Human Leukocyte Elastase Inhibitors: Analog Based Studies to Design Novel Lead Compounds for the Treatment of COPD

Human Leukocyte Elastase is a serine protease that is released from leukocytes upon inflammatory stimulus. HLE is thought to aid in the migration of neutrophils to extra vascular compartments throughdegradation of a number of structural proteins including elastin. Normally this enzyme is kept in check by indigenous inhibitors, most notably alpha-1 proteinase. However alpha-1 p1 may be damaged by cigarette smoking, or because of genetic defect, produced in insufficient quantities. in either case , the balance between HLE and alpha -1 p1 is upset which can lead to tissue damage which manifests itself in diseases such as chronic obstructive pulmonary diseases, asthma, emphysema ,cystic fibrosis, and adult respiratory distress syndrome. Consequently there has been concerted effort to develop low molecular weight inhibitors of HLE to treat these disease states.We have performed Pharmacophore and 3D-QSAR studies for developing novel HLE inhibitors using the Catalyst 4.7 and Cerius2 program suite respectively. QSAR equations has been generated for 58 HLE inhibitors employing Molecular Field Analysis (MFA) as well as Receptor surface Analysis (RSA) using Genetic function approximation (GFA) as regression method. The best equations with training set consisting 40 molecules, produced r2 value of 0.845 and r2cv value of 0.839 in MFA-model and r2 value of 0.880 & r2cv of 0.856 in the RSA-model. For the 18 test set molecules predicted activities have correlation of 0.845 and 0.880 for MFA and RSA with observed activities.Pharmacophore models were generated using 21 molecules as training set. The best quantitative pharmacophore model consists of hydrogen bond acceptor, hydrophobic aliphatic and hydrophobic aromatic feature. For the training set the accuracy in predicting active and inactive compounds was 80%. The best pharmacophore hypothesis yielded a RMS deviation of 1.06 and a Correlation coefficient of 0.943 with a Cost difference (Null cost minus Total cost) of 79.02. The obtained pharmacophore models were validated on 50 test molecules to give correlation value of 0.846.