Search

Novel heteroaryl-containing benzamide derivatives were synthesized and screened using an in vitro assay measuring increases in glucose uptake and glucokinase activity stimulated by 10mM glucose in rat hepatocytes. From a library of synthesized compounds, 3-(4-methanesulfonylphenoxy)-N-[1-(2-methoxy-ethoxymethyl)-1H-pyrazol-3-yl]-5-(3-methyl pyridin-2-yl)-benzamide (19e) was identified as a potent glucokinase activator with assays demonstrating an EC50 of 315nM and the induction of a 2.23 fold increase in glucose uptake. Compound 19e exhibited a glucose AUC reduction of 32% (50mg/kg) in an OGTT study with C57BL/6J mice compared to 28% for metformin (300mg/kg). Single treatment of the compound in C57BL/J6 and ob/ob mice elicited basal glucose lowering activity, while in a two-week repeated dose study with ob/ob mice, the compound significantly decreased blood glucose levels with no evidence of hypoglycemia risk. In addition, 19e exhibited favorable pharmacokinetic parameters in mice and rats and excellent safety margins in liver and testicular toxicity studies. Compound 19e was therefore selected as a development candidate for the potential treatment of type 2 diabetes.

Three dimensional quantitative structure-activity relationship studies for a series of isatin derivatives as a non-peptidyl caspase-3 enzyme inhibitor were investigated using comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA). The first approach of non-peptidyl small molecules by 3D QSAR may be useful in guiding further development of potent caspase-3 inhibitors.

Department of Chemistry, Inha University, Incheon 402-751, KoreaReceived June 30, 2004Ligand-based quantitative structure-activity relationship (QSAR) studies were performed on indolinonesderivatives as a potential inhibitor of the protein tyrosine kinase of fibroblast growth factor receptor (FGFR) bycomparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis(CoMSIA) implemented in the SYBYL packages. The initial X-ray structure of docked ligand (Su5402) toFGFR was used to minimize the 27 training set molecules using TRIPOS force field. Seven models weregenerated using CoMFA and CoMSIA with grid spacing 2 A. After the PLS analysis the best predictedCoMSIA model with hydrophobicity, hydrogen bond donor and acceptor property showed that a leave-one out(LOO) cross validated value (r

The generalized interaction properties function (GIPF) methodology developed by Politzer and coworkers, which calculated molecular surface electrostatic potential (MSESP) on a density envelope surface, was modified by calculating the MSESP on a much simpler van der Waals (vdW) surface of a molecule. In this work, vdW molecular surfaces were obtained from the fully optimized structures confirmed by frequency calculations at B3LYP/6-31G(d) level of theory. Multiple linear regressions for normal boiling point, heats of vaporization, heats of sublimation, heats of fusion, liquid density, and solid density were performed using GIPF variables from vdW model surface. Results from our model are compared with those from Politzer and coworkers. The surface-dependent β (and γ) values are dependent on the surface models but the surface-independent α and regression coefficients (r) are constant when vdW surface and density surface with 0.001 a.u. contour value are compared. This interesting phenomenon is explained by linear dependencies of GIPF variables. © 2004 Wiley Periodicals, Inc. J Comput Chem 25: 2073–2079, 2004

Receptor-based (RB) quantitative structure-activity relationship (QSAR) studies were performed on the peroxisome proliferator-activated receptors gamma (PPAR γ ) agonists with antihyperglycemic property using the comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) implemented in the SYBYL packages. The initial X-ray structure of PPAR γ with GI262570 agonist was used to sketch 36 training set molecules having the receptor bound conformation. These structures were subjected to subset energy minimization using TRIPOS force field and aligned against an α-helix of PPAR γ receptor. Four models were generated using CoMFA and CoMSIA with two different grid spacing, 2 A and 1 A. After PLS analysis, the best predicted CoMSIA model with grid spacing 1 A, Model 4, showed that a leave-one out (LOO) cross validated value (r 2 c v ) and non-cross validated conventional value (r 2 n c v ) are 0.758 and 0.994, respectively. From the CoMFA and CoMSIA contour maps, important interactions crucial to strong binding to the receptor can be identified and verified. This information on the pharmacophore models can be used to design new ligands. Traditional ligand-based (LB) QSAR studies were also performed using the global minima of the same training set molecules. Compared to LB QSAR study, advantage of using subset-minimized RB QSAR study is discussed in detail.

The gas-phase nucleophilic bimolecular substitutions at unactivated vinylic carbon (CH2CHCl, 1) by four nucleophiles (OH-, SH-, Cl-, and Br-) are investigated theoretically at the G2(+)(MP2) level. The results show that the stronger nucleophiles (OH- and SH-) substitute by an out-of-plane SN2 path with retention (SN π) but an in-plane SN2 path (SN σ) with inversion of configuration is preferred for the substitution by the weaker bases, Cl- and Br-. However, the elimination pathway is much more facile for OH- than any substitution process. We have considerd three factors, (i) the LUMO symmetries (σ*, or π*), (ii) the proximate σ−σ* charge-transfer interactions, and (iii) the electrostatic interactions in the transition state, as possible causes for preferred pathway for each nucleophile. The stability of the SN π transition state is predominantly influenced by the proximate σ−σ* type interactions, whereas electrostatic interactions are the major factor conducive to the energetic preference for the SN σ ove...

Three-dimensional quantitative structure-activity relationship (3D-QSAR) studies on 59 cinnamaldehyde analogues as Farnesyl Protein Transferase (FPTase) inhibitors were investigated using comparative molecular field analysis (CoMFA) with the PLS region-focusing method. Forty-nine training set inhibitors were used for CoMFA with two different grid spacings, 2A and 1A. Ten compounds, which were not used in model generation, were used to validate the CoMFA models. After the PLS analysis, the best predictive CoMFA model showed that the cross-validated value (r2 cv) and the non-cross validated conventional value (r2 nCv) are 0.557 and 0.950, respectively. From the CoMFA contour maps, the steric and electrostatic properties of cinnamaldehyde analogues can be identified and verified.