Protein degradation targeting chimera for EGFR-VHL system and application thereof

Abstract

The invention discloses a protein degradation targeting chimera for an EGFR-VHL system. The protein degradation targeting chimera comprises an EGFR binding molecule and a VHL binding molecule, the EGFR binding molecule is any one of proteins as shown in SEQ ID NO.1 to SEQ ID NO.3; the VHL binding molecule is any one of proteins as shown in SEQ ID NO. 4 to 37. The invention also discloses application of the protein degradation targeting chimera in preparation of drugs targeting EGFR and VHL. The invention provides a set of complete calculation design framework, a diversified candidate binding protein library is generated, and finally three high-quality EGFR binding proteins and 34 high-quality VHL binding proteins are obtained and show excellent prediction evaluation indexes. The invention establishes a general framework which is preliminarily verified and is suitable for the rational design of the next generation of protein PROTAC, and lays a foundation for a targeted protein degradation treatment strategy of tumors and other diseases.

Description

Technical Field

[0001] This invention relates to a protein degradation-targeting chimera for the EGFR-VHL system and its application, belonging to the field of protein degradation-targeting chimera technology. Background Technology

[0002] Protein degradation-targeting chimeras (PROTACs) are heterobifunctional molecules that bind to both a target protein ligand and an E3 ubiquitin ligase ligand via a linker. Their mechanism of action involves inducing the formation of a ternary complex between the target protein and the E3 ligase, leading to ubiquitination of the target protein followed by degradation by the 26S proteasome. PROTACs represent an emerging therapeutic strategy that selectively degrades pathogenic proteins by recruiting the endogenous ubiquitin-proteasome system, rather than simply inhibiting their function. Unlike traditional small-molecule inhibitors that only block protein activity, PROTACs catalytically induce target protein degradation, potentially improving efficacy, reducing drug resistance, and expanding access to previously considered "undruggable" targets. The effectiveness of PROTACs depends on the formation of a stable ternary complex. A key challenge in developing protein-based PROTACs is obtaining protein-binding molecules that simultaneously possess high affinity and high specificity, and can be geometrically assembled into feasible ternary complexes.

[0003] Epidermal growth factor receptor (EGFR) is a classic oncogenic driver, and its targeted therapy is often limited by acquired resistance mutations. The von Hippel-Lindau (VHL) protein, due to its well-defined binding pocket structure and favorable degradation kinetics, has become one of the most commonly used E3 ligases in PROTAC design. Summary of the Invention

[0004] In view of the above-mentioned prior art, the present invention provides a protein degradation-targeting chimera for the EGFR-VHL system and its application.

[0005] This invention is achieved through the following technical solution:

[0006] A protein degradation-targeting chimera for the EGFR-VHL system, comprising an EGFR-binding molecule and a VHL-binding molecule; wherein the EGFR-binding molecule is any one of the proteins shown in SEQ ID NO. 1 to 3; and the VHL-binding molecule is any one of the proteins shown in SEQ ID NO. 4 to 37.

[0007] Furthermore, the EGFR-binding molecule is the protein shown in SEQ ID NO.1, and the VHL-binding molecule is the protein shown in SEQ ID NO.4.

[0008] The application of the protein degradation targeting chimera in the preparation of drugs targeting EGFR and VHL.

[0009] This invention proposes a complete computational design framework for novel protein-binding molecules in EGFR and VHLE3 ligase systems. By integrating state-of-the-art deep learning tools such as ColabDesign, ProteinMPNN, and AlphaFold-Multimer, this invention generates a diverse library of candidate binding proteins and employs a rigorous multi-index screening process to systematically select designs that exhibit excellent performance in both predicted structure and energetics. This process ultimately yielded three high-quality EGFR-binding proteins and 34 high-quality VHL-binding proteins, all demonstrating superior predictive performance metrics. Based on these optimal designs, this invention further assembles the best EGFR-binding and VHL-binding proteins into a complete protein PROTAC construct in a computer, and uses this optimal construct as the experimental subject for subsequent cell experiments. The results show that this optimal construct can induce efficient and specific degradation of EGFR in cell models and significantly affect cell function. These results demonstrate that the computational design strategy proposed in this study can be used to construct functionally active protein-targeted degradation therapeutic molecules, providing a powerful tool for the development of targeted protein degradation drugs. In summary, this invention establishes a preliminarily validated general framework applicable to the rational design of next-generation protein-based PROTACs, laying the foundation for targeted protein degradation therapy strategies in the fields of oncology and other diseases.

