Ascomycin

Invasive fungal infections are a leading cause of death worldwide. Translating molecular insights into clinical benefits is challenging because fungal pathogens and their hosts share similar eukaryotic physiology. Consequently, current antifungal treatments have limited efficacy, may be poorly fungicidal in the host, can exhibit toxicity, and are increasingly compromised by emerging resistance. We have established that the phosphatase calcineurin (CaN) is required for invasive fungal disease and an attractive target for antifungal drug development. CaN is a druggable target, and there is vast clinical experience with the CaN inhibitors FK506 and cyclosporin A (CsA). However, while FK506 and its natural analog FK520 exhibit antifungal activity, they are also immunosuppressive in the host and thus not fungal-selective. We leverage our pathogenic fungal CaN-FK506-FKBP12 complex X-ray structures and biophysical data to support structure-based ligand design as well as structure–activity relationship analyses of broad-spectrum FK506/FK520 derivatives with potent antifungal activity and reduced immunosuppressive activity. Here, we apply molecular docking studies to develop antifungal C22- or C32-modified FK520 derivatives with improved therapeutic index scores. Among them, the C32-modified FK520 derivative JH-FK-44 ( 7 ) demonstrates a significantly improved therapeutic index compared to JH-FK-08, our lead compound to date. NMR binding studies with C32-derivatives are consistent with our hypothesis that C32 modifications disrupt the hydrogen bonding network in the human complex while introducing favorable electrostatic and cation–π interactions with the fungal FKBP12 R86 residue. These findings further reinforce calcineurin inhibition as a promising strategy for antifungal therapy.

Modulating protein-protein interactions (PPIs) is an attractive strategy in drug discovery. Molecular glues, bifunctional small-molecule drugs that promote PPIs, offer an approach to targeting traditionally undruggable targets. However, the efficient discovery of molecular glues has been hampered by the current limitations of conventional ensemble-averaging-based methods. In this study, we present a YaxAB nanopore for probing the efficacy of molecular glues in inducing PPIs. Using YaxAB nanopores, we demonstrate single-molecule-based, label-free monitoring of protein complex formation between mammalian target of rapamycin (mTOR) and FK506-binding proteins (FKBPs) triggered by the molecular glue, rapamycin. Owing to its wide entrance and adjustable pore size, in combination with potent electro-osmotic flow (EOF), a single funnel-shaped YaxAB nanopore enables the simultaneous detection and single-molecule-level quantification of multiprotein states, including single proteins, binary complexes, and ternary complexes induced by rapamycin. Notably, YaxAB nanopores could sensitively discriminate between the binary complexes or ternary complexes induced by rapamycin and its analogues, despite the subtle size differences of ∼122 or ∼116 Da, respectively. Taken together, our results provide proof-of-concept for single-molecule-based, label-free, and ultrasensitive screening and structure-activity relationship (SAR) analysis of molecular glues, which will contribute to low-cost, highly efficient discovery, and rational design of bifunctional modality of drugs, such as molecular glues.