Research Article 3D-QSAR ANALYSIS OF SERIES 3-GUANIDINOPROPIONIC ACIDS AS ANTIDIABETC AGENTS SATISH SAHU*, YOGESH POUNIKAR, LOPAMUDRA BANERJEE AND D V KOHLI Pharmaceutical Chemistry and Drug Design Research Laboratory, Dept. of Pharm. Sciences, Dr. H. S. Gour University, Sagar (M.P.), India- 469002. Email: satishbu@yahoo.co.in Received: 14 Aug 2012, Revised and Accepted: 30 Sep 2012 ABSTRACT The structures of series of 3-guanidinopropionic acids were submitted to molecular modeling software and after energy minimization and conformational analysis of the structures; a number of electronic, spatial and thermodynamic descriptors were calculated. Several statistical regression expressions were obtained using multiple regression analysis. Amongst them, one model was found to be best on various statistical criteria, involving the descriptor viz. steric parameters (molar refractivity) and hydrophobic parameters (hydrophobicity) with significant correlation coefficient. INTRODUCTION Non-insulin dependent diabetes mellitus (NIDDM) is a complex, chronic metabolic disorder characterized by a resistance of the peripheral target tissues to fully respond to the binding of insulin and insufficient insulin secretion by the pancreas to overcome this reduced response[1,2]. The result of these two pathologies is impaired glucose uptake and metabolism, leading to fasting hyperglycemia. The etiology of NIDDM is complex but is now generally accepted to entail the initial development of insulin resistance in the prediabetic state that leads to compensatory hyperinsulinemia. Eventually the β-cells of the pancreas can no longer maintain the hyperinsulinemic state, and the ensuing insulin deficiency leads to chronic hyperglycemia. Untreated NIDDM leads to several chronic diseases such as neuropathy, nephropathy and cardiovascular diseases[3]. The later lead to increase in mortality. At present, therapy for type II diabetes relies mainly on several approaches intended to reduce hyperglycemia itself: sulfonylureas, biaguanides, thiazolidinediones, α-glucosidase inhibitors, insulin sensitizer and insulin secretagogues. Meglasson et. al. reported that 3-guanidino-propionic acid possess both antihyperglycemic and antiobesity activity in KKA γ mouse, a rodent model of NIDDM.[4,5] Although the antidiabetic potential of lipophilic guanidine derivatives has been recognized by Bailey[6,7]. Survey of literature showed that biguanides have potential hypoglycemic agents. Though some research has been done on this molecule, off and on for the past century, the development of guanidine derivative (3- Guanidinopropionic acid) as a novel antidiabetic agent is yet to emerge. The extreme hydrophobicity of 3-guanidinopropionic acid (GPA) may offer an advantage over more lipophilic guanidine antidiabetic agents which have historically been associated with lactic acidosis, a potentially fatal overproduction of lactic acid resulting from inhibition of mitochondrial oxidative phosphorylation[8,9]. The higher observed incidence of lactic acidosis in patients receiving phenformin relative to that observed in patients treated with the closely related but markedly less lipophilic drug, metformin, tends strong clinical support the hypothesis that lipophilicity and toxicity are positively correlated.[9,10]. We therefore decided to study quantitative structure activity relationship (QSAR) of 3-guanidinopropionic acid analogues as antidiabetic agents[11]. MATERIALS AND METHODS The in-vitro transativation activity data of 3-guanidinopropionic acid analogues were taken from reported work of Larsen et al [11] (Table 1). The biological activity was converted to negative logarithm for QSAR analysis. For the present 3D-QSAR analysis Apex-3D expert system on a silicon graphics INDY-4000 was used. All molecular modeling and 3D-QSAR studies were performed on a silicon graphics INDY-4000 workstation employing molecular simulation software. A series of 55 compounds were taken as a training set. The molecular structure of all compounds were constructed in 2D using the sketch program in the builder module of INSIGHT-II software and then converted to 3D for optimization of their geometry (net charge 0.0) by selecting the forcefield potential action and charge action as fixed. The molecules structures were finally minimized using the steepest descent, conjugate gradients and Newton Raphson's algorithm followed by Quasi-Newton-Raphson. Optimization techniques implemented in Discover module (version 2.9) by energy tolerance value of 0.001 Kcal/mol and maximum number of iteration set at 1000. A total of 1193 conformers were generated for total molecules and lowest energy conformer of each cluster was selected by conformation clustering methodology. These conformations were subjected to different computational chemistry program including MOPAC 6.0 version (MNDO Hamiltonian) for the calculations of physicochemical parameters (-population, atomic charges, electron donor and acceptor indexes, HOMO and LUMO coefficient and hydrophobicity and molar refractivity based on atomic contributions) and quantum chemical parameters. The data was used by Apex-3D program for automated identification of biophores, superimposition of compounds and quantitative model building. Compounds present in the test have been predicted to check the validity of model. In addition to it "Leave One Out (LOO)" cross validation was also performed in which the objects were left out randomly but only once. On the basis of chance value, RMSA, RMSP, R and size, models have been selected which can be considered to be most robust model for the series. RESULT AND DISCUSSION Pharmacophore models with different size and arrangements were generated for the training set given in Table 2. Among several 3D biophoric models for all the molecules of training set, Model No. 50 was selected based on criterion and is given in Table 3. 1. R (Correlation coefficient) > 0.70 2. The difference between RMSA and RMSP < 0.02 3. Chance < 3 4. Number of variables < 5 5. Number of compounds as maximum as possible (n=49) N N O O H o o A B C 4 5 1 H ฀ = Biophoric site  = Secondary site International Journal of Pharmacy and Pharmaceutical Sciences ISSN- 0975-1491 Vol 5, Suppl 1, 2013 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