In contrast to most enveloped viruses, poxviruses produce infectious particles that do not acquire their internal lipid membrane by budding through cellular compartments. Instead, poxvirus immature particles are generated from atypical crescent-shaped precursors derived from host membranes. Two key viral proteins participate to this process. A17 inserts in the membrane and is proposed to induce curvature. D13 forms a scaffold remodelling membranes into a closed, spherical particle. Rifampicin reversibly inhibits the formation of viral crescents in a process unrelated to its antibiotic effect.
We have determined the crystal structure of vaccinia virus D13 in complex with rifampicin at a resolution of 2.7Å. This structure identifies the central channel of the D13 trimer as the rifampicin binding site. The antibiotic contacts three conserved phenylalanine residues (F168, F486 and F487) that form a hydrophobic belt delineating the binding site.
Using CPMG, a ligand-detected NMR technique, we showed that rifampicin directly binds D13 and competes with A17. Consistently, SPR and NMR analysis show that mutations within the phenylalanine belt of the rifampicin-binding site severely reduce A17 binding, evidencing an overlap of binding sites. On the other hand, to map the D13 binding site in A17, a set of A17-derived peptides were assessed for binding to D13 by CPMG. Residues Tyr6 and Tyr7 were identified as the binding determinants for A17 interaction with D13.
We then used a classical fragment-based drug design approach to target the A17/rifampicin binding site in D13, screening a library of 1137 compounds by STD and CPMG NMR. This allowed to identify 25 fragments that bind D13 (representing a 2.2% hit rate). Out of these, 2 molecules with unrelated structures were found to compete with both rifampicin and A17. These two molecules represent lead compounds suitable for optimisation towards the design of assembly inhibitors against poxviruses.