The role of Cys108 in Trigonopsis variabilis d-amino acid oxidase examined through chemical oxidation studies and point mutations C108S and C108D.

Abstract Oxidative modification of Trigonopsis variabilis d -amino acid oxidase in vivo is traceable as the conversion of Cys108 into a stable cysteine sulfinic acid, causing substantial loss of activity and thermostability of the enzyme. To simulate native and modified oxidase each as a microheterogeneity-resistant entity, we replaced Cys108 individually by a serine (C108S) and an aspartate (C108D), and characterized the purified variants with regard to their biochemical and kinetic properties, thermostability, and reactivity towards oxidation by hypochlorite. Tandem MS analysis of tryptic peptides derived from a hypochlorite-treated inactive preparation of recombinant wild-type oxidase showed that Cys108 was converted into cysteine sulfonic acid, mimicking the oxidative modification of native enzyme as isolated. Colorimetric titration of protein thiol groups revealed that in the presence of ammonium benzoate (0.12 mM), the two muteins were not oxidized at cysteines whereas in the wild-type enzyme, one thiol group was derivatized. Each site-directed replacement caused a conformational change in d -amino acid oxidase, detected with an assortment of probes, and resulted in a turnover number for the O 2 -dependent reaction with D -Met which in comparison with the corresponding wild-type value was decreased two- and threefold for C108S and C108D, respectively. Kinetic analysis of thermal denaturation at 50 °C was used to measure the relative contributions of partial unfolding and cofactor dissociation to the overall inactivation rate in each of the three enzymes. Unlike wild-type, C108S and C108D released the cofactor in a quasi-irreversible manner and were therefore not stabilized by external FAD against loss of activity. The results support a role of the anionic side chain of Cys108 in the fine-tuning of activity and stability of d -amino acid oxidase, explaining why C108S was a surprisingly poor mimic of the native enzyme.