Multilevel virtual screening method based on molecular docking and use of compounds screened thereby in preparation of anti-lung cancer drugs

Abstract

The present invention relates to a multi-level virtual screening method based on molecular docking and the use of the compounds screened therefrom in the preparation of anti-lung cancer drugs. By analyzing the types of small molecule compounds obtained by the first screening, ginsenoside compounds are selected as the second screening library; the ginsenoside compound library is then selected as the second screening ligand, and virtual screening is performed again to obtain the docking score results of TMPRSS2 and ginsenoside compounds, obtain the top five compounds, and perform molecular docking and activity research. The present invention uses multi-level virtual screening to build a secondary virtual screening ligand library, enriching the variety of ligands of this type while reducing the number of compounds in the ligand library. While effectively improving the accuracy of virtual screening, it reduces the number of experimental screening compounds, shortens the R&D cycle, and saves costs.

Description

Technical Field

[0001] The present invention relates to the field of anti-tumor, and in particular to a multi-level virtual screening method based on molecular docking and use of the screened compounds in the preparation of anti-lung cancer drugs. Background Art

[0002] TMPRSS2 (transmembrane serine protease 2) is a major member of the type II transmembrane serine protease family and possesses a single transmembrane domain. The human TMPRSS2 gene is located on chromosome 21 (21q22.3) and is most highly expressed in the prostate.

[0003] TMPRSS2 plays an important role in cancer development. Studies have shown that it plays a carcinogenic role in prostate cancer. Inhibition of TMPRSS2 expression has an inhibitory effect on the proliferation of prostate cancer cells (Glowacka I, Bertram S, Müller MA, Allen P, Soilleux E, Pfefferle S et al: Evidence that TMPRSS2 activates the severe acute respiratory syndrome coronavirus spike protein for membrane fusion and reduces viral control by the humoral immune response. J Virol 2011, 85(9): 4122-4134.). In addition, high expression of TMPRSS2 was also found in colorectal cancer and breast cancer (Wang H, Yang J. Colorectal Cancer that Highly Express Both ACE2and TMPRSS2, Suggesting Severe Symptoms to SARS-CoV-2 Infection. Pathol Oncol Res. 2021Apr 15;27:612969. doi:10.3389 / pore.2021.612969. PMID:34257580;PMCID:PMC8262219.)(Chi M, Shi X, Huo X, Wu X, Zhang P, Wang G: Dexmedetomidine promotes breast cancer cell migration through Rab11-mediated secretion of exosomal TMPRSS2. Ann Transl Med 2020, 8(8):531.). In lung cancer, high expression of TMPRSS2 is also associated with a worse prognosis in patients who do not respond to anti-PD-1 therapy. Therefore, the development of lead compounds that inhibit TMPRSS2 activity is of great significance for the treatment of cancer and other hyperproliferative diseases.

[0004] In recent years, computer-aided drug design has become an increasingly important part of drug research and development. It has accelerated the speed of new drug design, saved manpower and material resources in new drug development, and has become one of the conventional methods in drug design. It is widely used in the development and optimization of innovative drug lead compounds.

[0005] Traditional virtual screening can only select high-scoring compounds as target ligands from a diverse small molecule library. However, current virtual screening inevitably suffers from biases, which leads to discrepancies in the results of each repeated screening, and has a significant impact on the accuracy of the results.

[0006] This study compensated for this error through a strategy of multiple virtual screenings. After the first natural product compound library screening, the top-ranked compound types were analyzed and a virtual screening library of the compound was built for re-screening, which narrowed the target range and effectively improved the accuracy of virtual screening.

[0007] Therefore, we used computer-aided drug design, with TMPRSS2 gene protein as the target and the natural product small molecule database as the ligand, and a virtual screening method based on molecular docking to discover TMPRSS2 gene inhibitors through layer-by-layer screening, thereby discovering small molecule compounds for lung cancer. Summary of the Invention

[0008] The purpose of the present invention is to address the problems existing in the prior art and provide a multi-level virtual screening method based on molecular docking and the use of the screened compounds in the preparation of anti-lung cancer drugs.

