Mathematical models to correlate molecular topology with substrate affinity of the glycine antagonist in glutamate receptors

Abstract

Mathematical models that correlate chemical structure with biological activity have been useful in the design of new drugs and can be used to predict biological behavior of new, chemically related molecules. A mathematical model was generated to correlate the substrate affinities with variations in the molecular topology of glycine antagonists in NMDA sub-class glutamate receptor and, subsequently, to propose new molecules with antagonist activity. By use of molecular connectivity indexes, the electronic structure and atomic bonding patterns of 45 glycine antagonists were coded. Correlation between connectivity indexes and antagonist affinity was determined by regression analysis. The connectivity index that best described affinity behavior was 4Xvpc, which indicates the relative importance of heteroatoms, the vicinity of aromatic ring substitutes, and valency gradient. The equations generated predicted new antagonist affinities, and the model was able to suggest structural requirements for designating compounds with increased affinity. Twelve new molecules were proposed, from which three appeared promising-based of the affinities previously calculated by means of the new equations. Energetic interaction analysis was developed as a control for the mathematical methodology. Glycine antagonists structure were analyzed mathematically by means of connectivity indexes. The equations modeled receptor behavior and contributed useful information for new antagonist design.