QSAR analysis of antitumor activity of new quinoline-arylamidine hybrids

The discovery of new and more potent antitumor molecules that act simultaneously on multiple targets is one of the most active fields of research in chemotherapy. The use of methodologies based on cheminformatics, including Quantitative- Structure–Activity Relationship (QSAR) techniques play important role in pharmaceutical design of more efficient antitumor drugs. Also, QSAR analysis is a practical tool for prediction of antitumor activity of new and untested antioxidant. New hybrid molecules with 7- chloroquinoline and arylamidine moieties have been the most effective compounds against leukemia cell lines (K562 and Raji). Antitumor activity was modelled using mono-dimensional (1D), to three- dimensional molecular (3D) descriptors. In order to find a thriving QSAR models for antitumor activities, genetic algorithm- multilinear regression (GA-MLR) was used. The best obtained QSAR models include the following group of descriptors: RDF (Radial Distribution Function) ; GETAWAY (Geometry, Topology, and Atom-Weights AssemblY) descriptors, electronic charge, and information indices. Obtained QSAR models relieved in elucidation of important physicochemical and structural requirements for this biological activity. Highly potent molecules should have increased: three- dimensional distribution of mass in molecules calculated at the radius of 110 Å, atomic distribution at neighbourhood symmetry of 1- order, electric dipole moment along X-axis, as well as more atoms higher electronegativity distributed at the topological distance 3. Based on the conclusions given in the QSAR analysis, a new compounds with possible great antitumor activity were proposed and predicted by means of obtained QSAR models, giving a guidance for further synthesis of new effective quinoline- arylamidine hybrid molecules.