Human metapneumovirus (HMPV) is a major cause of acute respiratory infections, especially among children and immunocompromised patients, but until now, no antiviral therapy has been approved. Nucleocapsid (N) protein, an ultra-conserved protein critical to the encapsidation of RNA and replication, is an attractive antiviral target. Herein, we used structure-based drug repurposing through the integration of high-throughput virtual screening, binding energy estimation by the merged molecular mechanics and generalized born/surface area (MM/GBSA) method, and long-timescale (1 μs) all-atom all-hydrogen atom molecular dynamics (MD) simulations to identify new HMPV N protein inhibitors. Of an initial set of 106 candidate compounds, two candidates, namely F0840-0186 and F3165-0800, emerged with the best docking scores (–6.81 and –6.99 kcal/mol) and MM/GBSA binding energies (–74.42 and –58.83 kcal/mol) upon the first round of screening. MD simulations validated the stability of these protein–ligand complexes with average root-mean-square deviation (RMSD) values of 2.5 and 2.8 Å to outperform the reference compound, namely ginkgolic acid (RMSD: 4.2 Å; ΔG_bind: –59.20 kcal/mol). Additional analysis on the structure stability and compactness of the complex included the free energy landscape and the principal component analysis. These computational results give predictive clues for potential HMPV inhibitors, for practical confirmation of the antiviral activity and drug potential.
GRAPHICAL ABSTRACT