The hydroxylation mechanism of the molybdoprotein xanthine oxidoreductase (XOR) has been modelled using density-functional theory. High activation barriers are often obtained for models of this enzyme due to the absence of factors that stabilize the accumulation of charge on the substrate at the transition state. Xanthine provides much lower barriers than small model substrates such as formamide or imidazole due to charge delocalization to centers which appear to interact with key residues in the protein. Of the two mechanisms of stabilization discussed in the literature—tautomerization and protonation of xanthine—density-functional theory calculations suggest that proton transfer from Glu1261 to N9 reduces the activation barrier by ∼30 kcal mol−1 and leads to an intuitive product complex. Further, similar values for the activation barriers of methyl xanthine isomers lead to the conclusion that the wide variation in rates for substituted purines is due to interactions with key residues in the active site. In addition, the transition state for oxidation of xanthine can be superimposed over the X-ray structure of inhibitor-bound XO with high correlation suggesting that the enzyme active site orients the substrate into the ideal position for reaction. The activation barriers for models of a hypothetical tungsten-substituted XO are predicted to be ∼10 kcal mol−1 higher in energy due to the higher reduction potential of the metal and unfavourable electrostatic interactions for the hydride transfer.