[0009] The present invention analyzed the small molecule compounds obtained in the first screening and selected ginsenoside compounds as the second screening library. A library of ginsenoside compounds was selected from the MCE compound library website (https: / / www.medchemexpress.cn / ) and 39 SDF files were downloaded as ligands for the second screening. Virtual screening was performed again using Sailvina software to obtain docking scores for TMPRSS2 and ginsenoside compounds. The 2D structures of the top five compounds were drawn using ChemBioDrawUltra 12.0 software (https: / / www.chemdraw.com.cn / ir / ), and 3D structures were generated using Chem3D software for molecular docking and activity studies.

[0010] To achieve the above object, the present invention adopts the following technical solutions:

[0011] The present invention provides a multi-level virtual screening method based on molecular docking, comprising the following steps:

[0012] S1. Identify TMPRSS2 protein (PDB ID: 7MEQ) as the receptor protein from the PDB database; download small molecule compounds from the Pubchem database as natural compound ligand small molecules to complete the preparation of receptors and ligands;

[0013] S2. Use virtual screening software to perform a first virtual screening from a library of natural compound ligand small molecules to obtain docking scoring data for natural compound ligand small molecules that have the ability to inhibit TMPRSS2 activity;

[0014] S3. Based on the top-ranked compound types in S2, a virtual screening library was built and a second virtual screening was performed using virtual screening software to obtain docking scoring data for natural compound ligand small molecules that inhibit TMPRSS2 activity;

[0015] S4. Based on the scoring data from S3, multiple top-ranked natural compound ligand small molecules were screened, and then molecular docking was performed with the TMPRSS2 protein. The docking results were combined to form a complex, and the conformation with the best docking effect was screened based on the docking results.

[0016] Furthermore, the TMPRSS2 protein compound in step S1 is a protein compound with high resolution and complete amino acid peptide segments.

[0017] Furthermore, step S1 also includes receptor protein pretreatment, specifically including: removing the original ligand and water molecules, and then repairing the receptor structure.

[0018] Furthermore, in step S2, the first virtual screening specifically includes: selecting the site where TMPRSS2 directly interacts with the co-crystallized ligand nafamostat as the docking active site, determining the center coordinates and docking sphere radius based on the active site; setting the number of docking attempts, obtaining docking results, and ranking them according to the docking scores. The center coordinates of the docking are determined using the Center on Ligand method, and the virtual screening software includes software such as Sailvina.

[0019] Furthermore, the central coordinates of the active site are the central site of the active site in the co-crystallized ligand of TMPRSS2 and nafamostat, specifically: the TMPRSS2 protein has a good docking effect with the nafamostat small molecule, and its action site is located at SER-436, GLY-439, ASP-440, SER-441, GLN-438, CYS-437, GLY-464 and ASP-440 of the A chain; according to the action site, the central coordinate parameters of the original ligand (-9.771, -6.594, 20.191) are found in the original co-crystallized ligand complex, which is the central site.

[0020] Furthermore, in step S3, the virtual screening library is constructed by determining the compound type from the top-ranked compounds in S2 and collecting similar compounds of the same type to establish a virtual screening library. For example, if the top-ranked compound type is a ginsenoside compound, ginsenoside compounds of that type are collected.

[0021] Furthermore, in step S3, the second virtual screening specifically includes: setting the center coordinates, docking sphere radius, and docking times of the same active site as in step S2, obtaining docking results, and sorting them according to the docking scores. The virtual screening software includes software such as Sailvina.

[0022] Furthermore, in step S4, before molecular docking, the top-ranked natural compound ligand small molecule is used to generate a two-dimensional structure using drawing software, and Chem3D software is used to generate a three-dimensional structure sdf file for molecular docking.

[0023] Furthermore, in step S4, molecular docking is specifically as follows: setting the same active site center coordinates and docking sphere radius as in step S2, designing a docking box, and screening the conformation with the best docking effect according to the poses score value.

[0024] Furthermore, after step S4, the results of the complex are visualized, and the activity of the screened compounds is screened.