[0010] The various terms and phrases used in this invention have their general meanings known to those skilled in the art. Attached Figure Description

[0011] Figure 1 Statistical analysis results and comparison of the average pLDDT of EGFR-bound molecules and VHL-bound molecules.

[0012] Figure 2 Statistical analysis results and comparison of the average pLDDT of EGFR-bound molecules and VHL-bound molecules.

[0013] Figure 3 Statistical analysis and comparison of binding energies between EGFR-binding molecules and VHL-binding molecules.

[0014] Figure 4 Statistical analysis results and comparison of the average shape complementarity between EGFR-binding molecules and VHL-binding molecules.

[0015] Figure 5Statistical analysis results and comparison of the average number of interfacial hydrogen bonds between EGFR-bound molecules and VHL-bound molecules.

[0016] Figure 6 Statistical analysis and comparison of the average interface SASA% of EGFR-binding molecules and VHL-binding molecules.

[0017] Figure 7 Statistical analysis results and comparison of the average dSASA of EGFR-bound molecules and VHL-bound molecules.

[0018] Figure 8 The three-dimensional front view of the structural model of the EGFR–PROTAC–VHL ternary complex is shown. In the figure, the EGFR-binding molecule is represented in blue and the VHL-binding molecule is represented in green. The two are linked by optimized linker peptides to form stable complexes with EGFR (gray) and VHL (brown), respectively.

[0019] Figure 9 The three-dimensional side view of the structural model of the EGFR–PROTAC–VHL ternary complex is shown. In the figure, the EGFR-binding molecule is represented in blue and the VHL-binding molecule is represented in green. The two are linked by an optimized linker peptide to form stable complexes with EGFR (gray) and VHL (brown), respectively.

[0020] Figure 10 Immunofluorescence analysis images of the control group and experimental group. HUVECs were transfected with the empty vector (pCDH-Vehicle) or the PROTAC expression vector (pCDH-Binder), respectively, and EGFR (red) and cell nucleus (DAPI, blue) were stained. GFP (green) was used as a transfection marker.

[0021] Figure 11 Quantitative analysis of EGFR fluorescence signals. Data are expressed as mean ± standard deviation (mean ± SD). ***P < 0.001 indicates a highly significant decrease in EGFR levels.

[0022] Figure 12 Results of Western blot experiments.

[0023] Figure 13 Quantitative results of EGFR band gray values ​​after GAPDH normalization. Data are expressed as mean ± standard deviation (mean ± SD). ***P < 0.001 indicates a highly significant decrease in EGFR levels.

[0024] Figure 14 Representative images from the HUVEC scratch healing experiment.

[0025] Figure 15Representative images of EdU proliferation assay staining.

[0026] Figure 16 : Summary bar chart of scratch healing rate at specified time points, where all data are expressed as mean ± standard deviation; ***P < 0.001, ****P < 0.0001.

[0027] Figure 17 : Percentage of EDU-positive cells in HUVECs, where all data are expressed as mean ± standard deviation; ***P < 0.001, ****P < 0.0001.

[0028] Figure 18 HUVEC cell proliferation rate calculated after CCK-8 staining. All data are expressed as mean ± standard deviation; ***P < 0.001, ****P < 0.0001.

[0029] Figure 19 Flow cytometry scatter plot.

[0030] Figure 20 :EDU- / EDU + Subpopulation fluorescence histograms, where all data are expressed as mean ± standard deviation; ***P < 0.001, ****P < 0.0001. Detailed Implementation

[0031] The present invention will be further described below with reference to embodiments. However, the scope of the present invention is not limited to the following embodiments. Those skilled in the art will understand that various changes and modifications can be made to the present invention without departing from the spirit and scope thereof.

[0032] Unless otherwise specified, the instruments, reagents, and materials used in the following embodiments are all conventional instruments, reagents, and materials already available in the prior art and can be obtained through legitimate commercial channels. Unless otherwise specified, the experimental methods and detection methods used in the following embodiments are all conventional experimental methods and detection methods already available in the prior art.

