Fig. 1. ( A ) The crystal structure of the S1S2 ligand binding domain of the N -methyl-d-aspartate NR1 subunit showing the positions of the xenon binding sites predicted by our grand canonical Monte Carlo modeling. The published structure contains two nonidentical copies of the S1S2 domain (in different orientations), chain A ( lower panel ) and chain B ( upper panel ). The predicted xenon binding sites (superclusters) are represented by the gray surfaces . The red spheres represent xenon atoms at the center of the density clusters that comprise the binding sites. The predicted xenon binding sites are in equivalent positions in chains A and B, occupying the site normally occupied by glycine. ( B ) Crystal structure of the NR1 subunit showing the glycine binding site. The predicted xenon binding site in chain A is represented by the gray surfaces . The red spheres represent xenon atoms. The 11 amino acids identified from the modeling as being within 4 Å of the binding site ( table 2 ) are shown as stick models (atoms color coded: gray =carbon; blue =nitrogen; red =oxygen). ( C ) The xenon atoms occupy the same position as the competitive inhibitor dichlorokynurenic acid, shown here as orange sticks . These images were produced using PyMol (PyMol Molecular Graphics System; DeLano Scientific, Palo Alto, CA).