[0025] The present invention also provides a use of a ginsenoside compound in preparing a drug for treating lung cancer. The ginsenoside compound includes a compound Ginsenoside-Rf, a compound 20R-Ginsenoside Rg2, a compound Ginsenoside F1, a compound Ginsenoside Rg1, and a compound Ginsenoside Z-R1; wherein the structural formula of Ginsenoside-Rf is shown in Formula 1, the structural formula of the compound 20R-Ginsenoside Rg2 is shown in Formula 2, the structural formula of the compound Ginsenoside F1 is shown in Formula 3, the structural formula of the compound Ginsenoside Rg1 is shown in Formula 4, and the structural formula of the compound Ginsenoside Z-R1 is shown in Formula 5.

[0026] 5

[0028]

[0029] The innovation of the present invention is that in step S2, ginsenoside compounds are determined by the first screening of multiple natural product compounds. Therefore, in step S3, the small molecule compounds used for virtual screening are selected from a self-built ginsenoside compound database. The database selects 39 natural product compounds of ginsenosides, covering dammarane-type and oleanane-type ginsenosides, including not only mainstream research ginsenosides and their extracts and trace components, but also covers relatively less studied bacterial metabolites, R-type isomers, ginsenoside aglycones, glycosides, etc. The self-built library covers a rich variety of species, can improve the accuracy of virtual screening when screening a small number of ligands, and improve the precision of the present invention.

[0030] Another innovation of this study is that the receptor binding site of TMPRSS2 was first determined by co-crystallization of the ligand complex (e.g. Figure 2 The present invention selects the parameter coordinates of the central site of the small molecule ligand in the co-crystallized ligand of TMPRSS2 and nafamostat as the central coordinates for two virtual screenings and molecular docking. Compared with Blind docking, which uses the center of the protein itself as the screening or docking center and sets the docking box radius to surround the entire protein, virtual screening or molecular docking based on the central coordinates of the co-crystallized ligand greatly improves the computational efficiency and accuracy of the results.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] (1) The present invention is based on the concept of discovering known drugs and compounds as potential TMPRSS2 inhibitors. Structure-based virtual screening is performed using drug databases to discover new uses for existing drugs beyond their original medical indications. The present invention uses virtual screening strategies to seek drugs currently on the market as treatments for lung cancer, i.e., drug repurposing. Since known drugs generally have acceptable safety and pharmacokinetic properties, discovering new indications for these drugs will enable patients in need to benefit from potential new therapies earlier.

[0033] (2) Through multi-level virtual screening, a secondary virtual screening ligand library was built, which enriched the variety of ligands of this type while reducing the number of compounds in the ligand library. While effectively improving the accuracy of virtual screening, it also reduced the number of experimental screening compounds, shortened the R&D cycle, and saved costs.

[0034] (3) The polypeptide compound screened by the method of the present invention has a high activity of inhibiting TMPRSS2, and the natural product itself has the advantages of low drug resistance and high safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Other features, objects and advantages of the present invention will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0036] Figure 1 The protein crystal structure of TMPRSS2 (PDB ID: 7MEQ) of the present invention;

[0037] Figure 2 is the binding site of the receptor protein of the present invention;

[0038] Figure 3 Schematic diagram of molecular docking analysis between the receptor protein and the compound of the present invention;

[0039] Figure 4 This is the affinity curve of ginsenoside compounds with different concentrations and TMPRSS2 in Example 2;

[0040] Figure 5 This is a diagram showing the inhibitory effect of ginsenosides at different concentrations on TMPRSS2 enzyme in Example 2. DETAILED DESCRIPTION

[0041] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention; it is obvious that the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0042] Example 1

[0043] The method for multiple virtual screening of TMPRSS2 inhibitors comprises the following steps:

[0044] Step S1, obtaining the three-dimensional structure files of the receptor protein and the ligand small molecule, specifically including:

[0045] ① Receptor preparation: Download the complex crystal structure of TMPRSS2 protein (PDBID: 7MEQ) from the PDB database (https: / / www.rcsb.org / ). Use Pymol (https: / / pymol.org / 2 / ) software to process the receptor protein and remove the original ligand and water molecules. The protein crystal structure is as follows: Figure 1 Then, use the Prepare PDB file for docking programs website (https: / / swift.cmbi.umcn.nl / servers / html / prepdock.html) to repair the receptor structure and save it in pdb format in the receptor protein folder.