[0033] The experimental EGFR-VHL system utilizes a novel chimeric design based on the targeting protein degradation of the binding protein.

[0034] 1. Introduction

[0035] The key challenge in developing protein-based PROTACs lies in obtaining protein-binding molecules that simultaneously possess high affinity and high specificity, and can be geometrically assembled into feasible ternary complexes.

[0036] In recent years, breakthroughs in computational protein design, especially the rapid development of deep learning algorithms, have provided powerful new tools for developing protein-based PROTACs. Combining ColabDesign, used for novel backbone generation, with the high-precision sequence design model ProteinMPNN and the structure prediction engine AlphaFold-Multimer, allows for the rational design of protein-binding molecules with specific recognition spectra and ideal biophysical properties at the atomic level.

[0037] This invention constructs a de novo design computational pipeline for high-affinity protein-binding molecules targeting EGFR and VHL. By systematically exploring the protein design space and combining it with rigorous multi-parameter screening criteria, this invention obtains a set of high-quality binding proteins. Furthermore, the best-performing EGFR-binding molecule is assembled with a VHL-binding molecule into a functional PROTAC in a computer, and its ability to degrade EGFR is verified in cell experiments. This work provides a universal and robust platform for the development of next-generation protein-targeted degradation therapeutic molecules.

[0038] 2. Methods

[0039] 2.1 Computational Design and Modeling

[0040] The computational design process begins with the preparation of the target protein structure. The EGFR kinase domain (PDB: 2GS2) and the substrate recognition domain of the VHL protein (PDB: 4W9H) are selected as target regions for the binding molecule design. Subsequently, a four-stage, complex-oriented algorithm is used for de novo design. In the first stage, ColabDesign generates a novel protein backbone to fully cover diverse conformational spaces. In the second stage, ProteinMPNN is used for sequence optimization of the backbone; this deep learning model can design the expected high-fidelity folded amino acid sequence for a given backbone. To reduce the risk of nonspecific aggregation and incorrect disulfide bond formation, cysteine ​​residues are intentionally excluded during the design process.

[0041] After obtaining the initial design sequences, each design sequence was assembled with the corresponding target protein into binary complexes, and these complexes were then used to predict their structure and assess their quality using AlphaFold-Multimer. AlphaFold-Multimer provides a comprehensive set of metrics, including per-residue confidence (pLDDT) and interfacial prediction™ score (i_pTM). Subsequently, a rigorous screening process was implemented based on a multi-parameter comprehensive score; this screening process encompassed over 30 evaluation criteria, with key thresholds including: predicted average pLDDT > 0.85, i_pTM > 0.8, Rosetta binding energy < -40 REU, shape complementarity > 0.7, and at least 5 interfacial hydrogen bonds formed. Only candidates that met all screening criteria were included in subsequent assembly steps.

[0042] For the optimal EGFR and VHL binding molecules selected through screening, a complete protein PROTAC model was further assembled in a computer. Specifically, flexible linker peptides of different lengths [(GGGGS)n] and rigid linker peptides [(EAAAK)n] were used to link the EGFR-binding protein and the VHL-binding protein in the program, constructing a ternary complex model. The spatial configuration of the resulting complex was then analyzed, focusing on assessing whether it possessed a geometric conformation conducive to ubiquitin transfer and protein degradation, thereby confirming the structural feasibility of the designed PROTAC.

[0043] 2.2 Cell Culture and Plasmid Transfection

[0044] Human umbilical vein endothelial cells (HUVECs) were cultured in Dulbecco's Modified Eagle Medium (DMEM) containing 10% fetal bovine serum (FBS) and a 1% penicillin-streptomycin-gentamicin antibiotic mixture. Cells were maintained in 10 cm culture dishes at 37°C, 5% CO2, and saturated humidity.

[0045] In the transfection experiments, HUVECs were cultured to approximately 50%–60% confluence. Two plasmids were used: one expressing a binding protein-binding protein fusion construct (a construct assembled from the best-performing EGFR-binding molecule and VHL-binding molecule, i.e., the proteins shown in SEQ ID NO.1 and SEQ ID NO.4), and the other serving as an empty vector control. For each transfection, 10 μg of plasmid DNA was dissolved in 500 μL of... In the buffer solution, vortex for 10 seconds, then add 20 μL. The transfection reagent was incubated at room temperature for 10 min to form a transfection complex, and then added to the cells dropwise. The culture medium was replaced with fresh medium 24 h after transfection; 48 h after transfection, puromycin was added to a concentration of 0.5 μg / mL, and the cells were cultured for another 48 h to screen for stable transfected cell lines. Transfection efficiency was confirmed by observing GFP expression under a fluorescence microscope; the results showed that over 95% of the cells expressed GFP reporter protein.

