research and communication. Specifically, the category stratification is based on the type of immune cells within the tumor site , density, and location. The authors classified tumors according to immune infiltration rather than cancer type, and the scoring system ("Immunoscore") is based on the quantification of two lymphocyte populations (CD3 and CD8) at the tumor center and invasive margins. Score Ranges from I0 (low density, as in neither cell type is present in either area) to I4 (high density of immune cell types in both locations). I4 tumors are considered "hot" while I0 tumors are is "cold." It has been reported that tumor progression (T stage) and invasion (N stage) depend on this pre-existing adaptive intratumoral immunity. More often, researchers are now studying the nature of immune cells in tumors, Density, immune function direction and distribution.

As reported by Galon, the essential characteristics of thermal immune tumors are (i) high T cell and cytotoxic T cell infiltration and (ii) checkpoint activation or impaired T-cell function.

Furthermore, in order to confirm that the galactosyl group in lactose is the main functional group binding lectin and galectin, the present invention tested lactose, sucrose, and N-acetyllactosamine (LacNAc) through a lectin competition staining experiment. and lactose-BSA (bovine serum albumin). Since it does not contain a galactosyl moiety, sucrose does not exhibit competitive binding properties for ECL, sWGA and Gal-1. Compared with lactose, LacNAc can compete more effectively with β-galactosides in the cell surface glycome for ECL and Gal-1 binding, but can also bind to sWGA at higher concentrations. In mice, intravenous injection of LacNAc showed an inhibitory effect on MC38 tumor growth. Since lactose in the blood is quickly excreted from the body through the kidneys, in order to stabilize its concentration in the blood, the present invention couples lactose to BSA through high temperature, which may improve pharmacokinetics and delay its metabolism and excretion in the body.

Based on the above findings, this application provides the following technical solutions.

In one aspect, the present invention provides the use of galectin-1 (Gal-1, galectin-1) as a T cell immune checkpoint in the preparation of tumor immunotherapy pharmaceutical compositions.

The tumor is an immunoinflammatory tumor.

The application includes: selecting at least one galectin-1 targeted conjugate or a derivative thereof as the active ingredient of the drug; and/or selecting a natural galectin-1 targeted conjugate , allowing the tumor cells to overexpress the precursor of the targeted conjugate; and/or, determining the galectin-1 binding site on the surface of immune cells in the tumor microenvironment, so that the binding site is reduced or deleted or inactivated.

The pharmaceutical composition is selected from the following pharmaceutical compositions:

(1) The active ingredient of the pharmaceutical composition includes at least one galectin-1 targeted conjugate or a derivative thereof;

(2) The pharmaceutical composition includes an immune enhancer, which includes a vector that overexpresses a galectin-1 targeted conjugate, or a precursor that overexpresses a galectin-1 targeted conjugate. accelerator;

(3) The pharmaceutical composition includes a modified immune cell in which the galectin-1 binding site on the surface of the immune cell is reduced, deleted or inactivated.

The galectin-1 targeting conjugate is generally selected from polynucleotides, polypeptides, antibodies, lipids or carbohydrates. In certain advantageous embodiments, the galectin-1 targeting conjugate is in a soluble form.

Preferably, the tumor is a solid tumor. Preferably, the solid tumor is immunogenic. In some embodiments, the solid tumor has a high degree of T lymphocyte infiltration in the interior and/or edges. In some embodiments, the tumor is further characterized by impaired T lymphocyte function. Preferably, the T lymphocytes are CD3 + T lymphocytes and/or CD8 ⁺ T lymphocytes.

In some embodiments, the tumor is breast cancer, pancreatic head cancer, colon cancer, melanoma, etc.

In some embodiments, the tumor is a tumor that has become resistant to PD-1 and/or PD-L1 treatment.

Preferably, the dosage form of the pharmaceutical composition is an injection preparation, such as intravenous injection preparation, subcutaneous injection preparation, intradermal injection preparation, intramuscular injection preparation or intratumoral injection preparation, etc. In a specific embodiment, the dosage form is an intravenous formulation.

In some embodiments, the galectin-1 targeting conjugate is a carbohydrate or carbohydrate-containing molecule, including but not limited to: a disaccharide, non-limiting examples of which include lactose, lactulose, lactose sucrose, methyl β-Lactopyranoside, D-galactose, 4-O-β-D-galactopyranosyl-D-mannopyranoside, 3-O-β-D-galactopyranosyl-D-arabinose, 2'-O-methyllactose, lacto-N-disose, N-acetylactosamine, and thiodigalactopyranoside; larger sugars, such as molecules containing polylactosamine; and synthesis inhibitors, Examples include thiogalactopyranosides and sugar polymers. Advantageously, the carbohydrate or carbohydrate-containing molecule is not metabolizable in the host to which the composition is administered. In certain embodiments, the galectin-1 targeting conjugate is selected from the group consisting of N-acetyl-lactosamine, β-lactosyl-thioalbumin, citrus pectin, D-lactitol monohydrate, lactose Acid, benzyl 4-O-β-D-galactopyranosyl-β-D-glucopyranoside, methyl 4-O-β-D-galactopyranoside-β-D-glucopyranoside Glycoside, 2-methyl-β-D-galactose(1→4)D-glucose, methoxyethylthioethyl 2acetamido-2-deoxy-4-O-β-D galactopyranoside-β- D-glucopyranoside, carboxyethylthioethyl 2-acetamido-2-deoxy4-O-β-D galactopyranoside-β-D-glucopyranoside-BSA conjugate, 4-nitrate Phenyl 2-acetamido-2-deoxy-3-O-β-D-galactopyranoside-β-D-glucopyranoside and N-propyl-β-lactopyranoside.

Preferably, the carbohydrate or carbohydrate-containing molecule is not metabolizable in the host to which the composition is administered.

Preferably, the galectin-1 targeted binder is β-galactoside.

In some embodiments, the β-galactoside is selected from the group consisting of lactose, lactulose, lactosucrose, methyl β-lactoside, methyl β-lactoside, 4-O-β-D-galactopyranosyl -D-mannopyranoside, 3-O-β-D-galactopyranosyl-D-arabinose, 2'-O-methyllactose, lacto-N-biose, N-acetyllactosamine, β -D-Thiogalactopyranoside.

In some embodiments, the beta-galactoside is lactose; in some embodiments, the beta-galactoside is a lactose-peptide, such as lactose-BSA; in some embodiments, the beta-galactoside is LacNAc.

In some embodiments, the β-galactoside is lactose. The concentration of lactose in the pharmaceutical composition is 5mM-400mM.

In some embodiments, the medicament contains a mixture of lactose and derivatives thereof.

In some embodiments, the lactose is lactose and/or hydrates thereof. In some embodiments, the lactose is α-lactose and/or β-lactose; in other embodiments, the lactose is a hydrate of α-lactose; in some embodiments, the lactose is α-lactose. A mixture of lactose and α-lactose hydrate; in some embodiments, the lactose is β-lactose; in some embodiments, the lactose is one or two of α-lactose, β-lactose, and α-lactose hydrate. A mixture of more than one species.

The derivative of lactose is a pharmaceutically acceptable derivative, which may have the same parent core structure as the compound itself, and may become lactose or a molecule with similar functions to lactose during the administration of the drug , for example, the drug produces molecules with the same or similar activity as the original compound through reactions such as hydrolysis. The derivative can be a specific covalent modification product of lactose or other disaccharides or polysaccharides or small molecules with similar galactose functional groups.

Preferably, the pharmaceutically acceptable derivatives of lactose may particularly refer to its simple derivatives, and particularly refer to one of its lower esters, lower ethers, lower alkyl substituents, pharmaceutically acceptable salts and lower amides, That is, derivatives obtained by the condensation of carboxylic acids and alcohols. Amines having 1 to 6, preferably 2 to 6 or 2 to 4 carbon atoms with the parent compound.

In some embodiments, the pharmaceutically acceptable derivative of lactose is a salt thereof. The pharmaceutically acceptable salt of lactose can be synthesized from the parent compound by conventional chemical methods, such as using Pharmaceutical Salts: Properties, Selection and Use, P Salt synthesized by the method in Heinrich Stahl (Editor), Camille G. Wermuth (Editor), ISBN: 3-90639-026-8, Hardcover, 388 pages, August 2002. Typically, this can be done by making the free base of the compound an acid in a solution with water, an organic solvent, or a mixture of both; typically, a non-aqueous medium can be used, such as diethyl ether, ethyl acetate, ethanol, isopropanol, or acetonitrile.

Acid addition salts can be prepared with various acids (inorganic and organic). Examples of acid addition salts may include salts prepared from acids that may be selected from acetic acid, 2,2-dichloroacetic acid, adipic acid, alginic acid, ascorbic acid (e.g., L-ascorbic acid), L-aspartic acid Acid, benzenesulfonic acid, benzoic acid, 4-acetamidobenzoic acid, butyric acid, (+)-camphoric acid, camphorsulfonic acid, (+)-(1S)-camphor-10-sulfonic acid, capric acid, caproic acid Acid, caprylic acid, cinnamic acid, citric acid, cyclic acid, dodecyl sulfonic acid, ethane 1,2-disulfonic acid, ethanesulfonic acid, 2-hydroxyethanesulfonic acid, formic acid, fumaric acid, galactose Acid, gentisic acid, glucuronic acid, D-gluconic acid, glucuronic acid (such as D-glucuronic acid), glutamic acid (such as L-glutamic acid), alpha-ketoglutarate, glycolic acid, horseradish Uric acid, hydrobromic acid, hydrochloric acid, hydroiodic acid, isethionic acid, (+)-L-lactic acid, (+)-DL-lactic acid, lactobionic acid, maleic acid, malic acid, (-)-L- Mariic acid, malonic acid, (±)-DL-mandelic acid, methanesulfonic acid, naphthalene-2-sulfonic acid, naphthalene-1,5-disulfonic acid, 1-hydroxy-2-naphthoic acid, nicotinic acid, Nitric acid, oleic acid, orotic acid, oxalic acid, palmitic acid, palmitic acid, phosphoric acid, propionic acid, L-pyroglutamic acid, salicylic acid, 4-aminosalicylic acid, sebacic acid, stearic acid, succinic acid Acids, sulfuric acid, tannic acid ((+)-L-tartaric acid, thiocyanic acid, p-toluenesulfonic acid, undecenoic acid and valeric acid, and acyl amino acids.

The lactose or pharmaceutically acceptable derivatives thereof may contain more than one asymmetric center and may therefore exist in various stereoisomeric forms, for example as enantiomers and/or diastereomers. Thus, the lactose or pharmaceutically acceptable derivatives and combinations thereof of the present invention may be in the form of individual enantiomers, diastereomers or geometric isomers, or may be in the form of mixtures of stereoisomers.

The medicament contains lactose and/or its pharmaceutically acceptable derivatives. In some embodiments, lactose or its derivatives is the only active ingredient other than water in the drug. In some embodiments, the medicament further contains pharmaceutically acceptable excipients. In some embodiments, the drug is a pharmaceutical composition.

In some embodiments, the β-galactoside derivative is a conjugate of lactose and a polypeptide, the polypeptide is an antibody or antibody fragment capable of recognizing a tumor surface specific antigen, and/or, the polypeptide is with Lactose conjugates slow down the metabolism of lactose and prolong the maintenance time of the effective concentration of lactose in the body. Furthermore, conjugation of some polypeptides with lactose can further enhance the binding ability of lactose to Gal-1. An exemplary The peptide is BSA.

Preferably, the pharmaceutical composition is an injection preparation, and the pharmaceutical composition is injected through at least one route selected from the group consisting of intravenous, subcutaneous, intradermal, parenteral and intramuscular. In a specific embodiment, the pharmaceutical composition is an intravenous formulation.

In some embodiments, the immune enhancer comprises a vector overexpressing a galectin-1 targeting conjugate that is a galectin-1 antibody.

In some embodiments, the galectin-1 targeting conjugate precursor is alpha-lactalbumin (LALBA).

In some embodiments, the immune enhancer includes alpha-lactalbumin (LALBA) booster. The accelerator is selected from the group consisting of substances that increase LALBA levels, substances that enhance LALBA activity and/or substances that retard LALBA metabolism. For example, the accelerator is selected from: natural purified substances, modified natural purified substances, semi-synthetic substances, chemically synthesized substances; for example, the accelerator is derived from mammals; for example, the accelerator is derived from: humans, Non-human primates (such as orangutans, apes), rodents (such as rats, mice, guinea pigs), pets (such as cats, dogs), domestic animals (such as horses, cattle, sheep, pigs, rabbits).

The promoter is selected from: LALBA overexpression vector, nanoparticles carrying LALBA gene, viral vector carrying LALBA or its overexpression vector, PEG modified protein, protein microspheres, liposomes encapsulating LALBA, cells carrying LALBA External vesicle.

LALBA is a substance selected from the following group: LALBA gene, LALBA mRNA, cDNA, LALBA protein or active fragment of any of the foregoing;

The LALBA is derived from mammals, such as humans.

The enhancer is comprised in a pharmaceutical composition or kit, e.g. in a form suitable for administration by a means selected from the group consisting of oral administration, injection (e.g. direct naked DNA or protein injection, liposome encapsulation) DNA, RNA or protein injection method), gold-coated gene gun bombardment method, reproduction-deficient bacteria carrying plasmid DNA method, replication-deficient adenovirus carrying target DNA method or protein encoded by the target gene, electroporation, transvenous, lung, Mucosal, nasal, intraperitoneal, intracranial, intratumoral, sublingual, buccal, transdermal administration; and/or the pharmaceutical composition or kit further includes other anti-tumor active ingredients or is combined with other anti-tumor active ingredients When used in combination, for example, the other anti-tumor active ingredients are selected from: tumor immunotherapy agents and chemotherapy drugs; for example, the other anti-tumor active ingredients are selected from: ipilimumab and nivolumab. The chemotherapy drug is selected from the following drugs: cyclophosphamide, trabectedin, temozolomide, melphalan, dacarbazine, oxaliplatin, methotrexate, mitoxantrone, gemcitabine, 5- Fluorouracil (5-FU), bleomycin, doxorubicin, daunorubicin, epirubicin, idarubicin, valrubicin, paclitaxel, cabazitaxel, docetaxel, topote Can, irinotecan, etoposide, carboplatin, cisplatin, bortezomib, vinblastine, vincristine, vindesine, vinorelbine, serquinone, nitrogen mustard, mitomycin C, fluda Labine, cytosine arabinoside; and combinations thereof.

The immune enhancer is a tumor immunotherapy enhancer.

The modified immune cell surface galactosylation level is reduced, deleted or inactivated.

Preferably, the modified immune cells do not express active B4GALT1 protein.

More preferably, the B4GALT1 gene of the modified immune cells is knocked out or inhibited.

Further, the suppression of the B4GALT1 gene means that the B4GALT1 gene is silenced, inactivated or the expression is inhibited. Specifically, it can mean that the B4GALT1 gene does not initiate transcription, the transcription process is inhibited, the B4GALT1 mRNA is degraded, and the B4GALT1 protein is degraded, or It can be that the inhibitory gene of B4GALT1 is activated, the protein that inhibits the expression of B4GALT1 is activated, or the expression of factors or proteins that promote the expression of the B4GALT1 gene is inhibited.

In some specific embodiments, the immune cells comprise an agent that inhibits the transcription or expression of B4GALT1 and the agent that inhibits the B4GALT1 gene includes siRNA or shRNA that inhibits the transcription or expression of B4GALT1; in some specific embodiments, the immune cells comprise a gene encoding A base sequence of shRNA. Preferably, the shRNA can be processed by cells into siRNA that inhibits the transcription or expression of B4GALT1. Preferably, the base sequence is an introduced exogenous sequence. Preferably, the base sequence Integrated into the genome of the immune cells.

In a specific embodiment, the immune cells contain siRNA or shRNA that inhibits the transcription or expression of B4GALT1.

In some specific embodiments, the genome of the immune cell contains a base sequence encoding the shRNA.

In some embodiments, the immune cells are from humans or non-human mammals.

In some embodiments, the cell is a T cell, B cell, or NK cell.

In some embodiments, the immune cells are CD8 ⁺ T cells.

In some embodiments, the cells are TCR-T cells or TCR-NK cells, preferably TCR-T cells.

In one aspect, the present invention provides a tumor immunotherapy drug, the active ingredient of which includes at least one galectin-1 targeted conjugate as described above.

In one aspect, the present invention provides an immune enhancer, which is the immune enhancer described in the first aspect of the present invention. In one aspect, the present invention provides a tumor immunotherapy drug including the immune enhancer.

In one aspect, the invention provides a modified immune cell, which is the modified immune cell described in the first aspect of the invention. In one aspect, the present invention provides a tumor immunotherapy drug including the modified immune cells.

On the one hand, the present invention provides a new use of beta-1,4-galactosyltransferase1 (B4GALT1) gene/protein, and the new use is application in the preparation of tumor immunotherapy drugs .

Further, the drug is selected from the following reagents:

a) Reagents that knock out/inhibit the B4GALT1 gene or inhibit the B4GALT1 protein; and/or,

b) Immune cells in which the B4GALT1 gene has been knocked out or suppressed.

The immune cells are T cells, B cells or NK cells. In some embodiments, the immune cells are TCR-T or TCR-NK cells. Preferably, the immune cells are TCR-T.

Further, the knockout is complete knockout or knockdown of the B4GALT1 gene. Preferably, the knockout is complete knockout.

Furthermore, the knockout of the B4GALT1 gene refers to knocking out all or part of the B4GALT1 gene sequence in the genome of the target cell, so that it cannot encode the B4GALT1 protein or cannot correctly encode the functional B4GALT1 protein. For example, in some embodiments, the full length of the B4GALT1 gene is knocked out; in some embodiments, the sequence from the start codon to the stop codon of the B4GALT1 gene is knocked out or mutated; in some embodiments, B4GALT1 All or part of one or more exons of the gene are knocked out; in some embodiments, the transcriptional regulatory elements of the B4GALT1 gene, such as the promoter sequence, are knocked out.

In some embodiments, the reagent for knocking out the B4GALT1 gene includes a nucleic acid molecule targeting the B4GALT1 gene and a nuclease guided by the nucleic acid molecule. Preferably, the nucleic acid molecule is guide RNA (gRNA). In some embodiments, the nuclease is a CRISPR nuclease. Preferably, the nuclease is Cas9. Optionally, the reagent also contains a screening marker gene.

In some embodiments, the agent for knocking out the B4GALT1 gene includes a nucleic acid molecule targeting the B4GALT1 gene and an mRNA encoding an endonuclease. Preferably, the nucleic acid molecule is sgRNA. In some embodiments, the nuclease is a CRISPR nuclease. Preferably, the nuclease is Cas9.

In some embodiments, the reagent for knocking out the B4GALT1 gene includes a recombinant nucleic acid molecule encoding a protein targeting the B4GALT1 gene and encoding an endonuclease. In some embodiments, the protein encoding the B4GALT1 gene is a TAL protein, and in other embodiments, the protein encoding the B4GALT1 gene is a zinc finger protein. The nuclease is preferably FokI.

Preferably, the reagent for knocking out the B4GALT1 gene further includes a delivery vector, which is a gRNA and endonuclease targeting the B4GALT1 gene or its encoding mRNA as described above, or will target the B4GALT1 gene as described above. Delivery of recombinant nucleic acid molecules into target cells. The delivery carrier is selected from, but is not limited to, cells, viruses, liposomes, nanoparticles, vesicles, ferritin and other carriers. The cells are preferably erythrocytes or mesenchymal stem cells. The viral delivery vector is preferably an adenovirus or lentiviral delivery vector.

