Search

CN-121983131-A - Method for screening anti-type 2 diabetes multi-target drugs based on Boolean network

CN121983131ACN 121983131 ACN121983131 ACN 121983131ACN-121983131-A

Abstract

The invention discloses a method for screening anti-type 2 diabetes multi-target drugs based on a Boolean network, and belongs to the technical field of biological medicines. The method comprises the steps of constructing a dynamic network model, converting related signal paths into a calculation frame, recognizing a synergistic target combination (DPP-4 and GPR 40) through disturbance analysis, precisely positioning a minimum intervention node set by utilizing the characteristics of attractors, constructing a virtual screening model to screen candidate compounds to form a closed loop flow, and combining synchronous stabilizer attractor calculation with the minimum intervention node set standard to ensure that the analysis is more precise and the target combination is limited to ensure the synergistic effect. The system integrates four functional modules to realize automatic and efficient screening. In application, the candidate compound is used for preparing anti-type 2 diabetes medicines, and solves the existing treatment difficulty. The multi-target pharmaceutical composition avoids multi-drug combination and reduces risk. The drug combination product integrates a DPP-4 inhibitor and a GPR40 agonist, solves the problem of multi-drug combination and provides a new treatment choice.

Inventors

  • LI XIAOJIE
  • LI RONG
  • CHENG MIN
  • XIE WEIWEI
  • ZHAO WEI
  • LIU XUEYING
  • MA JIBIN
  • Sun Dingkang
  • SHU QUANYONG
  • LI ZONGLIN
  • WEI CHAO
  • ZHAI FAN
  • BAI YUJUN

Assignees

  • 陕西医药控股医药研究院有限公司
  • 西安培华学院
  • 商洛学院

Dates

Publication Date
20260505
Application Date
20260112

Claims (10)

  1. 1. The method for screening the anti-type 2 diabetes multi-target drug based on the Boolean network is characterized by comprising the following steps of: Step 1, constructing a signal path Boolean network model related to type 2 diabetes mellitus, wherein the signal path Boolean network model comprises a plurality of nodes and logic relations among the nodes, and the nodes represent proteins, genes or biological paths related to occurrence and development of the type 2 diabetes mellitus; Step 2, performing disturbance analysis on the signal path Boolean network model, and identifying a synergistic target combination capable of stably regulating and controlling a disease state network to a health state by simulating inhibition or activation states of different nodes or node combinations, wherein the synergistic target combination comprises DPP-4 and GPR40; And 3, screening candidate compounds capable of simultaneously acting on DPP-4 and GPR40 from a compound database based on the DPP-4 and GPR40 target combination identified in the step 2, and outputting a result as a multi-target drug for resisting type 2 diabetes.
  2. 2. The method for screening the anti-type 2 diabetes multi-target drugs based on the Boolean network according to claim 1, wherein in the step 2, the disturbance analysis comprises carrying out disturbance simulation on all double-node combinations in a signal path Boolean network model, evaluating the influence efficiency of each disturbance on the convergence of the network state to a healthy attractor, and screening out a target combination with a synergistic effect, wherein the synergistic target combination is a combination of fixing the state of a DPP-4 node to an inhibition state and fixing the state of a GPR40 node to an activation state.
  3. 3. The method for screening anti-type 2 diabetes multi-target drugs based on boolean network according to claim 2, characterized in that in step 3, the step of screening candidate compounds comprises: step 301, respectively constructing a pharmacophore model aiming at a DPP-4 inhibitor and a pharmacophore model aiming at a GPR40 agonist, and carrying out joint virtual screening on a compound database to obtain a primary candidate compound set; Step 302, performing molecular docking calculation on the compounds in the preliminary candidate compound set and target structures of proteins of DPP-4 and GPR40 respectively, and screening a candidate compound subset with predicted binding affinity to both DPP-4 and GPR40 targets based on docking scoring.
  4. 4. The method of claim 3, wherein the pharmacophore model of the DPP-4 inhibitor comprises structural features of a hydrogen bond receptor, a charge center and a hydrophobic center, and the pharmacophore model of the GPR40 agonist comprises structural features of two aromatic ring centers and two hydrogen bond receptors.
  5. 5. The method for screening the anti-type 2 diabetes multi-target drug based on the Boolean network according to claim 3, wherein in the molecular docking calculation, a DPP-4 target adopts a PDB ID: 3G0B protein crystal structure, a GPR40 target adopts a three-dimensional structure predicted based on AlphaFold, and the screening standard is that docking scores of candidate compounds and DPP-4 and GPR40 are respectively superior to or equal to docking scores of known positive drugs alogliptin and TAK-875.
  6. 6. The method for screening anti-type 2 diabetes multi-target drugs based on the boolean network according to claim 1, characterized in that the candidate compounds capable of acting on DPP-4 and GPR40 simultaneously are screened as any one of the following structural formulas; 。
  7. 7. The method for screening anti-type 2 diabetes multi-target drugs based on boolean networks according to any one of claims 1-6, characterized in that the step 3 further comprises the step of performing in vitro activity verification on the outputted candidate compounds capable of simultaneously acting on DPP-4 and GPR40, and confirming that the candidate compounds have DPP-4 inhibitory activity and GPR40 agonistic activity.
  8. 8. The method for screening anti-type 2 diabetes multi-target drugs based on the Boolean network according to claim 1, wherein the method further comprises the step of combining the screened candidate compounds capable of simultaneously acting on DPP-4 and GPR40 as active ingredients with pharmaceutically acceptable carriers to form the anti-type 2 diabetes multi-target drug composition.
  9. 9. The method of claim 1, wherein step 3 comprises screening at least two different active ingredients from a database of compounds, wherein the first active ingredient is a DPP-4 inhibitor and the second active ingredient is a GPR40 agonist, to form a multi-component pharmaceutical combination for the synergistic prevention and/or treatment of type 2 diabetes.
  10. 10. A system for performing the screening method of any one of claims 1 to 9, comprising: The model construction module is used for constructing and storing the signal path Boolean network model related to the type 2 diabetes; The target point identification module is used for carrying out disturbance analysis on the signal path Boolean network model and identifying a cooperative target point combination; A screening module for screening candidate compounds capable of acting on both DPP-4 and GPR40 from a compound database based on the identified synergistic target combinations; and the result output module is used for outputting the candidate compound information obtained by screening.

