A new and powerful analytical method for comparing molecular shapes by optimizing the overlap of molecular volumes has been developed. This shape comparison method provides both a quantitative measure of the shape similarity of molecules and a means to align molecules such that shape similarity if maximized. Our MSC method has been enhanced with an option to allow discrimination between groups with different chemical properties. Atoms or groups of atoms may be assigned to different classes based on specific properties such as electrostatic potential, hydrogen bonding ability, or hydrophobicity. This enables matches based on criteria such as alignment of hydrophobic groups or hydrogen bond acceptor groups. In this study, we report shape comparisons of angiotensin II (AII) receptor antagonists from two structural classes, 4-(biphenyl-4-ylmethoxy)-quinoline derivatives such as ICI D8731 and N-(biphenyl-4-ylmethyl)imidazole derivatives, such as DuP753. Starting with a list of low-energy conformations for the two molecules, each conformation of the first molecule is paired with each of the conformations of the second molecule. For each of these conformational pairs, an MSC comparison, which generates multiple MSC maxima, is initiated. Eight high scoring conformational pairings were found with shape matching based on the intersection of the total molecular volume, while nine high-scoring pairs were identified with matching by atom type. MSC identifies conformational pairs with high shape similarity, as measured by the intersection volume, and thus generates and prioritizes several alternative models for the AII antagonist pharmacophore.

On the basis of an extension of the literature lead 1, a series of benzimidazoles have been synthesized and shown to be angiotensin II (AII) receptor antagonists. The structure-activity relationships of these new antagonists have been explored and the key binding interactions defined. Molecular mechanics calculations were carried out on analogues of imidazole AII antagonists and conformationally restricted analogues were synthesized. The benzimidazole antagonists displaced AII in binding studies in vitro with IC50 values in the range 10(-5)-10(-7) M and antagonized the hypertensive effects of AII in vivo (rats) following intravenous administration with ED50 values in the range of 5-20 mg/kg.

Multiple R-groups (monovalent fragments) are implicitly accessible within most of the molecular structures that populate large structural databases. R-group searching would desirably consider pIC50 contribution forecasts as well as ligand similarities or docking scores. However, R-group searching, with or without pIC50 forecasts, is currently not practical. The most prevalent and reliable source of pIC50 predictions, existing 3D-QSAR approaches, is also difficult and somewhat subjective. Yet in 25 of 25 trials on data sets on which a field-based 3D-QSAR treatment had already succeeded, substitution of objective (canonically generated) topomer poses for the original structure-guided manual alignments produced acceptable 3D-QSAR models, on average having almost equivalent statistical quality to the published models, and with negligible effort. Their overall pIC50 prediction error is 0.805, calculated as the average over these 25 topomer CoMFA models in the standard deviations of pIC50 predictions, derived from the 1109 possible "leave-out-one-R-group" (LOORG) pIC50 contributions. (This novel LOORG protocol provides a more realistic and stringent test of prediction accuracy than the customary "leave-out-one-compound" LOO approach.) The associated average predictive r(2) of 0.495 indicates a pIC50 prediction accuracy roughly halfway between perfect and useless. To assess the ability of topomer-CoMFA based virtual screening to identify "highly active" R-groups, a Receiver Operating Curve (ROC) approach was adopted. Using, as the binary criterion for a "highly active" R-group, a predicted pIC50 greater than the top 25% of the observed pIC50 range, the ROC area averaged across the 25 topomer CoMFA models is 0.729. Conventionally interpreted, the odds that a "highly active" R-group will indeed confer such a high pIC50 are 0.729/(1-0.729) or almost 3 to 1. To confirm that virtual screening within large collections of realized structures would provide a useful quantity and variety of R-group suggestions, combining shape similarity with the "highly active" pIC50, the 50 searches provided by these 25 models were applied to 2.2 million structurally distinct R-group candidates among 2.0 million structures within a ZINC database, identifying an average of 5705 R-groups per search, with the highest predicted pIC50 combination averaging 1.6 log units greater than the highest reported pIC50s.

A novel analytical method for comparing molecular shapes by optimizing the intersection of molecular "SKINS" has been developed. This method provides a quantitative measure of the shape similarity by maximizing the intersection volume of molecular surfaces with a finite thickness; a molecular skin. We report shape matching of a small tripeptide inhibitor (DFKi) of elastase class proteins with the 56 residue turkey ovomucoid inhibitor (TOMI). To match a large elastase inhibitor such as TOMI with a small inhibitor or drug, we found that it is necessary to use a skin match rather than molecular volume. Skin based comparisons of TOMI protein with DFKi successfully found the alignment expected from comparison of their respective crystallographic complexes with elastase (i.e. HLE/TOMI complex and PPE/tripeptide complex). In the skin comparison of the tripeptide with the TOMI protein, blind searching for skin matches involved optimization of the skin intersection from 172 starting positions randomly selected from a set of 500 points on the TOMI van der Waals surface [within 9.5 A of the Leu-18 on the TOMI binding loop (1 point/A2)]. The tripeptide center of mass was placed at these points and its orientation was randomized before optimization was initiated. The best skin intersection, 86.4 A3, was found three times and corresponds to the experimental alignment. The next best skin intersection was 78.1 A3 giving a discrimination factor in this case of 10%. Searches over the entire surface of the TOMI protein did not identify any new matches with skin intersection greater than 78.1 A3. Matching the DFKi with a TOMI structure relaxed from its crystal conformation by molecular dynamics gives similar results.

Studies to assess the risks of revealing chemical structures by sharing various chemical descriptor data are presented. Descriptors examined include "Lipinski-like" properties, 2D-BCUT descriptors, and a high-dimensional "fingerprint-like" descriptor (MACCs-vector). We demonstrate that unless sufficient precautions are taken, de novo design software such as EA-Inventor is able to derive a unique chemical structure or a set of closely related analogs from some commonly used descriptors. Based on the results of our studies, a set of guidelines or recommendations for safely sharing chemical information without revealing chemical structure is presented. A procedure for assessing the risk of revealing chemical structure when exchanging chemical descriptor information was also developed. The procedure is generic and can be applied to any chemical descriptor or combination of descriptors and to any set of structures to enable a decision about whether the exchange of information can be done without revealing the chemical structures.