Further, the suppression of the B4GALT1 gene refers to silencing or inhibiting the B4GALT1 gene. The method includes activating the inhibitory gene of B4GALT1, activating the protein that inhibits the expression of B4GALT1, introducing siRNA that inhibits the transcription or expression of B4GALT1, activating microRNA that promotes the degradation of B4GALT1 mRNA, and introducing Molecules that promote B4GALT1 protein degradation and inhibit factors and protein expression that promote B4GALT1 expression.

The reagents that inhibit the B4GALT1 gene are selected from reagents that activate the inhibitory gene of B4GALT1, reagents that activate proteins that inhibit the expression of B4GALT1, siRNA and shRNA that inhibit the transcription or expression of B4GALT1, microRNA that promotes the degradation of B4GALT1 mRNA, molecules that promote the degradation of B4GALT1 protein, and Inhibit one or more than two factors and proteins that promote B4GALT1 expression.

In some embodiments, the agent that inhibits the B4GALT1 gene includes siRNA or shRNA that inhibits the transcription or expression of B4GALT1, or an siRNA expression vector; in some embodiments, the expression vector includes bases encoding an shRNA. sequence, preferably, the shRNA can be processed by cells into siRNA that inhibits the transcription or expression of B4GALT1. Preferably, the base sequence can be integrated into the host cell genome.

In some embodiments, the agent for inhibiting the B4GALT1 gene further includes a delivery vector. Preferably, the delivery vector is a plasmid or a viral vector.

In one embodiment, the immune cells contain siRNA or shRNA that inhibits the transcription or expression of B4GALT1.

In some embodiments, the genome of the immune cell contains a base sequence encoding the shRNA, and the shRNA can be processed by the cell into siRNA that inhibits the transcription or expression of B4GALT1.

In some embodiments, the agent that inhibits B4GALT1 protein activity includes a small molecule compound capable of inactivating B4GALT1 protein.

Preferably, the B4GALT1 gene is a human or non-human mammalian B4GALT1 gene.

Preferably, the drug is a liquid preparation. Further preferably, the dosage form of the drug is injection.

Preferably, the pharmaceutical composition is a liquid preparation. Further preferably, the dosage form of the pharmaceutical composition is an injection.

"Tumor" in the present invention includes benign tumors or malignant tumors, but preferably refers to malignant tumors. Optionally, the tumor can be a cold tumor or a hot tumor. Preferably, the tumor is a hot tumor.

Preferably, the tumor of the present invention is selected from the following group: hematological tumors, solid tumors, or combinations thereof. In another preferred embodiment, the blood tumor is selected from the following group: acute myeloid leukemia (AML), multiple myeloma (MM), chronic lymphocytic leukemia (CLL), acute lymphoblastic leukemia (ALL), diffuse leukemia B-cell lymphoma (DLBCL), or combinations thereof.

In some embodiments, the solid tumor is selected from the group consisting of: tumors, lung cancer, esophageal cancer, osteosarcoma, breast cancer, endometrial cancer, gastric cancer, colon cancer, rectal cancer, melanoma, prostate cancer, liver cancer, At least one of nasopharyngeal cancer, brain glioma, leukemia, lymphoma, oral cancer, laryngeal cancer, tongue cancer, bladder cancer, kidney cancer, penile cancer, pancreatic cancer, cervical cancer, and ovarian cancer.

In one embodiment, the solid tumor is colon cancer, and in another embodiment, the solid tumor is melanoma.

On the one hand, the present invention provides a combined pharmaceutical composition for tumor immunotherapy, the active ingredients of which at least include:

(1) One or a combination of two or more of the galectin-1 targeted conjugates or derivatives thereof, the immune enhancer and the modified immune cells as described above;

(2) At least one component that induces the transformation of cold tumors into hot tumors.

Preferably, the galectin-1 targeted conjugate is lactose. In some embodiments, the derivative is a glycopeptide. In some embodiments, the derivative is lactose and is pharmaceutically acceptable. Salt.

In some embodiments, the pharmaceutical composition further includes a pharmaceutically acceptable carrier.

The combined pharmaceutical composition further includes other anti-tumor active ingredients or is used in combination with other anti-tumor active ingredients. For example, the other anti-tumor active ingredients are selected from: tumor immunotherapy agents and chemotherapy drugs; for example, the other anti-tumor active ingredients Ingredients selected from: ipilimumab, nivolumab. The chemotherapy drug is selected from the following drugs: cyclophosphamide, trabectedin, temozolomide, melphalan, dacarbazine, oxaliplatin, methotrexate, mitoxantrone, gemcitabine, 5- Fluorouracil (5-FU), bleomycin, doxorubicin, daunorubicin, epirubicin, idarubicin, valrubicin, paclitaxel, cabazitaxel, docetaxel, topote Can, irinotecan, etoposide, carboplatin, cisplatin, bortezomib, vinblastine, vincristine, vindesine, vinorelbine, serquinone, nitrogen mustard, mitomycin C, fluda Labine, cytosine arabinoside; and combinations thereof.

The present invention provides a method for treating tumors, which method includes administering medicine to a subject to achieve the effect of inhibiting tumors or eliminating tumor cells, and the active ingredient of the medicine includes a galectin inhibitor. Preferably, the galectin inhibitor is a galectin-1 inhibitor. Further preferably, the galectin-1 inhibitor is a β-D-galactoside.

In one aspect, the present invention provides a tumor immunotherapy method, which method includes the following steps: delivering the galectin-1 targeted conjugate as described above into the tumor immune microenvironment of the subject.

In some embodiments, the drug is administered by injection. A therapeutically effective amount of a compound thereof includes pharmaceutical compositions.

Preferably, the pharmaceutical composition is activated by at least one route selected from the group consisting of intravenous, subcutaneous, intradermal, parenteral, intramuscular and intratumoral. More preferably, the Eurasia 5 is administered via intravenous injection.

In some embodiments, the pharmaceutical composition is administered in combination with at least one other treatment modality.

In some embodiments, the compounds of the other treatment modalities are selected from the group consisting of radiation therapy, chemotherapy, and surgery, and combinations thereof.

In one aspect, the present invention provides a tumor immunotherapy method, which method includes the following steps: contacting the immune enhancer as described above and/or the modified immune cells as described above with the subject's tumor cells. .

In some embodiments, the drug is administered by injection.

Preferably, the pharmaceutical composition is activated by at least one route selected from the group consisting of intravenous, subcutaneous, intradermal, parenteral, intramuscular and intratumoral. More preferably, the injection is an intratumoral injection.

In some embodiments, the pharmaceutical composition is administered in combination with at least one other treatment modality.

In some embodiments, the compounds of the other treatment modalities are selected from the group consisting of radiation therapy, chemotherapy, and surgery, and combinations thereof.

"Tumor" in the present invention includes benign tumors or malignant tumors, but preferably refers to malignant tumors. Optionally, the tumor can be a cold tumor or a hot tumor. Preferably, the tumor is a hot tumor.

Preferably, the tumor of the present invention is selected from the following group: hematological tumors, solid tumors, or combinations thereof. In another preferred embodiment, the blood tumor is selected from the following group: acute myeloid leukemia (AML), multiple myeloma (MM), chronic lymphocytic leukemia (CLL), acute lymphoblastic leukemia (ALL), diffuse leukemia B-cell lymphoma (DLBCL), or combinations thereof.

In some embodiments, the solid tumor is selected from the group consisting of: tumors, lung cancer, esophageal cancer, osteosarcoma, breast cancer, endometrial cancer, gastric cancer, colon cancer, rectal cancer, melanoma, prostate cancer, liver cancer, At least one of nasopharyngeal cancer, brain glioma, leukemia, lymphoma, oral cancer, laryngeal cancer, tongue cancer, bladder cancer, kidney cancer, penile cancer, pancreatic cancer, cervical cancer, and ovarian cancer.

In some embodiments, the tumor is a tumor that has developed resistance to PD-1/PD-L1 antibodies.

In one embodiment, the tumor is colon cancer; in one embodiment, the tumor is melanoma; in one embodiment, the tumor is gastric cancer.

Preferably, the tumor originates from a subject who has a poor response to immunotherapy or needs to improve the response to immunotherapy.

More preferably, the subject or the subject is selected from the following group: tumor patients with acquired immune deficiency or impairment; tumor patients for whom tumor immunotherapy is ineffective or has poor results or has poor expected results; patients who are receiving, Patients who will receive or have received tumor immunotherapy; or tumor patients with two or more of the aforementioned conditions; and/or the subjects are mammals, such as humans, non-human primates (such as orangutans, apes) , rodents (such as rats, mice, guinea pigs), pets (such as cats, dogs), livestock (such as horses, cattle, sheep, pigs, rabbits).

The invention has the following beneficial effects:

The present findings demonstrate that cancer cells can transfer abnormally high levels of galectin-1 to the surface of neighboring immune cells, particularly tumor-targeting cytotoxic T cells. Subsequently, tumor-derived galectin-1 interacts with galactosylated proteins on the surface of cytotoxic T cells to regulate T cell activation and function.

Targeting this new immune checkpoint will be a new strategy for tumor immunotherapy. On the one hand, knocking out B4GATL1 to reduce protein galactosylation can significantly reduce the level of Gal-1 transferred from tumors to T cells, thereby enhancing T cell activation and function. On the other hand, lactose and its derivatives can inhibit this immune checkpoint by competing for Gal-1 on the surface of cancer cells and immune cells.

Targeting this new immune checkpoint with small molecules will be a feasible strategy for tumor immunotherapy. Lactose and related derivatives containing galactosides can inhibit this immune checkpoint by competing for Gal-1 on the surface of cancer cells and immune cells. In particular, lactose, as a structurally simulated competitive galectin inhibitor, can effectively inhibit tumor growth through systemic intravenous injection into the circulation system of mice with different tumor burdens. The conjugation of macromolecules (such as BSA) and lactose can enhance the binding ability to Gal-1 and may help prolong the maintenance of effective lactose concentrations in the body. Lactose is a natural product in dairy products, with a content of about 4.8% in milk. Lactose can be prepared in large quantities from milk with mature technology, simple method and low cost. Currently, the annual output of high-purity lactose (more than 99%) in China is Tens of thousands of tons, it is mainly used as an additive and molding agent for drugs; the production method of lactose is much simpler than that of anti-PD1 antibodies, and the production cost is several orders of magnitude lower. It has great prospects for tumor immunotherapy.

In addition, overexpression of the lactose synthase component LALBA in mouse tumors can achieve de novo synthesis of lactose in the tumor microenvironment, thereby enhancing the immune response mediated by CD8 ⁺ T cells. Combining the anti-tumor efficacy and economic cost of lactose will provide a widely feasible idea for cancer treatment.

Description of the drawings

Figure 1 shows the results of in vitro and in vivo CRISPR/Cas9 screens to identify genes that regulate PD1 expression and CD8 ⁺ T cell function, where:

A. Schematic diagram of in vivo and in vitro CRISPR/Cas9 screening using mouse primary CD8 ⁺ T cells;

B. Volcano plot shows the results of CRISPR/Cas9 genome-wide in vitro screening;

C. Verification results of candidate genes in the screening results; flow cytometry was used to detect the mean fluorescence intensity (MFI) of PD1 on the cell surface and RT-qPCR was used to detect the expression of PD1 mRNA;

D. KEGG pathways enriched in the genome-wide screening of gene enrichment analysis (GSEA);

E. Volcano plot showing the results of in vitro CRISPR/Cas9 screening using a custom mini-library;

F. Volcano plot showing the results of in vivo CRISPR/Cas9 screening using a customized mini-library;

Figure 2 shows that knocking out B4GALT1 in T cells can enhance TCR signaling and T cell anti-tumor function.

A. The expression of PD1 increased after knocking out the B4GALT1 gene in CD8 ⁺ T cells; flow cytometry was used to detect the mean fluorescence intensity (MFI) of PD1 on the cell surface and RT-qPCR was used to detect the expression of PD1 mRNA;

B. Overexpression of the short or long isoform cDNA of B4GALT1 in mice can complement the high PD1 expression phenotype induced by B4GALT1 knockout;

C. After co-culture with B16F10-OVA cells, knocking out B4GALT1 in CD8 ⁺ T cells will increase the expression of IL2, TNFα and IFNγ; RT-qPCR was used to detect the expression of IL2 and IFNγ mRNA; enzyme-linked immunosorbent assay (ELISA) was used ) Detect the secreted TNFα and IFNγ content in the culture medium;

D. Knocking out B4GALT1 in T cells will enhance the ability to kill B16F10-OVA cells in vitro;

E. The heat map shows the differentially expressed genes between B4GALT1 knockout and control OT-1 cells after co-culture with tumor cells; genes involved in the TCR signaling pathway are marked on the left side of the heat map;

F. Volcano plot shows the significantly up-regulated and significantly down-regulated genes (pvalue<0.01) after B4GALT1 knockout OT-1 cells were co-cultured with tumor cells; the genes of the TCR signaling pathway are marked in dark gray in the figure; the expression is marked in the volcano plot The most significant differences and the names of some genes in the TCR pathway;

G. Volcano plot shows the significantly enriched KEGG pathways after B4GALT1 knockout OT-1 cells were co-cultured with tumor cells; the names of some of the significantly enriched pathways are marked in the volcano plot;

H.B4GALT1 knockout OT-1 cells have a stronger ability to inhibit B16F10-OVA tumor growth in vivo;

I. B16F10-OVA tumors were smaller in OT-1 mice injected with B4GALT1 knockout;

J. Knocking out B4GALT1 in OT-1 cells can increase its number in B16F10-OVA tumors;

The data in the figures are displayed as mean ± standard error (SEM);

Figure 3 demonstrates that B4GALT1 expression and tumor-infiltrating CD8 ⁺ T cells in the surface tumor microenvironment are associated with tumor patient prognosis, where:

A. Kaplan-Meier survival curve of B4GALT1 expression levels (left) and B4GALT1 expression levels normalized by CD8a (right) for all primary tumor samples in the TCGA cohort;

B. Association between CD8a expression levels and overall survival in patients with different B4GALT1 expression levels in all primary tumor samples in the TCGA cohort;

C. B4GALT1 expression levels in patients with different CD8a expression levels in all primary adrenocortical carcinoma (ACC), acute myeloid leukemia (LAML), lung adenocarcinoma (LUAD), and rectal gland (READ) tumor samples in the TCGA cohort Association with overall survival;

D. CD8a expression levels in patients with different B4GALT1 expression levels in all primary adrenocortical carcinoma (ACC), acute myeloid leukemia (LAML), lung adenocarcinoma (LUAD), and rectal gland (READ) tumor samples in the TCGA cohort Association with overall survival;

The p-value of all survival curves is calculated by two-sided Log-rank test;

Figure 4 shows that knocking out B4GALT1 in tumor cells can enhance the body’s immune surveillance response to tumor cells, where:

A. Knocking out the B4GALT1 gene in MC38 cells can inhibit their growth in wild-type mice;

B. Knocking out the B4GALT1 gene in MC38 cells does not affect their growth in immunodeficient NPG mice;

C. Growth curve of B4GALT1 knockout MC38 cells in wild-type mice depleted of CD8+T, CD4+T or NK cells;

D. Heat map shows differentially expressed genes between control MC38 cells and B4GALT1 knockout MC38 cells;

E. The volcano plot shows the significantly up-regulated and significantly down-regulated genes in MC38 cells after B4GALT1 knockout (pvalue<0.01); the genes of the interferon γ signaling pathway are marked in dark gray in the figure; the volcano plot marks the genes with the most significant expression differences. and the names of some genes in the interferon gamma pathway;

F. Volcano plot shows the significantly enriched HALLMARKER gene set in MC38 cells after B4GALT1 knockout; the name of the most significantly differentially expressed pathway is marked in the volcano plot;

G. Compared with control MC38 cells, the expression of B2m in B4GALT1 knockout MC38 cells increased after IFNγ treatment;

H. Compared with control MC38 cells, B4GALT1 knockout MC38 cells have an enhanced ability to present Ova peptides after IFNγ treatment;

I. Knocking out the B4GALT1 gene in MC38 cells can make them more susceptible to killing by OT-1T cells;

J. After co-culture with MC38 cells knocking out B4GALT1, the expression level of IFNγ in OT-1T cells increased significantly;

The data in the figures are displayed as mean ± standard error (SEM);

Figure 5 shows that intratumoral expression of α-lactalbumin (LALBA) can produce lactose in tumor cells and activate tumor immune responses, where:

A. Overexpression of LALBA in MC38 cells can promote the synthesis of lactose; LC-MS detects the content of lactose in cells;

B. LC-MS detects the lactose content in the culture medium;

C. Overexpression of LALBA in MC38 cells can inhibit its growth in wild-type mice;

D. Overexpression of LALBA gene in MC38 cells does not affect their growth in immunodeficient NPG mice;

E. Growth curve of MC38-GFP cells in mice with complete regression of MC38-LALBA tumors and uninjected wild-type mice of the same age;

F. Orthotopic injection of lentivirus encoding LALBA into MC38 tumors can inhibit tumor growth; the arrow indicates the time of virus injection;

After orthotopic injection of lentivirus encoding LALBA into G.MC38 tumors, the proportion of IFNγ-positive cells among tumor-infiltrating CD8+ T cells increased;

H. Re-inoculated MC38 cells cannot grow in mice whose tumors completely regressed after injection of LALBA-encoding lentivirus;

I. Orthotopic injection of AAV encoding LALBA into MC38 tumors can inhibit tumor growth; the arrow indicates the time of virus injection;

After orthotopic injection of AAV encoding LALBA into J.MC38 tumors, the proportion of PD1-positive cells among tumor-infiltrating CD8+ T cells increased;

After K.MC38 tumors were injected with AAV encoding LALBA in situ, the proportion of IFNγ-positive cells among tumor-infiltrating CD8+ T cells increased;

The data in the figures are displayed as mean ± standard error (SEM);

Figure 6 shows that tail vein injection of lactose solution can enhance the anti-tumor function of mice, where:

A. The effect of tail vein injection of lactose solution on the growth of MC38 tumors in wild-type mice;

B. After MC38 tumor-bearing mice were injected with lactose solution through the tail vein, the proportion of PD1-positive tumor-infiltrating CD8+ T cells increased;

C. After MC38 tumor-bearing mice were injected with lactose solution through the tail vein, the proportion of IFNγ-positive cells in tumor-infiltrating CD8+T cells increased;

D. The effect of tail vein injection of lactose solution on the growth of MC38 tumors in immunodeficient NPG mice;

E-G. Effect of tail vein injection of lactose solution on the growth of MC38 cells in wild-type mice depleted of CD8+T(E), CD4+T(F) or NK(G) cells;

H. Effect of tail vein injection of lactose solution on the growth of 4T1 tumors in wild-type BALB/c mice;

I. The effect of tail vein injection of lactose solution on the growth of CT26 tumors in wild-type BALB/c mice;

J. Effect of tail vein injection of lactose solution on the growth of B16F10 tumors in wild-type C57BL/6J mice;

K. Effect of tail vein injection of lactose solution on the growth of B16F10-OVA tumors in wild-type C57BL/6J mice;

L. Effects of tail vein injection of lactose solution SGC7901 on tumor growth in NPG mice with humanized immune systems;

After M.SGC7901 tumor-bearing humanized mice with immune systems were injected with lactose solution through the tail vein, the proportion of IFNγ-positive tumor-infiltrating human CD8+ T cells increased;

N. Effect of tail vein injection of lactose solution SGC7901 on tumor growth in NPG mice with defective immune system;

The data in the figures are displayed as mean ± standard error (SEM);

Figure 7 shows that a genome-wide screen identified the N-glycan synthesis pathway regulating PD1 expression in CD8+ T cells, where:

A. Schematic diagram of the N-glycan synthesis pathway; the genes identified in the genome-wide screen are marked in the figure;

B. The distribution of genes related to the N-glycan synthesis pathway in the genome-wide screening results; the solid line represents all genes; the long dotted line represents the genes in the N-glycan synthesis pathway; the short dotted line represents the negative control gRNA;

C. Flow cytometry was used to detect the expression of PD1 in OT-1 cells after knocking out B4GALT1, Mgat2, and Dpm3 genes respectively.

