RESUMO
A hypoxic environment occurs predominantly in tumors. During the growth phase of a tumor, it grows until it exceeds its blood supply, leaving regions of the tumor in which the oxygen pressure is dramatically low. They are virtually absent in normal tissues, thus creating perfect conditions for selective bioreductive therapy of tumors. To this aim, a novel series of cytotoxic radiosensitizer agents were synthesized by linking the nitroimidazole scaffold with oxadiazole or triazole rings. The majority of the compounds exhibited moderate to excellent antiproliferative activities toward HCT116 cell line under normoxic and hypoxic conditions. The structure-activity relationship study revealed that compounds containing the free thiol group either in the oxadiazoles 11a,b or the triazoles 21a,b-23a,b demonstrated the strongest antiproliferative activity, which proves that the free thiol group plays a crucial role in the antiproliferative activity of our compounds under both normoxic (half-maximal inhibitory concentration [IC50 ] = 12.50-24.39 µM) and hypoxic conditions (IC50 = 4.69-11.56 µM). Radiosensitizing assay of the four most active cytotoxic compounds 11b and 21-23b assured the capability of the compounds to enhance the sensitivity of the tumor cells to the DNA damaging activity of γ-radiation (IC50 = 2.23-5.18 µM). To further investigate if the cytotoxicity of our most active compounds was due to a specific signaling pathway, the online software SwissTargetPrediction was exploited and a molecular docking study was done that proposed cyclin-dependent kinase 2 (CDK2) enzyme to be the most promising target. The CDK2 inhibitory assay assured this assumption as five out of six compounds demonstrated a comparable inhibitory activity with roscovitine, among which compound 21b showed threefold more potent inhibitory activity in comparison with the reference compound. A further biological evaluation proved compound 21b to have an apoptotic activity and cell cycle arrest activity at the G1 and S phases. During the AutoQSAR analysis, the model demonstrated excellent regression between the predicted and experimental activity with r2 = 0.86. Subsequently, we used the model to predict the activity of the test set compounds that came with r2 = 0.95.
