| Summary: | Both mammalian and bacterial peroxidases contain novel covalent linkages. In the former a sulfonium linkage to a critical Met residue is thought to modify the normal planarity of the haem affecting the functional properties. Following on from the work of Metcalfe et al., 2004 which showed that such links could be engineered in ascorbate peroxidase, horseradish peroxidase (HRP) has been used as convenient model system to try and understand the structural and electronic effects of engineered covalent linkages. Previous work in the group (Cali, 2008) had shown that mutation of Ser167 to Met in HRP-C* resulted in autocatalytic cross-linking on incubation with hydrogen peroxide. In this thesis, two additional HRP variants, S167Y and S167W were studied and a new novel structural linkage discovered. The UV/Vis spectrum of the S167Y variant suggested a more 6-coordinate high spin character. The molar extinction coefficients were markedly increased, 180 mM-1 cm-1 for S167Y and 135 mM-1 cm-1 for S167W, compared to 100 mM-1 cm-1 for the WT enzyme, consistent with a more 6 coordinate high spin character normally seen in lignin peroxidase. In contrast, the dissociation constant (Kd) of the S167W variant mutant for the aromatic donor BHA was hardly affected, whilst that of the S167Y variant increased two-fold relative to the WT, implying a significant perturbation of the aromatic donor binding site and / or the associated haem-linked hydrogen bonding network. After peroxide treatment the haem group of the S167Y variant could not be extracted into acid butanone in contrast to the WT. Only a proportion of the haem could be extracted even from the untreated S167Y variant, implying that a substantial fraction of the protein had formed the haem-protein linkage during folding and purification. These results were confirmed during reverse phase HPLC and MALDI-TOF / ESI mass spectroscopy measurements. The haem and protein completely co-eluted in the case of peroxide treated S167Y, while only ~50% of the haem was linked to the protein in the untreated as isolated enzyme. The MALDI-TOF and ESI mass spectrum showed that there was a large increase (614 Da) in the mass of the linked S167Y protein, compared to that of the unlinked enzyme. Unlike the sulfonium linkage obtained earlier, treatment with hydrogen peroxide was unnecessary to observe this increase. Interestingly, the 100% unlinked S167Y protein could only be isolated if enzyme was prepared in the presence of an efficient peroxidase substrate as an antioxidant scavenger. It appears that a Tyr residue at position 167 is highly reactive with respect of the haem vinyl side chain forming a spontaneous covalent link not otherwise seen in nature. Pre steady-state comparison of the intermediates has shown that Compound I was formed essentially normally at near WT rates, however its stability was greatly affected in the S167Y variant (linked or unlinked), the life time being decreased to ~0.04 s, compared to of the WT enzyme, where it was ~80 s. The substrate preference of the cross-linked S167Y variant was also altered. Stopped-flow measurements of the individual rate constants for the partial reactions of the catalytic cycle with luminol as reducing substrate revealed an increase in the rate of reduction of Compound I to Compound II (k2). The X-ray crystal structure of S167Y variant was solved to 1.7 Å resolution and the structure has been modelled and determined by x-ray crystallography. The x-ray structure reveals an unanticipated linkage containing an additional ring structure bonded to the engineered Tyr. ESI mass measurements supported this structure.
|
|---|
