Gdf15 microbinders and uses thereof

Abstract

This invention discloses GDF15 micro-binding agents and their applications. This application provides a novel class of micro-binding agents specifically targeting the GDF15-GFRAL signaling pathway. These micro-binding agents possess extremely high binding affinity, excellent structural stability, and good in vitro and in vivo functional activity. Experiments show that these micro-binding agents can significantly reverse cancer-related cachexia in various tumor models, restore body weight, and prolong survival; simultaneously, they can restore sensitivity to PD-1 immune checkpoint inhibitors and enhance CD8... + The infiltration and function of T cells in the tumor microenvironment have broad clinical application prospects.

Description

Technical Field

[0001] This invention belongs to the field of biomedicine, specifically relating to GDF15 micro-binding agents and their applications. Background Technology

[0002] Cancer-associated cachexia is a severe syndrome characterized by progressive weight loss, reduced survival, and resistance to anticancer therapies. Growth differentiation factor 15 (GDF15), which functions by binding to its receptor GFRAL, has become a key mediator of cachexia, but effective and well-defined pathway neutralization strategies remain lacking. Given its widespread involvement in various diseases, GDF15 has become a highly anticipated therapeutic target. Several antibody therapies targeting GDF15 or GFRAL have entered clinical development, highlighting the translational value of this pathway. However, existing strategies rely almost entirely on monoclonal antibodies, which are inherently limited by their large molecular weight, limited tissue penetration, and insufficient flexibility in targeting specific protein-protein interaction interfaces. Furthermore, antibody optimization is often achieved through empirical screening rather than explicit structural-guided engineering of the GDF15-GFRAL interaction interface. These limitations underscore the need to develop alternative therapies that can precisely and mechanistically target the GDF15 signaling pathway.

[0003] Compared to antibodies, de novo-designed microconjugates offer several potential advantages, including smaller molecular weight, greater stability, and the ability to incorporate specific mechanistic designs at the binding interface. Therefore, designing suitable GDF15 microconjugates is crucial for the treatment of GDF15-GFRAL-related diseases. Summary of the Invention

[0004] To overcome the shortcomings of the prior art, the present invention provides GDF15 micro binder and its applications.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A first aspect of the present invention provides a micro-binding agent, the sequence of which is shown in SEQ ID NO:1.

[0006] Furthermore, the micro-binding agent may also include its modified product, or a detection reagent may be attached to the micro-binding agent.

[0007] Furthermore, the modifications include amination, methylation, amidation, hydroxylation, carboxylation, carbonylation, alkylation, acetylation, phosphorylation, sulfation, esterification, glycosylation, and cyclization.

[0008] Furthermore, the detection reagents include fluorescent dyes, radioactive labels, metal ions, and enzymes.

[0009] A second aspect of the present invention provides a nucleic acid molecule that encodes the micro-binding agent described in the first aspect of the present invention.

[0010] A third aspect of the present invention provides a carrier comprising the nucleic acid molecule described in the second aspect of the present invention.

[0011] Furthermore, the vector also includes a transcription promoter and / or an enhancer.

[0012] Furthermore, the vector also includes operatively linked nucleic acid molecules.

[0013] Furthermore, the operatively linked nucleic acid molecule includes a tag.

[0014] Furthermore, the label includes a localized epitope label and a label for purification.

[0015] A fourth aspect of the present invention provides a host cell comprising the nucleic acid molecule described in the second aspect of the present invention or the vector described in the third aspect of the present invention.

[0016] Furthermore, the host cells include prokaryotic cells and eukaryotic cells.

[0017] Furthermore, the prokaryotic cells include Escherichia coli.

[0018] Furthermore, the eukaryotic cells include protist cells, animal cells, or fungal cells.

[0019] Furthermore, the animal cells include mammalian cells, bird cells, and insect cells.

[0020] A fifth aspect of the present invention provides a medicament comprising the micro-binding agent of the first aspect of the present invention, the nucleic acid molecule of the second aspect of the present invention, the carrier of the third aspect of the present invention, or the host cell of the fourth aspect of the present invention.

[0021] Furthermore, the drug also includes a PD-1 antibody.

[0022] Furthermore, the drug also includes pharmaceutically acceptable excipients.

[0023] The sixth aspect of the present invention provides for any of the following applications: (1) Application of GDF15 inhibitors and / or PD-1 antibodies in the preparation of drugs for treating GDF15-related diseases; (2) Application of GDF15 inhibitors in the preparation of drugs that enhance the sensitivity to PD-1 antibody therapy; (3) The use of the micro-binding agent of the first aspect of the present invention, the nucleic acid molecule of the second aspect of the present invention, the carrier of the third aspect of the present invention, the host cell of the fourth aspect of the present invention, or the drug of the fifth aspect of the present invention in the in vitro detection of GDF15 or in the preparation of products for detecting GDF15.

[0024] Furthermore, the GDF15 inhibitor includes a GDF15 antibody, a micro-binding agent as described in the first aspect of the present invention, a nucleic acid molecule as described in the second aspect of the present invention, a vector as described in the third aspect of the present invention, a host cell as described in the fourth aspect of the present invention, or a drug as described in the fifth aspect of the present invention.

[0025] Furthermore, the GDF15-related diseases include cancer and its associated cachexia, metabolic diseases, and inflammatory diseases.

[0026] Furthermore, the cancer is selected from one or more of colon cancer, lung cancer, and lymphoma.

[0027] Furthermore, the lymphoma is a B-cell lymphoma.

[0028] A seventh aspect of the present invention provides a method for treating GDF15-related diseases, the method comprising administering to a subject in need the micro-binding agent of the first aspect of the present invention, the nucleic acid molecule of the second aspect of the present invention, the carrier of the third aspect of the present invention, the host cell of the fourth aspect of the present invention, or the drug of the fifth aspect of the present invention.

[0029] Furthermore, the method also includes administering a PD-1 antibody.

[0030] Furthermore, the GDF15-related diseases include cancer and its associated cachexia, metabolic diseases, and inflammatory diseases.

[0031] Furthermore, the cancer is selected from one or more of colon cancer, lung cancer, and lymphoma.

[0032] Furthermore, the lymphoma is a B-cell lymphoma.

[0033] An eighth aspect of the present invention provides a method for improving the sensitivity of PD-1 antibody therapy, the method comprising using the micro-binding agent of the first aspect of the present invention, the nucleic acid molecule of the second aspect of the present invention, the vector of the third aspect of the present invention, the host cell of the fourth aspect of the present invention, or the drug of the fifth aspect of the present invention.

[0034] A ninth aspect of the present invention provides a method for detecting GDF15, the method comprising using the micro-binding agent of the first aspect of the present invention, the nucleic acid molecule of the second aspect of the present invention, the vector of the third aspect of the present invention, the host cell of the fourth aspect of the present invention, or the drug of the fifth aspect of the present invention for detection.

[0035] Advantages and beneficial effects of the present invention: This application provides a novel type of micro-binding agent specifically targeting the GDF15-GFRAL signaling pathway. The micro-binding agent exhibits extremely high binding affinity, excellent structural stability, and good in vitro and in vivo functional activity. Experiments show that this micro-binding agent can significantly reverse cancer-related cachexia in various tumor models, restore body weight, and prolong survival; simultaneously, it can restore sensitivity to PD-1 immune checkpoint inhibitors and enhance CD8... + The infiltration and function of T cells in the tumor microenvironment have broad clinical application prospects. Attached Figure Description

