Abstract / Summary
Abstract The continuous mutation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) poses a serious challenge to existing antiviral drugs, underscoring the urgent need to develop broad-spectrum antiviral agents targeting conserved regions of the virus. Mac1 (Macrodomain 1) is a highly conserved domain of the SARS-CoV-2 nsp3 protein with ADP-ribose hydrolase activity, playing a key role in viral evasion of the host innate immune response, and has been validated as an antiviral drug target by the international CACHE Challenge. In this study, using LRH-0003 (IC₅₀ = 1.7 µM) as the lead compound and its co-crystal structure with Mac1 (PDB: 5SRY), we performed structure-guided derivative design and two rounds of iterative optimization. Through CB-Dock2 molecular docking, the binding affinities of 16 derivatives were evaluated, and L-5-A—a derivative in which a carboxyl group was introduced at the C6 position of the LRH-0003 core scaffold and the pyrrolidinone ring was expanded to a piperidinone ring—achieved a Vina score of − 9.3 kcal/mol, an improvement of 1.1 kcal/mol over the parent compound (− 8.2 kcal/mol). PLIP interaction analysis indicated that the improved binding affinity of L-5-A originates from a strong 2.75 Å hydrogen bond between the C6 carboxyl group and the backbone NH of PHE156, together with a newly formed aromatic–hydrophobic contact (3.03 Å). ADMET prediction showed that L-5-A satisfies all criteria of Lipinski’s Rule of Five (0 violations), with predicted GI absorption of “High” and blood–brain barrier permeability of “No”, suggesting favorable oral bioavailability and a low risk of central nervous system side effects. This study provides clear structural guidance for the further optimization of Mac1 inhibitors, and L-5-A, as a novel Mac1 inhibitor candidate with strong binding affinity and favorable drug-likeness, warrants further experimental validation of its antiviral activity.