Although tyrosinase was one of the first discovered monooxygenases, its crystallographic structure has not been yet elucidated. However, it can be assumed that tyrosinases, haemocyanins and catechol oxidases share similar binuclear copper sites considering the following features: (i) comparable valence and conformational change during oxygen binding process (Woolery et al. 1984, Della Longa et al. 1996), (ii) comparable spectroscopic and magnetic properties (Himmelwright et al. 1980), (iii) homologies in the primary sequence (Van Gelder et al. 1997), (iv) catechol oxidases showed monophenolase activity on 4-hydroxyanisole (Espin et al. 1998), (v) haemocyanins, which are oxygen-carriers, showed catecholase activity after in vitro sodium dodecyl sulphate or protease treatments (Decker and Rimke 1998, Jaenicke and Decker 2004), as in the case of fungal tyrosinases (Espin et al. 1999), probably resulting from a higher accessibility of phenolics to the active site.
The adjacent amino acid residues surrounding the copper ions are generally highly conserved among tyrosinases, catecholoxidases and mollusc haemocyanins (Fig. 2). Each of the two copper regions of these proteins contains three imidazole residues bound to copper ions (Lerch 1983, Jackman et al. 1992). All of the conserved copper ligands, namely HA1, HA2, HA3 for copper A, and HB1, HB2, HB3 for copper B, follow the general rules HA1-x(n)-HA2-x(8)-HA3 and HB1-x(3)-HB2-x(n)-HB3 (Fig. 2), where n is a variable number of residues (Garcia-Borron and Solano 2002). Considering crystallographic data for catecholoxidase (Klabunde et al. 1998) and structure predictions for N. crassa (Kupper et al. 1989) and A. bisporus (Van Gelder et al. 1997) tyrosinases, Garcia-Borron and Solano (2002) described the tyrosinases active site as a hydrophilic sphere delimited by a four-helix bundle containing the six imidazole residues. The hydrophilic sphere would be located in a hydrophobic shell which is aromatic and formed by highly conserved residues: F(HA3−4), Φ(HA3−1), and Φ(HA3+3) around CuA, F(HB3−4), and H(HB3−1) around CuB (with Φ meaning aromatic residues). In this arrangement, the configuration of the tyrosinases active site would essentially be maintained by electrostatic or cations-π interactions. According to Garcia-Borron and Solano (2002), these noncovalent binding forces are the result of interactions between several residues located at the proximity of the dicopper center: Φ(HA1−7), R(HA3+1), Φ(HA3+3), Φ(HA3+7), E(HA3+8), D(HB3−7), D(HB3+4), Φ(HB3+7), W(HB3+10). The comparison of fungal tyrosinases primary structures shows the conservation of the following amino acids (Fig. 2): Φ(HA1−7) (with Φ corresponding to F or Y), R(HA3+1), Φ(HA3+3) (with Φ corresponding to Y), E(HA3+8), D(HB3−7), D(HB3+4), and W(HB3+10). Another feature of the binuclear copper centre in fungal tyrosinases is the highly conserved C(HA2−2) which is involved in a covalent thio-ether bond with the copper A ligand HA2 (Lerch 1982). Such a covalent cysteine-histidine bridge was reported in molluscan haemocyanins subunits (Gielens et al. 1997) and in the catecholoxidase from Ipomea batatas (Klabunde et al. 1998). In the case of N. crassa tyrosinase, Lerch (1982) suggested that this unusual thio-ether linkage was formed during an intramolecular rearrangement of the enzyme, resulting in the activation of the protyrosinase, whereas Nakamura et al. (2000) suggested the involvement of the C(HA2−2) residue in the binding to the copper A atom in the A. oryzae tyrosinase. According to Klabunde et al. (1998), it seems likely that this structure optimizes the redox potential of the copper centre for ortho-diphenol oxidation, allowing for a rapid electron transfer in the redox processes.