In silico methods predict new blood-brain barrier permeable structure for the inhibition of kynurenine 3-monooxygenase

Abstract

Kynurenine 3-monooxygenase (KMO) regulates the levels of bioactive substances in the kynurenine pathway of tryptophan catabolism and its activity is tied to so many diseases that finding an appropriate inhibitor for KMO has become an urgent task. This especially proved to be difficult for the central nervous system related diseases due to the requirement that the supposed inhibitor should be both blood brain barrier permeable and should not cause hydrogen peroxide as a harmful side product. In this in silico study, we present our step-wise approach, whose starting point is based on the important experimental observations. To tackle the problem, a library of 7561938 structures was obtained from Zinc15 database utilizing the tranche browser. From this library, a subset of 501777 structures was determined with the considerations of their functional groups that constrain their applicability. Then, the binding affinity ranking of this set of structures was determined via virtual screening. Starting from the structures whose affinities are the highest among this subset, the ADMET properties were checked through in silico methods and the binding properties of the selected inhibitor candidates were further investigated via molecular dynamics simulations and MM/GBSA calculations. According to the computational results of this study, ZINC_71915355 has passed all the evaluations and is a potentially BBB permeable structure that can inhibit KMO. Additionally, ZINC_19827377 was identified as a new potential KMO inhibitor which may be more suitable for peripheral administration. From the in silico study presented herein, ZINC_71915355 and ZINC_19827377 structures, which showed high binding affinity without harmful H2O2 production, along with the tailored properties can now serve as powerful candidates for KMO inhibition and these hits are worth of further experimental validation.