Jul 2026· Journal of Medicinal Chemistry· Vol 69, pp. 16234 - 16243· 0 citations· 34 references
Medicine
TL;DR
A comparative, structure-driven analysis of PD-L1 complexes deposited in the Protein Data Bank is presented and it is demonstrated that distinct inhibitory modalities converge on the same functional CC′FG hotspot region while employing fundamentally different neutralization mechanisms.
Abstract
Protein–protein interactions dominated by large, flat interfaces are widely considered challenging drug targets. The programmed cell death protein-1/programmed death ligand-1 (PD-1/PD-L1) immune checkpoint exemplifies this problem, as the interaction is mediated by an extended β-sheet surface lacking deep pockets. Despite this, PD-L1 has been successfully inhibited by chemically distinct modalities, including antibodies, macrocyclic peptides, and small molecules. Here, we present a comparative, structure-driven analysis of PD-L1 complexes deposited in the Protein Data Bank and demonstrate a striking convergence: all effective inhibitors engage the same CC′FG β-sheet face of PD-L1. Antibodies directly occlude this surface, macrocyclic peptides such as pAC65 reproduce antibody-like surface coverage in a compact and preorganized scaffold, and biphenyl small molecules neutralize the same epitope indirectly by inducing PD-L1 homodimerization. This unified structural framework reveals modality-agnostic design principles for targeting flat immune checkpoint PPIs. This Perspective provides a unified structural framework for understanding PD-L1 inhibition across clinically tested antibodies, macrocyclic peptides, and small molecules. Both visualizing and quantitatively comparing interface overlap, hotspot conservation, and buried surface area, the work demonstrates that distinct inhibitory modalities converge on the same functional CC′FG hotspot region while employing fundamentally different neutralization mechanisms. These findings establish structure-guided principles for the rational design of next-generation PD-L1 modulators across diverse therapeutic modalities.
Fragment-based diffusion modeling is an efficient and interpretable strategy for the discovery of computationally prioritized PD-L1 small-molecule candidate inhibitors and offers a promising framework for tackling challenging targets in cancer immunotherapy.
Jun Liu, Yuxing Yi, Xiaoyan Wu et al.· Molecular diversity· 0 citations
The quaternary ammonium strategy represents an effective approach to developing water-soluble PD-1/PD-L1 inhibitors with dual immunomodulatory and antiangiogenic functions.
Jianwei Xu, Pan Yang, Xixiang Yang et al.· Journal of Medicinal Chemist...· 0 citations
The molecular basis of G4 recognition by CX-5461 and MTR-106 is elucidated, and a structural framework for the rational development of next-generation G4-targeted therapeutics with improved selectivity and efficacy is provided.
Yubo Li, Kefu Liu, Xiaodong Hu et al.· Nucleic Acids Research· 1 citation
AlphaFold 3 (AF3) and Boltz-2 are state-of-the-art AI-based tools for biomolecular structure prediction, but whether their predictions provide useful guidance for lead optimization, SAR interpretation, and virtual screening remains insufficiently characterized. We benchmarked their performance using newly determined soluble epoxide hydrolase co-crystal structures and matched activity data together with a curated post-training-cutoff dataset spanning kinases, allosteric modulators, covalent systems, PROTACs, molecular glues, fragments, membrane proteins, RNA binders, and activity-cliff pairs. Both models recovered canonical orthosteric enzyme and kinase complexes, including key DFG/αC conformational states, whereas allosteric, membrane-protein, and induced-proximity complexes remained challenging. Pharmacophore RMSD was often lower than overall ligand RMSD, indicating preservation of key recognition features despite imperfect whole-ligand alignment. AF3 minPAE correlated with pose accuracy, and very low minPAE values (<0.85 Å) were strongly enriched for accurate poses. Model confidence scores were not associated with experimental activity, whereas Boltz-2 predicted affinity captured relative activity trends and distinguished the activity-cliff pair, although its performance varied across ligand series.
Ke Chen, Zuo-Huang Qi, Omar Lozano Ramos et al.· bioRxiv· 0 citations
Mediation of protein–protein interactions with molecules that bind strongly and selectively to one of the partners at the protein interface is a promising therapeutic strategy for myriad diseases. One such approach is the rational design of non-peptidic scaffolds that reproduce the display of amino acid side chains from one face of a secondary structural element. We have previously disclosed proof-of principle syntheses of β-strand mimetics composed of alternating (hetero)aromatic and cyclic urea units, conformationally preorganised through dipolar repulsion in organic solvents, that are in good agreement with the i, i + 2, and i + 4 side-chain vectors of a canonical strand. Here we demonstrate sequence diversity of the approach through the incorporation of hydrophobic and hydrophilic side-chain mimics via an improved synthetic route. The scaffold is conformationally preorganised for target binding in aqueous media including buffer, is readily soluble, and thus is suitable for elaboration and deployment against specific protein targets.
R. Bannister, Emily F Jones, Jonathan E Ross et al.· Organic and biomolecular che...· 0 citations
Design and development of theranostic gold nanostructures functionalized with PD-L1-targeting peptides (PTP, sPTP, and rPTP) whose sequences were identified combining structural analysis of the PD-1/PD-L1 interaction interface and molecular dynamics simulations are reported.
Micaela Giannetti, Marina Gobbo, Lucio Litti et al.· ACS Applied Bio Materials· 0 citations