Figure 8 shows the effect of B4GALT1 knockdown on the function of OT-1 cells in tumors, where:

A. 24 hours after transplantation, detect the effect of B4GALT1 knockout on the ability of OT-1 cells to infiltrate tumors;

B. CFSE signal indicates that B4GALT1 knockout can enhance the proliferation ability of OT-1 cells in tumors;

The data in the figure are displayed as mean ± standard error (SEM); NS means no significant difference;

Figure 9 shows that the expression of B4GALT1 and tumor-infiltrating CD8+ T cells in the tumor microenvironment are related to the prognosis of tumor patients, where:

A. Heat map showing the association between B4GALT1 expression levels (left) and B4GALT1 expression levels (right) normalized to CD8a and overall survival in different primary tumor samples in the TCGA cohort; ALL-TCGA: represents the TCGA cohort. All primary tumor samples;

B. Heat map shows the association between B4GALT1 expression levels and overall survival in patients with different CD8a expression levels in primary tumor samples in the TCGA cohort (left, CD8a high expression; right, CD8a low expression); ALL-TCGA: represents all primary tumor samples in the TCGA cohort;

C. Heat map shows the association between CD8a expression levels and overall survival in patients with different B4GALT1 expression levels in primary tumor samples in the TCGA cohort (left, B4GALT1 high expression; right, B4GALT1 low expression); ALL-TCGA: represents all primary tumor samples in the TCGA cohort;

Figure 10 shows the flow cytometry test demonstrating the successful elimination of CD8+T (A), CD4+T (B) and NK (C) cells in wild-type mice;

Figure 11 shows that overexpression of LALBA in some MC38 cells can inhibit the growth of MC38 tumors in wild-type mice;

Figure 12 shows the effect of intratumoral injection of lentivirus encoding two LALBA mutants, LALBA-D107A (A) and LALBA-A126K (B), into MC38 tumors on the growth of MC38 tumors in wild-type mice;

Figure 13. Shows the effects of lectin staining to detect B4GALT1 knockout and LALBA overexpression on the overall N-glycome of cells, where:

A. Demonstrates that increasing the concentration of lactose in the solution can competitively block the staining of MC38 cells by ECL (left), but has no effect on the staining of sWGA (right);

B. Shows the effects of B4GALT1 knockdown and LALBA overexpression on the mean fluorescence intensity (MFI) of ECL and sWGA on MC38 cells;

C. Shows the effects of B4GALT1 knockdown and LALBA overexpression on ECL and sWGA mean fluorescence intensity (MFI) on CD8+ T cells;

Note: MC38 cells were scraped directly from the plate without trypsin digestion for ECL and sWGA staining experiments;

Figure 14 shows the effect of tail vein injection of different doses of lactose solution on the growth of MC38 tumors in wild-type mice.

Effects of tail vein injection of 50mM, 250mM and 400mM lactose solutions once every two days on the growth of MC38 tumors in wild-type mice;

Figure 15 shows the effect of tail vein injection of lactose solution and anti-PD1 (PDCD1) combined on the growth of MC38 tumors in wild-type mice;

Figure 16 shows the effect of tail vein injection of lactose solution and anti-PDL1 (CD274) combined on the growth of MC38 tumors in wild-type mice;

Figure 17 shows RNA-seq analysis of the effect of lactose treatment on the MC38 tumor transcriptome in wild-type mice, where:

A. The heat map shows the differentially expressed genes in tumors after lactose treatment; the genes of the interferon gamma signaling pathway are marked on the left side of the figure;

B. The volcano plot shows the genes that were significantly up-regulated and significantly down-regulated in MC38 tumors after lactose treatment (pvalue<0.05); the genes of the interferon γ signaling pathway are marked in dark gray in the figure; the volcano plot marks the genes with the most significant expression differences. and the names of some genes in the interferon gamma pathway;

C. Volcano plot shows the HALLMARKER gene set that is significantly enriched in MC38 tumors after lactose treatment; the name of the most significantly differentially expressed pathway is marked in the volcano plot;

D-F: Transcriptional profile analysis of control and lactose-treated tumor samples;

The data in the figures are displayed as mean ± standard error (SEM);

Figure 18 shows single-cell transcriptome analysis of tumor-infiltrating T cells after control and lactose treatment, where:

A. Single cell transcriptome analysis and TCR analysis flow chart;

B. 13839 CD3-positive T cells isolated from control and lactose-treated tumor samples can be divided into 8 groups (C0-C7) in the t-SNE plot;

C.t-SNE diagram shows the distribution of CD3-positive T cells isolated from control (left) and lactose-treated (right) tumor samples in 8 populations respectively;

D.t-SNE plot shows the expression of Cd8a (left) and Cd4 (right) in all cells;

E. Define marker genes for each T cell population;

F. Distribution of T cells isolated from control and lactose-treated tumor samples in different populations;

Figure 19 Single-cell TCR sequencing analysis of tumor-infiltrating T cells after control and lactose treatment, where:

A. Clone steady-state space uses CDR3 amino acid sequences for clonotype identification in control and lactose-treated samples;

B. Use the CDR3 amino acid sequence to identify clonotypes, and the relative proportion of specific clonotypes in control and lactose-treated samples;

C. Based on Shannon, InvSimpson, Chao and ACE in control and lactose treated samples

TCR diversity analysis of (abundance-basedcoverageestimator) index;

D. Distribution of the largest T cell clone, CN1, in control and lactose-treated samples;

E. Distribution of the largest T cell clone, CN1, among different populations in control and lactose samples;

Figure 20: Shows that in NPG mice with humanized immune systems, tail vein injection of lactose shows a similar effect to pembrolizumab (anti-humanPD1 antibody) in inhibiting the growth of SGC7901;

Figure 21A, B, and C evaluate the safety and toxicity of short-term and long-term intravenous lactose administration in mice, where

A. Short-term (24 hours after injection) effects of intravenous lactose on blood/serum biochemical and hematological parameters in wild-type mice;

B. Long-term effects of intravenous lactose on blood/serum biochemical and hematological parameters of wild-type mice (administered through 12 tail vein injections and measured after 24 days);

C. Monitor the effect of intravenous lactose or PBS injection on the body weight of mice within 23 days;

Figure 22 Structural diagram of lentiviral vector overexpressing LALBA;

Figure 23 Structural diagram of an adenoviral vector overexpressing LALBA;

Figure 24 Structural diagram of pMSCV-CD19 scFv-IRES-RFP plasmid;

Figure 25 CRISPR/Cas9 knockout of B4galt1 in OT-IT cells can change lectin binding;

Figure 26 Expression of Galectins in OT-IT cells and MC38 tumor cells;

Figure 27 The production of Gal-1 on the surface of OT-Ⅰ T cells does not depend on TCR activation;

Figure 28 Gal-1 on the surface of OT-IT cells comes from MC38 cells;

Figure 29 The transfer of Gal-1 from MC38 cells to OT-ⅠT cells requires close contact;

Figure 30B4 Antibody activation of galt1 knockout T cells is not affected by Gal-1;

Figure 31 Killing experiments of wild type and B4galt1 knockout OT-Ⅰ T cells and wild type and Gal-1 knockout MC38

Figure 32 Gal-1 is transferred from tumor cells to infiltrating CD8+ T cells in the tumor microenvironment in vivo;

Figure 33 The growth of Gal-1 knockout MC38 tumors is inhibited by CD8+ and CD4+ T cells of the host immune system;

Figure 34 Flow cytometry analysis of Gal-1 expression on the surface of CD8+ T cells in mouse spleen, peripheral blood and MC38 tumors;

Figure 35Gal-1 knockout MC38 tumor-infiltrating CD8+ T cells have reduced levels of Gal-1 on the surface;

Figure 36 Purification and activity testing of recombinant Gal-1 protein;

Figure 37 Identification and analysis of Gal-1 binding proteins in wild-type and B4galt1 knockout OT-ⅠT cells;

Figure 38 Western blotting verifies the substrates of CD8β and other B4galt in T cell membrane proteins;

Figure 39Gal-1 binds and regulates TCR-CD8 co-localization;

Figure 40B4 The roles of galt1 and CD8 in OT-Ⅰ T cells and hCD19-CAR T cells respectively;

Figure 41 Lactose can significantly enhance the specific killing of tumor cells by OT-IT cells;

Figure 42 Lactose treatment can remove Gal-1 on the surface of OT-ⅠT cells in the in vitro killing system;

Figure 43 Lactose reverses the inhibitory effect of Gal-1 on OT-ⅠT cell activity;

Figure 44 Lactose can remove Gal-1 on the surface of tumor-infiltrating T cells in vivo and enhance T cell toxicity;

Figure 45 Intravenous injection of lactose can inhibit the growth of MC38 tumors in wild-type mice;

Figure 46 The effect of intravenous lactose on different types of tumors;

Figure 47 Parallel artificial membrane permeability experiment of lactose;

Figure 48 Properties of lactose and its derivatives.

Detailed ways

The present invention will be further described below through specific embodiments, but the present invention is not limited to the scope of the described embodiments.

1. Experimental materials, reagents, and instruments

In order to explain the present invention more clearly, the following experimental materials, reagents, and instruments used in the present invention are hereby listed. Those not listed are all conventional materials, reagents, or instruments in this field, and can be obtained through normal commercial purchases. .

1.1 Experimental materials

(1) pKLV-U6-sgRNA-PGK-puro2ABFP (Addgene#50946) was purchased from Addgene.

(2) pMSCV-IRES-GFP (Addgene#20672) was purchased from Addgene.

(3) The pMSCV-CD19 scFv-IRES-RFP plasmid was from Shao Feng’s laboratory at the Beijing Institute of Life Sciences.

(4) The lenti-EF1α related plasmid is constructed by lenti-U6 sgRNA-EF1α-IRES-GFP or lenti-EF1α-puro empty.

(5) Mouse Gal-1 gene cDNA (mm39 ENSMUST00000089377.6) was amplified from the mouse T cell cDNA library.

(6) Mouse B4galt1 gene cDNA was amplified from the mouse T cell cDNA library.

(7) Trans1-T1 transfection competent cells were purchased from Beijing Quanshijin Biotechnology Co., Ltd., product number CD501-02.

(8) DH10B-Plus electroporated competent cells were purchased from Shanghai Vidy Biotechnology Co., Ltd., product number DE1072M.

(9) Human embryonic kidney HEK293T cell line: purchased from Life Technologies.

(10) Mouse hybridoma PK136 cell line: purchased from American Type Culture Collection, ATCC.

(11) C57BL/6J mice: purchased from Beijing Vitong Lever Company.

(12) BALB/c mice: purchased from Beijing Vitong Lever Company.

(13) B-NDG (NOD.CB17-Prkdc ^scid Il2rg ^tm1 /Bcgen) mice: purchased from Biocytogen (Beijing) Pharmaceutical Technology Co., Ltd.

(14) Peptide-N-glycosidase F (PNGase F): purchased from NEB Company, product number P0704S.

(15) Recombinant mouse interleukin 2 (rIL-2): purchased from BioLegend, Cat. No. 575404.

(16) Recombinant mouse interleukin 7 (rIL-7): purchased from BioLegend, product number 577804.

(17) Recombinant mouse interleukin 15 (rIL-15): purchased from BioLegend, Cat. No. 566304.

(18) OVA oligopeptide (SIINFEKL): purchased from Sangon Bioengineering (Shanghai) Co., Ltd., product number T510212.

(19) Polybrene: Purchased from Sigma Company, item number S2667.

(20) α-Lactose hydrate: purchased from Sigma Company, item number L3625 (L3625-1KG, Lot#SLCD8654).

(21) Collagenase IV: purchased from Sigma Company, product number V900893.

(22) Anti-mouse CD3e antibody anti-mouse CD3e (clone number: 145-2C11), purchased from eBioscience, product number 14-0031-82.

(23) Anti-mouse CD28 antibody Ultra-LEAF ^TM Purified anti-mouse CD28 (clone number: 37.51): purchased from BioLegend, product number 102116.

(24) Anti-mouse PD-1 antibody PE anti-mouse CD279 (clone number: RMP1-30): purchased from BioLegend, product number 109104.

(25) Anti-mouse PD-1 antibody APC anti-mouse CD279 (clone number: RMP1-30): purchased from BioLegend, product number 109112.

(26) Anti-mouse Gal-1 antibody PE conjugated: purchased from bio-techne R&D, product number IC1245P.

(27) Anti-mouse IFNγ antibody FITC anti-mouse IFNγ (clone number: XMG1.2): purchased from BioLegend, product number 505806.

(28) Anti-mouse CD8aAPC anti-mouse CD8a (clone number: 53-6.7): purchased from eBioscience, catalog number 17-0081-83.

(29) Anti-mouse TCR Vα2 monoclonal antibody PE conjugated (clone number: B20.1): purchased from BDBiosciences, product number 553289.

(30) Biotin anti-mouse CD8a monoclonal antibody (clone number: 53-6.7): purchased from BioLegend, product number 100704.

(31) Anti-mouse CD8β antibody (clone number: EPR22331-54): purchased from abcam, catalog number ab228965.

(32) Biotin ECL: purchased from Vector Laboratories, product number B-1145-5.

(33) Biotinylated succinyl wheat germ agglutinin (Biotin sWGA): purchased from Vector Laboratories, product number B-1025S-5.

(34) Biotinylated mouse galactosyl lectin (Biotin-rGal-1,): prepared in the laboratory (Gal-1 is the sequence shown in mm39ENSMUST00000089377.6).

(35) PE Streptavidin: Purchased from BioLegend, Cat. No. 405204.

(36) Anti-mouse NK1.1 antibody: purified using mouse hybridoma cell PK136.

(37) Anti-mouse CD8α antibody (clone number: 2.43): purchased from BioXCell, product number BE0061.

(38) Anti-mouse CD4 antibody (clone number: GK1.5): purchased from BioXCell, product number BE0003-1.

(39) Anti-mouse Ly9 antibody: purchased from abcam, catalog number ab252931.

(40) Anti-mouse Igf2γ antibody: purchased from Sino Biological, Cat. No. 107533-T40.

(41) Anti-mouse Itga1 antibody: purchased from Solarbio, Cat. No. K002895P.

(42) Anti-mouse Ighg1 antibody: purchased from Solarbio, Cat. No. K111394P.

(43) Anti-mouse Lnpep antibody: purchased from Santa Cruz, Cat. No. sc-365300.

(44) Anti-mouse Sell antibody: purchased from Santa Cruz, Cat. No. sc-390756.

(45) Anti-mouse IgG 680RD fluorescent secondary antibody: purchased from LI-COR, Cat. No. P/N 926-68072

(46) Anti-mouse IgG 800RD fluorescent secondary antibody: purchased from LI-COR, Cat. No. P/N 926-32210

(47) Anti-rabbit IgG 680RD fluorescent secondary antibody: purchased from LI-COR, Cat. No. P/N 926-68071

(48) Anti-rabbit IgG 800RD fluorescent secondary antibody: purchased from LI-COR, Cat. No. P/N 926-32211

(49) Anti-human CD8α APC anti-human CD8α: purchased from BioLegend, Cat. No. 300912.

(50) ImProm-Ⅱ ^TM Reverse Transcriptase: Promega Company, product number A3803.

(51) Random Hexamer Primer: purchased from Thermo Scientific, Cat. No. 51709.

1.2 Enzyme-linked immunosorbent assay (Elisa) reagent

(1) Mouse TNFα Elisa kit: ABclonal, product number RK00027.

(2) Mouse IFNγElisa kit: ABclonal, product number RK00019.

(3) Human TNFα Elisa kit: ABclonal, product number RK00030.

(4) Human IFNγElisa kit: ABclonal, product number RK00015.

1.3 Experimental instruments

(1) Flow cytometer: BDAccuriTM C6, BD FACSAria II, BD FACSAria III, BD FACSAriaFusion.

(2) Microplate reader: PerkinElmer EnSpire Multimode Plate Reader.

2. Experimental methods

In order to explain the present invention more clearly, the following experimental methods used in the present invention are listed here. Those not listed are routine experimental methods known to those skilled in the art.

(1) Construction of vector MSCV-U6 sgRNA-PGK-PURO-2A-BFP.

Backbone vector: MSCV-U6gRNA-PGK-Puro-2a-BFP, digested with BbsⅠ.

Forward primer: 5’-CACCG+20nt sgRNA sequence-3’

Reverse primer: 5’-AAAC+20nt sgRNA sequence+C-3’

Cloning method: T4 connection.

(2) Construct the vector PKLV-U6 sgRNA-PGK-PURO-2A-BFP.

Backbone vector: PKLV-U6gRNA-PGK-Puro-2a-BFP, digested with BbsⅠ.

Forward primer: 5’-CACCG+20nt sgRNA sequence-3’

Reverse primer: 5’-AAAC+20nt sgRNA sequence+C-3’

Cloning method: T4 connection.

(3) Construct the vector lenti-U6 sgRNA-EF1α-Gal-1-EGFP fusion.

Step one: Construct MSCV-EF1α-Gal-1-IRES-GFP.

Backbone vector: MSCV-EF1α-Lalba-IRES-GFP, digested with BsiWⅠ and EcoRⅠ.

PCR template: mouse T cell cDNA library.

Forward primer: 5’-CCATTTCAGGTGTCGTGACGTACGTGCCACCATGGCCTGTGGTCTGGTCG-3’(SEQID NO:6)

Reverse primer: 5’-ACGTTAGGGGGGGGGGCGGAATTCTCACTCAAAGGCCACGCAC-3’ (SEQ IDNO: 7)

Cloning method: Gibson assembly.

Gibson assembly system: 30ng digested vector, 10ng PCR product, 2.5μL 2×Gibson, add double distilled water to make up to 5μL, ligate at 50°C for 15 minutes.

Step 2: Construct lenti-U6 sgRNA-EF1α-Gal-1-IRES-GFP.

Backbone vector: lenti-U6 sgRNA-EF1α-IRES-GFP, digested with BamHI and MluI.

Fragment template: MSCV-EF1α-Galectin1-IRES-GFP, digested with BamHI and MluI.

Cloning method: T4 connection.

Step 3: Build the target vector.

Backbone vector: lenti-U6 sgRNA-EF1α-Gal-1-IRES-GFP, digested with XcmⅠ and MluⅠ.

First round PCR template: lenti-U6 sgRNA-EF1α-Gal-1-IRES-GFP

Forward primer: 5’-AGACTTCAAGATTAAGTGCGTGGCCTTTGAGATGGTGAGCAAGGGCGAGG-3’(SEQID NO:8)

Reverse primer: 5’-AATCCAGAGGTTGATTGTCG-3’ (SEQ ID NO: 9)

Second round PCR template: the product of the previous round of PCR.

Forward primer: 5’

-CCTCAACATGGAGGCCATCAACTACATGGCGGCGGATGGAGACTTCAAGATTAAGTGC-3’(SEQ IDNO:10)

Reverse primer: 5’-AATCCAGAGGTTGATTGTCG-3’ (SEQ ID NO: 11)

Cloning method: Gibson assembly.

(4) Construction vector: lenti-EF1α-puro-2A-OVA

Backbone vector: lenti-EF1α-puro empty, digested with BamHI and MluI.

PCR template 1: MSCV-U6 sgRNA-PGK-PURO-2A-BFP

Forward primer: 5’-TGGTGCATGACCCGCAAGCC-3’ (SEQ ID NO: 12)

Reverse primer: 5’-TGGGCCAGGATTCTCCTCCA-3’ (SEQ ID NO: 13)

PCR template 2: fully synthetic OVA oligopeptide sequence vector.

