The molecular chaperone cpn60 binds many unfolded proteins and facilitates their proper folding. Synthetic peptides have been used to probe the question of how cpn60 might recognize such a diverse set of unfolded proteins. Three hybrid peptides were synthesized encompassing portions of the bee venom peptide, apamin, and the sequence KWLAESVRAGK from an amphipathic helix in the NH2-terminal region of bovine rhodanese. Two disulfides connecting cysteine residues hold the peptides in stable helical conformations with unobstructed faces oriented away from the disulfides. Peptides were designed to present either a hydrophobic or hydrophilic face of the amphipathic helix that is similar to the one near the amino terminus of rhodanese. Aggregation of these peptides was detected by measuring 1,1'-bis(4-anilino)napthalene-5,5'-disulfonic acid (bisANS) fluorescence at increasing peptide concentrations, and aggregation was not apparent below 2 microM. Thus, all experiments with the peptides were performed at a concentration of 1 microM. Reducing agents cause these helical peptides to form random coils. Fluorescence anisotropy measurements of fluorescein-labeled peptide with the exposed hydrophobic face yielded a Kd = approximately 106 microM for binding to cpn60, whereas there was no detectable binding of the reduced form. The peptide with the exposed hydrophilic face did not bind to cpn60 in either the oxidized or reduced states. Fluorescence experiments utilizing bisANS as a probe showed that binding of the helical hydrophobic peptide could induce the exposure of hydrophobic surfaces on cpn60, whereas the same peptide in its random coil form had no effect. Thus, binding to cpn60 is favored by a secondary structure that organizes and exposes a hydrophobic surface, a feature found in amphipathic helices. Further, the binding of a hydrophobic surface to cpn60 can induce further exposure of complementary surfaces on cpn60 complexes, thus amplifying interactions available for target proteins.
The proliferation of multi-drug-resistant bacteria underscores an urgent need for novel antibiotics. Traditional discovery methods face challenges due to limited chemical diversity, high costs, and difficulties in identifying structurally novel compounds. Here, we explore the integration of small molecule high-throughput screening with a deep learning-based virtual screening approach to uncover new antibacterial compounds. Leveraging a diverse library of nearly 2 million small molecules, we conducted comprehensive phenotypic screening against a sensitized Escherichia coli strain that, at a low hit rate, yielded thousands of hits. We trained a deep learning model, GNEprop, to predict antibacterial activity, ensuring robustness through out-of-distribution generalization techniques. Virtual screening of over 1.4 billion compounds identified potential candidates, of which 82 exhibited antibacterial activity, illustrating a 90X improved hit rate over the high-throughput screening experiment GNEprop was trained on. Importantly, a significant portion of these newly identified compounds exhibited high dissimilarity to known antibiotics, indicating promising avenues for further exploration in antibiotic discovery.
Phenotypic drug discovery has yielded many first-in-class small-molecule drugs by discovering modulators of disease phenotypes in physiologically relevant cellular systems. However, high-content phenotypic assays lack the ultra-high-throughput scalability of target-based screens. Recent advances in virtual screening present an opportunity to address this bottleneck, but have been limited to simple phenotypes like viability, restricted to small repurposing libraries, or lack in-depth biological validation. Here, we present PhenoCompass, a multimodal co-embedding model that aligns compound structures and high-content phenotypic imaging to enable virtual phenotypic screening over billion-compound libraries. Following training on the Joint Undertaking in Morphology dataset with more than 100,000 Cell Painting compound profiles, retrospective validation with historical biochemical high-throughput screening data demonstrates that PhenoCompass ranks compounds according to their biochemical target engagement. Leveraging PhenoCompass, we performed a prospective screen of 3.8 billion Enamine REAL compounds for inhibitors of PI3K/mTOR pathway, a critical signaling cascade whose aberrant activation is a common tumor driver. This search identified 11 novel compounds with pathway-consistent Cell Painting readout and diverse scaffolds, a 54-fold enrichment over the training set. Orthogonal validation experiments using a FOXO3A reporter assay and direct kinase inhibition confirmed seven structurally novel inhibitors with distinct mechanisms of action. These results highlight the convergence of diverse molecular target profiles onto a shared morphological pathway signature and establish PhenoCompass as a robust framework for high-content phenotypic virtual screening.