[0046] 2.3 Immunofluorescence staining

[0047] In immunofluorescence experiments, stably transfected HUVECs (pCDH-Binder group, pCDH-Vehicle group) and untransfected control groups were seeded on 6-well plate slides and cultured to 70%–80% confluence before treatment. Cells were first washed with PBS and fixed with 4% paraformaldehyde for 20 min. Subsequently, cells were permeabilized with 0.3% Triton X-100, and non-specific binding sites were blocked with PBS containing 3% BSA at room temperature for 2 h.

[0048] Subsequently, the cells were incubated overnight at 4°C in anti-EGFR primary antibody solution (YA468, MCE, company, dilution 1:100). After washing, the cells were incubated at room temperature for 60 min in the dark, and then Alexa... 594-labeled goat anti-rabbit IgG H&L secondary antibody (ab150080, Abcam, dilution 1:100). Finally, ProLong containing DAPI was used. TM Gold anti-quenching mounting medium (S36939, Invitrogen) was used for mounting the slides for nuclear counterstaining. All images were acquired under a fluorescence microscope.

[0049] 2.4 Western Blot

[0050] To extract total protein, HUVEC cells were lysed in RIPA lysis buffer (Catalog No. PC101, Epizyme Biotech) containing protease and phosphatase inhibitors (Catalog No. P1260, Solarbio, Beijing, China). Protein concentration was determined using a BCA protein quantification kit (Catalog No. P0009, Beyotime, Haimen, China). Equal amounts of protein (20 μg) were separated by electrophoresis on a 10% polyacrylamide gel and subsequently transferred to a PVDF membrane (Catalog No. ISEQ00010, Millipore Sigma, Burlington, MA, USA).

[0051] To reduce nonspecific binding, the PVDF membrane was first blocked in TBST solution containing 5% skim milk powder, and then incubated overnight at 4°C in primary antibody solution, which included anti-EGFR antibody (catalog number HY-P80648, MedChemExpress; dilution 1:1000) and anti-GAPDH antibody (catalog number HY-P80137, MedChemExpress; dilution 1:2000). After incubation with HRP-labeled secondary antibody (catalog number HY-P8001, MedChemExpress; dilution 1:5000), the membrane was developed using an ECL chemiluminescence kit (SQ201, Epizyme Biotech, Cambridge, MA, USA), and the band signals were recorded.

[0052] 2.5 Scratch Healing Test

[0053] HUVEC cells were loaded at a rate of 2 × 10 5 Cells were seeded at a density of cells / well in 12-well plates and cultured for 48 hours to form a dense monolayer. Then, a straight scratch was made in the center of the cell layer using a sterile 10 μL pipette tip, and the cells were washed with PBS to remove detached cells and debris.

[0054] Images were acquired at 0, 24, and 48 hours using an inverted optical microscope (Olympus IX-81, Olympus Optical Co., GmbH) and a CCD camera (XC-30, Olympus). All experiments were repeated in triplicate. Image acquisition was performed using AnalySISgetIT software (Olympus Soft Imaging Solutions, GmbH), and quantitative analysis of scratch area was performed using ImageJ software.

[0055] Cell migration rate (%Migration) is defined as the percentage of the initial scratch area covered by migrating cells at a specified time point, where A0 represents the initial scratch area at 0h, and At represents the remaining scratch area at t (24h or 48h). The migration rate is calculated using the following formula: %Migration = (A0 - At) / A0 × 100%.

[0056] 2.6 Cell viability assay (CCK-8 assay)

[0057] To evaluate the proliferation status of HUVEC cells after EGFR degradation, cell viability and proliferation capacity were assessed using a CCK-8 assay kit. Cells were seeded at a density of 5000 cells per well in 96-well plates and cultured for 12 h. The culture medium was then replaced with serum-free medium containing CCK-8 reagent (final concentration 0.5 mg / mL). After 3 h of incubation, absorbance was measured at 450 nm using a microplate reader (spaces between values ​​and units; microplate reader model: Bio-Rad Model 550, Bio-Rad, USA). The absorbance at 450 nm was positively correlated with the number of surviving cells.