Description

Method for screening anti-type 2 diabetes multi-target drugs based on Boolean network Technical Field The invention belongs to the field of biological medicine, and in particular relates to a type 2 diabetes multi-target drug constructed based on a Boolean network and a screening method thereof. Background Type 2 diabetes mellitus (T2 DM) is a chronic metabolic disease caused by multiple factors, and its pathogenesis is complex, involving multiple links such as insulin secretion disorder, insulin resistance, islet beta cell apoptosis, etc. The current clinical treatment of T2DM is mainly divided into two main categories, injection and oral. Injections include insulin and glucagon-like peptide-1 (Glucagon-LIKE PETIDE, GLP-1) receptor agonists, and oral administration includes biguanides, alpha-glucosidase inhibitors, sulfonylureas, thiazolidinediones, dipeptidylpeptidase-4 (DIPEPTIDYLPEPTIDASE 4, DPP-4) inhibitors, sodium-glucose cotransporter 2 (Sodium-DEPENDENT GLUCOSE TRANSPORTERS, SGLT-2) inhibitors, and the like. Although these drugs can control blood glucose through different mechanisms, due to the complex pathological features of T2DM multi-channel interaction, long-term stability of blood glucose is often difficult to achieve by single drug therapy, and combined drug administration is often required in the clinical treatment process. However, multi-Drug combinations often face side effects, a superposition of toxicities, low patient compliance, and unnecessary Drug interactions (DDI). Aiming at the dilemma, a multi-target drug design strategy is developed, and the strategy synchronously regulates and controls a plurality of pathological targets through a single molecule, so that the defect of the traditional multi-drug combination is avoided while the blood glucose reducing synergistic effect is enhanced, and the novel direction of T2DM treatment is realized. However, how to screen out key target combinations with synergistic therapeutic effects from complex disease networks is a core challenge for multi-target drug design. In recent years, the development of cross fusion of multiple subjects such as system biology, computational biology, network pharmacology and the like provides a new theoretical basis and technical means for multi-target drug research and development. However, the application of network pharmacology in multi-target drug research and development is still in an exploration stage at present, and a mature and effective multi-target drug screening method and strategy are still lacking, particularly in the field of multi-target drug research and development of type 2 diabetes mellitus, and a technical method capable of accurately identifying key target pairs and efficiently screening candidate compounds is urgently required to be developed. Chinese patent (publication No.) discloses network pharmacology and protein-protein interaction (PPI) network analysis for correlating compounds with potential targets. However, the method focuses on static association or target recognition of specific compounds, fails to deeply simulate the dynamic process of interaction among targets along with state evolution, and lacks the capability of systematically discovering brand-new synergistic target combinations, and although the Chinese patent (publication number: I) focuses on curative effect evaluation of multi-target combinations, the method for actively discovering key synergistic target pairs from the dynamic state of a disease network is not provided fundamentally. The Boolean network is used as a dynamic biological network modeling tool, can convert a static biomolecule regulation and control relation into a dynamic calculation model, and reveals key regulation and control nodes and potential treatment target combinations of disease states through simulation, disturbance and attractor analysis. In recent years, the boolean network has been successfully applied to mechanism research and target discovery of complex diseases such as psoriasis, T-cell leukemia, etc. However, the application of the method in the field of T2DM multi-target drug discovery is still blank, and particularly how to construct a Boolean network model close to the pathology of T2DM, so how to construct a type 2 diabetes disease network based on the Boolean network modeling, accurately identify key target pairs with synergistic therapeutic effects, and find candidate compounds for regulating and controlling the target combinations through virtual screening, and still face a plurality of technical challenges and difficulties. The development of the multi-target drug screening method for the type 2 diabetes constructed based on the Boolean network has important theoretical significance and practical application value. Disclosure of Invention Aiming at the key bottleneck existing in the research and development of the existing type 2 diabetes (T2 DM) multi-target drugs, namely how to systematically and dynamically identify the opt