[0036] Figure 1 This is a design diagram of a micro-binding agent targeting GDF15-GFRAL. Among them, 1A is an overall structural diagram of the GDF15 / GFRAL complex, 1B is an enlarged view of the GDF15 and GFRAL binding interface, and 1C is a rotated view of the GDF15 / GFRAL complex. Figure 2 This is an optimization diagram of micro-binding agents based on pAE interaction scores. 2A is a flowchart of the micro-binding agent structure screening process, 2B is a graph showing the calculated indices of the top five candidate micro-binding agents, and 2C is... Figure 2 The structural diagram of candidate micro-binding agents in B, 2D is a statistical graph of the average pLDDT value of candidate binding agents, 2E is a statistical graph of the average pAE value of candidate binding agents, 2F is a statistical graph of the average RMSD value of candidate binding agents, and 2G is a sequence analysis diagram of the binding agents. Figure 3These are biochemical characterization diagrams of the GDF15 micro-binding agent. Among them, 3A is a diagram showing the binding of candidate micro-binding agent clones to GDF15; 3B is a distribution diagram of micro-binding agents selected by yeast display; 3C is a size exclusion chromatogram of micro-binding agent 05_25_1; 3D is a size exclusion chromatogram of micro-binding agent 05_41_14; 3E is a surface plasmon resonance (SPR) sensing image of micro-binding agent 05_41_14 bound to GDF15; 3F is a SPR sensing image of micro-binding agent 05_25_1 bound to GDF15; 3G is a structural diagram of the 05_25_1-GDF15 complex predicted by AlphaFold3; 3H is a structural diagram of the 05_41_14-GDF15 complex predicted by AlphaFold3; and 3I is a secondary structure diagram of the micro-binding agent in different temperature ranges. Figure 4 This is a diagram showing the competitive blocking of the GDF15–GFRAL interaction by micro-binding agents 05_41_14 and 05_25_1. Specifically, 4A is a superimposed structural diagram of the micro-binding agent-GDF15 complex and the GDF15-GFRAL complex; 4B is a diagram of the competitive binding of the two candidate micro-binding agents to GDF15 detected by ELISA; 4C is a heatmap showing the effect of GDF15 residue amino acid substitutions on the affinity of 05_41_14 and GDF15; and 4D is a diagram of GDF15–GFRAL interaction. The heatmap shows the effect of amino acid substitution at residue F15 on the affinity of 05_25_1 and GDF15; 4E is the heatmap showing the effect of saturation mutation at residue 05_41_14 on the affinity of 05_41_14 and GDF15; 4F is the heatmap showing the effect of saturation mutation at residue 05_25_1 on the affinity of 05_25_1 and GDF15; 4G and 4H are diagrams showing the functional neutralization effect of charge-complementary rescue mutations that can restore the function of 05_41_14 and 05_25_1. Figure 5This is a graph showing the inhibition of downstream signaling and reversal of cancer-related cachexia by GDF15 microbinding agents. Specifically, 5A is an immunoblot image of HEK293 cells stably expressing GFRAL after treatment with GDF15 alone, or in combination with ponsegromab or GDF15 microbinding agents; 5B is a graph showing changes in the expression levels of early genes (FOS, JUN, EGR1) after stimulation by the above three methods; 5C is a graph showing the inhibitory effect of ponsegromab or GDF15 microbinding agents on the GDF15-GFRAL interaction; 5D is a graph showing the weight changes in MC38 colon cancer (MC38-hGDF15) tumor-bearing mice expressing human GDF15; 5F is a graph showing the weight changes in LLC1 lung cancer tumor-bearing mice expressing human GDF15; and 5H is a graph showing the A-20 expression of human GDF15. Figure 5E shows the weight change of B-cell lymphoma-bearing mice, while 5G shows the survival analysis of MC38 colon cancer-bearing mice expressing human GDF15, 5G shows the survival analysis of LLC1 lung cancer-bearing mice expressing human GDF15, and 5I shows the survival analysis of A-20 B-cell lymphoma-bearing mice expressing human GDF15. Figure 6 It is the neutralization effect of micro-binding agents on GDF15, and in the form of CD8 + The graph shows the T-cell-dependent recovery of sensitivity to PD-1 antibody therapy. Figure 6A shows the tumor growth curves of MC38-hGDF15 tumor-bearing mice after treatment with four methods: solvent control, GDF15 microbinding agent, PD-1 antibody, and PD-1 antibody + GDF15 microbinding agent. Figure 6B shows the Kaplan-Meier survival analysis of MC38-hGDF15 tumor-bearing mice after receiving the above four treatments. Figure 6C shows the CD8+ sensitivity of MC38-hGDF15 tumor-bearing mice after receiving the above four treatments. + The percentage of T cells in the total number of tumor-infiltrating cells is shown in the graph, with 6D representing intratumoral GZMB. + CD8 + The percentage of T cells in total T cells is shown in the graph. 6E represents intratumoral IFNγ. + CD8 + The graph shows the percentage of T cells among total T cells. 6F is the immunofluorescence image of MC38-hGDF15 tumor sections after treatment with PD-1 antibody monotherapy or PD-1 antibody + GDF15 microconjugate combination therapy. 6G is CD8+. + The growth curve of MC38-hGDF15 tumor under T cell depletion conditions. 6H is a schematic diagram of the coordination of metabolism and immune function reconstruction by GDF15 microbinding agents. Detailed Implementation

[0037] The following provides definitions for some of the terms used in this specification. Unless otherwise stated, all technical and scientific terms used herein generally have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0038] The present invention provides a micro-binding agent, the sequence of which is shown in SEQ ID NO:1.

[0039] In some implementations, micro-binder, GDF15 binder, and binder are used interchangeably and have the same meaning.

[0040] In some implementations, the human gene corresponding to GDF15 has a Gene ID of 9518. When the subject is a non-human species, the GDF15 gene refers to the ortholog of the human GDF15.

[0041] In some embodiments, the microbinding agent further includes a functional variant having at least 80%, 85%, 90%, 95%, 98%, or 99% sequence identity with the sequence shown in SEQ ID NO:1.

[0042] In some embodiments, identity / similarity refers to the similarity or identity between two or more nucleic acid sequences, or two or more amino acid sequences, expressed based on the identity or similarity between sequences. Sequence identity can be measured by percentage identity; the higher the percentage, the more consistent the sequences. When aligned using standard methods, homologs or orthologs of nucleic acid or amino acid sequences have a relatively high level of sequence identity / similarity. The microbinding agents disclosed in this application are not limited to the precise sequences shown, and as those skilled in the art will recognize, the sequences can be modified if necessary without significantly affecting the ability of the microbinding agents to function.

[0043] The micro-binding agent may also include its modified products, or a detection reagent may be attached to the micro-binding agent.

[0044] In some embodiments, the detection reagent can be any substance having detectable physical or chemical properties. Such detectable reagents are well-developed in the field of immunoassays, and generally, most of any labeling useful in such methods can be applied to the provided methods. Therefore, the detection reagent can be any composition detectable by spectroscopic, photochemical, biochemical, immunochemical, electrical, optical, or chemical methods. Detection reagents include, but are not limited to, fluorescent dyes (e.g., fluorescein isothiocyanate, Texas red, rhodamine, etc.), and radiolabeling (e.g., 3 H, 125 I, 35 S, 14 C or 32P), in particular, radioactive labels (e.g., 157 Gd, 55 Mn, 162 Dy、 52 Cr and 56 Fe), metal ions (e.g., 111 In、 97 Ru、 67 Ga、 68 Ga、 72 As、 89 Zr and 201 Tl), enzymes (e.g., horseradish peroxidase, alkaline phosphatase, and other enzymes commonly used in ELISA), electron transfer agents (e.g., including metal-binding proteins and compounds), luminescent and chemiluminescent labels (e.g., luciferin and 2,3-dihydrophtahlazinediones, such as luminol), magnetic beads (e.g., DYNABEADS™), and colorimetric labels such as colloidal gold or colored glass or plastic beads (e.g., polystyrene, polypropylene, latex).

[0045] In some embodiments, the subject refers to any animal, including but not limited to humans, non-human primates, rodents, etc. Generally, the terms "subject" and "patient" are used interchangeably in this application when referring to human subjects.

[0046] The present invention provides a nucleic acid molecule that encodes the aforementioned micro-binding agent.

