Rational design of Tamiflu derivatives targeting at the open conformation of neuraminidase subtype 1.

Neuraminidase is an attractive therapeutic target for standing against influenza virus, such as the threatening avian influenza virus H5N1. A recently discovered cavity near the well-known catalytic site on neuraminidase subtype 1 (N1) provides a good possibility to develop dual-site-binding inhibitors, which may achieve improved activities and selectivities against N1. We have designed some derivatives of Tamiflu with such features through a fragment-based approach combining multiple computational methods. Over 1000 FDA-approved small-molecule drugs were computationally screened targeting at the open conformation of N1 with the GOLD program in combination with the X-Score scoring function. Some chemical fragments on the top-scored hits, which were able to fit into the 150-cavity, were transplanted onto the core structure of Tamiflu to produce a total of 30 new molecules. Then, binding of these designed molecules to N1 was evaluated by molecule docking. The promising ones were further subjected to molecular dynamics simulation of 3 ns long, and their binding free energies were computed by using the MM-PB/SA method. Some of our designed molecules were predicted to have comparable or even better binding affinities than that of Tamiflu. We report our results herein so that other researchers who have the necessary chemical and biological resources can utilize them in the development of new N1 inhibitors. In addition, our study actually suggests a practical strategy for optimizing a given lead compound based on the outcomes of a standard virtual screening trial.