A novel view of modelling interactions between synthetic and biological polymers via docking

Multipoint interactions between synthetic and natural polymers provide a promising platform for many topical applications, including therapeutic blockage of virus-specific targets. Docking may become a useful tool for modelling of such interactions. However, the rigid docking cannot be correctly applied to synthetic polymers with flexible chains. The application of flexible docking to these polymers as whole macromolecule ligands is also limited by too many possible conformations. We propose to solve this problem via stepwise flexible docking. Step 1 is docking of separate polymer components: (1) backbone units (BU), multi-repeated along the chain, and (2) side groups (SG) consisting of functionally active elements (SG(F)) and bridges (SG(B)) linking SG(F) with BU. At this step, probable binding sites locations and binding energies for the components are scored. Step 2 is docking of component-integrating models: [BU](m), SG = SG(F)-SG(B), BU-SG, BU-BU(SG)-BU, BU(SG)-[BU](m)-BU(SG), and [BU(var)(SG(var))](m). Every modelling level yields new information, including how the linkage of various components influences on the ligand-target contacts positioning, orientation, and binding energy in step-by-step approximation to polymeric ligand motifs. Step 3 extrapolates the docking results to real-scale macromolecules. This approach has been demonstrated by studying the interactions between hetero-SG modified anionic polymers and the N-heptad repeat region tri-helix core of the human immunodeficiency virus type 1 (HIV-1) envelope glycoprotein gp41, the key mediator of HIV-1 fusion during virus entry. The docking results are compared to real polymeric compounds, acting as HIV-1 entry inhibitors in vitro. This study clarifies the optimal macromolecular design for the viral fusion inhibition and drug resistance prevention.