Forward primer: 5’-ACGTGGAGGAGAATCCTGGCCCAAGCACCAGGACACAAATAAATAAGG-3’(SEQID NO:14)

Reverse primer: 5’-CCAGAGGTTGATTGTCGACTTAACGCGTTTAAGGGGAAACACA TCTGCCAAAG-3’(SEQ ID NO:15)

Cloning method: Gibson assembly.

(5) Construction vector: MSCV-U6 sgB4galt1-PGK-PURO-2A-B4GALT1 CDS (the two transcripts are cloned separately, and the PCR template 2 forward primers are different)

Backbone vector: MSCV-U6 sgB4GALT1-PGK-Puro-2a-BFP, digested with BamHI and SalI.

PCR template 1: MSCV-U6 sgB4GALT1-PGK-Puro-2a-BFP

Forward primer: 5’-GTCCTAGCAATTTTTTTGGATCCAATTCT-3’ (SEQ ID NO: 16)

Reverse primer: 5’-TGGGCCAGGATTCTCCTCC-3’ (SEQ ID NO: 17)

PCR template 2: cDNA library of mouse T cells.

Forward primer: 5’-TGGAGGAGAATCCTGGCCCAATGAGGTTTCGTGAGCAGTTCC-3’ (long transcript) (SEQ ID NO: 18)

5’-TGGAGGAGAATCCTGGCCCAATGCCGGGCGCGACCCTG-3’ (short transcript) (SEQ ID NO: 19)

Reverse primer: 5’-CCTTTGGAGTATTGGATGCAAATAATACAGCCAGT-3’ (SEQ ID NO: 20)

PCR template 3: cDNA library of mouse T cells.

Forward primer: 5’-CTGGCTGTATTATTTGCATCCAATACTCCAAAGG-3’ (SEQ ID NO: 21)

Reverse primer: 5’

-TCGATAAGCTTGGCTGCAGGTCGACCTATCTCGGTGTCCCGATGTCCACTGTG-3’ (SEQ ID NO: 22)

Cloning method: Gibson assembly.

(6) Construction vector: pQE-80-Gal-1opti

Backbone vector: pQE-80-Tn5, digested with BamHI and HindII.

PCR template: fully synthetic codon-optimized mouse Gelctin-1 sequence vector.

Forward primer: 5’-TCGCATCACCATCACCATCACGGATCCATGGCCTGCGGCCTGGTGG-3’ (SEQ IDNO: 23)

Reverse primer: 5’-GGAGTCCAAGTCCAGCTAATTAAGCTTTTATTCAAAGGCCACGCATT-3’ (SEQ IDNO: 24)

Cloning method: Gibson assembly.

(7) Construction vector: pMSCV-CD19 scFv(FMC63)-IRES-RFP-U6 sgRNA

Backbone vector: pMSCV-CD19 scFv-IRES-RFP, digested with ClaⅠ.

PCR template: MSCV-U6 sgB4galt1-PGK-PURO-2A-B4GALT1 CDS.

Forward primer: 5’-CGTCGACCTGCAGCCAAGCTTATCGATGAGGGCCTATTTCCCATGAT-3’(SEQ IDNO:25)

Reverse primer: 5’-CTAAATAAAATCTTTTATTTTATCGATcAAAAAAATTGCTAGGACCGGC-3’

(SEQ ID NO:26)

Cloning method: Gibson assembly.

(8) Construction vector: lenti-SFFV-NYESO-1-T2A-BFP-U6 shRNA

Backbone vector: lenti-SFFV-NY-ESO-1-T2A-BFP-U6 sgRNA, digested with BamBI and XhoI.

PCR template 1: MSCV-CD19 scFv-IRES-RFP-U6 shRNA.

Forward primer: 5’-GTCGACCTGCAGCCAAGCTT-3’ (SEQ ID NO: 27)

Reverse primer: 5’-GATTGTCGACGGATCCTCTAGACTCGAGATCGCCATTTGTCTCGAGGT-3’

(SEQ ID NO:28)

PCR template 2: lenti-SFFV-NY-ESO-1-T2A-BFP-U6 sgRNA.

Forward primer: 5’-CCAACGGCCCTGTGATGCA-3’ (SEQ ID NO: 29)

Reverse primer: 5’-AAGCTTGGCTGCAGGTCGACTCAATTAAGCTTGTGCCCCAG-3’ (SEQ ID NO: 30)

Cloning method: Gibson assembly.

(9) Construction vector: pMSCV-Biggest clone TCRA-2A-TCRN-IRES-GFP

Backbone vector: pMSCV-IRES-RFP, digested with EcoRI and XhoI.

Fragment template: fully synthetic Biggest clone vector, digested with EcoRⅠ and XhoⅠ.

Cloning method: T4 connection.

(10) Construction of B16F10-OVA cell line

The lenti-EF1α-puro-2A-OVA vector is used to package the lentivirus, and the virus liquid can be directly infected into the B16F10 cell line without concentration. 48 hours after infection, add 2 μg/mL puromycin for positive screening for infection. Since dead cells and debris are produced during the screening process, the culture medium needs to be replaced or passaged in time. After 4 to 5 days of screening, the presentation of OVA oligopeptides on the B16F10 surface can be detected by flow cytometry. After about 7 days, the B16F10-OVA cell line can be sufficiently amplified and can be aliquoted and frozen in sufficient quantities for later use.

(11) Construction of MC38-Cas9 cell line

The lenti-EF1α-Cas9-2A-blast vector is used to package lentivirus. Since the vector is large and the virus titer is low, the virus needs to be centrifuged and concentrated before infecting the MC38 cell line. 48 hours after infection, add 10 μg/mL blasticidin for positive screening for infection. During the same screening process, dead cells and debris require timely replacement of the culture medium or passage. After about 10 days of screening, the MC38-Cas9 cell line can be sufficiently amplified and can be aliquoted and frozen for later use.

(12) Construction of MC38-B2m KO, Gal-1 KO, B2m/Gal-1 DKO cell lines

pKLV-U6 sgB2m-PGK-PURO-2A-BFP, pKLV-U6 sgGal-1-PGK-PURO-2A-BFP, pKLV-U6sgB2m/sgGal-1-PGK-PURO-2A-BFP and pKLV-U6 sgEmpty-PGK -PURO-2A-BFP (without targeting sgRNA empty as a control), a total of 4 vectors package lentivirus respectively, and the virus liquid can directly infect the MC38-Cas9 cell line without concentration. 48 hours after infection, add 5 μg/mL puromycin and 10 μg/mL blasticidin for positive screening for infection. During the same screening process, the production of dead cells and debris requires timely replacement of the culture medium or passage. After about 7 days of screening, each knockout cell line and control cell line can be sufficiently amplified, and sufficient amounts can be aliquoted and frozen for later use.

(13) Construction of MC38-Gal-1-EGFP and MC38-IRES-EGFP cell lines

Two vectors, lenti-U6 sgRNA-EF1α-Gal-1-EGFP fusion and lenti-U6 sgRNA-EF1α-IRES-GFP (control), were used to package the lentivirus respectively. The virus liquid was directly infected with the MC38-Cas9 cell line without concentration. 48 hours after infection, GFP-positive cells were sorted using flow cytometry. After about 7 days of sorting, Gal-1 expressing cell lines and control cell lines can be sufficiently amplified, and sufficient amounts can be aliquoted and frozen for later use.

(14) Activation and adoptive transfer of memory T cells

Days 2 to 7 are the differentiation and culture stage of memory T cells. If the infected virus contains puromycin resistance, puromycin will be added for screening starting from 48 hours after infection (day 3) at a concentration of 3 μg/mL; If the infected virus does not contain puromycin resistance but expresses fluorescent proteins, flow cytometry can be used to perform fluorescent labeling sorting at 48 hours after infection (day 3) or at a later time. After 4 days of screening of infected positive cells (day 7), T cells can be co-cultured with tumor cells presenting OVA oligopeptide (no need to add puromycin), that is, activation of memory T cells, in which T cells and tumor cells The ratio is 3:1 or other specified ratio. Before the adoptive transfer of T cells, CFSE was used to label T cells to distinguish them from T cells in mice. The labeled cells were injected into B16F10-OVA tumor-bearing mice through the tail vein (on the 14th day of tumor-bearing). Mice were implanted with 2 × 10 ⁶ cells.

(15) In vitro killing experiment: OT-IT cells target tumor cells

B16F10-OVA cells and B16F10 cells were labeled with CFSE ^hi (2.5mM) and CFSE ^lo (50nM) respectively, and then co-cultured with T cells in a 96-well plate at the specified ratio. Samples containing only tumor cells and no T cells served as controls. After 24 hours of culture, flow cytometry was used to detect the ratio of CFSE ^hi and CFSE ^lo populations. Specific killing rate calculation formula: [1-(Sample containing T cells CFSE ^hi /CFSE ^lo )/(Sample containing only tumor cells CFSE ^hi /CFSE ^lo )]×100%. For the MC38 cell killing experiment, before collecting the cells, add OVA oligopeptide to some MC38 cells to a final concentration of 100ng/mL and incubate at 37°C for 3 hours. The subsequent steps are basically the same as the B16F10 cell killing experiment. MC38 cells that have been incubated with OVA oligopeptide, namely OVApulsedMC38, and MC38 cells that have not been incubated with OVA oligopeptide are marked as CFSE ^hi and CFSE ^lo respectively, and then mixed with T cells in proportion. Perform co-culture.

(16) Killing experiment of OT-Ⅰ T cells treated with lactose

Prepare conditioned medium: 50% cytokine-free fresh T cell medium and 50% MC38 conditioned medium. Use the above conditioned medium to prepare a medium containing 0mM and 20mM lactose, and then use the prepared medium for the killing experiment of OT-ⅠT cells targeting tumor cells.

(17) Construction of antigen-specific T cells: Construction of Anti-CD19-CART cells

CD8 ⁺ T cells were purified from the spleens of Cas9-EGFP knock-in mice using a mouse CD8 ⁺ T cell isolation kit.

On day 0, CD8 ⁺ T cells were stimulated with anti-CD3 (1 μg/mL) and anti-CD28 (0.5 μg/mL) for 24 hours in RPMI 1640 complete medium containing 20 ng/mL IL-2. Day 1 Activated CD8 ⁺ T cells were enriched through Percoll isolation solution, and the enriched T cells were centrifugally infected with retrovirus carrying anti-CD19-CAR and sgRNA expression units in a 24-well plate. Add 8 μg/mL Polybrene, 2000g Centrifuge at 30°C for 1 hour, with the rising acceleration as 1 and the falling acceleration as 0. On the 3rd day, the infection-positive (RFP ⁺ ) cells were flow cytometrically sorted and further cultured. The culture conditions were the same as the aforementioned memory T cells. Day 7 T cells expressing anti-CD19-CAR were mixed with Nalm6 cells to perform in vitro killing experiments.

(18) Construction of antigen-specific T cells: Construction of NY-ESO-1 specific TCRT cells

The construction process of NY-ESO-1-specific TCR T cells refers to the isolation, culture and infection methods of human primary T cells. Three days after infection, the infection-positive (BFP ⁺ ) cells were flow sorted and cultured and expanded. Seven days after transduction, T cells expressing NY-ESO-1-specific TCR were mixed with A375 cells for in vitro killing experiments.

(19) Construction of antigen-specific T cells: Construction of Biggest clone TCR T cells

The Biggest clone TCR T cell construction process is consistent with the anti-CD19-CAR T cell construction process. During infection, enriched T cells are centrifugally infected with retrovirus carrying the largest clone TCR and sgRNA expression unit. On the third day, the infection-positive (GFP ⁺ ) cells were flow sorted and further cultured. The culture conditions were the same as the aforementioned memory T cells. On the 7th day, T cells expressing the largest clone TCR were mixed with MC38 cells for in vitro killing experiments.

(20) Establishing a mouse subcutaneous tumor model

Collect the cells, wash them twice with DPBS to remove residual culture medium, filter the cells with a 40 μm filter, and aliquot according to the required number of cells.

Table 1 Number of cells implanted in mouse subcutaneous tumors

The hair on one side of the mouse's back was shaved one day before tumor transplantation. When transplanting tumors, the cell suspension needs to be mixed before aspirating. Use a 1mL syringe to absorb 100μL of cell suspension. Pick up the skin on the hairless part of the back and insert it 0.5cm for injection. A small bag can be seen under the skin. This needs to be done before pulling out the needle. Rotate the pillow half a circle to prevent cell fluid from leaking out. About 7 days after tumor transplantation, use electronic vernier calipers to measure the length and width of the tumor every 2 to 3 days. Tumor volume calculation formula: 1/2×length×width×width.

One day after tumor transplantation, prepare a 250mM lactose solution with DPBS and filter it with a 0.22μm filter. Each tumor-bearing mouse was injected with 200 μL of lactose solution through the tail vein using a 1mL syringe, and the control group was injected with 200 μL of LDPBS once every two days until the 21st day. For LacNAc treatment, LacNAc (20 mM, 200 μL) in DPBS was injected intravenously every other day, and the control group was injected with 200 μL DPBS until day 21.

(21) Elimination of NK cells, CD4+T cells, and CD8+T cells in mice

Depletion of NK cells: Dilute anti-NK1.1 antibody with DPBS to a final concentration of 100μg/200μL. Taking the day of tumor transplantation as day 0, mice were intraperitoneally injected with 100 μg of diluted antibody on days -3, -1, 6, and 8. Depletion of CD4 ⁺ T cells: Dilute anti-CD4 antibody with DPBS to a final concentration of 400μg/200μL. Taking the day of tumor transplantation as day 0, mice were intraperitoneally injected with 400 μg of diluted antibody on days -2, -1, 6, and 7. Depletion of CD8 ⁺ T cells: Dilute anti-CD8a antibody with DPBS to a final concentration of 250μg/200μL. Taking the day of tumor transplantation as day 0, mice were intraperitoneally injected with 250 μg of diluted antibody on days -2, -1, 6, and 7.

(22) Purification of recombinant Gal-1 protein:

Cloning expression test: Transform the prokaryotic expression plasmid pQE-80-Gal-1opti into the BL21 competent state, and spread it evenly on an agarose plate containing ampicillin. Pick 5 single clones and put them into LB medium containing ampicillin, and culture them overnight on a shaking table at 37°C. The bacterial solution was inoculated into LB medium containing ampicillin at a ratio of 1:100 and cultured on a shaking table at 37°C. In about 2 hours, detect the 600nm OD value of the bacterial solution, which is about 0.7 to 0.8, and then stop culturing. Cool down on ice for 1 to 2 minutes, then add IPTG to a final concentration of 0.25mM, and incubate in a shaker at 18°C for 17 hours. After induction of expression, the cells were collected, washed with DPBS, resuspended in E. coli lysis buffer and disrupted by sonication. Take a small amount of lysed bacterial solution, add 5×SDS loading buffer, cook at 100°C for 10 minutes, and then perform protein electrophoresis.

Purification of recombinant Gal-1 protein: The selected monoclonal strain is expanded and cultured to 500mL. The activation and induction expression steps are the same as those for cloning expression test. After inducing expression for 17 hours, centrifuge at 9600g for 15 minutes at 4°C to collect the cells. From this step onwards, operate on ice. (1) Wash once with pre-cooled DPBS, add 80 mL of E. coli lysate and resuspend, conduct ultrasonic disruption, turn on for 15 seconds, off for 15 seconds, and sonicate for half an hour to one hour until the bacterial solution becomes clear. Be sure to cool down in a mixture of ice and water during this period. Transfer the lysed bacterial solution to an ultracentrifuge tube and centrifuge at 21600g for 20 minutes at 4°C. During centrifugation, wash 2mL Ni-NTA twice with lysis buffer. Low-speed centrifugation is required to prevent damage to the nickel-labeled agarose beads. Mix the supernatant of the lysed bacterial solution and Ni-NTA, and incubate for 2 hours at 4°C with slow inversion and mixing. Transfer the incubated Ni-NTA to the chromatography column. When the liquid is almost gone, wash it with ten times the column volume of nickel column cleaning solution. After the liquid has drained out, add 10 mL of nickel column eluent. During this period, use a 1.5 mL centrifuge tube to receive the eluate, 400 to 500 μL per tube. Using Nanodrop, detect protein concentration in each tube sequentially. The liquids in the centrifuge tubes containing higher concentrations of protein are collected together and put into dialysis bags. Dialysis was performed three times, the first two were two hours each, and the third was overnight. Dispense the dialyzed protein solution into 1.5 mL centrifuge tubes, determine the concentration using the Bradford protein assay, and store at 4°C for later use or frozen at -80°C.

Recombinant Gal-1 protein activity test: Collect mouse primary T cells cultured in vitro and aliquot them into 1.5 mL centrifuge tubes, with 5 × 10 ⁵ cells in each tube. Add the recombinant Gal-1 protein to be detected, mix evenly, and incubate at room temperature for 10 minutes. Wash twice with DPBS. Anti-Gal-1 staining was performed according to conventional cell surface antigen and antibody staining methods. For flow cytometry detection, the first indicator is the staining intensity of anti-Gal-1, and the second indicator is the degree of T cell clumping displayed by FSC or SSC.

(23)Mass spectrometry and analysis

The mass spectrometer was an LTQ ORBITRAP Velos mass spectrometer (Thermo Fisher Scientific, San Jose, CA, USA) equipped with a nano-ESI ion source. The mass spectrometer was operated in data-dependent mode, with one MS scan per cycle followed by ten HCD (High Energy Collision Dissociation) MS/MS scans. Use Proteome Discoverer (version1.4, URL: https://www.thermofisher.com/hk/en/home/industrial/mass-spectrometry/liquid-chromatography-mass-spectr ometry-lc-ms/lc-ms-software /multi-omics-data-analysis/proteome-discoverer-software.html) processed the original data to obtain the original intensity table of peptide spectrum matches (PSM) of all identified proteins, and used the R package DEqMS for further analysis. The raw intensity values were log2 transformed and the median log2 intensity was subtracted from each PSM to obtain the relative log2 ratio. The protein differential expression was then calculated using the Bayesian algorithm. Differentially expressed proteins were screened according to P＜0.05 standard. These proteins were subjected to functional enrichment analysis using the R package ClusterProfiler (version 3.12.0).

(24)Gal-1 transfer assay

Gal-1 transfer experiments were performed by co-culture of T cells and MC38 cells. OT-ⅠT cells were compared with wild-type or Gal-1 knockout MC38 cells in conditioned medium (50% fresh T cell medium without cytokines and 50% wild-type or Gal-1 knockout MC38 cells conditioned medium ) in 2:1 or specified ratio for 8 hours, and then the level of Gal-1 on the cell surface was detected by flow cytometry. Gal-1 transfer experiments were performed in Boyden chamber. OT-IT cells were seeded in the upper compartment of Boyden chamber, MC38 cells were seeded in the lower compartment, or OT-IT cells and MC38 cells were seeded in the lower compartment together, and the cell surface was detected by flow cytometry after 8 hours of culture. Gal-1 levels. Gal-1 transfer experiments were performed in low-melting agarose gels. Inoculate MC38 cells in the well plate in advance, add low melting point agarose dissolved in the culture medium after they are attached, and after solidification, add OT-IT cells to the agarose layer, or add them directly to the agarose layer without covering it. On MC38 cells, the level of Gal-1 on the cell surface was detected by flow cytometry after 8 hours of culture.

(25) Fluorescence resonance energy transfer (FRET)

(1) Recombinant Gal-1 treatment: Treat 1.2×10 ⁶ wild-type or B4galt1 knockout memory T cells with 2.5 μg recombinant Gal-1 in 100 μL DPBS, incubate at room temperature for 10 minutes, and then wash twice with DPBS.

(2) Lactose treatment: 20mM lactose was added during recombinant Gal-1 treatment while incubating.

(3) Antibody incubation: After treatment with recombinant Gal-1 and lactose, the T cells were incubated with PE-anti-Vα2 and APC-anti-CD8α antibodies for 30 minutes at 4°C. Each sample needs to be set up with four groups of staining, including one group for each of the two antibodies, a group stained with two antibodies at the same time, and a group without antibodies.

