Editorial: Computational Drug Discovery for Targeting of Protein-Protein Interfaces

Protein-protein and protein-peptide recognition are central in most physiological and pathological cellular processes. Unveiling the molecular basis of these mechanisms of interaction could be the key to understand fundamental biological functions and the underlining causes of complex diseases. In the development of novel therapeutics, targeting specific pathways has become a preferred strategy to overcome issues of specificity and side effects of drugs, but to this end a thorough understanding of the molecular details of PPI is essential. The pandemic emergency brought by the SARS-CoV-2 Virus is the most present-day example of how molecular details on protein-protein interactions and protein binding surfaces are critical to develop treatments for this class of diseases. Theoretical and computational methods have gained increasing importance in molecular recognition studies. They offer faster cost-effective strategies and can provide directions for experimental design and interpretation of results. For this reason, we believe that the development and application of new theoretical methods to study protein-protein and protein peptide recognition is a timely and exciting topic. These motivations gave us the cue to write this Research Topic that received a total of 7 contributions: 1 review, 1 perspective, and 5 original researches. The review of Shechter et al. compares and contrasts conventional and in silico High Throughput Screening approaches to identify agents targeting Nuclear Import Inhibitors for Venezuelan Equine Encephalitis Virus Capsid Protein. They conclude that both methods could be a good starting point to enhance the other approach: HTS could give a good starting point for in silico drug design process and in silico techniques could direct the expensive HTS to more promising leads. Two novel methods were proposed, one by Santini and Zacharias and one by Iannuzzi et al.. As underlined by Santini and Zacharias, sharp inhibition of pathological PPIs is of significant clinical relevance. The Authors propose a general methodology to quickly develop potential inhibitors by designing cyclic peptides that are able to mimic the epitope, i.e., a portion of the surface responsible for the PPI. They propose an automatic procedure to find the inhibitor within a pre-build cyclic peptide database. The database was built from the analysis of protein-protein complexes present in the Protein Data Bank (www.rcsb.org). The study of Iannuzzi et al. is focused on a specific protein-protein interaction of essential biomedical relevance, namely the T-cell-antigen interactions. They develop an in silico peptide library for the study of in vitro or in vivo T-cell reactivity (Iannuzzi et al.). The library was built from the docking results of 216 generated peptide against some T cell receptors. In the remaining four original research contributions we can appreciate how computational methods like molecular dynamics, meta-dynamics, and in silico screening can be integrated with experiments to find inhibitors of protein-protein interactions or to elucidate the mechanism of binding of a peptide to a protein. As remarked by Yun et al. in their paper the hedgehog (Hh) signaling pathway responsible of embryogenic and tissue homeostasis is overactivated in many cancers. So, the inhibition of this signaling pathway could stop the proliferation of cancer cells. The target along this pathway is Sonic hedgehog (Shh) and in particular its interaction with 12-transmembrane glycoprotein patched (Ptch) protein. The authors were able to identify by virtual screening some inhibitors with an experimental activity on cells, these compounds are a good starting point for further development. Dailing et al. discuss the inhibition of a ternary protein complex r