Importance of retention data from affinity and reverse-phase high-performance liquid chromatography on antitumor activity prediction of imidazoacridinones using QSAR strategy.

Quantitative structure-activity relationships (QSAR) studies for prediction of cytotoxic and antitumor activity of imidazoacridinones (IA) based on experimentally obtained high-performance liquid chromatography (HPLC) retention data and calculated parameters using computational (molecular modeling) medicinal chemistry methods were proposed. The RP-HPLC and affinity-HPLC chromatographic techniques with four diversified HPLC systems applying columns with octadecylsilanes (C18), phosphatidylcholine (IAM), as well as α(1)-glycoprotein (AGP) and albumin (HSA) were used for the determination of the retention constants logk and logk(w) which characterize lipophilicity and protein affinity of IA. Moreover, molecular modeling studies were performed using HyperChem and Dragon software's, and structural descriptors were calculated and subsequently used. The QSAR equations using multiple linear regression (MLR) analysis method were derived which indicated that in vivo antileukemia activity of IA depends on cytotoxic activity against leukemia cells, whereas this cytotoxic activity depends on logk and logk(w) parameters obtained on all HPLC systems. Moreover, the QSRR equations were derived and indicated that logk and logk(w) parameters depend on calculated non-empirical structural parameters. The predictive power of obtained QSAR and QSRR equations allowed the prediction of cytotoxic and antitumor activity of IA and also their HPLC retention parameters. Finally, the equations can be used for prediction of antileukemia activity of IA without the necessity of carrying out experimental measurements.