(4) After staining, use cell fixative to fix at 4°C for 10 minutes, wash once and then use flow cytometer for detection.

(5) Detect the average fluorescence intensity value of PE, APC and FRET channels, and calculate the TCR-CD8 FRET unit through the formula.

(26) Mouse tumor model, drug treatment and adoptive transfer experiments

4 × 10 ⁵ B16F10-OVA cells were subcutaneously injected into female C57BL/6J mice; 7 days after injection, tumor-bearing mice with similar tumor sizes were randomly divided into two groups, and then 2 × 10 cells were injected through the tail vein. ^6. OT-1T cells transfected with B4GALT1sgRNA or control RNA; then use a vernier caliper to measure the tumor size every two days; tumor volume calculation method: width × width × length × 1/2;

Lactose treatment: MC38 (1×10 ⁶ cells), B16F1O (4×10 ⁵ cells), B16F10-OVA (4×10 ⁵ cells), B16F10-Control (4×10 ⁵ cells), 4T1 (4×10 ⁵ cells) ), CT26 (4×10 ⁵ cells) and SGC7901 (5×10 ⁶ cells) cells were injected subcutaneously into female C57BL/6J or NPG mice; unless otherwise specified, starting from the day after tumor transplantation, For each tumor-bearing mouse, a dose of 200 μl of 250 mM lactose solution dissolved in PBS was injected through the tail vein every other day (injection to day 21);

Combination therapy: On days 6, 9, and 12 after tumor implantation, intraperitoneally inject 200 μl of anti-PD1 (CloneRMP1-14, BioXCell, BE0146) or PDL1 (Clone10 F.9G2, BioXCell, BioXCell, BE0101) Antibody;

Depletion of CD8 ⁺ T cells, CD4 + T cells and NK cells: 3 days and 1 day before tumor transplantation, intraperitoneally inject 200 μl of NK antibody diluted with PBS for a total amount of 100 μg; 2 days and 1 day before tumor transplantation On the first day, 200 μl of CD8 antibody (Clone2.43, BioXCell, BE0061) diluted in PBS was intraperitoneally injected to a total amount of 250 μg; 2 days and 1 day before tumor transplantation, 200 μl of CD4 diluted in PBS to a total amount of 400 μg was injected intraperitoneally. Antibody (CloneGK1.5, BioXCell, BE0003-1);

MC38 secondary inoculation tumor model: 1×10 ⁶ MC38-GFP cells or MC38-LALBA overexpressing cells were subcutaneously injected into female C57BL/6J mice; 21 days after tumor transplantation, the MC38-LALBA tumors completely disappeared Mice were inoculated twice with 1 × 10 ⁶ MC38-GFP cells; the control group was female C57BL/6J mice of the same age that had not been inoculated with MC38-GFP cells;

MC38-GFP/MC38-LALBA mixed tumor model: MC38-GFP and MC38-LALBA cells were mixed at a ratio of 9:1 or 1:1, and then injected subcutaneously into female C57BL/6J mice with a total implantation volume of 1× 10 ⁶ cells; the control group was mice of the same age injected subcutaneously with 1×10 ⁶ MC38-GFP cells;

(27)Humanized mice

NPG mice aged 3-4 weeks were irradiated with a semi-lethal dose of 120 cGy, and then human CD34+ umbilical cord blood cells were injected through the tail vein; 12 weeks after transplantation, human CD45+ cells in peripheral blood were detected to determine the artificial hematopoietic system. Reconstruction; 13 weeks after hematopoietic stem cell implantation, 5×10 ⁶ SGC7901 cells were subcutaneously implanted into mice. Starting from the second day after tumor implantation, each tumor-bearing mouse was injected through the tail vein every other day. One dose of 200μl 250mM lactose solution dissolved in PBS, the control group was injected with an equal volume of PBS buffer; 10 days after tumor transplantation, two doses of anti-human PD1 antibody (Pembrolizumab, Selleck, Cat#A2005) (10mg/kg) were injected weekly; Tumor size was measured every 2-3 days.

(28) Culture, infection and adoptive transfer of memory T cells in vitro

On day 0, spleen cells were isolated from 6-8 week old Cas9-EGFP/OT-1 female mice and cultured in RPMI complete medium containing IL2 (10ng/ml) and SIINFEKL peptide (10ng/ml). RPMI 1640, 10% FBS, 20mMHEPES, 1mM sodium pyruvate, 100μg/ml penicillin) for 24 hours; on day 1, activated Cas9-EGFP/OT-1T cells were enriched by Percoll density gradient centrifugation. Then add cell suspension, retrovirus and 8μg/ml polybrene to the 24-well plate and centrifuge the infection (2,000g, 30℃, 1h, the acceleration of rising and falling speed is minimum); after centrifugation, keep the culture plate at a constant temperature of 37℃ Cultivate in a CO ₂ incubator for 5 hours, and then change the medium to RPMI complete medium containing IL2 (2ng/ml), IL7 (2.5ng/ml) and IL15 (10ng/ml). The cell density is 3×10 ⁵ / ml; two days after infection (day 3), select with 3 μg/ml puromycin for 4 days; on day 7, cells are ready for co-culture or adoptive transfer experiments.

(29) Isolation of tumor-infiltrating lymphocytes (TIL)

Add 5 ml of RPMI medium containing 2% FBS and 50 U/ml type IV collagenase (Invitrogen, V900893) to a six-well plate, and then cut the B16F10-OVA, MC38, and SGC7901 tumors into small pieces in the above solution and incubate at 37°C. Incubate for 1 hour; filter the digested suspension with a 70 μm filter membrane, wash it three times with PBS, and then use it for antibody staining and flow cytometric analysis.

(30) Infiltration and proliferation of T cells in tumors

After 4 days of puromycin selection, the infected OT-1T cells were labeled with CFSE (carboxyfluorescein diacetate succinimidylester, Invitrogen), and then 2 × 10 ⁶ labeled cells were injected into the tail vein of mice bearing tumors (B16F10-OVA) for 14 days; The intensity of CFSE was detected by flow cytometry at 24 h and 6 days after implantation.

(31) Retroviral whole genome CRISPR/Cas9 library construction

The whole genome CRISPR knockout library (1000000096) was purchased from Addgene; the primer pair F: 5'-GGCTTTATATATCTTGTGGAAAGGACGAAACACCG-3' (SEQ ID NO: 3) and R: 5'-CTAGCCTTATTTTAACTTGCTATTTCTAGCTCTAAAAC-3' (SEQ ID NO: 4) The gRNA region was amplified by PCR and constructed into the MSCV-gRNA-PGK-PURO-2A-BFP vector through Gibson reaction.

(32)Construction of customized mini libraries of retroviruses

Based on the genome-wide knockout screening results, a total of 1398 genes were screened; based on the performance of gRNA, an average of 3 gRNAs were selected from the initial library for each gene, and a total of 105 intergenic control gRNAs were added; oligonucleotides containing gRNA sequences were synthesized acid (customized sequence chip), amplified by PCR, and cloned into the MSCV-gRNA-PGK-PURO-2A-BFP vector through Gibson reaction.

(33)CRISPR/Cas9 screening

For in vitro PD1 expression screening, on the 6th day of OT-1T-cell culture, 3x10 ⁶ B16F10-OVA cells were spread in a 15cm culture dish containing DMEM (DMEM+10% FBS+Pen/Strep); on the 7th day On the next day, the puromycin-selected OT-1T cells were resuspended in complete RPMI 1640 medium containing IL2 (2ng/ml), IL7 (2.5ng/ml) and IL15 (10ng/ml) to a final concentration of 6x10 per ml. ⁵ cells were added to B16F10-OVA cells at a T cell:B16F10-OVA cell ratio of 4:1; cells were co-cultured at 37°C overnight; CD8 ⁺ T cells were collected and used with PE anti-PD1 in 2% FBS/PBS Stain on ice for 30 minutes; BD FACSAria sorting cells with the highest and lowest 5% PD1+ expression intensity;

For in vivo screening of custom secondary minilibraries, save ^2x10 infected OT-1T cells as starting input sample (approximately 269X cell coverage per sgRNA) after 4 days of puromycin screening; library Infected OT-1 T cells ( ^2x10 cells per recipient mouse) were transferred intravenously into B16F10-OVA tumor-bearing mice; a total of 24 mice were used as recipients; 7 days after transplantation, B16F10-OVA tumors were digested into single-cell suspensions, and biotin anti-mouse CD8a (Biolegend, 100704) and streptavidin magnetic beads were used to isolate the infiltrating CD8 ⁺ T cells in the tumors; at the same time, single cells from each tumor were removed 1/20 volume of the suspension was stained for OT-1 to estimate the number of infiltrating OT-1T cells in each tumor; a total of 1x104 to 1x10 ⁵ OT-1T cells were collected from each tumor.

(34) Sequencing library preparation

Genomic DNA was extracted using phenol/chloroform extraction method; Titanium Taq DNA polymerase (Clontech, 639242) was used to perform the first round of PCR to amplify the sgRNA region; the second round of PCR added adapter sequences and sequencing tags to each sample; finally, Nova- seq150-bp paired-end sequencing (Illumina).

(35) Screening result analysis

The raw data were preprocessed using sequence sorting tools, and the tool Cutadapt (version 3.4) was used to remove adapter sequences and the tool FLASH (version 1.2.11) was used to merge paired-end sequencing sequences; first, MAGeCK (version 0.5.9.5) was used to perform CRISPR/Cas9 To filter the data for analysis, the MAGeCK "count" command generates gRNA counts for the samples and merges their raw reads into a count matrix; next, use the default settings of the MAGeCK "test" command (available from https://sourceforge.net/projects /mageck/obtained) to identify the top forward or reverse gRNAs or genes; gene set enrichment analysis was performed using the default parameters of the GSEA() function of the R language clusterProfiler package (version 3.12.0); KEGG The database is available in the R language package msigdbr package (version

7.5.1)(https://igordot.github.io/msigdbr/) found in the "C2" category.

(36)Flow cytometry

For cell surface antigen staining, cells were stained in 2% FBS/PBS on ice for 30 min; intracellular staining was performed using a fixation/permeabilization kit (BD Biosciences) according to the manufacturer's instructions; the following antibodies were used: APC anti-mouse CD8a (Invitrogen, 17-0081-83), PE anti-mouse CD279 (PD-1) (BioLegend, 109104), APC anti-mouse CD279 (PD-1) (BioLegend, 109112), FITC anti-mouse IFNγ (BioLegend , 505806), APC anti-mouse β2-microglobulin (BioLegend, 154506), APC anti-mouse H2Db (BioLegend, 111513), APC anti-mouse OVA257-264 (SIINFEKL) (Invitrogen, 17-5743-82), FITC anti-mouse CD3 (BioLegend, 100204), PE-streptavidin (peolegend, 405203), biotinylated Erythrina lectin (B-1145-5), biotinylated succinyl wheat germ agglutinin (B -1025S-5), APC anti-human CD8a (BioLegend, 300912), PE anti-human IFNγ (BioLegend, 506507), FITC anti-mouse TCRVα2 (BioLegend, 127806); use BD FACSAria instrument for flow cytometry experiments, and FlowJo to analyze the data .

(37)Enzyme-linked immunosorbent assay

OT-1T cells (3x10 ⁵ /ml) and B16F10-OVA cells were co-cultured in a medium containing IL2 (2ng/ml), IL7 (2.5ng/ml) and IL15 (10ng/ml) at 37°C for 8 hours; Collect the supernatant after co-culture; ELISA kit detects TNFα and IFNγ in the co-culture supernatant (ABclona, Cat#RK00027, Cat#RK00019); each sample has two replicates.

(38) In vitro killing experiment

For the in vitro killing experiment of OT-1T cells, B16F10-OVA cells and B16F10 cells were labeled with CFSEhi (2.5mM) and CFSElo (50nM) respectively, and then co-cultured with OT-1T cells in a 96-well plate at the specified ratio; the control group was not Add OT-1T cells to tumor cells; after 24 hours of incubation, use flow cytometry to detect the ratio of two groups of cells, CFSEhi and CFSElo;

For MC38 in vitro killing experiments, MC38 cells were treated with 100ng/ml Ovapeptide (SIINFEKL) at 37°C for 3 hours, and then co-cultured with OT-1T cells;

The specific killing rate calculation formula is [1-(CFSEhi/CFSElo containing T cells)/(CFSEhi/CFSElo containing only tumor cells)] × 100%.

(39) Lactose quantification using liquid chromatography-mass spectrometry (LC-MS)

Cell sample preparation: Collect cells and wash them twice with PBS; add 400 μl of methanol/acetonitrile mixed solution (methanol: acetonitrile: water = 4:4:2) for every 2×10 ⁶ cells, mix thoroughly, and place at -80°C for 1 hour , then centrifuge at 14,000g for 15 minutes at 4°C; transfer the centrifuged supernatant to a new tube, repeat the above steps, and the final supernatant can be used for mass spectrometry quantitative analysis.

Cell culture medium sample preparation: collect the cell supernatant and filter it with a 0.45 μm filter to remove cell debris, then add 2.5 times the volume of methanol, mix thoroughly for 2 minutes and centrifuge at 13,000 rpm for 15 minutes; the final supernatant can be used for mass spectrometry quantification analyze.

The 100mM lactose solution was diluted with 50% methanol aqueous solution to 100, 200, 500, 1000, 2000, 5000, 10000nM and used as a standard; the standard solution and the sample prepared above were transferred to an LC sample bottle for analysis; LC -MS analysis was performed on a ThermoVanquish UHPLC equipped with a Thermo Q Exactive HF-X hybrid quadrupole-Orbitrap mass spectrometer in negative ESI mode; using a WatersAcquity UPLC BEHAmide column (1.7 μm, 2.1 × 100 mm) at 35°C column temperature , separation was achieved under isocratic elution of 30% mobile phase A (5mM ammonium formate aqueous solution) and 70% mobile phase B (acetonitrile); the flow rate was 0.3ml/min, the injection volume was 10μl, and each injection run The time is 5min.

Full scan mass spectrometry acquisition was performed over the m/z 66.7 to 1000 range using the following ESI source settings: spray voltage: 2.5kV; auxiliary gas heater temperature: 380°C; capillary temperature: 320°C; sheath gas flow rate: 10 units; MS1 scan parameters included resolution 60000, AGCtarget3e6 and maximum injection time 200ms; Thermo Xcalibur software (version 4.2) was used for data processing and [M+FA]-adduct of lactose (m/z387.1138) for quantitation .

(40)RT-QPCR

use Reagent (Ambion, Cat#15596018) was used to extract RNA; the ImProm-II ^TM reverse transcriptase system (Promega, Cat#A3801) was used to synthesize cDNA, and 100ng RNA was used in each reaction; using /> qPCR reactions were performed using PremixExTaq ^™ (Takara, Cat#RR420A), using 1 μl cDNA per 20 ul reaction volume; relative gene expression levels were normalized with GAPDH.

(41) TCGA (Cancer Genome Map) data analysis

Download the transcriptome expression profiles and clinical data of all tumor types from the TCGA official website (https://gdcportal.nci.nih.gov/); in order to normalize the B4GALT1 gene expression by the CD8a gene, the present invention The value is divided by the expression value of CD8a in the corresponding sample sample; subsequent survival analysis and HR value calculation are implemented through the R language survival package (version 3.3.1) and survminer package (version 0.4.9), and the median gene expression amount is used to group the patients. Number was used as the grouping standard, and survival curves were compared using the two-sided Log-rank test and Kaplan-Meier curves were drawn.

(42)RNA-seq data analysis

Transcriptome paired-end 150bp sequencing was performed on the NovaSeq6000S4 platform; raw data were mapped to the mouse genome (mm10) using TopHat (version 2.1.1)5; gene reads were calculated using HTSeq (version 1.99.2) and cufflinks ( Version 2.2.1) for normalization calculation; use R language DESeq2 package (version 1.22.2) to calculate differentially expressed genes, and use P-value 0.05 or 0.01 as the screening criterion; use R language ClusterProfiler package (version 3.12.0 )enricher() function performs functional enrichment analysis on differentially expressed genes; characteristic gene sets are estimated using the average gene expression levels of genes in the corresponding samples; Supplementary Table 7 lists the complete list of genes involved in each characteristic gene set .

(43)Single cell RNA-seq sample preparation

Use the CD3-positive gene to screen MC38 tumor cells in three pairs of samples treated with PBS or lactose; mix the same number of CD3-positive cells from each sample and perform flow cytometry screening; use 5×10 ⁵ -1×10 ⁶ cells/ml CD3-positive cells were resuspended at a density of 85%, and the cell survival rate was 85%; single-cell libraries were prepared using 5'V(D)J and gene expression platforms according to the methods provided by the 10×Genomics official website, and then sequenced on the NovaSeq6000S4 platform.

(44)TCR data quality control and analysis

The original data were analyzed and processed using the latest mouse reference genome (refdata-gex-mm10-2020-A, 10×Genomics) and the default parameters of the CellRangervdj command (version 7.0.0); then the R language scRepertoire package (version 1.2.0 ) process the intermediate results and generate the output file (filtered_contig_annotations.csv). Finally, there are 14,851 positive cells defined by the present invention, of which 6,914 are from PBS samples and 7,937 are from LAC samples. The results of further analysis are shown in Figure 19.

(45) Quality control, filtering and analysis of single-cell RNA-seq data

Assembly analysis of raw data was performed using the latest mouse reference genome (refdata-gex-mm10-2020-A, 10×Genomics) and CellRanger command (version 7.0.0) default parameters; sequencing saturation was approximately 60%, per cell The sequencing depth is about 25,000 sequences; as described in previously published articles, the present invention will eliminate cells with abnormal gene numbers and abnormal mitochondrial gene counts; ultimately, 19,334 cells were retained from the two libraries, of which 13,839 cells have both T-cell receptor sequencing data and single-cell transcriptome sequencing data; R language Seurat package (version 2.3.4) was used for further analysis; for cell clustering results, the FindClusters() function was used at a resolution of 0.5 Cells were divided into 10 groups.

(46)Blood sample blood biochemical analysis

Blood samples (500 μl) were collected in EDTA-coated test tubes; a portion of the blood was taken for evaluation of hematological parameters: white blood cells (WBC), lymphocytes (LYM), and intermediate cells were measured in a hematology analyzer (TEK-IIMINI hematology analyzer). (MID), granulocyte (GRA), red blood cell (RBC), hemoglobin (HGB), hematocrit (HCT), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red blood cells Distribution width-standard deviation (RDW-SD), red blood cell distribution-coefficient of variation (RDW-CV), platelets (PLT), procalcitonin (PCT), mean platelet volume (MPV), platelet distribution width (PDW); remaining The sample was centrifuged (1500g, room temperature for 10 minutes), and the supernatant was taken for biochemical parameter analysis: alanine aminotransferase (ALT), aspartate aminotransferase (AST), and alkaline phosphate were measured with a fully automatic clinical biochemical analyzer (BeckmanCoulterAU5800). enzyme (ALP), creatinine (CR), blood urea nitrogen (BUN), lactate dehydrogenase (LDH), creatine kinase (CK), glucose (GLU), inorganic phosphorus (Pi).

(47)Data analysis

All statistical analyzes were performed using R language version 4.1.0; significance analyzes were performed using two-tailed paired T tests to determine statistical significance (*P<0.05, **P<0.01; ***P<0.001; NS Not significant) and the function t.test(); two-factor analysis of variance is calculated using the aov() function; the average and standard error (SEM) are shown in the figures, where the standard error value is calculated using the function sem().