[0046] ② Preparation of natural compound ligands: Download 4320 small molecule sdf files from the Pubchem database (https: / / pubchem.ncbi.nlm.nih.gov / ), convert the sdf files into pdb files in batches using Open Babel (http: / / openbabel.org / wiki / Main_Page) software, and save them in the ligand small molecule folder.

[0047] Step S2, using Sailvina software to perform the first virtual screening, includes the following steps:

[0048] ①Input the folder where the ligand and receptor are located in the sailvina (https: / / github.com / beikwx / SailVina) software, use the Center on ligand method to determine the center coordinates of the docking, and use the site where TMPRSS2 directly interacts with the co-crystallized ligand nafamostat as the docking active site. The binding site is as follows: Figure 2 As shown in the original complex txt file, the center coordinates of the active site (-9.771, -6.594, 20.191) are obtained and set, and the docking sphere radius parameter is set to

[0049] ②Batch convert the pdb files of the ligand small molecule into pdbqt format in the sailvina software and save them in the original ligand folder;

[0050] ③ Use sailvina software for virtual screening, set the docking number to 1, obtain the docking results and output them in txt format, and sort the docking scores.

[0051] Table 1 shows the top ten natural product compounds in the first screening, of which six are triterpenoid compounds, so triterpenoid ginsenoside compounds were selected for the next virtual screening.

[0052] Table 1 Top ten compounds in the first screening

[0053]

[0054] Step S3, constructing a ginsenoside compound library and performing a second virtual screening, includes the following steps:

[0055] ① Download 39 ginsenoside compound sdf files from the MCE compound library website (https: / / www.medchemexpress.cn / ), use Open Babel (http: / / openbabel.org / wiki / Main_Page) software to batch convert the sdf format files into pdb format files, and save them in the ligand small molecule folder.

[0056] ②In the sailvina (https: / / github.com / beikwx / SailVina) software, enter the folder where the ligand and receptor are located, use the center coordinates of the active site screened in step S2 (-9.771, -6.594, 20.191), and set the docking sphere radius parameter to

[0057] ③ In the Sailvina software, batch convert the pdb files of the ligand small molecule into pdbqt format and save them in the original ligand folder;

[0058] ④ Use sailvina software for virtual screening, set the docking number to 1, obtain the docking results and output them in txt format, sort the docking scores of ginsenoside compounds, and select the top five ginsenoside compounds for further research.

[0059] Table 2 shows the top ten ginsenoside natural product compounds in the second screening. Since the types of ginsenosides were expanded and the scope of ligand screening was narrowed, the results of the second screening can more accurately reflect the docking effect of small molecule compounds with TMPRSS2, thereby finding the optimal compound for the TMPRSS2 target.

[0060] Table 2 Top ten compounds in the second screening

[0061]

[0062] Step S4: draw the two-dimensional structures of the top five ginsenoside compounds and generate three-dimensional structures for molecular docking.

[0063] The following steps are involved:

[0064] ① Use ChemBioDraw Ultra 12.0 software (https: / / www.chemdraw.com.cn / ir / ) to draw the two-dimensional structures of the top five ginsenoside compounds, preserving the in-plane and out-of-plane directionality of their hydrogen bonds;

[0065] ② Import the two-dimensional image into Chem3D software and generate a three-dimensional sdf file after modifying the structural deviation;

[0066] ③ Use Open Babel (http: / / openbabel.org / wiki / Main_Page) software to convert the sdf format file into a pdb format file for molecular docking;

[0067] Step S5: Using the autodock software, the top five compounds obtained in step S4 were docked with the treated TMPRSS2 protein. A docking box was designed using the binding sites of the drug nafamostat with the protein, SER-436, GLY-439, ASP-440, SER-441, GLN-438, CYS-437, GLY-464, and ASP-440. The conformation with the best docking effect was selected based on the poses score. The steps include:

[0068] ① Import the TMPRSS2 protein after removing water molecules and original ligands into the macromolecular protein window, and use the autodocktools module to perform hydrogenation on the receptor protein to calculate the Gasteiger charge;

[0069] ② Import the top five compounds generated in step S4 into the small molecule ligand window one by one, use the autodock tools module to perform hydrogenation calculations on the ligands, and save them in pdbqt format;

[0070] ③ Preparation of docking box: Existing studies have shown that TMPRSS2 protein has a good docking effect with the small molecule nafamostat, and its action sites are located at SER-436, GLY-439, ASP-440, SER-441, GLN-438, CYS-437, GLY-464 and ASP-440 of the A chain. Since both the small molecule and nafamostat screened in this study are inhibitors of TMPRSS2, a docking method based on co-crystallized ligands was chosen instead of a docking box that wraps the entire protein. The center coordinate parameters of the original ligand (-9.771, -6.594, 20.191) were found in the txt file of the original co-crystallized ligand complex and set as the docking site in Autodockvina. The docking parameters were set using Grid--Grid BOX, and the box radius was Save the relevant parameters in a file in gpf format.

[0071] ④ Molecular docking: Run the Grid module to generate a map file. Select semi-flexible docking, set the number of dockings to 50, and output a .dpf file. Run Autodock to generate the corresponding .dlg file and view the docking results.

[0072] This application selected five small molecule compounds based on the scoring results. The docking analysis diagram is shown in the following figure. Figure 3 The scoring is shown in Table 3.

[0073] Ginsenosides are steroid compounds, also known as triterpenoid saponins, primarily found in ginseng herbs. Ginsenosides all share a similar basic structure, consisting of a sterane steroid nucleus composed of 30 carbon atoms arranged in four rings. They are divided into two groups based on the structure of their glycoside moieties: dammarane and oleanane.

[0074] For this study, 39 natural small-molecule ginsenoside compounds were selected from the MCE compound library (https: / / www.medchemexpress.cn / ) as a ligand library for the second virtual screening. This library covers the vast majority of natural ginsenoside compounds, including both dammarane- and oleanane-type ginsenosides, including agonists, inhibitors, and modulators of signaling pathways. This self-built library includes both mainstream research ginsenosides such as Rh, Rg, and Rb, as well as aglycones and metabolites of ginsenosides such as (20S)-Protopanaxatriol and Neopanaxadiol. The five compounds discovered in this study all belong to the dammarane-type ginsenoside class.

[0075] The compound with the highest docking score, Ginsenoside-Rf, is a trace component of Panax ginseng root that inhibits N-type Ca2+ channels. At saturating concentrations, Ginsenoside-Rf rapidly and reversibly inhibits N-type and other high-threshold Ca2+ channels in rat sensory neurons to the same extent as a maximal dose of an opioid.

[0076] The second-highest docking-scoring compound, 20R-Ginsenoside Rg2, is a ginsenoside isolated from the stems and leaves of Panax ginseng. 20R-Ginsenoside Rg2 exhibits inhibitory effects on lung cancer NCI-H1650 cells and exhibits anticancer activity.

[0077] The third-ranked compound, Ginsenoside F1, is a metabolite of ginsenoside Rg1 and is known to have anti-aging, antioxidant, anti-cancer, and keratocyte-protective effects. Ginsenoside F1 competitively inhibits CYP3A4 with a Ki value of 67.8±16.2μM and has a weaker inhibitory effect on CYP2D6.

[0078] Ginsenoside Rg1, the compound ranked fourth in docking score, is one of the main active ingredients of ginseng. It is known to quickly relieve fatigue, improve learning and memory, and delay aging. It works by lowering the level of Aβ in the brain, has the effect of stimulating the central nervous system, and inhibiting platelet aggregation.

[0079] Ginsenoside Z-R1, the fifth-ranked compound, is a triterpenoid saponin that exhibits excellent anti-tumor and anti-angiogenic activities. It has some anti-HIV-1 activity and inhibits the uptake of 2-deoxy-D-glucose (2-DG) in EAT cells (IC50 = 91.3 μM).

