Semantic-Aware Graph Embedding Approach Uncovers LC-61, a Potent Anti-Leishmania infantum Compound

Abstract

Visceral leishmaniasis caused by Leishmania infantum remains a lethal disease with few therapeutic options, necessitating innovative computational methods approaches to accelerate drug discovery. Here, we present a semantic-aware graph neural network (GNN) framework that features holistic mechanisms to capture long-range molecular interactions and the chemical semantics of antileishmanial compounds. Across two classificatory antileishmanial datasets, our holistic GNNs demonstrated significant improvements in predictive performance, with area under the receiver operating characteristic curve (AUROC) increases of 2.2-29.2% on the unbalanced dataset (1 {micro}M threshold) and 3.4-22.5% on the balanced dataset (10 {micro}M threshold) compared to default GNNs. Subsequently, the framework was applied to screen a library of approximately 1.3 million compounds, pinpointing LC-61 as a potent antileishmanial agent with nanomolar activity against intracellular L. infantum (IC50 = 0.076 {micro}M) and minimal cytotoxicity to macrophages (THP-1 CC50 = 157 {micro}M). A comprehensive in vitro ADME profiling revealed that LC-61 combines high solubility at both acidic and physiological pH (>28 {micro}g/mL), balanced lipophilicity (eLogD = 4.07), and favorable passive permeability (PAMPA = 4.86 x 10-6 cm/s), while exhibiting lower microsomal stability. Overall, our semantic-aware GNN framework effectively accelerated the discovery of LC-61, a novel and biologically validated hit suitable for hit-to-lead optimization.