Susan Firbank, Peter Knowles, Mike McPherson, Simon Phillips.

Galactose oxidase (GO) is a monomeric copper containing enzyme that catalyses the oxidation of primary alcohols to their corresponding aldehydes. The enzyme has a wide range of substrates, but is strictly stereo-specific; whilst D-galactose is a good substrate, galactose oxidase shows no activity with L-galactose or D-glucose. In addition to the copper, the two electron reaction requires a second redox active centre, which is provided by a radical situated at a tyrosine residue (Tyr 272). The crystal structure of the mature enzyme revealed the presence of a novel thioether bond - a covalent link between Tyr 272 and a cysteine (Cys 228) at the active site (Ito et. al. (1991), Nature, 350, 87-90). To investigate how this thioether bond forms, current work in Leeds is investigating the processing of this enzyme from its precursor.

Galactose oxidase is thought to undergo several steps during processing. Cleavage of the signal sequence gives rise to a form that contains a seventeen amino-acid N-terminal pro-sequence, in addition to the mature sequence. This pro-form does not yet have the thioether bond at the active site. Generation of mature enzyme requires at least two further processing events - cleavage of the pro-sequence and formation of the thioether bond. Both of these events are autocatalytic, only requiring the addition of copper and molecular oxygen¹. To try and elucidate these processes, the precursor of galactose oxidase (containing the pro-sequence and no thioether bond) has been purified and crystallised.

X-ray diffraction data were collected on a crystal of the precursor, and the structure solved by molecular replacement using the mature structure as a model. The structure has been refined to a resolution of 1.4Å, and shows several differences between the precursor and mature form. Five regions of mainchain differ significantly between the two forms. The position of four of these regions can be attributed to the presence of the pro-sequence, whilst the position of the fifth appears to be due to differences at the active site. The most noticeable changes at the active site are the arrangement of the two residues that form the thioether bond, and also of the tryptophan that stacks over the thioether bond in the mature form. There are also slight differences in the positions of the other copper ligands. It is hoped that this structure will provide a starting point for modelling and understanding thioether bond formation.

Work is currently underway to obtain structures of intermediates between the precursor and mature forms. Since both pro-sequence cleavage and thioether bond formation require only copper and oxygen, anaerobic conditions can be used to try and trap intermediates. Copper has been soaked into a crystal of the precursor under such oxygen free conditions, and this structure is currently being refined to 2.4Å. We are also in the process of trying to purify and crystallise a form of the protein in which the pro-sequence has been cleaved, but the thioether bond has not yet formed. These structures, along with the solution work carried out by our collaborators, should give us more of an insight into the starting point for thioether bond formation.

M.S. Rogers and D.M. Dooley. Department of Biochemistry and Chemistry, Montana State University, Bozeman, Montana 59717.

Rogers, M.S., Baron, A.J., McPherson, M.J., Knowles, P.F., and Dooley, D. (2000) Galactose oxidase pro-sequence cleavage and cofactor assembly are self processing events. J. Am. Chem. Soc. 122, 990-991

We acknowledge the support of BBSRC.