3D QSAR AND MOLECULAR DOCKING STUDIES OF STRUCTURALLY DIVERSE ESTROGEN RECEPTOR LIGANDS Research Article MD ATAUL ISLAM, ARUP MUKHERJEE, ACHINTYA SAHA Department of Chemical Technology, University of Calcutta, 92, A.P.C.Road, Kolkata 700009, India. Email: achintya_saha@yahoo.com * Received: 28 Oct 2011, Revised and Accepted: 22 Dec 2011 ABSTRACT Hormone-responsive breast cancer is one of leading cause of cancer death world wide in the women community. The female sex hormone estrogen primarily controls the development of female sex characteristics including division of breast cell. This hormone exerts its effects after binding to estrogen receptor (ER), which is nuclear-activated transcription factor. The present study is considered to explore important pharmacophore signals for binding affinity of estrogen ligands using molecular field (CoMFA) and similarly analyses (CoMSIA), substantiated with molecular docking study. Both CoMFA (R 2 =0.974, se=0.240, Q 2 =0.589, R 2 pred =0.612) and CoMSIA (R 2 =0.997, se=0.088, Q 2 =0.703, R 2 pred Keywords: Estrogen, CoMFA, CoMSIA, Molecular docking. =0.624) models suggest that steric and electrostatic factors are crucial for binding affinity. Further, the similarity analysis and docking studies revealed that hydroxyl and alkyl groups are important for formation of potential interactions at the active site cavity of the ER. INTRODUCTION The estrogen belongs to the sex steroid hormones, secreted by the ovaries and testis with involvement of placenta, adipose tissue, and adrenal glands 1 . Among the several structurally related forms 17β- estradiol is found as predominant. Estrogen plays crucial role in female reproductive system and also exerts important effects on nonreproductive targets such as bone, cardiovascular system and neural sites involved in cognition 2 . It is also reported that estrogen influence the brain centers that maintain body temperature, and enable the vaginal lining to stay thick and lubricated 3 . Due to loss of estrogen production after menopause, hot-flushes, vaginal atrophy and sleeping disturbance arise, and also rise of low-density lipoprotein (LDL) that progressively increases the chance of coronary and osteoporosis diseases 1 . The hormone replacement therapy (HRT) in which synthetic estrogens are administered into the body that reduce osteoporotic fractures and improve severe menopausal symptoms 4 , but on other hand malevolent aspect of HRT is increasing chance of breast and uterin cancers 5,6 . Presently there are three strategy for treatment of hormone-responsive breast cancer, such as inhibition estrogen from binding to its main target estrogen receptor (ER) using antiestrogen, e.g. tamoxifen 7 ; preventing its synthesis using aromatase inhibitor 8 ; and down-regulating ER protein level using pure antiestrogen, e.g. fulvesteron 9 Estrogen mediates its biochemical mechanism in target tissues after binding to intracellular receptor proteins ER . 10,11 , which is a nuclear ligand-activated transcription factor 12 . ER constituted similar architecture to the other 50-60 members of the steroid/thyroid hormone receptor family 12-14 and comprises six distinct domains A–F. The ligand binding domain (LBD) consisting E/F domain at the carboxy terminal and responsible for ligand binding, receptor dimirization, nuclear translocation and transactivation of target gene expression via activation function – 2 (AF-2) 13,14 . The AF-2 region comprises of 12α- helices, which form a hydrophobic pocket responsible for binding of ligand 15 and fundamental in distinguishing between agonist and antagonist functions 16 of ER. Knowledge of ER has permitted the modeling of estrogenic activity using different chemometric techniques. The structural requirement for binding of steroid to ER is essential both for design of new drug and to evaluate the health risk of chemical of ER affinity 17 . The chemometric drug design (CDD) is widely used to design lead molecules involving two important techniques, ligand-based and structure-based approaches. When the properties of the ligands are analyzed without any information of receptor site is known as ligand-based drug design (LBDD), while the ligands are designed with help of receptor site is called structure-based drug design (SBDD). Researchers are devoted to search potent molecules for treatment of post-menopausal diseases using both LBDD and SBDD approaches. Our group has explored the prime pharmacophore signals for estrogen mediated bioactivities of different groups of structural congeneric compounds through Quantitative Structure Activity Relationships (QSAR) studies 18-21 . Molecular docking and QSAR studies of estrogen ligands are also explored for potent antiestrogens in several studies 22,23 . On the availability of the crystal structure of active site both ligand-based and structure-based studies will be powerful methods to design lead compound. The present work is considered to explore both approaches for a set of structurally diverse compounds 24,25 MATERIALS AND METHODS with respect to binding affinity to ER. The main objective of the work is to find out correlation between chemical structure with biological activity of the molecules using Partial Least Square (PLS) method and potential interactions between ligand and active site of the receptor molecule. The binding affinity for QSAR study is expressed as kRBA=log 10 (100xRBA). The dataset (Table 1) is randomly divided into training set (n tr =25) containing most and least active compounds, and test set (n ts =10) to validate the derived models in QSAR study. The different control parameters are used to check the superiority of 3D QSAR models are: R 2 (correlation coefficient), se (standard error of estimate), Cross-validated variance (CVV) Q 2 (Leave-One-Out (LOO) cross-validated 26 correlation), F (variance ratio) with df (degree of freedom), R 2 bs (bootstrapped correlation coefficient) and s b (standard error of bootstrapped correlation). To evaluate the predictive power of the model, R 2 pred and s p (standard error of prediction) of the test set are also estimated. In case of docking study GlideScore 27 3D QSAR and interactions between ligand and receptor are considered for best pose selection. QSAR is mathematical robust model which attempts to find a statistically significant correlation between chemical structure and biological activity 28 . 3D QSAR is a ligand-based approach which International Journal of Pharmacy and Pharmaceutical Sciences ISSN- 0975-1491 Vol 4, Suppl 1, 2012 A A c c a a d d e e m mi i c c S Sc c i i e e n n c c e e s s