[0058] 2.7 EdU proliferation assay and flow cytometry (FCM) analysis

[0059] To further assess cell proliferation levels, control, empty vector, and HUVEC cells transfected with pCDH-Binder plasmid were seeded at a density of 50,000 cells / well in 12-well plates and incubated in complete culture medium under standard conditions. After 24 h, EdU incorporation experiments were performed using an EdU cell proliferation kit (C0078S, Beyotime, Shanghai, China). In short, 50 mM EdU was added to the culture system, and after 4 h of incubation, cells were fixed, permeabilized, and stained with EdU according to the kit instructions. Subsequently, cell nuclei were stained with Hoechst 33342 (Sigma, 1 mg / mL) for 20 min. The proportion of cells containing EdU was counted and analyzed under a fluorescence microscope.

[0060] To further analyze the proliferation activity of HUVEC cells at the population level, cells were washed with PBS after EdU staining, and approximately 1×10⁶ cells were collected from each group. 6 Cells were resuspended in 200 μL PBS and immediately analyzed by flow cytometry to obtain EDU. + Distribution of / EDU-cell subsets.

[0061] 3. Results

[0062] 3.1 Design Process Performance and Binding Molecular Characterization

[0063] The computational design pipeline screened 5000 initial design trajectories targeting EGFR and VHL, respectively, and obtained 3 high-quality EGFR-binding proteins (amino acid sequences as shown in SEQ ID NO. 1, 2, and 3) and 34 high-quality VHL-binding proteins (amino acid sequences as shown in SEQ ID NO. 4–37). Specific information is shown in Tables 1 and 2. This corresponds to a design success rate of approximately 0.06% for EGFR and approximately 0.68% for VHL, reflecting the difference in the ease of de novo binding molecule design based on different target protein surface conformations. The selected EGFR-binding molecules showed excellent performance in terms of predicted structure and energy indices, with an average pLDDT of 0.88 ± 0.02 and an average Rosetta binding energy of -268.7 ± 20.1 REU; the VHL-binding molecules showed even more outstanding predictive indices, with an average pLDDT of 0.91 ± 0.02 and an average binding energy of -290.3 ± 17.7 REU. Both types of binding molecules exhibit high shape complementarity at their designed interfaces and form extensive hydrogen bond networks, suggesting that their binding interfaces are stable and highly specific.

[0064] Table 1

[0065]

[0066] Table 2

[0067]

[0068]

[0069]

[0070] 3.2 Statistical analysis combining molecular design characteristics

[0071] To gain a more detailed understanding of the design results, this invention conducted a statistical analysis of the key biophysical and structural indicators of the final screened binding molecules. This analysis compared the differences in the distribution of seven core parameters between the EGFR and VHL binding molecule populations, including: prediction confidence (mean pLDDT and mean i_pTM), binding energy (mean Binder Energy Score), shape complementarity (mean Shape Complementarity), number of interfacial hydrogen bonds (mean N_InterfaceHbonds), and interfacial solvent-accessible surface area characteristics (Average Interface SASA% and Average dSASA).

[0072] The relevant statistical summary results are as follows Figures 1 to 7As shown in the figure, the results reveal distinctly different characteristic distributions between the two types of binding molecules. Overall, VHL-binding molecules outperform EGFR-binding molecules in all metrics: they exhibit higher average confidence scores (pLDDT: 0.91 vs. 0.88; i_pTM: 0.80 vs. 0.74), more favorable binding energies (-290.3 REU vs. -268.7 REU), and a greater number of interfacial hydrogen bonds (10.6 vs. 8.2). Although both types of binding molecules meet stringent screening criteria, the statistical results suggest that VHL surfaces are more likely to acquire high-affinity de novo binding partners under this computational approach.

[0073] 3.3 PROTAC Assembly and Structural Modeling

[0074] The best-performing EGFR-binding molecule (the protein shown in SEQ ID NO.1) and the best-performing VHL-binding molecule (the protein shown in SEQ ID NO.4) were assembled into a complete PROTAC construct in a computer using flexible and rigid linker peptides, and the structure of the resulting ternary complex was modeled.