[0047] In some embodiments, a nucleic acid molecule or polynucleotide refers to a polymeric compound comprising covalently linked nucleotides, which may consist of native subunits (e.g., purine or pyrimidine bases) or non-native subunits (e.g., a morpholine ring). Purine bases include adenine, guanine, hypoxanthine, and xanthine, and pyrimidine bases include uracil, thymine, and cytosine. Polynucleotides include polynucleotides (RNA), including mRNA, microRNA, siRNA, viral genomic RNA, and synthetic RNA, and polydeoxyribonucleic acid (DNA), including cDNA, genomic DNA, and synthetic DNA, any of which may be single-stranded or double-stranded. If single-stranded, the polynucleotide may be a coding strand or a non-coding (antisense) strand. Polynucleotides encoding amino acid sequences include all nucleotide sequences encoding the same amino acid sequence. Some versions of the nucleotide sequence may also include introns to such an extent that the introns will be removed by co-transcriptional or post-transcriptional mechanisms. In other words, different nucleotide sequences may encode the same amino acid sequence due to redundancy or degeneracy of the genetic code, or through splicing.

[0048] In some embodiments, nucleic acid molecules / polynucleotides can be synthesized, for example, by standard chemical synthesis and / or recombinant methods, or produced semi-synthetically, for example by combinatorial chemical synthesis and recombinant methods. The ligation of the coding sequence with transcriptional regulatory elements and / or other amino acid coding sequences can be performed using established methods, such as restriction enzyme digestion, ligation, and molecular cloning.

[0049] The present invention provides a carrier comprising the above-mentioned nucleic acid molecules.

[0050] In some embodiments, a vector refers to a medium into which a nucleic acid molecule encoding a microbinding agent can be operatively inserted to induce expression of the microbinding agent. Vectors can be used to transform, transduce, or transfect host cells to express carried genetic elements within the host cells. Vectors may contain a variety of elements for controlling expression, including linker sequences, promoter sequences, transcription initiation sequences, enhancer sequences, selection elements, and reporter genes. Additionally, vectors may contain an origin of replication. An origin of replication is a sequence that initiates replication when present in the vector. An origin of replication may be recognized by a replication initiation factor or alternatively by a DNA helicase. Vectors may also include materials that facilitate their entry into cells, including but not limited to viral particles, liposomes, or protein envelopes. Vectors may also include additional nucleotide sequences operatively linked to the linked nucleic acid molecule, such as epitope tags for localization, like 6-HIS tags or MYC tags, or tags for purification, such as GST fusions; and sequences for guiding protein secretion and / or membrane association.

[0051] In some implementations, examples of vectors include, but are not limited to, retroviruses (including lentiviruses), adenoviruses, adeno-associated viruses, herpesviruses (e.g., herpes simplex virus), poxviruses, baculoviruses, papillomaviruses, multivoxoviruses (e.g., SV40), λ phages and M13 phages, and plasmids. Specific examples of vectors include, but are not limited to, pcDNA3.3, pMD18-T, pOptivec, pCMV, pEGFP, pIRES, pQD-Hyg-GSeu, pALTER, pBAD, pcDNA, pCal, pL, pET, pGEMEX, pGEX, pCI, pEGFT, pSV2, pFUSE, pVITRO, pVIVO, pMAL, pMONO, pSELECT, pUNO, pDUO, Psg5L, pBABE, pWPXL, pBI, p15TV-L, pPro18, pTD, pRS10, pLexA, pACT2.2, pCMV-SCRIPT, pCDM8, pCDNA1.1 / amp, pcDNA3.1, pRc / RSV, PCR2.1, pEF-1, pFB, pSG5, pXT1, pCDEF3, pSVSPORT, and pEF-Bos.

[0052] The present invention provides a host cell, wherein the host cell comprises the above-mentioned nucleic acid molecule or the above-mentioned vector.

[0053] In some embodiments, the host cell is a cell used to receive, maintain, replicate, and amplify the vector. The host cell may also be used to express the polypeptide encoded by the vector. When the host cell divides, the nucleic acids contained in the vector replicate, thereby amplifying the nucleic acids. In some embodiments, the host cell serves as genetic packaging that can be induced to express variant microbinding agents on its surface. In another embodiment, the host cell is infected with genetic packaging.

[0054] In some embodiments, the host cell can be virtually any cell available for the expression vector. This includes prokaryotic cells and eukaryotic cells, with the prokaryotic cells including, but not limited to, eubacteria such as Gram-negative or Gram-positive organisms, such as Enterobacteriaceae, such as Escherichia, for example, *Escherichia coli* (DH5α, BL21DE3, BL21DE3pLysS, JM109, TOP10, HB101, SCS110, *E. coli* JM110); *Enterobacter*; *Erwinia*; *Klebsiella*; *Proteus*; *Salmonella*, for example, *Salmonella typhimurium*; and *Serratia*, for example, *Serratia marcescens*. The genera *Bacillus marcescans*, *Shigella*, and *Bacilli*, such as *B. subtilis* and *B. licheniformis*; *Pseudomonas*, such as *P. aeruginosa*; and *Streptomyces*.

[0055] Eukaryotic cells include, but are not limited to, protist cells, animal cells, or fungal cells. Animal cells include mammalian cells, avian cells, and insect cells. Mammalian cells include, but are not limited to, CHO cells, F2N cells, CSO cells, BHK cells, Bowes melanoma cells, HeLa cells, 911 cells, AT1080 cells, A549 cells, 293T cells, and 293F cells.

[0056] This invention provides the use of GDF15 inhibitors and / or PD-1 antibodies in the preparation of medicaments for treating GDF15-related diseases.

[0057] In some embodiments, the GDF15-related diseases are GDF15-positive diseases, including but not limited to cancer and its associated cachexia, inflammatory diseases, metabolic diseases, and cardiovascular diseases. Specifically, cancers include but are not limited to colon cancer, lung cancer, lymphoma (such as B-cell lymphoma), gastric cancer, pancreatic cancer, liver cancer, prostate cancer, ovarian cancer, bladder cancer, breast cancer, and osteosarcoma; inflammatory diseases include but are not limited to systemic lupus erythematosus, rheumatoid arthritis, ankylosing spondylitis, multiple sclerosis, chronic obstructive pulmonary disease, bronchial asthma, ulcerative colitis, acute pancreatitis, glomerulonephritis, and sepsis; metabolic diseases include but are not limited to obesity and metabolic syndrome, diabetes, sarcopenia, metabolic bone disease, thyroid dysfunction, and polycystic ovary syndrome; and cardiovascular diseases include but are not limited to myocardial infarction, heart failure, myocarditis, and coronary heart disease.

[0058] In some embodiments, the combined administration of GDF15 inhibitors and PD-1 antibodies has a synergistic effect compared to the administration of either the GDF15 inhibitor or the PD-1 antibody alone. When the two are administered in combination, the drug may be a single compound preparation or a combination of two single preparations.

[0059] In some embodiments, the compound formulation refers to a formulation made of two or more active pharmaceutical ingredients. In this application, the compound formulation is a compound formulation containing a GDF15 inhibitor and a PD-1 antibody. The single-ingredient formulation refers to a formulation made of a single active pharmaceutical ingredient. In this application, the combination of single-ingredient formulations is a combination of a single-ingredient formulation containing a GDF15 inhibitor and a single-ingredient formulation containing a PD-1 antibody. The two single-ingredient formulations in the combination are administered simultaneously or sequentially.

[0060] In some implementations, the GDF15 inhibitor and PD-1 antibody in the combined administration can be administered simultaneously, separately, or sequentially. Simultaneous administration means that the two drugs are administered concurrently. If not simultaneous, they are administered sequentially within a time frame so that both can be therapeutically effective within the same time frame. Therefore, sequential administration allows for the administration of one drug 5 minutes, 10 minutes, 15 minutes, 30 minutes, 1 hour, 2 hours, or several hours after administering one drug, provided that the circulating half-life of the first administered drug allows for a simultaneously therapeutically effective amount of both. The time delay between administrations of the components will vary depending on the exact nature of the components, their interactions, and their respective half-lives. This differs from simultaneous or sequential administration, where the interval between administering one drug and another is significant, meaning that when the second drug is administered, the first administered drug may no longer be present in the bloodstream at a therapeutically effective amount.

[0061] The drug also includes pharmaceutically acceptable excipients.

