Abstract
Members of the ecto-nucleoside triphosphate diphosphohydrolase (E-NTPDase) family play a pivotal role in hydrolyzing nucleoside triphosphates and diphosphates, modulating purinergic and pyrimidinergic signaling pathways. The NTPDases have therapeutic potential; gaining structural insights into NTPDase-substrate complexes would be valuable for optimizing these enzymes for therapeutic applications. However, such insights remain limited, posing challenges for effective optimization. Molecular docking often fails to capture experimentally characterized substrate conformations, leading to biologically irrelevant models. To address this, we developed a computational strategy that preserves experimentally observed substrate features while leveraging the active site's conservation across NTPDases. Our method identifies a canonical linear-like substrate conformation encompassing the phosphate tail and nucleobase ring conserved across experimental NTPDase structures. This approach enabled the modeling of Homo sapiens (Hs) NTPDases (HsNTPDase1-8) complexed with ATP, ADP, GTP, GDP, UTP, and UDP, accurately positioning metal ion cofactor and catalytic water molecules. The resulting models offer a reliable framework for studying enzyme-substrate interactions, paving the way for rational enzyme engineering and therapeutic exploration.