Quantitative Prediction Of Drug Dissociation Rate Constants(K_off)

The dissociation rate constant is the key factor affecting drug potencies.The traditional methods for determing dissociation rate constant(k_off)may cost a lot of manpower and resources,and cannot meet the requirements of high-throughput screening.Therefore,the development of an effective prediction method of dissociation rate constant is of great significance for drug development and mechanism researches.By using quantitative structure-activity relationship?QSAR?and molecular dynamics simulation?MD?methods,the dissociation rate constants of adenosine receptor?A₁AR?agonists/antagonists,heat shock protein?Hsp90?inhibitors,and mitogen-activated protein kinase?p38 MAPK?inhibitors were quantitatively predicted with satisfying results.The prediction methods proposed in this paper can provide important methodological reference for the prediction of dissociation rate constants.The main contents and achievements are as follows:?1?Quantitative prediction of the dissociation rate constants of A₁AR agonists/antagonists.A total of 49 A1AR agonists/antagonists,characterized by three-dimensional molecular interaction field descriptors?Volsurf?and physicochemical descriptors,were used to construct QSAR models of dissociation rate constants by stepwise regression and partial least squares?PLS?methods.The results show that the physicochemical and Volsurf descriptors can characterize the key structural information related to the dissociation rates of A₁AR agonists/antagonists,and the models obtained have strong predictive capabilities and high interpretabilities.In addition,the physicochemical descriptors CH3R,nH,nS and Volsurf descriptors W5-N3+and POL are proved be the key determinants of the dissociation rate constants of A₁AR agonists/antagonists.They characterize the CH3R fragment information,number of hydrogen atoms,number of sulfur atoms,hydrophilic regions and molecular polarizability of A₁AR agonist/antagonist,respectively.?2?Quantitative prediction of the dissociation rate constants of Hsp90 inhibitors.A total of 52 Hsp90 inhibitors,characterized by Volsurf and physicochemical descriptors,were used to construct the QSAR models of the dissociation rate constant.By forward variable selection and PLS modeling,the interpretable QSAR models were established with strong predictive capabilities.The results show that physicochemical descriptor Sp,nCIR and Volsurf descriptor V-OH2,D8-DRY are the key determinants of the dissociation rate of the Hsp90 inhibitors.They characterize the atomic polarizabilities,number of circuits,molecular volume and hydrophobic regions of Hsp90 inhibitors,respectively.?3?Quantitative prediction of the dissociation rate constants of p38 MAPK inhibitors.Characterized by the Volsurf and physicochemical descriptors,28 p38MAPK inhibitors were used for constructing QSAR models of the dissociation rate constants by stepwise regression and PLS modeling methods.The results show that:physicochemical descriptor t-veVSA and Volsurf descriptor V-OH2 are the crucial factors affecting the dissociation rates of p38 MAPK inhibitors.They characterize the surface area and molecular volume of p38 MAPK inhibitors,respectively.?4?Quantitative prediction of the dissociation rate constants of p38 MAPK inhibitors by protein-ligand interaction fingerprints.Molecular dynamics simulations and energy decomposition techniques were used to extract protein-ligand interaction?PLI?energy fingerprints along the ligand dissociation pathway.Then,the 90 fingerprint descriptors were used to establish prediction models of the dissociation rate constants by PLS modeling combined by genetic algorithm?GA?method.The results show that the protein-ligand interaction energy fingerprint can characterize the key ligand-receptor interaction information affecting the dissociation rates.The results show that 11?,18?,and 26?are the key positions affecting the dissociation rates of p38 MAPK type II inhibitors,and polar or charged residues along the dissociation pathway such as Arg⁶⁷,Arg⁷⁰,Arg¹⁴⁹,His¹⁹⁹,Glu³²⁸ and Ser³²⁹ have important influences on the ligand dissociation processes.