[0062] In some embodiments, pharmaceutically acceptable excipients refer to non-toxic materials that do not interact with the active ingredient of the drug of this application. The pharmaceutically acceptable excipients refer to natural or synthetic, organic or inorganic components that, when used in combination with the active ingredient, facilitate application.

[0063] In some embodiments, the pharmaceutically acceptable excipients include, but are not limited to, diluents, binders, surfactants, humectants, adsorbents, lubricants, fillers, and disintegrants.

[0064] In some embodiments, the diluent includes, but is not limited to, lactose, sodium chloride, glucose, urea, starch, and water. The binder includes, but is not limited to, starch, pregelatinized starch, dextrin, maltodextrin, sucrose, gum arabic, gelatin, methylcellulose, carboxymethylcellulose, alginate and alginate, xanthan gum, and hydroxypropylcellulose. The surfactant includes, but is not limited to, sodium lauryl sulfate, glyceryl monostearate, and hexadecyl alcohol. The humectant includes, but is not limited to, glycerol and starch. The adsorbent carrier includes, but is not limited to, starch, lactose, bentonite, and soap clay. The lubricant includes, but is not limited to, zinc stearate, talc, calcium and magnesium stearate, polyethylene glycol, polyoxyethylene monostearate, monolaurate, and magnesium lauryl sulfate. The filler includes, but is not limited to, mannitol, xylitol, sorbitol, maltose, glucose, lactose, sucrose, dextrin, and starch. The disintegrant includes, but is not limited to, crosylvinylpyrrolidone, sodium carboxymethyl starch, low-substituted hydroxypropylmethyl, crosylcarboxymethyl cellulose sodium, and soybean polysaccharides.

[0065] The dosage forms of the drug include gastrointestinal dosage forms and non-gastrointestinal dosage forms.

[0066] In some embodiments, the gastrointestinal dosage form includes tablets, granules, capsules, solutions, dry suspensions, powders, sustained-release formulations, effervescent tablets, emulsions, suspensions, syrups, drops, and chewable tablets.

[0067] In some embodiments, the non-gastrointestinal dosage form includes injectable dosage form, respiratory dosage form, cavity dosage form, mucosal dosage form, and skin dosage form.

[0068] In some embodiments, the injectable dosage forms include, but are not limited to, various injectables such as intravenous injections, intramuscular injections, subcutaneous injections, intradermal injections, and intracavitary injections; the respiratory dosage forms include, but are not limited to, sprays, aerosols, and powder inhalers; the cavity dosage forms include, but are not limited to, suppositories, aerosols, effervescent tablets, drops, and pills, for use in the rectum, vagina, urethra, nasal cavity, and ear canal; the mucosal dosage forms include, but are not limited to, eye drops, nasal drops, ointments, mouthwashes, sublingual tablets, adhesive tablets, and films; and the skin dosage forms include, but are not limited to, topical solutions, lotions, liniments, ointments, plasters, pastes, and patches.

[0069] The invention is further illustrated below with reference to specific embodiments. It should be understood that the specific embodiments described herein are by way of example and are not intended to limit the invention. The main features of the invention can be used in various embodiments without departing from the scope of the invention.

[0070] Example I. Experimental Methods 1. Two complementary scaffold design strategies for GDF15 micro-binders Based on the GDF15-GFRAL complex structure (PDB:5VZ4), RFdiffusion was used to generate a backbone complementary to the GDF15 receptor binding surface. ProteinMPNN was used to design sequences, and AlphaFold was used to predict the structure and binding mode. After multiple rounds of screening, several high-confidence micro-binding agents were obtained.

[0071] 2. Expression and purification of GDF15 binder and recombinant protein The gene encoding the computationally designed GDF15 binder was synthesized and cloned into the pET21b vector. The resulting GDF15 binder-His6 recombinant plasmid was transformed into *E. coli* BL21(DE3) cells and cultured in LB medium supplemented with 100 μg / ml ampicillin until the culture reached its OD (Occurrence Limit). 600The pH value reached 0.6-0.8. Protein expression was induced by adding 0.4 mM IPTG, followed by incubation at 20°C for 16 hours. Cells were collected by centrifugation and resuspended in lysis buffer containing 50 mM Tris-HCl (pH 8.0), 150 mM NaCl, and 30 mM imidazole, followed by sonication on ice. Cell debris was removed by centrifugation at 15,000×g for 30 min, and the clear supernatant was loaded onto Ni-NTA agarose resin (Sangon Biotech) for His-tag affinity purification. The bound protein was eluted with lysis buffer containing 500 mM imidazole. The eluent was further purified by size exclusion chromatography (SEC) using a Superdex® 75Increase 10 / 300 GL column (catalog number GE29-1487-21, Cytiva). The purified protein was concentrated to 1 mg / ml, aliquoted, and stored at -80°C.

[0072] For mammalian cell expression, the genes encoding GDF15 and GFRAL were cloned into mammalian expression vectors. The GDF15 construct included an N-terminal FLAG tag and a His6 tag, while the GFRAL construct contained only a His6 tag; in both constructs, a thrombin protease cleavage site was introduced between the tag and the target protein. The recombinant plasmid was transfected into Expi293F cells (Catalog No. A14527, Thermo Fisher Scientific; cell density 2 × 10⁶) using polyethyleneimine (PEI) (catalog number 24765-2, Polysciences). 6 Cells were cultured at 37°C and 8% CO2 with shaking for 5 days, and expression enhancers were added 18-22 hours after transfection. After expression, the culture supernatant was collected and centrifuged to obtain soluble secreted protein for purification. His-tag affinity purification was performed on the clarified supernatant using Ni-NTA agarose resin (catalog number C600791, Sangon Biotech). The resin was pre-equilibrated with binding / washing buffer (20 mM Tris-HCl, pH 8.0, 150 mM NaCl, 30 mM imidazole), followed by loading the supernatant and incubation to achieve binding. After washing with buffer containing 30 mM imidazole to remove non-specifically bound proteins, the target protein was eluted with 500 mM imidazole. The Ni-NTA elution buffer was replaced to adjust to suitable digestion conditions, followed by the addition of thrombin to remove the affinity tag. Digestion was performed overnight at 4°C under mild conditions. The reaction mixture was passed through Ni-NTA resin again to remove residual His-tagged fragments and uncut proteins. The fraction corresponding to the SEC main peak was collected, concentrated to 5 mg / ml using ultrafiltration centrifuge tubes, aliquoted, and stored at -80°C.

[0073] 3. Yeast surface display Genes of the top 1000 binding agent candidates were synthesized (Twist Bioscience) and cloned into the yeast display vector pETCON, subsequently transformed into *Saccharomyces cerevisiae* strain EBY100 via lithium acetate / PEG transformation. Transformants were screened by culturing on SD-Ura-Trp (SD-UT) plates at 30°C for 48 hours to construct the initial library. During screening, the yeast library was amplified in SD-UT medium, induced to express on the surface in galactose-containing SG-UT medium, and incubated with FLAG-labeled GDF15 (1 μM) at 4°C for 1 hour. Cells were labeled with anti-HA and anti-FLAG antibodies, respectively, to assess display efficiency and target binding ability. Fluorescence-activated cell sorting (FACS) was performed on a BD FACSAria II flow cytometer to screen for HA / FLAG double-positive cells. Approximately 1 × 10⁶ cells were collected per round. 5 Cells were expanded and subjected to multiple rounds of induction and screening. After 3-4 rounds of screening, enriched clones were isolated and sequenced, and clones for downstream expression, purification, and affinity characterization were selected.

