Abstract
<jats:p>A series of novel cyclic alkene template (CAT) isosteres (CAT-0 to CAT-7) were designed and evaluated through an integrated in silico workflow combining molecular docking, density functional theory (DFT)-based electronic structure analysis, ADMET prediction, CYP450 profiling, and preliminary synthetic feasibility assessment. Two biologically relevant targets, HSP90AA1 and PTPN1, were investigated to explore potential anticancer-related activity profiles, with PTPN1 considered in light of its context-dependent and still debated role in cancer biology. Docking studies using a validated protocol on the 3ERT system confirmed reliable reproduction of the co-crystallized ligand pose (RMSD < 2 Å) and identified CAT-3, CAT-4, and CAT-7 as the most promising candidates in terms of predicted binding affinity across both targets. Electronic structure analysis suggested that variations in electrostatic potential distribution and frontier orbital characteristics may contribute to differences in binding behavior within the series, supporting the isosteric design strategy. ADMET and toxicity predictions indicated generally favorable drug-like properties, with most compounds exhibiting acceptable gastrointestinal absorption and low predicted CYP450 inhibitory activity, suggesting a limited risk of CYP-mediated drug–drug interactions. However, variability in blood–brain barrier permeability and model-dependent toxicity endpoints highlights the inherent uncertainty of in silico ADMET predictions and warrants cautious interpretation of absolute values. Among the designed compounds, CAT-3 emerged as the most balanced lead candidate, combining favorable docking performance, acceptable predicted pharmacokinetic properties, and improved synthetic accessibility relative to CAT-4 and CAT-7, which, despite strong predicted affinities, present greater synthetic constraints. Its structural relationship to trodusquemine further supports its relevance as a promising scaffold. Overall, CAT-3 represents the most favorable compromise between predicted biological activity and practical feasibility within the series. Nevertheless, given the limitations of empirical scoring functions, rigid receptor approximations, and predictive ADMET models, these findings should be regarded as hypothesis-generating. Further validation through molecular dynamics simulations, MM-GBSA free energy calculations, and experimental assays will be required to confirm the predicted biological relevance of the CAT series.</jats:p>