Predicting PROTAC-mediated ternary complex structures through in silico approaches

Degradation of proteins by PROteolysis TArgeting Chimera (PROTAC) technology has recently emerged as a promising new approach with more advantageous over occupancy-based inhibitors. PROTACs are heterobifunctional compounds featuring two binders connected by a chemical linker: one binds the protein of interest (POI) and the other recruits an E3 ubiquitin ligase. Hijacking POI in close proximity to E3 ligase, a PROTAC molecule promotes the degradation of the target protein by the ubiquitin-proteasome system (UPS). In this mechanism, the formation of a stable ternary complex is essential to have an efficient POI ubiquitination and therefore, the knowledge of its 3D structure is a key aspect for the rational design. To date, structural data (e.g., by X-ray) of few ternary complexes are available and they are limited to small number of target proteins, e.g., BRD4, SMARCA2/4, BTK, and WDR5. We here propose an innovative computational methodology to predict PROTAC-mediated ternary complex structures. Our method uses protein-protein docking experiments to determine all possible interactions between POI and E3 ubiquitin ligase, then inserts the PROTAC molecule so that each ligand reproduces its binding mode in the respective target binding pocket. Finally, the obtained ternary complex models are refined by multiple rounds of metadynamic simulations, which show the conformational changes of the two proteins and PROTAC occurring until they converge to a thermodynamically favoured state (Figure 1). The applicability and the accuracy of this approach is evaluated reconstructing a series of experimentally determined ternary structures, which differ by both target protein and E3 ubiquitin ligase. This computational methodology could be a useful tool to guide and accelerate the rational design of new PROTACs related to proteins undruggable or not extensively studied. In addition, it could elucidate the mechanism of actions of already developed PROTACs whose 3D structures have not been resolved, aiding the design of novel derivatives with improved pharmacological features.