[0074] 4. SPR determination combined with kinetics Binding kinetics were determined using a Biacore T200 system equipped with a CM5 Series S sensor chip. Recombinant GDF15 dimers were immobilized on the sensor chip surface via amine coupling using 10 mM sodium acetate (pH 5.5) as the coupling buffer. Immobilization levels were controlled within a low RU (resonance unit) range to minimize mass transfer limitations. HBS-EP+ buffer was used as both the run and sample buffer. Analytes (including GDF15 microbinding agents 05_41_14 and 05_25_1) were diluted in HBS-EP+ and injected into the GDF15-immobilized sensor chip at concentrations ranging from 3.125 to 50 nM (3.125, 6.25, 12.5, 25, and 50 nM). The injection flow rate was 35 μL / min, with a binding time of 350 seconds followed by a 350-second dissociation phase. After each cycle, the chip surface was regenerated with 10 mM glycine (pH 2.0). Binding and dissociation curves were fitted using a 1:1 Langmuir binding model to determine the binding rate constant (k_on) and dissociation rate constant (k_off), and the equilibrium dissociation constant (K_D = k_off / k_on) was calculated accordingly. All data analyses were performed using Biacore Insight Evaluation software.

[0075] 5. Site-directed saturation mutation of GDF15 microbinding agent Site-directed saturation mutagenesis (SSM) was employed to investigate the interfacial residues of the binding agent protein and its binding interaction with GDF15. First, a site-directed saturation mutant library was constructed by introducing mutations at selected interfacial residues between GDF15 and the binding agent protein using degenerate codons. This library was subjected to display-level and negative screening to remove non-functional clones, followed by a third round of affinity-based screening, using progressively decreasing concentrations of O5_41_14 and O5_25_1 (from 5 nM to sub-nanomoles / L). After each round of screening, the selected population was analyzed by deep sequencing using the Illumina NextSeq platform. Sequencing data were assembled and analyzed using PEAR software, the ΔlogKD value for each mutation was calculated, and heatmaps were generated to visualize the impact of individual mutations on binding affinity. Based on the SSM heatmap results, rescue mutations were introduced to restore binding affinity, and KD and IC50 were quantified using SPR and luciferase reporter gene assays. 50 The values ​​were used to evaluate the effectiveness of the functional neutralization. This verified the consistency between the design results and the actual results.

[0076] 6. Circular dichroism (CD) determination CD measurements were performed using a JASCO J1500 spectrometer to assess the secondary structural stability of the binding protein. Measurements were conducted using 1 mm path length cuvettes within a temperature range of 25–95 °C. The wavelength range was set to 200–260 nm, covering the characteristic absorption peaks of the peptide backbone in the far-UV region. Proteins were prepared in PBS buffer (pH 7.4) at a concentration of 0.4 mg / mL. Samples were equilibrated for 5 minutes prior to measurement at each temperature to ensure thermal equilibrium. CD spectra were recorded in 3 °C increments from 25 °C to 95 °C. Each spectrum was averaged after three scans to improve the signal-to-noise ratio. Thermal denaturation profiles were obtained by measuring the CD signal (a characteristic peak of α-helix content) at 222 nm as the temperature increased. The melting temperature (Tm), the temperature at which the protein undergoes significant unfolding, was determined using thermal transition profiles. All measurements were performed on a JASCO J1500 spectrometer equipped with a temperature-controlled Peltier system for precise temperature control.

[0077] 7. Cell line and GDF15 response signal model In vitro GDF15-GFRAL signal transduction assays were performed using HEK293 cells stably expressing human GFRAL and its co-receptor RET, while also carrying the SRE-Luc2 luciferase reporter gene. This system is widely used and mature, in which GDF15 can potently activate downstream ERK signaling and transcriptional responses. This system sensitively and reproducibly reflects GDF15 pathway activity and is commonly used for the assay of functional GDF15 neutralization. Mouse tumor cell lines used in in vivo studies included MC38 (colon cancer, GDC0671, China Center for Type Culture Collection), LLC1 (lung cancer, GDC0670, China Center for Type Culture Collection), and A20 (B-cell lymphoma, ZY-M073, US Type Culture Collection). All cell lines were cultured in DMEM (MC38, LLC1) or RPMI-1640 (A20) medium supplemented with 10% fetal bovine serum and penicillin-streptomycin, and were confirmed to be free of mycoplasma contamination prior to the experiments.

[0078] 8. GDF15 Stimulation Conditions In biochemical and transcriptional assays, the final concentration of recombinant human GDF15 was 50 ng / mL.

[0079] Western blot (pERK / ERK): After starving cells for 12-16 hours, stimulated with GDF15 for 15 minutes.

[0080] Real-time quantitative PCR (i.e., early gene sequencing): RNA was extracted from cells 60 minutes after stimulation with GDF15.

[0081] Luciferase reporter gene assay: Luciferase activity was measured 6 hours after cell treatment with GDF15. The microbinding agent or control reagent should be pre-incubated with GDF15 at room temperature for 30 minutes before being added to the cells.

[0082] 9. Western blot analysis of downstream phosphorylation (pERK / ERK) After 4 hours of cell starvation, cells were stimulated with recombinant GDF15 (± microbinding agent or anti-GDF15 antibody) at specified time intervals. Cells were lysed with ice-cold RIPA buffer supplemented with protease / phosphatase inhibitors. The lysate was centrifuged to clarify and protein concentration was determined by the BCA method. Equal volumes of protein were separated by SDS-PAGE electrophoresis, transferred to a PVDF membrane, blocked with 5% BSA in TBST solution, and then incubated overnight at 4°C with primary antibodies against antiphosphorylated ERK1 / 2 (Thr202 / Tyr204; CST catalog number 4370), total ERK1 / 2 (CST catalog number 9102), and α-tubulin (CST catalog number 3873). After incubation with horseradish peroxidase (HRP)-conjugated secondary antibody, the membrane was developed using ECL luminescent substrate.

[0083] 10. Real-time quantitative PCR analysis of early genes (IEGs) Cells were stimulated with GDF15 (± microbinding agent or control) for 1 hour. Total RNA was extracted using a silica gel column kit or a TRIzol-based method, treated with DNase, and then reverse transcribed into cDNA using a standard reverse transcription kit. Quantitative PCR was performed on a real-time PCR system using the SYBR Green dye method. The expression level of the target gene was normalized using the housekeeping gene β-actin as an internal control. Relative expression levels were calculated using the 2^-ΔΔCt method and expressed as a fold change relative to the solvent control.

[0084] 11. Luciferase reporter gene assay for GDF15-GFRAL pathway activity HEK293 cells stably expressing human GFRAL and its co-receptor RET and carrying the SRE-Luc2 luciferase reporter gene were used. For the luciferase assay, cells were seeded in 96-well plates at optimized density to ensure they were in logarithmic growth phase at stimulation. After overnight attachment, cells were starved for 4 hours, followed by stimulation with recombinant human GDF15 (50 ng / mL) with or without GDF15 microbinding agent or control reagent. Unless otherwise specified, GDF15 should be pre-incubated with the microbinding agent at room temperature for 30 minutes before cell addition. Six hours after stimulation, luciferase activity was measured using the luciferase assay kit according to the manufacturer's instructions. Luminescence intensity was recorded on a microplate reader, and raw values ​​were normalized to the corresponding solvent-treated control. Data are expressed as relative luciferase activity, reflecting the level of SRE-driven transcription downstream of the GDF15-GFRAL signal.

[0085] 12. Competitive binding assay based on ELISA The inhibitory effect of the GDF15-GFRAL interaction was quantified using a competitive binding assay in ELISA format. Brief procedure: A binding chaperone (GFRAL extracellular domain) was immobilized on a high-binding plate, followed by incubation with GDF15 in the presence of serially diluted micro-binding agents. The binding of GDF15 was detected using the appropriate detection antibody, and the inhibition curve was fitted using a four-parameter logistic model to calculate the IC50. 50 Value (logarithmic scale). In charge-complementary rescue experiments, the IC50 values ​​of wild-type and mutant combinations of GDF15 and the microbinding agent were determined. 50 The values ​​show that the charge-complementary double mutants exhibit partial recovery of neutralizing potency compared to the single mutants.

[0086] 13. Laboratory mice All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) and conducted strictly in accordance with institutional guidelines. Female or male mice (8 weeks old) were used uniformly, with strains matched to tumor cell lines (MC38 / LLC1 corresponding to C57BL / 6 mice; A20 corresponding to BALB / c mice). All mice used in this study were purchased from Charles River Laboratories and housed in a specific pathogen-free (SPF) environment.