(48)Data acquisition

The original sequencing data used in this article has been saved to the National Genome Science Data Center, the project number is PRJCA010494, and the login website is: https://ngdc.cncb.ac.cn/search/? dbId=&q=PRJCA010494; transcriptome data and clinical data of all tumor types were downloaded from the National Cancer Institute and Human Genome Research Institute Data Center (https://gdc-portal.nci.nih.gov/).

Lactose treatment of tumor-bearing mice: α-lactose was purchased from Sigma-Aldrich (L3625-1KG, Lot#SLCD8654); preparation of 250mM α-lactose: Dissolve every 0.09g lactose powder with 1mL PBS (C14190500BT, LOT#8121493) until completely dissolved Then filter with a 0.22μm filter; MC38 (1×10 ⁶ cells), B16F1O (4×10 ⁵ cells), B16F10-OVA (4×10 ⁵ cells), B16F10-Control (4×10 ⁵ cells), 4T1 (4×10 ⁵ cells), CT26 (4×10 ⁵ cells) and SGC7901 (5×10 ⁶ cells) cells were subcutaneously injected into female C57BL/6J or NPG mice; starting from the day after tumor transplantation, For each tumor-bearing mouse, a dose of 200 μl of 250 mM lactose solution dissolved in PBS was injected through the tail vein every other day (injection to day 21).

Others not listed are conventional experimental methods.

Example 1: Identification of genes and pathways regulating PD1 expression and TCR activation in CD8 ⁺ T cells through in vitro and in vivo CRISPR/Cas9 screening

The inventors established an in vitro genome-wide CRISPR/Cas9 screening system to identify genes and pathways that regulate PD1 expression in mouse primary CD8 ⁺ T cells (Figure 1A); briefly, ovalbumin (Ova) was used in advance The peptide stimulates splenic CD8 ⁺ T cells from Cas9-EGFP/OT-1 mice, and then uses retrovirus to infect the cells with a guide RNA (gRNA) library targeting the whole genome; after puromycin selection, the cells are treated with B16F10 -OVA cell co-culture to activate memory CD8 ⁺ T cells; flow cytometry was used to enrich cells with high PD1 expression and low PD1 expression; next-generation sequencing revealed these cell subpopulations as well as individual gRNAs in the starting cells in the whole genome library The distribution in Verification (Figure 1C); Gene set enrichment analysis (GSEA) identified several KEGG (Kyoto Encyclopedia of Genes and Genomes) pathways, which are significantly involved in regulating the expression of PD1 in CD8 ⁺ T cells (Figure 1D); known genes in the TCR activation pathway Genes (such as CD3d, Zap70 and Lat) were ranked high in the list of candidate genes for PD1 regulation; unexpectedly, the results showed genes involved in protein trafficking pathways (such as Srp14, Srp68, Sec16A) and genes involved in aminoacyl tRNA biosynthesis (such as Mars, Hars2, Eprs) were also significantly enriched in the PD1 low expression group; finally, the inventors screened and verified various components in N-glycan biosynthesis, including B4GALT1, Mgat2 and Dpm3, which can negatively Regulate the expression of PD1 (Figure 1C-D and Figure 7).

To high-throughput validate candidate genes and test their potential functions in the tumor microenvironment (Figure 1A), the inventors synthesized a custom gRNA library containing 4617 gRNAs targeting preferred candidate genes derived from a genome-wide screen and 105 intergenic control gRNA; gRNA library-infected Cas9-EGFP/OT-1 memory CD8 ⁺ T cells were reactivated in vitro or transplanted into wild-type C57BL/6J mice subcutaneously inoculated with B16F10-OVA tumors. In vivo screening for genes that regulate CD8 ⁺ T cell function; T cells were collected from tumors after 7 days for gRNA sequencing; gRNAs that inactivate Socs1, Regnase-1 (Zc3h12a), Rc3h1, and CD36 caused B16F10-OVA tumors in syngeneic mice The number of infiltrating OT-1 T cells increased (Figure 1F); Figure 1E-F shows that inactivation of B4GALT1 encoding β-1,4-galactosyltransferase 1 showed significant phenotypes in both in vitro and in vivo screens.

Example 2: Knocking out B4GALT1 in CD8 ⁺ T cells can activate TCR signaling and enhance T cell-mediated tumor immunotherapy

The cell surface PD1 protein and PD1 mRNA levels of Cas9-EGFP/OT-1CD8 ⁺ T cells infected with different gRNAs targeting B4GALT1 were significantly increased before and after co-culture with B16F10-OVA cells (Figure 2A and Figure 7C); these The phenotype could be complemented by overexpressing the short or long isoform cDNA of mouse B4GALT1 (Fig. 2B), indicating that the response to inhibition of TCR activation is not due to B4GALT1 ligand-induced signaling but rather to its biosynthetic function. ; In addition, after co-culture with B16F10-OVA, B4GALT1 knockout OT-1T cells showed enhanced expression of T cell activation and cytotoxicity markers IFN, IL2 and TNF, as well as enhanced target cell killing activity in vitro (Figure 2C- D); Whole-genome RNA sequencing analysis confirmed that B4GALT1 knockout CD8 ⁺ T cells had enhanced TCR activation after co-culture with B16F10-OVA cells (Figure 2E-G); CD8 ⁺ T cells infected with control gRNA and targeting B4GALT1 gRNA Gene set enrichment analysis (GSEA) of differentially expressed genes (DEGs) showed that the TCR signaling pathway was at the forefront of significantly altered pathways (Figure 2G).

B16F10-OVA cells were inoculated subcutaneously into wild-type mice, and B4GALT1gRNA-infected OT-1T cells had significantly higher tumor killing activity than control gRNA-infected cells (Figure 2H-I); tumor-infiltrating lymphocytes (TILs) analysis showed that , there were more infiltrating OT-1 T cells in tumors when infected with B4GALT1gRNA (Fig. 2J); these results suggest that the inhibitory protein N-glycome can enhance the function of CD8 ⁺ T cells and that B4GALT1 may be a potential target to enhance CAR-T cell activity. point.

In B16F10-OVA tumors, the number of infiltration of B4GALT1gRNA-infected OT-1T cells 24 hours after intravenous injection was similar to that of control gRNA-infected cells (Figure 8A), indicating that B4GALT1 has no significant effect on the infiltration of OT-1T cells; another On the other hand, CFSE (carboxyfluorescein succinimide ester) analysis at a later time point (6 days after injection) showed that B4GALT1 inactivation enhanced intratumoral OT-1 T cell proliferation (Fig. 8B).

Example 3: Experiment on the correlation between the expression of B4GALT1 in the tumor microenvironment and tumor-infiltrating CD8 ⁺ T cells and the prognosis of human patients

To investigate the potential clinical relevance of B4GALT1 in human cancer patients, the inventors analyzed cancer samples in The Cancer Genome Atlas. Although the expression level of B4GALT1 was not associated with an overall survival benefit in all primary tumor samples collected by TCGA, after normalizing the expression level of CD8a, Kaplan-Meier curves showed that patients with low B4GALT1 expression had significantly longer survival (Figure 3A and Figure 3 9A); On the other hand, the inventor first divided all tumor samples into B4GALT1 low expression and B4GALT1 high expression groups; high expression of CD8a in the B4GALT1 low expression group was positively correlated with longer overall survival, while the B4GALT1 high expression group was There was no such correlation (Figure 3B and Figure 9C); in the data sets of adrenocortical carcinoma (ACC), acute myeloid leukemia (LAML), lung adenocarcinoma (LUAD), and rectal adenocarcinoma (READ), with the CD8a low expression group Compared with the patients in the CD8a high expression group, it can be seen that the low expression level of B4GALT1 is positively correlated with longer overall survival (Figure 3C and Figure 9B); on the contrary, compared with the patients in the B4GALT1 high expression group, Among patients in the B4GALT1 low expression group, higher CD8a expression levels were positively correlated with better overall survival (Figure 3D and Figure 9C); the results showed that the expression level of B4GALT1 in the tumor microenvironment and the CD8 infiltration in the tumor ⁺ T cells are associated with prognosis in cancer patients.

Example 4: Inhibiting the activity of B4GALT1 in tumor cells can enhance the IFNγ signaling pathway and its immune surveillance response

The inventor believes that targeting the same gene or pathway in immune cells and tumor cells can work synergistically to enhance the effect of tumor immunotherapy; in addition, the expression of B4GALT1 in tumors mainly depends on malignant tumor cells rather than CD8 ⁺ T cells; unexpectedly Unexpectedly, when the inventors used the CRISPR/Cas9 system to conduct in vivo screening experiments in MC38 tumors for genes and pathways related to the resistance and sensitivity of tumor cells to immune surveillance, the inventors found that B4GALT1 could be screened out. ; In wild-type C57BL/6J mice, compared with control cells, the growth rate of subcutaneously implanted MC38 tumors with B4GALT1 knockout was significantly reduced, while in immunodeficient NPG (NOD-Prkdcscid IL2rgnull) mice, the growth rate was significantly reduced. There is no obvious difference (Figure 4A-B); if specific antibodies are used to eliminate different immune cells in wild-type mice and then implanted subcutaneously, the inventors found that depleting CD8 ⁺ and CD4+ T cells can significantly restore B4GALT1-deficient MC38 tumor growth rate, but depletion of natural killer (NK) cells did not restore its growth rate (Figure 4C, Figure 10); the above results indicate that in wild-type mice, CD8 ⁺ T cells play an important role in controlling B4GALT1 knockout MC38 tumors. plays a major role in the growth process.

The inventors performed RNA sequencing analysis on B4GALT1-knocked-out MC38 cells and control cells. The results showed that the IFNα, IFNγ, and TNFα signaling pathways were all significantly up-regulated (Figure 4D-F); after IFNγ treatment, compared with control cells, B4GALT1 In knockout MC38 cells, B2m expression level and Ova antigen presentation were significantly enhanced (Figure 4G-H); at the same time, knockout of the B4GALT1 gene in MC38 cells can enhance the specific in vitro killing function of OT-1T cells (Figure 4I) ; Notably, IFNγ expression levels in OT-1T cells were significantly increased when co-cultured with B4GALT1 knockout MC38 cells compared with co-culture with control MC38 cells (Figure 4J).

Example 5: Exogenous expression of α-lactalbumin (LALBA) can produce lactose in tumor cells and activate tumor immune response

In addition to its binding function with N-glycosyl groups, B4GALT1 can also form heterodimers with LALBA in breast tissue to synthesize lactose; the inventors hypothesized that exogenous overexpression of LALBA in tumor cells can change the expression of endogenous B4GALT1 in breast tissue. catalytic activity in lactose biosynthesis; therefore, LALBA may be able to competitively reduce the normal function of B4GALT1 in tumor cells; in addition, de novo synthesis of lactose can also inhibit the activity of B4GALT1 and interfere with tumor cells and intratumoral CD8 ⁺ T cells. N-glycosylation function; Lactose is a disaccharide connected by galactose and glucose, which can simulate galactosylated oligosaccharides or proteins; in vitro experiments show that lactose can compete and interfere with glycoproteins and glycolipids. Functions of β-galactopyranoside, such as interaction with lectins.

The inventor constructed two LALBA overexpression vectors (LALBAOE), in which the LALBA gene was selected from the cDNA region of the sequence with ID number 16770 in Genebank: 1918-2349. The structure of the constructed lentiviral vector is shown in Figure 22. Its base The base sequence is shown in SEQ ID NO: 1. The structure of the constructed adenoviral vector overexpressing LALBA is shown in Figure 23, and its base sequence is shown in SEQ ID NO: 2.

Lentivirus is packaged by co-transfection of HEK-293T cells with lentiviral expression vector, psPAX2 and pMD2.G; 48 hours after transfection, the cell supernatant was collected and filtered with 0.45 μm to remove cell debris; the virus can be used directly infection, it can also be concentrated by high-speed centrifugation (centrifuged at 107,000g for 2.5 hours at 4°C) and then resuspended in PBS and used for in vivo experiments; at the same time, the following controls were also set: lentivirus carrying site-directed mutation of D107A or A126K on the LALBA gene , the site-directed mutation of A126K is A126K:GCC→AAG; the site-directed mutation of D107A is D107A:GAT→GCC.

Adeno-associated virus type 8 (AAV) is prepared using an AAV viral vector expressing lentiviral expression vector (LALBA-IRES-GFP) or GFP; first, HEK-293T is co-transfected with the AAV vector, packaging plasmid and adeno-associated virus helper plasmid After 72 hours, the culture medium and cells were collected for precipitation and lysis respectively, followed by density gradient centrifugation with iodixanol; finally, the purified virus was ultracentrifuged and diluted with PBS for subsequent experiments.

1×10 ⁶ MC38 cells were subcutaneously injected into female C57BL/6J mice; 9 days after tumor transplantation, mice with tumors of similar size were injected with intratumoral viruses at 9, 12, and 15 days after tumor transplantation. days; lentivirus and adeno-associated virus titers were determined by serial dilutions in MC38 cells; in each injection, the same unit of lentivirus or adeno-associated virus (a virus that can infect 8 × 104 MC38 cells in vitro amount) will be injected into the tumor; tumor size will be measured every 3 days until day 24.

As shown in Figure 5A-B, overexpression of LALBA in MC38 cells using lentivirus can promote the de novo synthesis and secretion of lactose, and this process depends on the expression of B4GALT1; if D107A or A126K site-directed mutation is performed on the LALBA gene, Lactose synthase will be inactivated; when MC38 cells overexpressing LALBA were implanted subcutaneously in wild-type mice, similar to B4GALT1 knockout MC38 tumors, the tumor growth rate was significantly lower than that of control cells (expressing only GFP) ( Figure 5C); Importantly, when MC38 tumors overexpressed LALBA, the inventors saw complete tumor regression (CR standing for Complete response) in more than 50% of mice; in contrast, in immunodeficient mice In NPG mice, there was no significant difference in tumor growth rate between MC38 tumors overexpressing LALBA and control cells expressing only GFP (Figure 5D); subsequently, the inventors re-implanted MC38 controls in mice whose tumors had completely regressed. Compared with uninjected wild-type mice of the same age, mice with complete regression of MC38-LALBA tumors were completely resistant to MC38 control cells, indicating that tumor-specific immune memory has been established (Figure 5E); at the same time , the inventor also subcutaneously implanted a mixture of overexpressed LALBA and control MC38 cells in wild-type mice, with a mixing ratio of 1:1 or 1:9; as shown in Figure 11, LALBA overexpressed MC38 was added to the control cells. Can significantly inhibit tumor growth (among 1:1 mixed tumors, 6/22 tumors completely regressed; among 1:9 mixed tumors, 3/22 tumors completely regressed); the above results show that it is partially or completely overexpressed in tumor cells LALBA, both can enhance immune suppression against tumors and form immune memory.

In order to achieve overexpression of LALBA in tumors, the inventors injected lentivirus encoding GFP control or LALBA into tumors; as shown in Figure 5F, after three intratumoral injections of LALBA lentivirus, the growth rate of MC38 tumors in wild-type mice increased. was significantly inhibited (4/10 tumors completely regressed), while mutant LALBA (D107A and A126K) had no obvious inhibitory effect (Figure 12, the arrow indicates the time of virus injection); flow cytometry analysis showed that tumors injected with LALBA lentivirus Within, the number of IFNγ+CD8+T cells increased significantly (Figure 5G); in addition, mice whose tumors completely regressed after LALBA lentivirus injection were completely resistant to control MC38 cells (Figure 5H); finally, when adeno-associated virus was used When (AAV) overexpressed LALBA in MC38 tumors, the inventors could observe similar tumor inhibitory effects (Figure 5I-K); flow cytometry analysis showed that tumor-infiltrating CD8 ⁺ T cells increased in AAV-LALBA-injected tumors. Can express higher levels of IFNγ and PD1.

The inventors performed flow cytometry staining with biotin-labeled lectins, including Erythrina cristagalli lectin (ECL) and succinyl-wheatgerm lectin (sWGA), which can respectively measure terminal β-galactosides on the cell surface. enzyme and β-N-acetylglucosamine expression levels. For MC38 cells, lactose can block the staining of ECL, but has no effect on the staining of sWGA (Figure 13A); B4GALT1 knockout MC38 cells will express a slightly higher level of ECL on the surface, while the expression of sWGA will be significantly increased; The results showed that the expression of ECL and sWGA on the surface of LALBA-overexpressing MC38 cells significantly increased (Figure 13B); on the other hand, knocking out B4GALT1 in CD8 ⁺ T cells significantly reduced the expression of ECL and enhanced the expression of sWGA. (Figure 13C).

Example 6: As a competitive inhibitor that structurally mimics the N-glycosyl group, lactose can enhance tumor immune surveillance in vivo

The above results show that in CD8 ⁺ T cells and tumor cells, interfering with the function of the N-glycosyl group with lactose can activate the tumor immune response in a synergistic manner; after subcutaneously implanting MC38 tumors in wild-type mice, the inventors Lactose was injected into the tail vein every two days; compared with PBS control, lactose could significantly reduce the growth rate of MC38 tumors and at the same time enhance the expression of PD1 and IFNγ on the surface of CD8 ⁺ T cells within the tumor (Figure 6A-C, Figure 14); when When the inventors performed the same treatment with immunodeficient NPG mice, the growth rate of MC38 tumors injected with lactose was not significantly different from that of the PBS control (Figure 6D); after depleting CD8 ⁺ T cells in wild-type mice, lactose lost its ability to control tumor growth. The inhibitory effect of MC38 tumors, while depletion of CD4 ⁺ T cells or NK cells did not affect the inhibitory effect of lactose on tumor growth (Figure 6E-G); furthermore, in MC38 tumors treated with anti-PD1 or anti-PDL1 antibodies , tail vein injection of lactose will not further enhance the inhibitory effect on tumors (Figure 15, Figure 16).

The inventor performed cell population transcriptome sequencing analysis on MC38 subcutaneous tumors of wild-type mice after treatment with lactose or control (PBS), and finally found 585 significantly up-regulated genes and 528 significantly down-regulated genes (Figure 17A-B); Unexpectedly, the most significantly enriched HALLMARK pathways (Figure 17C), such as IFNγ, hypoxia pathway, TNFα signaling pathway via NFκB, and Myctargets (V1andV2), were significantly enriched in B4GALT1 knockout MC38 cells. The pathways (Figure 4F) were consistent; gene expression signature analysis (Figure 17E-F) showed that lactose-treated tumors had significantly high expression of Tfh cell signatures, T cell positive enrichment signatures, and cytotoxic T cell signatures.

In order to further analyze the potential impact of lactose on the tumor microenvironment, especially tumor-infiltrating T cells, the inventors sorted out CD3-positive T cells from MC38 tumor samples treated with lactose and PBS, and then performed single-cell transcriptome sequencing and TCR sequencing (Figure 18, Figure 19); the results showed that after lactose treatment, the proportion of effector/exhaustive memory CD8 ⁺ T cells (C0: CD8 ⁺ PD1 + PRF1 +) increased from 26.0% to 40.1%, while progenitor T cells /Undifferentiated CD8 ⁺ T cells (C2: CD8 ⁺ TCF7+SELL+) and regulatory CD4 ⁺ T cells (C1: CD4+AREG+FOXP3+) dropped from 19.7% to 11.8% and from 23.5% to 14.2% respectively. (Figure 18F); TCR clone type analysis results showed that lactose-treated tumors had significantly increased TCR clones compared with the control group (Figure 19A-C); the largest TCR clone was CN1 (CDR3: TRA; TRB = CAIDPYNQGKLIF; CASSPGQGYAEQFF), this clone can target the specific neoantigen of MC38 cells (ASMTNMELMin theAdpgk gene). After lactose treatment, its proportion increased from 2.3% to 35.1% (Figure 19D-E); in summary, Both cell population and single-cell transcriptome sequencing analysis showed that lactose treatment promoted the proliferation and enhanced the function of tumor-specific CD8 ⁺ T cells.