[0075] The structural model of the EGFR–PROTAC–VHL ternary complex is as follows: Figure 8 , Figure 9 As shown in the figure. Modeling results show that these ternary complexes generally possess geometric conformations favorable to ubiquitination reactions, and the linking peptides can effectively bridge two target proteins without introducing significant steric hindrance. Model predictions indicate that the PROTAC designed in this invention can stably induce the formation of the EGFR–PROTAC–VHL ternary complex, which is a key prerequisite for achieving targeted protein degradation.

[0076] 3.4 Experimental verification of PROTAC-mediated EGFR degradation

[0077] To verify the functional effects of the superior PROTAC candidate molecules, this invention conducted a series of experiments in human umbilical vein endothelial cells (HUVECs) to evaluate their ability to induce endogenous EGFR degradation in the cellular environment.

[0078] (1) First, the expression level of EGFR was observed at single-cell resolution using immunofluorescence microscopy. HUVECs were stably transfected with a plasmid expressing dominant PROTAC, which also co-expressed GFP as a transfection marker.

[0079] Immunofluorescence analysis images of the control group and experimental groups (pCDH-Binder group, pCDH-Vehicle group) are as follows: Figure 10 As shown, the quantitative analysis of EGFR fluorescence signals is as follows: Figure 11 As shown.

[0080] Figure 10 The results showed that, compared to surrounding untransfected cells, a significant decrease in EGFR-related red fluorescence signal was observed only in GFP-positive cells (cells successfully expressing PROTAC) in the pCDH-Binder group, suggesting targeted degradation. In contrast, no significant change in EGFR red fluorescence was observed in control cells transfected with the empty vector, regardless of whether the population was GFP-positive or GFP-negative. Meanwhile, Figure 11 The quantitative fluorescence results showed that, compared with the control group, EGFR-related signals were significantly decreased in PROTAC-expressing cells, with a highly statistically significant difference (P < 0.001). These results provide intuitive and clear evidence for the effective scavenging of endogenous EGFR by PROTAC in cells.

[0081] (2) To further corroborate the immunofluorescence results and quantitatively assess the degree of protein downregulation, the present invention performs Western Blot analysis on HUVEC cell lysates treated in the same manner.

[0082] HUVECs were transfected with either the empty vector (pCDH-Vehicle) or the PROTAC expression vector (pCDH-Binder). The lysates were analyzed by Western blotting to detect EGFR and GAPDH (internal control) protein levels. The results are as follows: Figure 12 As shown, the quantitative results of EGFR band gray values ​​after GAPDH normalization are as follows: Figure 13 As shown.

[0083] Figure 12 , Figure 13 The results showed that the intensity of the EGFR band was significantly lower in cells expressing the PROTAC construct compared to the empty vector control group and the untransfected group; while the expression of the internal control GAPDH was basically consistent among the groups, proving that the protein loading amount was comparable and the degradation effect had target specificity. The band grayscale analysis results showed that the decrease in total EGFR protein level induced by PROTAC was statistically significant (P < 0.001), further validating the immunofluorescence data from a quantitative perspective.

[0084] Based on the above experimental results, it can be confirmed that the computational design process of this study successfully obtained PROTAC molecules with strong degradation activity at the cellular level.

[0085] (3) Previous studies have shown that the proliferation and migration functions of endothelial cells largely depend on the normal expression of EGFR. To systematically evaluate the relationship between the degree of protein degradation and endothelial cell function, this invention conducted scratch healing assays, CCK-8 assays, and EdU proliferation assays on HUVEC cells under the same batch treatment conditions.

[0086] Representative images of the HUVEC scratch healing assay after transfection with the PROTAC expression vector (pcdh-binding-liu) are shown below. Figure 14 As shown, at 24 h and 48 h after scratch treatment, the migration ability of PROTAC-expressing cells was significantly lower than that of the empty vector control group and the untransfected group. Representative images of EdU proliferation assay staining are shown below. Figure 15 As shown in the staining results, the proliferation ability of cells expressing PROTAC was inhibited. A summary bar chart of scratch healing rates at specified time points is shown below. Figure 16 As shown, the proportion of EDU-positive cells in HUVEC cells is, for example, Figure 17 As shown, Figure 16 , Figure 17 Quantitative data further showed that the migration and proliferation activities of the PROTAC-treated group were significantly lower than those of the control group (P < 0.001). The HUVEC cell proliferation rate calculated after CCK-8 staining was as follows: Figure 18 As shown, the CCK-8 results also confirmed that the cell proliferation rate in the PROTAC-treated group was significantly reduced (P < 0.001).