[0087] 14. Tumor inoculation and monitoring Cells in the logarithmic growth phase were collected, washed, and resuspended in sterile PBS. Subcutaneous tumors were established by lateral ventral injection of 100 µL of cell suspension. MC38 cells: 2 × 10⁶ cells per mouse. 6 LLC1 cells: 1 × 10-1 per mouse; 6 A20 cells: 1 × 10⁶ cells per mouse; 6 Measure tumor size 2-3 times per week using a digital caliper. The tumor volume is calculated using the formula (length × width²) / 2. Simultaneously measure body weight at the same frequency, expressing it as an absolute change in body weight or as a percentage change relative to baseline (on the randomization day).

[0088] 15. Cachexia Intervention and Experimental Endpoint Mice were randomized to groups when tumors were palpable or reached a predetermined volume (typically 50–150 mm³). Microbinding agents (and corresponding controls) were administered every two days. Cachexia-related endpoints included progressive weight loss, clinical score, and survival. Predefined humanitarian endpoints were used (tumor volume >2000 mm³, weight loss >25%, severe morbidity). Kaplan-Meier curve analysis was employed for survival.

[0089] 16. Tumor sample processing for flow cytometry Tumors were excised at designated time points and immediately placed in pre-chilled RPMI-1640 medium. The tumor tissue was finely minced using sterile scissors and digested with gentle shaking at 37°C for 30 minutes in RPMI-1640 medium containing type IV collagenase (1 mg / mL) and DNase I (100 U / mL). After digestion, the cells were mechanically dissociated by gentle pipetting and filtered through a 70 µm cell sieve to obtain a single-cell suspension. Cells were washed with ice-cold staining buffer (PBS supplemented with 2% fetal bovine serum), and erythrocytes were lysed using potassium ammonium chloride (ACK) lysis buffer if necessary. Cell counts and viability assessment were performed on the cell suspension prior to staining.

[0090] Surface staining: Cells were incubated with Fc receptor blocking antibody (anti-CD16 / 32) at 4°C for 10 minutes, followed by incubation with fluorescein-conjugated surface marker antibodies (including CD8α) at ​​4°C in the dark for 30 minutes. All flow cytometry antibodies were purchased from BioLegend.

[0091] Intracellular cytokine staining: Cells were restimulated in vitro with phorbol 12-myristate-13-acetate (PMA, 50 ng / mL) and iomycin (500 ng / mL), while protein transport inhibitors (brefeldsin A or monensin) were added, and incubated at 37°C for 4 hours. Cells were fixed and permeabilized using a commercial fixation / permeabilization kit according to the manufacturer's instructions, followed by intracellular staining with BioLegend's anti-IFN-γ and granzyme B antibodies. Data were acquired by flow cytometry and analyzed using FlowJo software. Flow cytometry gating strategy: Single cells, live cells, lymphocytes, and CD8+ were screened sequentially. + T cell and cytokine-positive population.

[0092] 17. Immunofluorescence staining Tumor tissue was collected and fixed (either by 4% paraformaldehyde fixation followed by cryoprotection, or formalin fixation and paraffin embedding, depending on the experimental protocol), and sections were prepared (usually 5-10 µm thick). Antigen retrieval was performed if necessary. Sections were blocked with blocking solution (5% normal serum + 0.1% Triton X-100), incubated overnight at 4°C with anti-CD8α primary antibody (CST catalog number 85336), washed, and then incubated with fluorescein-conjugated secondary antibody. Cell nuclei were stained with DAPI. Images were acquired using a fluorescence microscope or confocal microscope under identical exposure conditions for each group. CD8 infiltration was quantified using a blinded method, with indicators including CD8... + Area proportion and / or CD8 per field of view + Cell count.

[0093] 18. Statistical Analysis Data are expressed as mean ± standard deviation (sd) unless otherwise specified. Comparisons between two groups were performed using the two-tailed Student's t-test; comparisons among multiple groups were performed using one-way or two-way ANOVA with appropriate multiple comparison corrections. Survival analysis was performed using the log-rank test (Mantel-Cox test). IC 50 The value is obtained through nonlinear regression (four-parameter logistic model).

[0094] II. Experimental Results 1. Structure-guided de novo design of micro-binding agents targeting the GDF15-GFRAL interface To effectively block GDF15-GFRAL signal transduction, we first performed structure-guided analysis of the GDF15-GFRAL interaction interface using existing high-resolution structural information. Analysis of the GDF15-GFRAL complex (GDF15: green, GFRAL: cyan) revealed a well-defined receptor-binding surface on GDF15, characterized by a concave topology and aggregated interaction hotspots, indicating that this surface is suitable for the design of targeting binders. Figure 1 A, B). Mapping this interface structure onto the experimentally determined GDF15 structure (PDB ID: 5VZ4) allows for the precise identification of key residues (purple) involved in receptor binding, laying the foundation for subsequent computational design work. Figure 1 C).

[0095] Subsequently, we established a combined computational pipeline for generating de novo micro-binding agents selectively targeting the GDF15-GFRAL interaction interface. AlphaFold-based predictions were used for structural analysis and complex modeling, providing a consistent reference standard for the interface geometry. Main-chain structures were generated via RFdiffusion, enabling the construction of diverse binding scaffolds geometrically complementary to the target surface. Candidate main-chain structures were then sequence-optimized using ProteinMPNN to obtain sequences predicting stable design binding conformations. To ensure structural realism and binding feasibility, design schemes were screened based on multiple criteria, including prediction alignment error (PAE) and structural deviation indices.

[0096] Simultaneously, we implemented an iterative design process that combined mainchain generation, sequence co-design, and structure prediction to progressively optimize microbinding agent candidates. The monomer stability and interface compatibility of each design path were evaluated using AlphaFold-based structure prediction, thus eliminating unstable or poorly folded candidates as early as possible. This multi-stage screening strategy yielded a set of highly reliable microbinding agent designs predicted to bind to the GDF15 receptor-binding surface with high structural complementarity, laying the foundation for downstream experimental validation.

[0097] 2. Computational screening and structural characterization of high-reliability GDF15 micro-binders Following structure-guided targeted design of the GDF15-GFRAL interface, we generated a large number of de novo micro-binding agent candidates through an integrated computational design pipeline. To systematically enrich structurally stable and interface-compatible binders, the candidates underwent a multi-stage screening process, including criteria such as secondary structure complexity, monomer folding confidence, predicted composite quality, and energy advantage. Figure 2A). Specifically, designs containing three or more secondary structural elements are retained to favor compact and stable folding. Poorly folded candidates (pLDDT>80) are then eliminated through monomer screening based on EsmFold. The complex form with GDF15 is further evaluated using AlphaFold3, and screening is conducted based on local confidence (complex pLDDT>80) and interface geometry (predicted alignment error PAE<10). Finally, designs are ranked according to predicted binding energies, retaining candidates with the most favorable interaction energies (ΔddG<-30).

[0098] This tiered screening strategy yielded a subset of highly reliable micro-binding agents that exhibited consistent structural quality across multiple metrics. The top-ranked designs demonstrated high predictive folding confidence and interface accuracy, with representative candidates showing an average pLDDT value of approximately 90, low prediction alignment error, and minimal deviation between the designed and predicted structures. Figure 2 (B, DF). It is worth noting that the consistency of multiple independent quality metrics indicates that these designs have stable sequence-structure compatibility, rather than being overfitted to a single evaluation criterion.

[0099] Structural analysis of the top-ranking microbinding agents revealed well-defined binding modes on the GDF15 receptor interaction surface. Representative AlphaFold3 predicted complexes showed that different microbinding agents docked to the target interface with high shape complementarity and persistently bound key hotspot residues. Figure 2 C, GDF15 is light blue, candidate micro-binding agents are green; key interacting residues are highlighted in purple. Despite the diversity of main chain topologies, these designs are all focused on the same binding surface, indicating that the computational process successfully captured the main geometric and chemical features required for GDF15 recognition.

