Funneling modulatory peptide design with generative models: Discovery and characterization of disruptors of calcineurin protein-protein interactions

摘要

Design of peptide binders is an attractive strategy for targeting "undruggable" protein-protein interfaces. Current design protocols rely on the extraction of an initial sequence from one known protein interactor of the target protein, followed by in-silico or in-vitro mutagenesis-based optimization of its binding affinity. Wet lab protocols can explore only a minor portion of the vast sequence space and cannot efficiently screen for other desirable properties such as high specificity and low toxicity, while in-silico design requires intensive computational resources and often relies on simplified binding models. Yet, for a multivalent protein target, dozens to hundreds of natural protein partners already exist in the cellular environment. Here, we describe a peptide design protocol that harnesses this diversity via a machine learning generative model. After identifying putative natural binding fragments by literature and homology search, a compositional Restricted Boltzmann Machine is trained and sampled to yield hundreds of diverse candidate peptides. The latter are further filtered via flexible molecular docking and an in-vitro microchip-based binding assay. We validate and test our protocol on calcineurin, a calcium-dependent protein phosphatase involved in various cellular pathways in health and disease. In a single screening round, we identified multiple 16-length peptides with up to six mutations from their closest natural sequence that successfully interfere with the binding of calcineurin to its substrates. In summary, integrating protein interaction and sequence databases, generative modeling, molecular docking and interaction assays enables the discovery of novel protein-protein interaction modulators. Author summaryPeptides that efficiently bind a target protein and interfere with its native protein-protein interactions are attractive reagents for basic research and therapeutic applications. However, rational peptide design remains challenging, as i) exhaustive exploration of the vast sequence space is impossible, ii) generically, there is a mismatch between selection criteria and target objectives, and iii) additional constraints such as low toxicity are frequently critical. Here, we present an integrative peptide design protocol based on a sequence generative model trained on native protein interactors of the target. We tested our protocol on Calcineurin, a serine/threonine phosphatase involved in multiple health and disease pathways. We showed that the generative model i) enables extensive exploration of the sequence space, ii) approximates well binding affinity to the target, and iii) yields highly diverse candidate sequences. After further selection via molecular docking and high-throughput binding assay, we found that 70% of the designed peptides successfully interfered with Cn-substrate interactions. Our integrative protocol could thus be broadly applicable to the rational design of protein-protein interaction disruptors.