The complex with the 8-AZA inhibitor, no equivalent signature of cyanide could be observed and the perfect spherical density at the cyanide location indicates that the catalytic water is still present with its centroid corresponding to the nitrogen atom of the cyanide. We then conclude that the substrate and the cyanide ion are both necessary for the formation of a stable and non-productive intermediate state. Compared to the structure of a previously solved non-cyanide 8-AZA complex [UOX/8-AZA] (2iba from PDB), both complexes, the ternary [UOX/UA/CN] and the [UOX/8AZA] crystallized in cyanide, show very little deviation on coordinates with an average r.m.s.d of 0.19 ?over all the main chain atoms. However, an important feature distinguishes complexes crystallized with cyanide from the others. At pH above 9, that invariably occurs upon addition of a cyanide excess, a flip of the His 98 is observed with a 1 angle moving from -60?to 180? leading to a complete rearrangement of the hydrogen bond network in this region. In all previously solved structures at pH below (or close to) 9, the imidazole group of His 98 was observed as a positively charged group because it is hydrogen bonded on both sides to carboxylates of Asp 100 and Glu 131. XAV-939 web Increasing the pH above 9 by cyanide addition PubMed ID: favors an uncharged His 98 at the origin of this drastic rearrangement. The different structures already solved with urate-like inhibitors have shown that the ligand is stacked with phenylalanine 159 and always hydrogen-bonded to the enzyme through a molecular tweezers (the “ligand tweezers”) composed by the two invariant residues arginine 176 and glutamine 228 [7,8]. In [UOX/UA/CN], the urate substrate (Figure 1) is exactly positioned as observed in all other inhibited UOX structures (this includes 8-azaxanthine, xanthine, 8-nitroxanthine, 9-methyl uric acid, oxonic acid and di-aminouracil). On top of the ligand, at a distance of 3.3 ? the catalytic water molecule W1, usually hydrogen bonded by the “reagent tweezers” built by the side-chains of asparagine 254 and threonine 57* from a symmetric subunit [7-9], is now replaced by a cyanide anion. When the “reagent tweezers” holds a single water molecule, the distance between the O and N2 atoms of the tweezers is remarkably constant (5.53 ?0.05 ? as shown in structures 1r4u, 1wrr, 1xxj, 1xt4, and 2iba from the PDB. In the [UOX/8-AZA] structure crystallized in cyanide, this distance falls within these limits (5.48 ?, an additional argument defending the lack of cyanide in the electron density. In the case of the ternary [UOX/UA/CN] complex, the cyanide insertion now significantly increases this distance up to 5.89 ? close to what is observed in theResultsIn normal reactive conditions, UOX crystals grown in the presence of uric acid lead to a complex between the protein and the final product of degradation, S-allantoin showing the high affinity of UOX for the last product of the reaction cascade [9]. The electron density map of the ternary [UOX/UA/CN] crystal structure shows a density clearly corresponding to the natural substrate that was never observed before because of its rapid degradation by the enzyme. The elongated density observed at stacking distance (3.3 ? of the mean-plane of urate, and attributed to the expected cyanide anion, fills a site where either a dioxygen or a catalytic water molecule have been previ-Page 2 of(page number not for citation purposes)BMC Structural Biology 2008, 8:http://www.biomedce.