Understanding the Functioning of the Copper Binding Domain of the Amyloid Precursor Protein

 

G.K.W. Kong,a,b W.J. McKinstry,a J.J. Adams,a G. Polekhina,a N.A. Williamson,b

D. Cappai,b K.J. Barnham, b R. Cappaib and M.W. Parker a

 

aBiota Structural Biology Laboratory, St. VincentÕs Institute of Medical Research, Fitzroy, Victoria 3065, Australia;  bDepartment of Pathology, The University of Melbourne, Victoria 3010, Australia (g.kong1@pgrad.unimelb.edu.au)

 

Redox active metal ions, like copper (Cu), are abundant in human brains and may contribute to the pathogenesis of various neurodegenerative diseases [1]. In AlzheimerÕs Disease (AD), the amyloid plaques, the pathological hallmark of the disease formed by the over-production and aggregation of peptides called Ab, bind and thus concentrate Cu ions in the vicinity of brain neurones. As Cu(II) ions are highly oxidising and can react with oxygen to produce reactive oxygen species, they can inflict oxidative damage to the neurones [1]. The clearance of metal ions from Ab and/or plaques thus represents an important means of intervention in AD.

The Ab peptide is derived from the cleavage of the amyloid precursor protein (APP), a single transmembrane protein found on the surface of brain neurones, by enzymes called secretases. The APP contains, near the N-terminus, a copper binding domain (CuBD) which binds Cu(II) with high affinity [2]. The binding of Cu(II) to APP helps to clear the ion from the extracellular environment [3] and influences APP metabolism so as to lower Ab secretion [4]. Cu(II)-binding also causes changes to the structure of the CuBD and may subsequently modulate APP-APP interactions [5]. Structural studies of the CuBD will help to understand the Cu(II)-binding process and may enable the development of CuBD agonists that disfavour the production of Ab.

Well-diffracting crystal forms of the CuBD have been grown in a few conditions and attempts to solve the phase problem involved using an existing NMR model for molecular replacement (MR) [6]. With a large number of molecules per asymmetric unit in the initial crystal forms (between 6 to 12), MR was further complicated by errors in the NMR model. Eventually a crystal form was found containing only one molecule in the asymmetric unit and a 2  resolution data set collected. After trying various MR programs, a solution was obtained by AMoRe [7]. The current, preliminary crystal structure is free of any metal ions, and like the NMR model, the fold is based on a helix packed against a b sheet of 3 strands. The focus is now on defining the Cu binding site through obtaining a Cu-bound structure of the CuBD.

 

References

1           Bush, A.I. (2000) Curr. Opin. Chem. Biol. 4: 184-191.

2           Multhaup, G. et al. (1997) Biochem. Pharmacol. 54: 533-539.

3           White et al (1999) Brain Res. 84: 439-444.

4           Borchadt, T. et al. (1999) Biochem. J. 344: 461-467.

5           Beher, D. et al. (1996) J. Biol. Chem. 271: 1613-1620.

6           Barnham, K.J. et al. (2002), submitted.

7           Navaza, J. (1994) Acta Cryst. A 50: 157-163