[0080] Table 3 Docking scores of five compounds

[0081]

[0082] Example 2

[0083] Materials and equipment: Human lung adenocarcinoma A549 cell line (ATCC@XY-XB-1411), TMPRSS2 fluorescence assay kit (CAT#78083, BPS Bioscience, San Diego, CA, USA)

[0084] Experimental procedures

[0085] The cell thermal migration assay was used to detect the affinity of each compound to the receptor in A549 cell line.

[0086] 2×107 cells were seeded in 10 cm 2 The cells were cultured in F-12K medium containing 10% fetal bovine serum for 24 hours. The cells were treated with 5 μM of the five ginsenoside compounds or DMSO for 3 hours as drug-exposed cells and control cells. The cells were collected and washed twice with phosphate-buffered saline (PBS). The cells were trypsinized and centrifuged. The treated cells were resuspended in 900 μL of PBS and evenly divided into 8 parts, with 100 μL of cells per part placed in a 0.2 mL PCR tube. Pairs consisting of a control group and an experimental group were heated at 52, 53, 54, 56, 58, 60, 61, and 62°C for 3 minutes. 1 μL of a protease and phosphatase inhibitor cocktail tablet was added to each tube (total: 101 ul). The samples were subjected to three freeze-thaw cycles. In each cycle, they were exposed to liquid nitrogen for 10 minutes and placed in a 37°C heating block for 10 minutes. Centrifuged at 14,000g for 20 minutes at 4°C. Transfer 80 μL of the sample to a microcentrifuge tube, add 20 μL of 5× loading buffer, and heat at 98°C for 10 minutes. Finally, place the sample on ice. Perform Western blotting and quantify the bands.

[0087] The results are as follows Figure 4 Compared with DMSO, ginsenoside compounds of different concentrations have high affinity to receptors.

[0088] TMPRSS2 fluorescence assay to evaluate the inhibition of TMPRSS2 enzyme activity by compounds

[0089] The TMPRSS2 fluorescence assay kit (CAT#78083, BPS Bioscience, San Diego, CA, USA) was used to assess the inhibitory activity of compounds against the TMPRSS2 enzyme. Briefly, 30 μL of TMPRSS2 (5 ng / μL) was added to 10 μL of various compound concentrations (0.007, 0.013, 0.026, 0.053, and 0.1054 μM). After incubation at room temperature for 30 minutes, 10 μL of TMPRSS2 substrate (50 μM) was added, and fluorescence intensity was measured in the dark using a microplate reader (Synergy H1, Biotek Ltd., Winooski, VT, USA) with emission and excitation wavelengths of 383 and 455 nm, respectively. Camostat mesylate (10 μM) was used as a positive control, while reactions without inhibitor and enzyme served as negative controls.

[0090] The results are as follows Figure 5 Different concentrations of ginsenosides had a significant inhibitory effect on TMPRSS2 enzyme compared with the negative control.

[0091] Comparative Example 1

[0092] To verify the accuracy of the two virtual screening results of this study, a control compound was set as a benchmark to evaluate the results of the present invention, including the following steps:

[0093] ① Based on previous research articles on the TMPRSS2 target, compound N-0385 (Shapira T, Monreal IA, Dion SP, Buchholz DW, Imbiakha B, Olmstead AD, Jager M, Désilets A, Gao G, Martins M, Vandal T, Thompson CAH, Chin A, Rees WD, Steiner T, Nabi IR, Marsault E, Sahler J, Diel DG, Van de Walle GR, August A, Whittaker GR, Boudreault PL, Leduc R, Aguilar HC, Jean FA TMPRSS2 inhibitor acts as a pan-SARS-CoV-2 prophylactic and therapeutic. Nature. 2022 May; 605(7909): 340-348. doi: 10.1038 / s41586-022-04661-w.) and carmustine

[0094]

[0095] (Egyházi S, Bergh J, Hansson J, Karran P, Ringborg U. Carmustine-induced toxicity, DNA crosslinking and O6-methylguanine-DNA methyltransferase activity in two human lung cancer cell lines. Eur J Cancer. 1991; 27(12): 1658-62. doi: 10.1016 / 0277-5379(91)90440-o.) Two compounds served as control groups;

