Xanthine dehydrogenases from various sources are proteins of molecular mass 300 kDa and are composed of two identical independent subunits; each subunit contains one molybdopterin, two non-identical 2Fe/2S centres and FAD [ 1,2]. T h e mammalian enzyme exists originally as xanthine dehydrogenase, but is converted into xanthine oxidase during extraction or purification procedures [3]. Xanthine dehydrogenase is characterized by high xanthine/NAD+ activity and low xanthine/Oz activity, whereas xanthine oxidase is characterized by high xanthine/OZ activity and negligible xanthine/NAD+ activity. This conversion occurs either reversibly by thiol oxidation of the protein molecule or irreversibly by proteolytic cleavage [4]. In contrast with the mammalian enzyme, xanthine dehydrogenase from chicken liver is not converted into xanthine oxidase by either thiol oxidant or proteinase treatment. T h e full amino acid sequences of the enzymes from human liver [5], rat liver [6], mouse liver [7], bovine milk [8], chicken liver [9] and Drosophifu [ 101 have been determined by cDNA cloning. All known xanthine-oxidizing enzymes consist of approx. 1330 residues, except for the chicken enzyme which consists of 1358 residues because of insertion of 23 amino acid residues in the N-terminal region. Although the amino acid sequences are highly homologous among the rat, mouse, bovine and human enzymes, with about 90% identity [5-81, weaker homology was observed between the mammalian enzyme and the Drosophila enzyme, with 52% identity [6,10], and between the chicken liver and mammalian enzymes, with 70% identity [9]. By limited proteolysis of the rat enzyme with trypsin, the enzyme is converted into a xanthine oxi-dase type with concomitant cleavage into three fragments (20, 40 and 85 kDa) [6]. These fragments are only dissociated under denaturating conditions such as in the presence of high concentrations of guanidinium chloride [6], suggesting that the three fragments associate closely with each other. Determination of the N-terminal amino acid sequences of each fragment led to the assignment of the 20 kDa fragment to the N-terminal portion, the 85 kDa fragment to the C-terminal portion and the 40 kDa fragment to the intermediate portion. The chicken enzyme was also cleaved into three fragments in a similar way, but the enzyme is not converted into the oxidase [9]. Comparative alignment of the deduced protein sequences has been attempted in order to identify the possible locations of the redox active centres. Sequence comparison [5-91 and chemical-modification studies suggested that each redox centre is located in different domains, i.e. the two iron-sulphur centres are in the 20 kDa domain, the FAD is in the 40 kDa domain and the molybdopterin is in the 85 kDa domain [6,9]. These conclusions are consistent with the results obtained from X-ray-crystallographic analysis of aldehyde oxidoreductase from Desulfovibrio g&as which is another molybdenum-containing iron-sulphur protein [ 1 11. There is high sequence homology between xanthine ox...