[0100] Further sequence variability analysis of the high-confidence design scheme revealed a conservative positional preference between the micro-binding core region and the interaction interface contact region. Figure 2 (G) Sequence identification maps from the screened design set show a high enrichment of specific residue types at structurally confined sites, consistent with the stability of the designed folds and the optimization of interfacial interactions. Conversely, solvent-exposed regions exhibit greater sequence diversity, indicating that these regions are tolerant of variation without compromising structural integrity.

[0101] In summary, these results demonstrate that our combinatorial design and screening process can efficiently enrich de novo-designed microbinding agents. These computationally selected microbinding agents exhibit high structural realism, favorable binding energies, and concentrated recognition capabilities on the GDF15 receptor binding surface. The microbinding agent library obtained through this computational screening lays the foundation for subsequent experimental validation and functional characterization.

[0102] 3. Experimental verification and identification of high-affinity GDF15 micro-binder To experimentally validate the computationally designed GDF15 microbinding agent, we expressed a screened candidate library on the surface of yeast cells and evaluated the target binding capacity and surface expression level by flow cytometry. Yeast surface display showed a well-defined clonal population exhibiting strong GDF15 binding capacity and high surface expression levels, effectively distinguishing the functional binder from poorly expressed or non-binding variants. Figure 3 A).

[0103] Sequencing of the enriched yeast population revealed a highly uneven distribution of microbinding clones, with only a few designs dominating the screening population. Figure 3 B). Notably, several top-ranking designs (including clones 05_41_14 and 05_25_1) constitute a significant proportion of the entire clone population, indicating that specific micro-binding agent structures are highlighted based on selection pressure.

[0104] To further characterize the biochemical properties of these candidate micro-binding agents, representative binding agents were recombinantly expressed and purified to homogenate. Size exclusion chromatography was used to verify their purity. Figure 3 (C, D). Both binders eluted with a single symmetrical peak, consistent with the monomer form, indicating that they were well folded and had solution stability.

[0105] Subsequently, the binding kinetics of the micro-binders were measured using surface plasmon resonance (SPR) technology. Both candidate micro-binders (05_41_14 and 05_25_1) bound GDF15 with sub-nanomolar affinity, exhibiting kinetic characteristics of rapid binding and slow dissociation. Figure 3 E, F). The equilibrium dissociation constant (K_D) of clone 05_41_14 (SEQ ID NO:1) is 2.5 × 10⁻⁶. - ¹ 0 M, the K_D of clone 05_25_1 is 3.1×10 - ¹ 0 M, confirming that de novo-designed micro-binding agents can achieve high affinity recognition of GDF15.

[0106] 05_41_14 sequence: SAKESAAAAILDFGKKVGKEEEAEKAAKEVLEAESREEAIRIAQEALKKME (SEQID NO: 1) Structural analysis of the micro-binding agent-GDF15 complex predicted by AlphaFold3 showed that the designed binding mode was highly consistent with the predicted binding mode. Figure 3 Both microbinding agents bind to the receptor-binding surface on GDF15 with highly shape complementary properties, and the interfacial residues are consistently localized. Quantitative analysis showed low root mean square deviation (RMSD≈1 Å), low prediction alignment error (PAE<6), and high pLDDT score (>90), supporting the structural realism of the design.

[0107] Finally, circular dichroism spectroscopy analysis showed that the micro-binder has a stable secondary structure within a certain temperature range. Figure 3 I). The pyrolysis folding curves show that there is almost no loss of structural integrity at temperatures up to 95°C, highlighting the excellent thermal stability of the candidate micro-binder.

[0108] In summary, these results demonstrate that computational prediction can generate de novo designed micro-binding agents that are not only structurally accurate but also biochemically stable, capable of binding GDF15 with picomolar affinity, providing a solid foundation for downstream functional and in vivo studies.

[0109] 4. Micro-binders competitively block the GDF15-GFRAL interaction through well-defined interfacial interactions. Given the high affinity and structural stability of the lead micro-binding agents, we further investigated whether these molecules could directly interfere with the GDF15-GFRAL interaction. AlphaFold3's predicted structural superposition of the complex showed that both micro-binding agents bound to the receptor-binding region of GDF15, occupying the interface where GFRAL interacts. Figure 4 A, GDF15 is shown in light blue, GFRAL in pink, micro-binding agent 05_41_14 in yellow, and micro-binding agent 05_25_1 in green. The predicted binding patterns show extensive spatial overlap between the micro-binding agents and GFRAL, suggesting that the micro-binding agents may prevent GFRAL from binding to its receptor through a competitive inhibition mechanism.

[0110] To quantitatively assess the function-blocking effect, we measured the ability of the micro-binding agents to inhibit GDF15-induced signal transduction using a dose-response assay. Both 05_41_14 and 05_25_1 effectively inhibited GDF15 activity in a concentration-dependent manner, with a half-maximal inhibitory concentration (IC50) of [value missing]. 50 ) is in the lower nM range ( Figure 4 B). Conversely, residue mutations at the predicted interfaces of GDF15 or microbinding agents significantly reduced repressive efficacy, confirming the importance of the designed interaction interfaces.

[0111] To gain a detailed understanding of the molecular determinants of the binding interface, we performed saturation mutations on the interface residues of GDF15 and evaluated their impact on microbinding. Heatmap analysis revealed the presence of different residue clusters whose mutations selectively impaired the binding of 05_41_14 or 05_25_1, indicating that their binding sites on GDF15 partially overlap but are not entirely identical. Figure 4 C and D (red indicates reduced binding affinity, blue indicates extremely low binding affinity). Notably, mutations in key hotspot residues lead to loss of repression, highlighting their central role in microbinding.

[0112] The reverse saturation mutation of the micro-binding interface residues further validated the designed binding mode. Figure 4 (E, F). Mutations disrupting the core interface lead to a significant reduction in inhibitory activity, while substitutions at peripheral sites are more readily tolerated. These data suggest that microbinding agents rely on a well-defined set of interfacial residues to achieve high-affinity binding and functional blockade.

[0113] To further verify whether the neutralizing effect of the microbinding agent depends on specific electrostatic interactions predicted at the interface, we designed charge-reversal mutations on GDF15 and the microbinding agent, and quantified the functional neutralization effect using ELISA-based competitive binding assays. Single charge-reversal mutations on either GDF15 or the microbinding agent significantly reduced the neutralizing efficacy, resulting in IC50. 50 The value shifted towards higher concentrations ( Figure 4 (G, H). It is noteworthy that introducing charge-complementary rescue mutations, i.e., pairing the GDF15 mutation with a corresponding compensating mutation on the microbinding agent, can partially restore the neutralizing potency. Compared to non-complementary single-mutation combinations, IC50... 50 Value shifted to the left ( Figure 4 (G, H). This “charge exchange” rescue provides strong functional evidence that the micro-binding agent binds to GDF15 through a pre-determined, structurally well-defined interface, rather than non-specifically, and supports its role at the GDF15-GFRAL interacting surface via a competitive inhibition mechanism.

[0114] 5. Micro-binding agent-mediated neutralization of GDF15 inhibits downstream signaling and reverses cancer-related cachexia in vivo. To determine whether the de novo GDF15 microbinding agent could functionally inhibit downstream signaling, we first examined the activation of the classical GDF15-GFRAL pathway in vitro. Dose-response curves showed that the microbinding agent (05_41_14) had higher binding affinity and a wider concentration window; therefore, we used 05_41_14 for subsequent biological effect validation. Recombinant GDF15 treatment strongly induced ERK1 / 2 phosphorylation, while simultaneous treatment with the anti-GDF15 antibody ponsegromab (catalog number: GM-Tg-mg-T33245-Ab-2, purchased from Genemedi Biological Technology CO., LTD.) or the designed microbinding agent (05_41_14) significantly reduced pERK1 / 2 levels without affecting total ERK expression. Figure 5 A). These data indicate that the microbinding agent (05_41_14) effectively inhibits proximal signaling of GDF15. Consistent with the inhibition of ERK pathway activation, GDF15 stimulation strongly upregulated early response genes, including FOS, JUN, and EGR1. Real-time quantitative RT-PCR analysis showed that both ponsegromab and the microbinding agent significantly attenuated the GDF15-induced transcriptional response, restoring expression levels to near baseline. Figure 5 B). These findings confirm that the microbinding agent can effectively block GDF15 signaling at both the signal transduction and transcriptional levels.