[0096] ② The two control compounds and 39 ginsenoside compounds in the self-built library were used as ligand small molecule libraries, and virtual screening was performed again using Sailvina. The same center coordinates (-9.771, -6.594, 20.191) and docking parameters were set to obtain the combined binding energy score;

[0097] ③ The screening results are shown in Table 4. The five ginsenoside small molecule compounds screened in this study ranked higher than N-0385 and carmustine in the comparative study, indicating that they have lower docking binding energy with TMPRSS2, the formed complexes are more stable, and the docking effect is better, verifying the accuracy of the two virtual screening results of this study.

[0098] Table 4 Binding energy ranking of comparative examples

[0099]

[0100]

[0101] In the absence of conflict, the above embodiments and features in the embodiments may be combined with each other.

[0102] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A multi-level virtual screening method based on molecular docking, characterized in that: The steps include: S1. Identify TMPRSS2 protein as a receptor protein; download small molecule compounds as natural compound ligand small molecules and complete the preparation of receptors and ligands; S2. Use virtual screening software to perform a first virtual screening from a library of natural compound ligand small molecules to obtain docking scoring data for natural compound ligand small molecules that have the ability to inhibit TMPRSS2 activity; S3. Based on the top-ranked compound types in S2, a virtual screening library was built and a second virtual screening was performed using virtual screening software to obtain docking scoring data for ginsenoside ligand small molecules that inhibit TMPRSS2 activity; S4. Based on the scoring data from S3, multiple top-ranked ginsenoside ligand small molecules are screened, and then molecular docking is performed with the TMPRSS2 protein. The docking results are combined to form a complex, and the conformation with the best docking effect is screened based on the docking results, which is the small molecule compound for multi-level virtual screening; In step S2, the first virtual screening specifically includes: taking the site where the TMPRSS2 protein directly interacts with the co-crystallized ligand nafamostat as the docking active site, determining the center coordinates and the docking sphere radius based on the active site; setting the docking number, obtaining the docking results, and sorting them according to the docking score; The central coordinates of the active site are the central site of the active site in the co-crystallized ligand of TMPRSS2 protein and nafamostat. Specifically, the TMPRSS2 protein and the nafamostat small molecule have a good docking effect, and its action sites are located at SER-436, GLY-439, ASP-440, SER-441, GLN-438, CYS-437, GLY-464 and ASP-440 of the A chain. According to the action sites, the central coordinate parameters of the original ligand are found in the original co-crystallized ligand complex, which are the central coordinates of the active site. In step S4, molecular docking is specifically as follows: setting the same active site center coordinates and docking sphere radius as in step S2, designing the docking box, and screening the conformation with the best docking effect based on the poses score value.

2. The multi-level virtual screening method according to claim 1, characterized in that Step S1 also includes receptor protein pretreatment, specifically including: removing the original ligand and water molecules, and then repairing the receptor structure.

3. The multi-level virtual screening method according to claim 1, characterized in that In step S3, the self-built virtual screening library specifically includes: determining the type of compound from the top-ranked compounds in S2; and collecting similar compounds of the type to establish a virtual screening library.

4. The multi-level virtual screening method according to claim 1, characterized in that In step S3, the second virtual screening specifically includes: setting the center coordinates, docking sphere radius, and docking times of the same active site as in step S2, obtaining docking results, and sorting them according to the docking scores.

5. The multi-level virtual screening method according to claim 1, characterized in that In step S4, before molecular docking, the top-ranked natural compound ligand small molecule is used to generate a two-dimensional structure using drawing software, and the three-dimensional structure sdf file is generated using Chem3D software for molecular docking.

6. A use of a ginsenoside compound screened by the multi-level virtual screening method according to claim 1 in the preparation of a drug for treating lung cancer, wherein the ginsenoside compound comprises Ginsenoside-Rf as shown in Formula 1, 20R-Ginsenoside Rg2 as shown in Formula 2, Ginsenoside F1 as shown in Formula 3, Ginsenoside Rg1 as shown in Formula 4, and Ginsenoside Z-R1 as shown in Formula 5: 。

Images