New quinazoline derivatives with anti-inflammatory activity and computationally predicted PDE7A binding: synthesis,...

Quinazoline scaffolds are widely recognized in drug discovery for their diverse pharmacological profiles, and several quinazoline derivatives have been reported as phosphodiesterase 7A (PDE7A) inhibitors with potential anti-inflammatory effects. This study aimed to synthesize novel quinazoline derivatives, evaluate their anti-inflammatory activity and investigate their computationally predicted binding to PDE7A. A series of quinazoline hydrazone (6a-j) and benzylthioquinazolinone (7a-c, 8a,b) derivatives were synthesized and structurally characterized using IR, ¹H/¹³C NMR, mass spectrometry, and elemental analysis. Anti-inflammatory activity was evaluated in vivo using the carrageenan-induced rat paw edema model. An acute oral toxicity assessment identified the highest tested doses without observed mortality or overt clinical toxicity of 1000-1200 mg/kg, from which anti-inflammatory test doses of 100-120 mg/kg were selected. Compound 6f produced the highest numerical inhibition among the synthesized derivatives, with 79.39% inhibition at 100 mg/kg. Compounds 6a, 6b, and 8a produced 61.85%, 61.65%, and 60.12% inhibition, respectively, at 120 mg/kg, whereas diclofenac produced 68.21% inhibition at 7.5 mg/kg. Molecular docking, Prime MM-GBSA calculations, and 100-ns molecular dynamics simulations were used to evaluate predicted PDE7A interactions and comparative dynamic behavior. Compound 6f showed the most favorable Glide GScore and IFDScore, persistent Tyr211 and Gln413 hydrogen bonds, and the clearest apparent late-stage stationary root-mean-square deviation regime. In silico ADMET analysis indicated acceptable predicted drug-likeness and absorption-related properties, although moderate-to-low aqueous solubility and predicted hERG liability require experimental evaluation. Overall, compound 6f was prioritized as the leading anti-inflammatory candidate within the evaluated series; however, PDE7A involvement remains computationally inferred and requires direct biochemical and selectivity validation.