In wild-type mice, tail vein injection of lactose did not have a significant effect on the growth of "cold tumors" such as 4T1 and CT26 (Figure 6H-I); compared with the control group, exogenous overexpression of Ova in B16F10 cells The sensitivity of tumors to lactose treatment can be significantly enhanced (Figure 6J-K). This is due to the expression of Ova that transforms the original B16F10 "cold tumor" into a "hot tumor".

At the same time, the inventor subcutaneously implanted SGC7901 human gastric cancer cells into NPG mice transplanted with human immune cells, and then injected lactose or PBS into the tail vein; similar to anti-PD1 antibody-treated mice (Figure 20), lactose can significantly Reduced the growth rate of SGC7901 tumors in humanized mice (Figure 6L); after lactose treatment, SGC7901 tumors contained a higher proportion of IFNγ+ tumor-infiltrating human CD8 ⁺ T cells (Figure 6M); in contrast, In NPG mice not transplanted with human immune cells, lactose did not exert a tumor suppressive effect (Fig. 6N).

Finally, the inventors evaluated the safety of wild-type C57BL/6J mice after lactose injection through the tail vein; as shown in Figure 21A, B, and C, short-term (24 hours) or long-term (24 days) once every two days None of the lactose injections produced any obvious toxic side effects.

Example 7: B4galt1 regulates the killing ability of CD8 ⁺ T cells through tumor-derived Gal-1

1.B4galt1 catalyzes CD8 galactosylation and regulates TCR-CD8 co-localization through Gal-1

In order to further analyze the molecular mechanism of B4galt1 regulating T cell activation, the present invention first used biotin-labeled erythritol agglutinin (ECL) and succinyl wheat germ agglutinin (sWGA) for flow cytometry staining, and analyzed the cell surface βGal respectively. and expression of βGlcNAc. The ECL staining intensity of OT-Ⅰ T cells with B4galt1 knockout was slightly reduced, and the sWGA staining intensity was significantly increased (Figure 25A-B). Surprisingly, the results of staining with biotin-labeled recombinant galectin-1 (rGal-1) showed very significant differences between control and B4galt1 knockout OT-Ⅰ T cells (Figure 25C) .

In Figure 25, A.B4galt1 knockout cells and control OT-IT cells were incubated with biotinylated ECL and then stained with streptavidin-PE. The dashed line indicates cells that were not incubated with lectin and stained only with streptavidin-PE. The mean fluorescence intensity value (MFI) of each sample was measured using flow cytometry. B. B4galt1 knockout cells and control OT-IT cells were incubated with biotinylated sWGA and then stained with streptavidin-PE. C. B4galt1 knockout cells and control OT-IT cells were incubated with biotinylated recombinant Gal-1 and then stained with streptavidin-PE. The dashed line indicates cells that were not incubated with lectin and stained only with streptavidin-PE.

RNA sequencing analysis showed that galectin-1, -3, and -9 (Gal-1, Gal-3, Gal-9) were the most expressed galectins in OT-ⅠT cells and MC38 cells (Figure 26A) . Intracellular flow staining confirmed that all these lectins were highly expressed in OT-ⅠT cells and MC38 cells. However, cell surface staining showed that Gal-1, -3 and -9 were only expressed on the surface of MC38 cells and did not exist on the surface of OT-ⅠT cells (Figure 26B). By comparing the intensity of MC38 cell surface staining and intracellular staining, it was found that Compared with Gal-3 and Gal-9, the distribution of Gal-1 on the surface of MC38 cells was more significant (Fig. 26C).

In Figure 26, A.FPKM represents the expression levels of various galectin mRNAs in OT-IT cells and MC38 cells, derived from RNA sequencing analysis. B. Flow cytometric staining and analysis of surface and intracellular galectin-1, -3, and -9 of OT-ⅠT cells and MC38 cells. C. Average fluorescence intensity values of MC38 cell surface and intracellular galectin-1, -3, and -9 staining and calculate the ratio of surface to intracellular intensity.

2. Gal-1 on the surface of CD8 ⁺ T cells comes from adjacent tumor cells

In order to explore the source of Gal-1 on the surface of CD8 ⁺ T cells, after co-culture of OT-Ⅰ T cells with OVA oligopeptide-treated or untreated MC38 cells, the Gal-1 staining on the surface of OT-Ⅰ T cells increased significantly without significant difference. This result suggests that the interaction between the TCR receptor and the MHC-I/OVA complex, or the activation of TCR, is not necessary for the appearance of Gal-1 on the surface of T cells (Figure 27A). On the other hand, in the absence of MC38 cells, using anti-CD3/anti-CD28 antibodies alone to activate OT-IT cells did not increase their surface Gal-1 staining (Figure 27A). Consistent with this, in the absence of MC38 cell co-culture, B4galt1 knockout OT-Ⅰ T cells did not exhibit significantly enhanced activity upon stimulation with anti-CD3/anti-CD28 antibodies compared with wild-type controls. T cell activation markers (Figure 27B). In summary, the increased Gal-1 staining on the surface of OT-IT cells is likely to come from co-cultured tumor cells.

In Figure 27, A. OT-Ⅰ T cells activated or untreated with anti-CD3/anti-CD28 antibodies, and OT-Ⅰ T cells co-cultured with OVA oligopeptide-pretreated MC38 cells or untreated MC38, respectively treated with anti- Gal-1 antibody staining. Dashed line indicates isotype control antibody staining. B. B4galt1 knockout cells and control OT-Ⅰ T cells were activated by anti-CD3/anti-CD28 antibodies. Quantitative RT-qPCR was used to detect the expression levels of Ifnγ and Tnfα mRNA.

In order to determine that Gal-1 on the surface of OT-IT cells comes from co-cultured tumor cells, the present invention co-cultured OT-IT cells with wild-type and Gal-1 knockout MC38 cells, and found that only co-cultured with wild-type MC38 cells Gal-1 staining on the surface of cultured OT-IT cells increased significantly (Fig. 28A). Then, the present invention uses the conditioned medium that has been cultured with wild-type or Gal-1 knockout MC38 cells as the medium for co-culture of OT-ⅠT cells and MC38. It is found that only in the presence of wild-type MC38 cells can OT-ⅠT cells be increased. Gal-1 staining on the surface was not possible with only wild-type conditioned medium (Fig. 28B). By setting co-culture conditions with different ratios of MC38 cells and OT-ⅠT cells, it was found that the intensity of Gal-1 staining on the surface of OT-ⅠT cells increased with the increase in the proportion of MC38 cells (Figure 28C-D).

In Figure 28, A. OT-IT cells co-cultured with Gal-1 knockout MC38 cells, co-cultured with wild-type MC38 cells, and not co-cultured with MC38 cells were stained with anti-Gal-1 antibodies respectively. B. OT-IT cells and wild-type or Gal-1 knockout MC38 cells were co-cultured in different MC38 cell conditioned media at a cell ratio of 2:1 for 8 hours. Stain with anti-Gal-1 antibody and detect the average fluorescence intensity of Gal-1 on the cell surface using flow cytometry. C. OT-ⅠT cells and MC38 cells were co-cultured at the ratio of 0:10, 1:10, 1:2, 2:1, 10:1 and 10:0, and then stained with anti-Gal-1. D. Flow cytometry detects the average fluorescence intensity of Gal-1 on the surface of OT-ⅠT cells and MC38 cells after co-culture at different ratios.

Further exploration revealed that when OT-IT cells and MC38 cells were physically separated by a separator in the Boyden chamber (Figure 29A) or a thin layer of low-melting point agarose membrane (Figure 29B), the Gal-1 protein could not effectively Transferred from MC38 cells to OT-IT cells, this experimental evidence indicates that close contact between cells is necessary for the transfer of Gal-1. Different from previous cell-to-cell transfer mechanisms of surface proteins, such as cytosis, exosome transport and nanotube shuttling, the present invention names this process as proximity-dependent intercellular protein spreading (PDICPS).

In Figure 29, A. OT-IT cells were seeded in the upper compartment (i) of Boyden Chamber, or MC38 cells were seeded in the lower compartment (ii) at the same time, or together with MC38 cells in the lower compartment (iii). After 8 hours, the cells were stained with anti-Gal-1 antibody and flow cytometry was used to detect the average fluorescence intensity of Gal-1 on the cell surface. B. After MC38 cells attach to the bottom of the culture dish, add medium containing low melting point agarose. OT-IT cells were then added to the bottom of the culture dish (i), or above the agarose layer (ii), or directly onto MC38 cells (iii). Stain with anti-Gal-1 antibody and detect the average fluorescence intensity of Gal-1 on the cell surface using flow cytometry.

3.B4galt1 regulates the killing ability of CD8 ⁺ T cells through Gal-1

By detecting the RNA levels of cytotoxic factors TNFα and IFNγ, the present invention found that adding exogenous recombinant Gal-1 protein to wild-type OT-Ⅰ T cells can significantly inhibit the activation of T cells by anti-CD3/anti-CD28 antibodies. However, there was no inhibitory effect in B4galt1 knockout OT-IT cells (Figure 30A). In the absence of OVA oligopeptide pretreatment, OT-ⅠT cells co-cultured with Gal-1 knockout MC38 cells showed anti-CD3/anti- Enhancement of CD28 antibody activation (Figure 30B). Similarly, B4galt1 knockout OT-Ⅰ T cells did not show enhanced T cell activation levels under this condition.

In Figure 30, A. Recombinant Gal-1 treatment (2.5 μg/mL) can significantly reduce Tnfα in wild-type OT-Ⅰ T cells (1×10 ⁵ cells/mL) after TCR activation by anti-CD3/anti-CD28 antibodies. and IfnγmRNA expression levels, but had no significant effect on B4galt1 knockout OT-ⅠT cells. The p-value comes from the two-tailed T test; B. Compared with co-culture with wild-type MC38 cells, co-culture with Gal-1 knockout MC38 cells can increase the number of wild-type cells activated by anti-CD3/anti-CD28 antibodies. The expression levels of Tnfα and Ifnγ mRNA in OT-ⅠT cells had no significant effect on B4galt1 knockout OT-ⅠT cells.

From the perspective of tumor cell killing function, the present invention compared the specific killing degree of wild-type and B4galt1 knockout OT-Ⅰ T cells against wild-type MC38 and Gal-1 knockout MC38 respectively (Figure 31A), and found that the specific killing effect of wild-type MC38 and Gal-1 knockout MC38 was similar to that of Gal-1. Compared with knocking out MC38, wild-type MC38 cells are more sensitive to specific killing mediated by B4galt1 knockout OT-Ⅰ T cells than wild-type OT-Ⅰ T cells (Figure 31B). These results prove that Gal-1 transferred from tumor cells to the surface of T cells plays an important role in regulating their cell activation and cytotoxicity. Only in the presence of Gal-1 can the knockout of B4galt1 play a role.

In Figure 31, A. B4galt1 knockout and wild-type OT-IT cells were used to kill wild-type (left picture) and Gal-1 knockout (right picture) MC38 cells pretreated with OVA oligopeptide, respectively. p-value comes from two-tailed T test; B. Calculate the fold change of the killing ability of B4galt1 knockout OT-ⅠT cells and the killing ability of wild-type OT-ⅠT cells when the target cells are wild-type and Gal-1 knockout MC38 cells respectively. . Data were calculated using a ratio of 2:1 (T cells:MC38 cells).

Subsequently, the present invention wanted to know whether the transfer of Gal-1 from tumor cells to the surface of infiltrating T cells and inhibiting their immune response can also occur in the tumor microenvironment in vivo. Therefore, the present invention overexpressed Gal-1-EGFP fusion protein in MC38 cells to track the metastasis of Gal-1 in tumors in vivo. Fluorescence was detected by flow cytometry. Compared with control tumors, the present invention found that CD8 ⁺ T cells infiltrating in Gal-1-EGFP MC38 tumors had significantly enhanced total GFP signals (Figure 32A) and surface GFP signals (Figure 32B ).

In Figure 32, A. MC38 cells exogenously expressing Gal-1-EGFP fusion protein or control EGFP protein were inoculated subcutaneously into wild-type C57BL/6J mice. After 3 weeks, tumor-infiltrating CD8 ⁺ T cells showed significant GFP signals only in Gal-1-EGFP tumors. B. Only in Gal-1-EGFP tumors, anti-GFP antibody staining on the surface of tumor-infiltrating CD8 ⁺ T cells showed obvious GFP signals.

Furthermore, subcutaneous tumor growth of Gal-1 knockout MC38 cells progressed extremely slowly in wild-type C57BL/6J mice compared with wild-type MC38 cells (Fig. 33A), but in immunodeficient NPG (NOD-Prkdc ^scid Il2rg Growth was normal in ^null ) mice (Fig. 33B). By clearing different types of immune cells in mice with specific neutralizing antibodies (Figure 33C), and then subcutaneously inoculating Gal-1 knockout MC38 cells into immune-deficient mice, the present invention found that CD8 ⁺ and CD4 ⁺ T cells Depletion significantly restored the growth of Gal-1 knockout MC38 tumors, but depletion of NK cells had no effect (Fig. 33D). Knockout of MHC-I complex component B2m also restored the growth of Gal-1 knockout MC38 tumors in wild-type mice (Fig. 33E). These results indicate that the growth of Gal-1 knockout MC38 tumors is inhibited by CD8 ⁺ and CD4 ⁺ T cells of the host immune system, which means that Gal-1 may mediate the suppression of T cell activity in tumors in vivo.

In Figure 33, A. Gal-1 knockout MC38 cells and wild-type MC38 cells were subcutaneously inoculated into wild-type C57BL/6J mice and the tumor growth size was monitored. p-value comes from two-factor analysis of variance; B. Gal-1 knockout MC38 cells and wild-type MC38 cells were subcutaneously inoculated into immunodeficient NPG (NOD-Prkdc ^scid Il2rg ^null ) mice and the tumor growth size was monitored. p-value comes from two-factor analysis of variance; C. Flow cytometry analysis shows that different neutralizing antibodies successfully removed CD8 ⁺ T cells (left) and CD4 ⁺ T cells (middle) in peripheral blood samples of wild-type mice. and NK cells (right); D. Gal-1 knockout MC38 cells were subcutaneously inoculated into wild-type mice lacking CD8 ⁺ T cells, CD4 ⁺ T cells, and NK cells, and tumor growth size was monitored. The p-value comes from two-factor analysis of variance; E. Gal-1/B2m double knockout and Gal-1 single knockout MC38 cells were subcutaneously inoculated into wild-type C57BL/6J mice and the tumor growth size was monitored. The p-value comes from a two-way ANOVA.

Based on the results of previous in vitro experiments, it is speculated that the distribution of endogenous Gal-1 on immune cells in different tissues in the body may be different. Therefore, the present invention detects endogenous Gal-1 staining on the surface of CD8 ⁺ T cells in wild-type mice. Spleens, peripheral blood and subcutaneous tumors were taken from wild-type C57BL/6J mice, stained with anti-CD8 and anti-Gal-1 or isotype control antibodies, and flow cytometry was used to detect Gal- on the surface of CD8 ⁺ T cells in different tissues. 1 expression, the results are shown in Figure 34. The presence of surface Gal-1 is almost undetectable on CD8 ⁺ T cells in the peripheral blood and spleen of wild-type mice, while the tumor-infiltrating CD8 ⁺ T cells are mainly PD- 1 positive, that is, exhausted CD8 ⁺ T cells, whose surface stains positive for Gal-1.

Furthermore, B2m knockout and Gal-1/B2m double knockout MC38 cells were subcutaneously inoculated into wild-type C57BL/6J mice. After 3 weeks, tumors were harvested and stained with anti-CD8a and anti-Gal-1 antibodies. The results showed that compared with B2m knockout MC38 tumors, in Gal-1/B2m double knockout MC38 tumors, the Gal-1 signal on the surface of CD8 ⁺ T cells was significantly reduced, suggesting that MC38 tumors are caused by the Gal-1 signal on the surface of CD8 ⁺ T cells in vivo. -1 source (Figure 35), a conclusion similar to the in vitro experimental results was obtained.

Example 7 demonstrated in in vivo and in vitro experiments that galectin-1 (Gal-1) can be transferred from tumor cells to the surface of adjacent CD8 ⁺ T cells and inhibit their activation and tumor killing ability. However, when the B4GALT1 gene in T cells is deleted, After survival, Gal-1 transfer was significantly reduced and its ability to suppress CD8 ⁺ T cells decreased. The above results prove a new immune checkpoint mechanism. The present invention may use small molecules to inhibit the binding of tumor-derived Gal-1 to the surface of adjacent CD8 ⁺ T cells, thereby enhancing tumor immune function.

Example 8: B4galt1 regulates TCR-CD8 co-localization through cell surface Gal-1

In order to identify the substrate of B4galt1 on the surface of CD8 ⁺ T cells, the present invention wants to use the affinity of β-galactoside and Gal-1 to collect galactosylated proteins on the cell membrane through Pulldown, and use protein mass spectrometry to Methods: To identify and compare the protein abundance of wild-type and B4galt1 knockout CD8 ⁺ T cells to find the substrate of B4galt1.

First, a large amount of recombinant Gal-1 protein needs to be purified. The overexpression vector with His tag is transferred into the expression strain BL21 using the E. coli expression system. After the small-scale expression test, single clones with considerable expression amounts are selected, and then large-scale purification is performed. , Figure 36A, the lanes from left to right are the expression strain lysis supernatant, Ni-NTA affinity purification column effluent, Ni-NTA affinity purification column cleaning solution, protein eluate, protein Marker and quantitative control protein BSA. The size of the target protein was consistent with the prediction and the band was obvious. A large amount of recombinant Gal-1 (~16kD) is eluted and dialyzed into a storage buffer or coupling buffer for later use.

Figure 36B. 0.5 μg of commercial recombinant Gal-1 protein or laboratory purified recombinant Gal-1 protein was incubated with OT-ⅠT cells respectively, and then stained with anti-Gal-1 antibody and detected by flow cytometry. Figure 36C. OT-IT cells were incubated with recombinant Gal-1 proteins from different sources and in different amounts and then stained with anti-Gal-1 antibodies to detect the average fluorescence intensity.

Before the formal experiment, the present invention conducted an activity test on the purified recombinant Gal-1 protein, taking advantage of its characteristics of binding to T cell surface proteins and being detectable by anti-Gal-1 antibodies. Incubate 0.5 μg of commercial recombinant Gal-1 protein or laboratory-purified recombinant Gal-1 protein with OT-ⅠT cells respectively, then stain with anti-Gal-1 antibody, and detect by flow cytometry. According to the flow cytometry staining results (Figure 36B ), OT-ⅠT cells were incubated with recombinant Gal-1 proteins from different sources and in different amounts and then stained with anti-Gal-1 antibodies. The results showed that the binding activity of laboratory-purified recombinant Gal-1 was comparable to that of commercial recombinant Gal-1. -1 is similar (Fig. 36B-C). Subsequently, NHS-activated Sepahrose was used to incubate and couple with purified recombinant Gal-1 protein to obtain recombinant Gal-1-Sepharose.

1.B4galt1 catalyzes CD8 galactosylation and regulates TCR-CD8 co-localization through Gal-1

Recombinant Gal-1-Sepharose was used to enrich Gal-1 binding proteins from membrane protein extracts of OT-IT cells, and then eluted with lactose solution. The resulting proteins were subjected to mass spectrometry analysis to identify the species and abundance. Analysis of proteins with significant differences between wild-type and B4galt1 knockout OT-Ⅰ T cells found that TCRα/β (OT-Ⅰ) and CD8α/β were both at the forefront (Figure 37A). KEGG analysis showed that the TCR signaling pathway was significantly enriched (Figure 37B). This shows that in B4galt1 knockout OT-Ⅰ T cells, the galactosylation of TCR and CD8 protein is significantly reduced.