[0087] In addition, flow cytometry analysis was performed on the HUVEC cell population treated with EdU. The flow cytometry scatter plot is shown below. Figure 19 As shown, EDU- / EDU + Subgroup fluorescence histograms as shown Figure 20 As shown. Under the condition of the same total number of events, it can be observed that EDUs expressing PROTAC... + The cell proportion decreased by approximately 26.9%, and the corresponding fluorescence intensity distribution shifted to the left. These multiple pieces of evidence from both the molecular and cellular functional levels collectively demonstrate that the PROTAC constructed in this study can not only effectively induce the degradation of the target protein EGFR, but also significantly inhibit the migration and proliferation of endothelial cells.

[0088] 4. Discussion

[0089] This study demonstrates the successful application of a complete and feasible computational framework for de novo design of protein-like PROTACs in the EGFR–VHL system. By integrating deep learning tools for backbone generation, sequence design models, and structure prediction engines, and supplementing them with a rigorous multi-parameter screening process, high-quality binding molecules were obtained between a challenging kinase target and a classic E3 ligase. Subsequent cell experiments validated that the PROTACs designed in this invention can achieve efficient and specific degradation of EGFR in cell models, thus proving that this computational workflow has the ability to translate in silico design into in vitro functional realization.

[0090] The difference in success rates between EGFR and VHL design activities suggests a crucial role of target protein surface topology in de novo design. Statistical analysis quantitatively supports this conclusion: VHL-binding molecules generally exhibit higher confidence scores and more favorable binding energies, indicating that the VHL substrate recognition interface is more receptive to novel binding partners under these design conditions. In contrast, the more restricted spatial structure of the EGFR kinase domain increases design difficulty, resulting in a lower yield of high-quality binding molecules. This finding provides important reference for subsequent design work, suggesting that target selection and interface hotspot definition are key factors in the success or failure of computational binding molecule design.

[0091] More importantly, the experimental results provide strong support for the potential therapeutic applications of the PROTAC designed in this invention. Significant EGFR degradation, confirmed by both immunofluorescence and Western blotting, is highly consistent with the structural and energetic characteristics predicted in the computational model. The control group demonstrates that EGFR degradation is strictly dependent on the expression of a specific chimeric PROTAC molecule, thus exhibiting high targeting specificity—a crucial characteristic for any potential therapeutic molecule.

[0092] 5. Conclusion

[0093] In summary, this invention constructed and validated a de novo protein-binding molecule computational design pipeline for PROTAC applications. This pipeline successfully generated high-affinity protein-binding molecules targeting EGFR and VHL, and assembled them into functional PROTACs capable of inducing potent and specific degradation of EGFR in cancer cells. This study not only provides a promising therapeutic candidate molecule but also establishes a robust design platform that can be extended to various disease-related protein targets, offering new insights for the development of protein drugs and targeted degradation technologies.

[0094] The above embodiments are provided to those skilled in the art to fully disclose and describe how the claimed implementations can be carried out and used, and are not intended to limit the scope of the disclosure herein. Modifications that will be obvious to those skilled in the art will be within the scope of the appended claims.

Claims

1. A protein degradation-targeting chimera for the EGFR-VHL system, characterized in that: It includes EGFR-binding molecules and VHL-binding molecules; the EGFR-binding molecule is any one of the proteins shown in SEQ ID NO. 1 to 3; the VHL-binding molecule is any one of the proteins shown in SEQ ID NO. 4 to 37.

2. The protein degradation-targeting chimera for the EGFR-VHL system according to claim 1, characterized in that: The EGFR-binding molecule is the protein shown in SEQ ID NO.1, and the VHL-binding molecule is the protein shown in SEQ ID NO.

4.

3. The use of the protein degradation targeting chimera according to claim 1 or 2 in the preparation of drugs targeting EGFR and VHL.

Images