Philippine Science Letters
vol. 6 | no. 2 | 2013
published online November 29, 2013


Multiple linear regression model of Shiga toxin inhibitory activity of dihydroquinazolinone derivatives of Retro-2cycl

by Liza T. Billones1 and Junie B. Billones2*

1Department of Epidemiology and Biostatistics, College of Public Health
2Institute of Pharmaceutical Sciences, National Institutes of Health
 and Department of Physical Sciences and Mathematics, College of Arts and Sciences
 University of the Philippines Manila
 Taft Avenue, Ermita, Manila, Philippines 1000



Quantitative Structure-Activity Relationship (QSAR) study was carried out on dihydroquinazolinone derivatives of Retro-2cycl using multiple linear regression methodology. The model shows that the activities of these compounds against Shiga toxin (Stx) are dependent on the partial charges on N10 and C3, absolute hardness (η), and a lipophilicity parameter milogP. Stx tend to be inhibited by a lipophilic and electronically soft compound with electron-rich N10 and electron deficient C3 atoms. Furthermore, a retro-2cycl derivative is likely to be active against Stx if its AM1-level HOMO energy is above -8.74 kcal/ mol and its topologic polar surface area (tPSA) does not exceed 35.0 ┼2. These data are crucial in structure optimization of Retro- 2cycl-based Shiga toxin inhibitors.

*Corresponding author
Email Address:
Submitted: July 25, 2013
Revised: September 16, 2013
Accepted: September 26, 2013
Published: November 29, 2013
Editor-in-charge: Gisela P. Padilla-Concepcion