Background: Analysis of different types of cancer such as breast, prostate and ovarian carcinomas
have shown that protein kinase B is overexpressed, making this an important target in drug design
studies. We use various models of structure-based drug design to investigate the PKB target
screening half a million compound databases. Genetic algorithms and grid generation are the bases used
in the docking procedures in order to select the best hits that exhibit chemical, electronic and structural
characteristics that promote the appropriate interactions with the desired target. Toxicity is predicted
using databases implemented in software that create models based on chemical structures taking into
consideration the toxicophoric groups. A wide variety of pharmaceutical relevant properties is determined
in order to make decisions about molecular suitability. The molecular dynamics simulations indicate
that the inhibitor is stable in the docking site. After screening and analysis, we are able to suggest
the most promising lead compound.
Methods: The Structure-based virtual screening focuses on the three-dimensional information of the
therapeutic target. Docking procedures with the crystal structure of PKB inhibitor (PDB code 2X3D) in pH
= 7.0 using GOLD 5.1 (www.ccdc.cam.ac.uk) and GLIDE 2014-3 (www.schrodinger.com) select the
compounds that have the best interaction with the target to find new potential compounds, which are able
to inhibit PKB. The study works with different databases, such as Maybridge, Zinc and Chembridge.
Simulations of molecular dynamics with Discovery Studio (www.accelrys.com), ADME predictions with
QikProp (www.schrodinger.com) and toxicity prediction with Derek (www.lhasalimited.com) were the
next and final steps.
Results: From a chemical universe of half a million compounds from several databases it was possible
to select 4 top chemical structures with good inhibiting profile for PKB, with suitable ADME/Tox properties,
using a low cost strategy made possible by structure-based virtual screening techniques
Conclusion: Virtual screening strategies, such as docking have shown to be an important tool in the
search for new promising drugs in cancer. ADME/Tox analysis allows discarding problematic inhibitors
at an early stage of the Drug Discovery process. From a chemical universe of half a million compounds
from several databases it was possible to select 4 top chemical structures with good inhibiting profile
for PKB, with suitable ADME/Tox properties, using a low cost strategy made possible by structurebased
virtual screening techniques.