[0115] We measured the neutralizing potency of the microbinding agent using an ELISA-based competitive binding assay to assess its inhibitory effect on the GDF15-GFRAL interaction. The microbinding agent 05_41_14 exhibited potent, dose-dependent inhibition of downstream signaling. Figure 5 (C), supporting its efficacy as a functional GDF15 antagonist.

[0116] We then evaluated the physiological responses of model mice after neutralizing GDF15 in several tumor-associated cachexia models. In MC38 colon tumor-bearing mice expressing human GDF15, administration of the GDF15 microbinding agent 05_41_14 significantly slowed weight loss in the model mice, restoring the weight change trajectory to levels close to those of control MC38 tumor-bearing mice that do not express GDF15. Figure 5 D). Similar protective effects were also observed in LLC1 lung cancer and A20 B-cell lymphoma models, indicating that microbinding agent-mediated cachexia reversal is not limited to a single tumor type. Figure 5F, H). Importantly, reversal of cachexia improved survival in tumor-bearing mice. Kaplan-Meier analysis showed that tumor-bearing mice expressing GDF15 had significantly lower survival rates compared to controls, while GDF15 microbinding therapy significantly prolonged survival in all three tumor models (F, H). Figure 5 (E, G, I). Notably, a significant proportion of mice treated with the microbinding agent remained alive at the experimental endpoint, highlighting the lasting physiological benefits of GDF15 neutralization.

[0117] In summary, these results demonstrate that de novo-designed GDF15 microbinding agents can effectively inhibit GDF15-GFRAL signaling, reverse tumor-induced cachexia, and confer a survival advantage in various cancer models.

[0118] 6. GDF15 neutralizes via CD8 + T-cell-dependent mechanisms restore sensitivity to PD-1 antibody therapy Given recent evidence linking GDF15 to resistance to immune checkpoint inhibitors, we next investigated whether neutralizing GDF15 with the microbinding agent 05_41_14 could restore sensitivity to PD-1 antibody therapy in a GDF15-driven tumor model. Using an MC38 tumor-bearing mouse model expressing human GDF15 (Charles River Laboratory), we observed that PD-1 antibody [29F.1A12] (catalog number: CP159, purchased from BioX Cell) monotherapy only moderately delayed tumor growth, consistent with immunotherapy resistance. Conversely, combination therapy with the PD-1 antibody and the GDF15 microbinding agent 05_41_14 significantly inhibited tumor growth, demonstrating significantly superior efficacy compared to either monotherapy. Figure 6 A).

[0119] This synergistic anti-tumor effect translated into significant survival benefits. Kaplan-Meier analysis showed that mice treated with a combination of PD-1 antibody and GDF15 microbinding agent had significantly longer survival rates compared to the solvent control, PD-1 antibody monotherapy, or microbinding agent monotherapy groups. Figure 6 B). Notably, most mice in the combined treatment group were still alive at the experimental endpoint, indicating that it has a durable therapeutic effect.

[0120] To investigate the immunological basis of this synergistic effect, we analyzed tumor-infiltrating lymphocytes using flow cytometry. Compared with all other treatment groups, the combination therapy (PD-1 antibody and GDF15 microbinding agent 05_41_14) significantly increased CD8+ in the tumor microenvironment. + The proportion of T cells ( Figure 6 C). Furthermore, CD8 in tumors receiving combination therapy+ T cells exhibited enhanced function, characterized by increased expression of granzyme B (GZMB) and interferon-γ (IFNγ). Figure 6 (D, E). Consistent with these findings, immunofluorescence staining of tumor sections showed that, compared with PD-1 antibody monotherapy, the combination therapy group had significantly higher levels of CD8+ in tumors. + Increased T cell infiltration ( Figure 6 F), supporting GDF15 neutralization can promote cytotoxic T cell infiltration into the tumor.

[0121] To directly verify whether the anti-tumor effect of combination therapy depends on CD8 + For T cells, we performed an antibody-mediated CD8 depletion assay. Under CD8 depletion conditions, the therapeutic benefit of combined therapy with PD-1 antibody and GDF15 microbinding agent was completely eliminated, and tumor growth kinetics were not significantly different from the control group. Figure 6 G). These data indicate that the recovery of PD-1 antibody efficacy neutralized by GDF15 is strictly dependent on CD8. + T cells.

[0122] To elucidate the metabolic and immunological effects of GDF15 neutralization, we constructed a schematic diagram summarizing how tumor-derived GDF15 affects systemic physiological functions and anti-tumor immunity, and how de novo-designed micro-binding agents intervene in this process. Figure 6 H). In tumors with high GDF15 expression, circulating GDF15 levels are elevated, suppressing appetite and promoting weight loss through its brainstem receptor GFRAL, and driving cancer-related cachexia by activating downstream MAPK signaling. Simultaneously, tumor-derived GDF15 acts locally in the tumor microenvironment, limiting CD8... + T-cell infiltration leads to decreased intratumoral effector function and impaired sensitivity to immune checkpoint blockade therapy. By directly targeting the GDF15-ligand binding interface with a de novo-designed micro-binding agent, GDF15-GFRAL signaling can be effectively blocked, alleviating cachexia-related physiological decline. Simultaneously, GDF15 neutralizes and restores CD8... + The increased ability of T cells to enter the tumor leads to increased infiltration of cytotoxic T cells producing granzyme B and IFN-γ within the tumor. This weakens the immunosuppressive microenvironment of the tumor and reverses the resistance of GDF15-overexpressing tumors to immune checkpoint inhibitors such as PD-1 antibodies.

[0123] In summary, these results indicate that GDF15-driven cachexia is associated with suppressed antitumor immunity and resistance to immune checkpoint inhibitors, and that de novo-designed GDF15 microbinding agents can be activated via CD8. + T-cell-mediated mechanisms restore sensitivity to PD-1 antibody therapy.

[0124] The above description of the embodiments is only for understanding the method and core ideas of the present invention. It should be noted that those skilled in the art can make various improvements and modifications to the present invention without departing from the principles of the invention, and these improvements and modifications will also fall within the protection scope of the claims of the present invention.

Claims

1. A micro-binder, characterized in that, The sequence of the micro-binding agent is shown in SEQ ID NO:

1.

2. A nucleic acid molecule, characterized in that, The nucleic acid molecule encodes the micro-binding agent as described in claim 1.

3. A carrier, characterized in that, The carrier comprises the nucleic acid molecule as described in claim 2.

4. A host cell, characterized in that, The host cell comprises the nucleic acid molecule of claim 2 or the vector of claim 3.

5. A drug, characterized in that, The drug comprises the micro-binding agent of claim 1, the nucleic acid molecule of claim 2, the carrier of claim 3, or the host cell of claim 4.

6. The drug according to claim 5, characterized in that, The drug also includes a PD-1 antibody.

7. Any of the following applications: (1) The use of GDF15 inhibitors and PD-1 antibodies, or GDF15 inhibitors in the preparation of drugs for treating GDF15-related diseases; (2) Application of GDF15 inhibitors in the preparation of drugs that enhance the sensitivity to PD-1 antibody therapy; (3) The use of the micro-binding agent of claim 1, the nucleic acid molecule of claim 2, the carrier of claim 3, the host cell of claim 4, or the drug of claim 5 in the in vitro non-diagnostic detection of GDF15 or in the preparation of products for detecting GDF15; The GDF15 inhibitor is selected from the micro-binding agent of claim 1, the nucleic acid molecule of claim 2, the vector of claim 3, the host cell of claim 4, or the drug of claim 5.

8. The application according to claim 7, characterized in that, The GDF15-related diseases include cancer and its associated cachexia, metabolic diseases, and inflammatory diseases.

9. The application according to claim 8, characterized in that, The cancer is selected from one or more of the following: colon cancer, lung cancer, and lymphoma.

Images