Figure 37A. Volcano plot showing Gal-1 binding proteins identified in wild-type and B4galt1 knockout OT-IT cell membrane proteins. Proteins in the TCR signaling pathway are underlined. The p-value is from Limma in DEqMS (version 1.8.0). Figure 37B. Bar graph showing the KEGG pathways that were significantly changed in B4galt1 knockout OT-I T cells.

Western immunoblotting experiments verified that the binding of Gal-1 to CD8 in B4Galt1 knockout T cells was reduced (Figure 38A). Interestingly, the migration rate of CD8β protein in SDS-PAGE was different between membrane proteins of wild-type and B4galt1 knockout T cells, indicating that CD8β is a direct substrate of B4galt1 (Fig. 38A). In fact, upon treatment with peptide-N-glycosidase F (PNGase F) to cleave all glycosylation on the protein, the migration difference of the CD8β protein disappeared (Figure 38B).

Figure 38A. Western blotting experiment validates the top ten genes in pulldown-MS data analysis. Compared with wild-type OT-Ⅰ T cells, the migration rate of CD8b, Itgal, Ly9 and Lnpep proteins was significantly faster in B4galt1 knockout OT-Ⅰ T cells. Figure 38B. After PNGase F digests protein sugar chains, the migration rates of CD8b, Itgal, Ly9 and Lnpep proteins in wild-type and B4galt1 knockout OT-Ⅰ T cells are consistent.

The interaction between TCR and CD8 was measured by fluorescence resonance energy transfer (FRET) experiments (Figure 39A). As shown in Figure 39B, when recombinant Gal-1 was added to wild-type OT-Ⅰ T cells, the FRET signal between TCR-CD8 was significantly reduced, and lactose treatment could reverse this phenotype. For OT-Ⅰ T cells with B4galt1 knockout, the effect of recombinant Gal-1 on TCR-CD8 FRET was not significant.

Figure 39A. Schematic diagram of detecting fluorescence resonance energy transfer between TCR-CD8; Figure 39B. Relative fold change of FRET unit after wild-type and B4galt1 knockout OT-Ⅰ T cells were treated with recombinant Gal-1 or recombinant Gal-1+lactose.

2. B4galt1 regulates T cell killing dependent on CD8

Through cell killing experiments, the present invention found that the inactivation of B4galt1 affects the target cell killing mediated by TCR-T cells, but does not affect the target cell killing mediated by CAR-T. Similarly, CD8 knockout does not affect CAR-T. killing, which suggests that the phenotype of B4galt1 regulating T cell tumor killing function through Gal-1 is dependent on CD8 (Figure 40A-D).

Figure 40A. CRISPR/Cas9 knockout of B4galt1 in OT-ⅠT cells can significantly improve the specific killing activity against B16F10-OVA cells in vitro. Figure 40B. CRISPR/Cas9 knocking out CD8a in OT-ⅠT cells can significantly reduce the specific killing activity against B16F10-OVA cells in vitro. Figure 40C. CRISPR/Cas9 knockout of B4galt1 in hCD19-CAR T cells does not affect the in vitro killing activity of Nalm6 cells. Figure 40D. CRISPR/Cas9 knockout of CD8a in hCD19-CAR T cells does not affect the in vitro killing activity of Nalm6 cells.

Example 8 By comparing the phenotypes of wild-type and B4GALT1 knockout CD8 ⁺ T cells in recombinant Gal-1 pulldown-MS and TCR-CD8 FRET, the present invention confirmed the co-localization between CD8 as a catalytic substrate of B4GALT1 and TCR Affected by Gal-1 on the surface of T cells, B4GALT1 inhibits TCR activation mediated by TCR-CD8 interaction through Gal-1.

Example 9: Lactose enhances the anti-tumor ability of CD8 ⁺ T cells by competing for Gal-1 on the cell surface

1. Lactose, as a competitive inhibitor of Gal-1, can enhance CD8 ⁺ T cell activity

In the mammary gland, B4GALT1 interacts with LALBA to form lactose synthase, which transfers galactose to glucose to generate lactose. In vitro experiments show that the galactose part of lactose can compete with and interfere with the function of β-galactoside in glycoproteins and glycolipids. Therefore, the present invention hypothesizes that lactose and its derivatives may be a new class of immune checkpoint inhibitors that enhance the anti-tumor immune function of T cells by competing for Gal-1 on the surface of CD8 ⁺ T cells.

Under in vitro culture conditions, lactose can significantly enhance the specific killing function of OT-IT cells against MC38 cells pretreated with OVA oligopeptide (Figure 41A). Moreover, in this in vitro killing system, lactose significantly increased the expression of cytotoxic factors such as Ifnγ and Tnfα in OT-I T cells, indicating enhanced T cell activation (Figure 41B). On the other hand, in the presence of anti-CD3/anti-CD28 antibody stimulation but in the absence of exogenous Gal-1, lactose treatment could not significantly affect the expression of Ifnγ and Tnfα in OT-I T cells (Fig. 41C).

Figure 41A. Lactose treatment increases the in vitro specific killing activity of OT-IT cells against MC38 cells pretreated with OVA oligopeptide. Figure 41B. Lactose treatment increases the expression of Tnfα and Ifnγ in OT-ⅠT cells after co-culture with OVA oligopeptide-pretreated MC38 cells. Quantitative RT-qPCR was used to detect the relative expression of Tnfα and Ifnγ mRNA. Figure 41C. Lactose treatment has no significant effect on the expression of Tnfα and Ifnγ mRNA in OT-Ⅰ T cells stimulated by anti-CD3/anti-CD28 antibodies. Quantitative RT-qPCR method was used to detect the relative expression of Tnfα and Ifnγ mRNA.

In the above-mentioned in vitro killing system, the present invention confirmed that lactose treatment can remove Gal-1 from the surface of OT-IT cells through anti-Gal-1 antibody staining, and at the same time, it also reduced the level of Gal-1 on the surface of MC38 cells (Figure 42A-C).

Figure 42A. Representative flow cytometry detection chart of OT-IT cells and MC38 cells co-cultured at a ratio of 2:1 for 8 hours. Figure 42B. After co-culture of OT-IT cells and MC38 cells, lactose was added for incubation, stained with anti-Gal-1 antibody, and flow cytometry was used to detect the average fluorescence intensity of Gal-1 on the surface of MC38 cells. Figure 42C. After co-culture of OT-IT cells and MC38 cells, lactose was added for incubation, stained with anti-Gal-1 antibody, and flow cytometry was used to detect the average fluorescence intensity of Gal-1 on the surface of OT-IT cells.

In order to further verify the mechanism of action of lactose, the present invention allows OT-ⅠT cells to obtain Gal-1 by adding exogenous recombinant Gal-1 or co-culturing with MC38. It is found that the activity of OT-ⅠT cells will be significantly inhibited. Treatment with lactose can reverse the inhibitory effect of Gal-1 on OT-ⅠT cell activity (Figure 43A-B).

Figure 43A. Recombinant Gal-1 treatment reduces the expression of TNFα in OT-Ⅰ T cells after anti-CD3/anti-CD28 antibody stimulation, and this inhibitory effect can be reversed by lactose treatment. Quantitative RT-qPCR method was used to detect the relative expression of TnfαmRNA. Figure 43B. After co-culture of OT-IT cells with wild-type or Gal-1 knockout MC38 cells, anti-CD3/anti-CD28 antibody stimulation was performed with or without the addition of lactose. Quantitative RT-qPCR was used to detect the relative expression of Tnfα and Ifnγ mRNA.

In order to explore the role of lactose in the body, the present invention used lactose solutions of different concentrations to treat cells digested from MC38 tumors in vivo, and then through anti-Gal-1 staining and flow cytometry, the present invention found that lactose can also significantly reduce infiltrating T Gal-1 levels on the cell surface (Figure 44A). Moreover, lactose could significantly increase the expression of Ifnγ and Tnfα in exhausted CD8 ⁺ PD-1 ⁺ T cells isolated from tumors (Fig. 44B).

Figure 44A. Cells isolated from MC38 tumors were treated with different concentrations of lactose to compete for Gal-1 on the surface of MC38 tumor-infiltrating CD8 ⁺ T cells. The average fluorescence intensity of Gal-1 was measured using flow cytometry. Figure 44B. Cells isolated from MC38 tumors were treated with different concentrations of lactose, and anti-CD3/anti-CD28 antibodies were added to stimulate the expression of Tnfα and Ifnγ in MC38 tumor-infiltrating CD8 ⁺ T cells. Quantitative RT-qPCR was used to detect the expression levels of Tnfα and Ifnγ mRNA.

2. Lactose can enhance the activity of CD8+ T cells in the body and inhibit tumor growth

In order to explore the effect of lactose on tumors in vivo, the present invention injected lactose solution into the tail vein of wild-type mice subcutaneously inoculated with MC38 tumors every two days and monitored tumor growth. The results showed that compared with the PBS control group, lactose injection significantly inhibited the growth of MC38 tumors (Figure 45A) and increased the number of IFNγ-positive CD8 ⁺ T cells in the tumors (Figure 45B). On the other hand, when the experimental mice were immunodeficient NPG mice, the tumor-inhibitory effect of lactose was absent (Fig. 45C). Depleting CD8 ⁺ T cells in wild-type mice with neutralizing antibodies can significantly reduce the effect of lactose injection on MC38 tumors. When neutralizing antibodies are used to deplete CD4 ⁺ T cells and NK cells, the degree of reduction in the anti-tumor effect of lactose is smaller. Small (Fig. 45D). For B2m knockout MC38 tumors, lactose treatment also had no obvious effect (Figure 45E). This suggests that the role of lactose in tumors in vivo is largely dependent on CD8 ⁺ T cells.

In Figure 45, A. The effect of intravenous lactose on the growth of subcutaneous MC38 tumors in C57BL/6J wild-type mice. B. Intravenous lactose increases the percentage of MC38 tumor-infiltrating IFNγ ⁺ CD8 ⁺ T cells. C. Effect of intravenous lactose on subcutaneous MC38 tumor growth in immunodeficient NPG (NOD-Prkdc ^scid Il2rg ^null ) mice. D. Effect of intravenous lactose on MC38 tumor growth in C57BL/6J wild-type mice lacking CD8 ⁺ T cells, CD4 ⁺ T cells or NK cells. E. Effect of intravenous lactose injection on B2m knockout MC38 tumor growth in C57BL/6J wild-type mice.

In order to detect whether lactose has an effect on other different types of tumors, the present invention subcutaneously inoculated the breast cancer cell line 4T1 and the mouse colon cancer cell line CT26 in wild-type syngeneic mice, and then injected lactose intravenously. The results showed that lactose treatment and The growth of 4T1 and CT26 tumors was not significantly affected (Figure 46A-B). These two tumors are considered immune "cold" tumors and are less responsive to immunotherapy. Subsequently, the present invention subcutaneously inoculated two melanoma cell lines, B16F10-Puro and B16F10-OVA, into wild-type syngeneic mice, using B16F10-Puro as a control cell line. Exogenous overexpression of OVA protein caused B16F10 cells to undergo "cold The tumors transformed into "hot" tumors (the latter are thought to be more responsive to immunotherapy), thereby significantly increasing sensitivity to lactose treatment (Figure 46C-D).

In Figure 46, A. Effect of lactose on subcutaneous 4T1 tumor growth in BALB/c wild-type mice. B. Effect of lactose on subcutaneous CT26 tumor growth in BALB/c wild-type mice. C. Effect of lactose on subcutaneous B16F10-Puro tumor growth in C57BL/6J wild-type mice. D. Effect of lactose on subcutaneous B16F10-OVA tumor growth in C57BL/6J wild-type mice.

3. Biological properties of lactose and its derivatives

To further explore the possibility of lactose as a new immunotherapeutic drug, the present invention conducted preliminary tests on the functionality of lactose and its galactosyl group. The present invention found that in the absence of mammalian functional transporters, lactose cannot effectively enter cells (Fig. 47), and the chemical properties of lactose are stable in the circulation system. Therefore, the present invention concludes that the inhibition of tumor immune checkpoints by intravenous injection of lactose is not due to its degraded monosaccharide products (galactose and glucose). Figure 47 Detection of lactose permeability using parallel artificial membrane permeability assay (PAMPA). Testosterone and methotrexate served as positive and negative controls, respectively.

In order to confirm that the galactosyl group in lactose is the main functional group that binds lectins and galectin, the present invention tested lactose, sucrose, N-acetyllactosamine (LacNAc) and lactose-BSA ( bovine serum albumin) (Figure 48A). Since it does not contain a galactosyl moiety, sucrose does not exhibit competitive binding properties for ECL, sWGA and Gal-1 (Figure 48B). LacNAc competes more efficiently with β-galactosides in the cell surface glycome than lactose for ECL and Gal-1 binding, but is also able to bind to sWGA at higher concentrations (Figure 48B). In mice, intravenous injection of LacNAc showed an inhibitory effect on MC38 tumor growth (Figure 48C). Since lactose in the blood will be quickly excreted through the kidneys, in order to stabilize its concentration in the blood, the present invention couples lactose to BSA through high temperature (Figure 48A), which may improve pharmacokinetics and delay its release in the body. Metabolism and excretion.

In Figure 48, A. Structures of lactose, sucrose, N-acetyllactosamine (LacNAc) and lactose-bovine serum albumin (BSA). Coomassie brilliant blue staining of SDS-PAGE gel showed that BSA was efficiently coupled to lactose. B. Competitive binding of lactose, sucrose, N-acetylactosamine and lactose-BSA with ECL, sWGA and Gal-1 on the surface of MC38 cells. C. Effect of intravenous injection of LacNAc (20mM/250μLDPBS) on MC38 tumor growth in C57BL/6J wild-type mice.

Through in vitro T cell tumor killing experiments, cytotoxic factor detection and in vivo tumor models, the present invention proves that lactose can enhance the co-localization of TCR-CD8 and TCR activation by competing for Gal-1 on the cell surface, thereby improving the activity and activation of T cells. Anti-tumor ability. From the competitive binding experiments of lactose and its derivatives with lectins, it is known that the galactosyl group in these molecules is the main functional group that binds Gal-1, which provides a basis for further exploring the potential of lactose and its derivatives as anti-tumor drugs. Provides a molecular basis.

The above are only preferred specific embodiments of the present invention; however, the protection scope of the present invention is not limited thereto. Any person familiar with the technical field who is familiar with the technical field shall make equivalent substitutions or changes based on the technical solutions and improvement concepts of the present invention within the technical scope disclosed in the present invention, and shall be covered by the protection scope of the present invention.

Claims (27)

1. The application of galectin-1 as an immune check point in preparing tumor immunotherapy medicaments.

2. The use according to claim 1, wherein the tumor is a tumor with immunoinflammatory properties.

3. The use of claim 2, comprising: selecting at least one galectin-1 targeting conjugate or derivative thereof as an active ingredient of the drug; and/or selecting a native galectin-1 targeting conjugate, allowing cells of said tumor to overexpress a precursor of the targeting conjugate; and/or determining a galectin-1 binding site on the surface of an immune cell in the tumor microenvironment, such that the binding site is reduced, deleted or inactivated.

4. The use according to claim 3, wherein the galectin-1 targeting conjugate is lactose and the immune cell is CD8 ⁺ T lymphocytes.

5. A tumor immunotherapeutic agent, the active ingredient of which comprises at least one galectin-1 targeting conjugate, said galectin-1 targeting conjugate being a β -galactoside which is soluble and non-metabolizable outside the gastrointestinal tract of a host.

6. The tumor immunotherapeutic agent of claim 5, wherein the tumor is a tumor having immunoinflammatory properties.

7. The tumor immunotherapeutic agent of claim 5, wherein the β -galactoside is selected from lactose, lactulose, lactose sucrose, methyl β -lactoside, 4-O- β -D-galactopyranosyl-D-mannopyranoside, 3-O- β -D-galactopyranosyl-D-arabinose, 2' -O-methyl lactose, milk-N-disaccharide, N-acetyllactosamine, or β -D-thiogalactopyranoside, and the derivative of the galectin-1 targeting conjugate is a conjugate of the β -galactoside with a polypeptide or a pharmaceutically acceptable salt of the β -galactoside.

8. The tumor immunotherapeutic agent of claim 5, wherein the galectin-1 targeting conjugate is lactose or N-acetyllactosamine.

9. The tumor immunotherapeutic agent of claim 5, wherein the pharmaceutical composition is in the form of an injectable preparation.

10. The tumor immunotherapeutic agent of claim 5, wherein the galectin-1 targeting conjugate is lactose and the concentration of lactose in the agent is 50mM-400mM.

11. A tumor microenvironment immunopotentiator comprising an expression vector or a lalab promoter of at least one galectin-1 antibody.

12. The tumor microenvironment immunopotentiator of claim 10, wherein the tumor is a tumor with immunoinflammation.

13. The tumor microenvironment immunopotentiator of claim 10, wherein the lala enhancer is selected from the group consisting of: substances that increase the level of LALBA, substances that enhance activity of LALBA, and/or substances that retard metabolism of LALBA.

14. The tumor microenvironment immunopotentiator of claim 10, wherein the lala enhancer is selected from the group consisting of: the kit comprises an over-expression vector of LALBA, nano particles carrying LALBA genes, virus vectors carrying LALBA or the over-expression vector of the LALBA, PEG modified proteins, protein microspheres, liposome wrapping the LALBA and extracellular vesicles carrying the LALBA; LALBA is a substance selected from the group consisting of: the LALBA gene, the mRNA, cDNA, LALBA protein of LALBA or an active fragment of any of the foregoing.

15. The tumor microenvironment immunopotentiator of claim 14, wherein the viral vector carrying LALBA or an overexpression vector thereof is an adenovirus vector or a lentiviral vector.

16. A modified immune cell which is a T lymphocyte or NK cell, the surface of which has a reduced, deleted or inactivated level of galactosylation.

17. The modified immune cell of claim 16, which is CD8 ⁺ T lymphocytes.

18. The modified immune cell of claim 16, which is a TCR-T cell or a TCR-NK cell.

19. The modified immune cell of claim 16, wherein the immune cell has significantly reduced galactosylation, a deletion, or an inactivation.

20. A modified exempt from according to claim 16Epidemic cells, characterized in that the immune cells are CD8 ⁺ T lymphocytes.

21. The modified immune cell of claim 16, wherein the immune cell does not express an active B4GALT1 protein.

22. The modified immune cell of claim 16, wherein the B4GALT1 gene of the immune cell is knocked out or inhibited.

23. A pharmaceutical composition for tumor immunotherapy, the active ingredients of which at least comprise: the galectin-1 targeting conjugate of any one of claims 5-10, and/or the tumor microenvironment immunopotentiator of any one of claims 11-15, and/or the modified immune cell of any one of claims 16-22.

24. The tumor immunotherapeutic pharmaceutical composition of claim 23, wherein the active ingredient further comprises at least one substance that induces the conversion of a cold tumor to a hot tumor.

25. The tumor immunotherapeutic pharmaceutical composition of claim 23, further comprising a pharmaceutically acceptable adjuvant.

26. A combination composition for use in tumor immunotherapy comprising the tumor immunotherapy pharmaceutical composition of any one of claims 23-25 and at least one additional anti-tumor active ingredient selected from the group consisting of a tumor immunotherapeutic agent and a chemotherapeutic agent.

27. The combination composition of claim 26, wherein the chemotherapeutic agent is selected from the group consisting of: cyclophosphamide, trabectedin, temozolomide, melphalan, dacarbazine, oxaliplatin, methotrexate, mitoxantrone, gemcitabine, 5-fluorouracil (5-FU), bleomycin, doxorubicin, daunorubicin, epirubicin, idarubicin, valrubicin, paclitaxel, cabazitaxel, docetaxel, topotecan, irinotecan, etoposide, carboplatin, cisplatin, bortezomib, vinblastine, vincristine, vindesine, vinorelbine, filoquinone, nitrogen mustard, mitomycin C, fludarabine, cytosine arabinoside; and combinations thereof.