Computational docking and structure–activity relationship analysis of triazine analogues as selective PDE4B and PDE4D inhibitors

Phosphodiesterase-4 (PDE4) enzymes are well-validated therapeutic targets implicated in inflammatory, neurological and oncological disorders. Among the PDE4 isoforms, PDE4B and PDE4D are of particular interest; however, their high structural similarity poses a major challenge for the development of isoform-selective inhibitors. To develop a ligand-based framework for predicting PDE4B and PDE4D selectivity in triazine analogues through systematic structure–activity relationship analysis and in silico pharmacokinetic profiling. An integrated in silico drug design approach was employed to evaluate thirty-four triazine-based ligands against PDE4B (PDB ID: 1XMU) and PDE4D using the GOLD 5.2 suite with the CHEMPLP fitness function. Docking analyses focused on binding scores, hydrogen-bond networks, interaction energies and hydrophobic fitting points to identify stable binding conformations. Docking protocol reliability was confirmed by redocking the co-crystallised ligand, yielding RMSD values below 2.0 Å. Binding modes were further examined using three-dimensional and two-dimensional interaction mapping. The most promising ligands were additionally evaluated for absorption, distribution, metabolism, excretion and toxicity (ADMET) properties using SwissADME and pkCSM. For PDE4B, docking scores ranged from 139.84 to 158.84, with ligands Aj11, Aj16, Aj24, Aj25 and Aj31 emerging as top candidates. Aj11 and Aj31 showed enhanced stability through multiple interactions with HIS278 and ASP348, while Aj25 exhibited the strongest hydrogen bonding with HIS234 (2.2 Å) and the most compact hydrophobic accommodation. In the case of PDE4D, the highest docking scores were observed for Aj5 (121.59), Aj32 (113.22), Aj31 (112.11), Aj11 (106.13) and Aj8 (105.05). Aj32 formed dual interactions with ASP318 and THR271, whereas Aj31 displayed the strongest hydrogen bonds with ASP318 (2.321–2.521 Å) and favourable hydrophobic fitting. Overall, Aj25 (PDE4B-selective) and Aj31 (PDE4D/dual inhibitor) emerged as the most promising candidates based on high docking scores, optimal hydrogen-bond geometry, compact hydrophobic accommodation, and favourable predicted ADMET profiles. These findings highlight triazine derivatives as promising scaffolds for further optimisation toward selective PDE4 inhibition.