Protein mis-folding and disease: the mechanism of amyloid formation

Dr R A Staniforth Sheffield 222 2761 R.A.Staniforth@sheffield.ac.uk

Until very recently, our understanding of the causes behind neurodegenerative diseases has been very limited. The recent outbreak of BSE (commonly referred to as mad cow disease) and the rise of Alzheimer’s as the third cause of death in North America has meant that there is increasing pressure to obtain a cure. These two diseases are characterised by the accumulation of fibrous material called "amyloid" in the brain. This substance has been shown to consist of naturally occurring proteins assembled into this unnatural mass. Whether or not this protein-based material is the direct cause for deterioration of brain tissue in patients suffering from these diseases is a matter of some controversy and has even been the subject matter of a recent TV program. However, it is clear to all that these proteinaceous fibres are closely associated with the pathology: mutations in genes coding for these proteins cause the early onset of the disease and premature death. Another cause for hope is that this pattern of events may be common to almost all neurodegenerative diseases, including Parkinson’s and Huntington’s, and a large number of other diseases including certain forms of cancer, diabetes and stroke. It is exiting to think that understanding the mechanism of amyloid formation may be a step towards a universal cure since, so far, therapies have only been targeted at alleviating symptoms but cannot prevent or halt the onset of these devastating diseases.

As a biochemist, my interest is in studying the assembly of "normal" proteins into amyloid fibres at a molecular level (fig.1). A first consideration when tackling such a problem is that, unlike other polymers, protein molecules exist as unique three-dimensional structures in vivo (referred to as their folded state). In amyloidogenesis, proteins with totally different folds form fibres with very similar properties suggesting there may be common, structurally analogous intermediates on the assembly route. This also means that considerable conformational changes (unfolding) of the protein molecules are likely to occur either before or after association.

A well-characterised protein with a number of advantageous properties is cystatin C, a protease inhibitor which is associated with a disease causing recurrent stroke (Human Cystatin C Amyloid Angiopathy). I have identified a number of different conformers of this protein which could all be candidate precursors for the assembly process. These include partially folded states of the monomeric protein and a "domain-swapped" dimer. The latter species (fig. 2) is the result of two molecules of cystatin coming together and exchanging part of their chains by unfolding then refolding into a conformation made up of two cystatin-folds, where each protein chain contributes half its constituent amino acids to one fold and half to the other. This is the first time this kind of intertwining of protein chains is observed for an amyloidotic protein. Such a process of assembly results in the formation of extremely stable oligomers and has often been speculated as a convenient mechanism for amyloid fibre formation. Structural characterisation of the different forms of cystatin from monomer to dimer to fibre stretches the boundaries of current technology. We are using a combination of techniques including dynamic NMR (nuclear magnetic resonance) spectroscopy and high-resolution electron microscopy.

Identifying different states of the protein is only the first step in characterising the mechanism of amyloid formation: each species must be put in context within a sequence of molecular events, i.e. a pathway needs to be defined. To do this, we have set up a screen to identify optimum conditions for observing fibre formation in vitro. Time courses will be recorded when the reaction is initiated from a number of different states of the protein including the dimeric form. This will not only establish a mechanism for the first time but it will also provide an experimental system which can be used to screen for potential therapeutic drugs.

Figure 1 - Three different proteins, (a) prion protein, (b) cystatin C and c) transthyretin, with three different three-dimensional structures which are known to form amyloid fibres in vivo. On the right is a proposed model for the structure of an amyloid fibril.

Figure 2. Cystatin monomer (A) and dimer (B). Cystatin dimerises via a domain-swapping mechanism which involves exchange of strand 1, helix and strand 2 between two molecules. The resulting "interwoven" structure is shown where 1 molecule is coloured in blue and the other in red.

Selected Publications A method for the reversible trapping of proteins in non-native conformations. Milanesi L, Jelinska C, Hunter CA, Hounslow AM, Staniforth RA, Waltho JP. Biochemistry 2008 Dec 23;47(51):13620-34.

Exclusion of the native alpha-helix from the amyloid fibrils of a mixed alpha/beta protein. Morgan GJ, Giannini S, Hounslow AM, Craven CJ, Zerovnik E, Turk V, Waltho JP, Staniforth RA. J Mol Biol. 2008 Jan 11;375(2):487-98.

Amyloid fibril formation by human stefin B: influence of pH and TFE on fibril growth and morphology. Zerovnik E, Skarabot M, Skerget K, Giannini S, Stoka V, Jenko-Kokalj S, Staniforth RA. Amyloid. 2007 Sep;14(3):237-47.

Essential role of proline isomerization in stefin B tetramer formation. Jenko Kokalj S, Guncar G, Stern I, Morgan G, Rabzelj S, Kenig M, Staniforth RA, Waltho JP, Zerovnik E, Turk D. J Mol Biol. 2007 Mar 9;366(5):1569-79.

A thiol labelling competition experiment as a probe for sidechain packing in the kinetic folding intermediate of N-PGK. Cliff MJ, Alizadeh T, Jelinska C, Craven CJ, Staniforth RA, Waltho JP. J Mol Biol. 2006 Dec 8;364(4):810-23.

The denatured state under native conditions: a non-native-like collapsed state of N-PGK. Reed MA, Jelinska C, Syson K, Cliff MJ, Splevins A, Alizadeh T, Hounslow AM, Staniforth RA, Clarke AR, Craven CJ, Waltho JP. J Mol Biol. 2006 Mar 24;357(2):365-72.

Folding and amyloid-fibril formation for a series of human stefins' chimeras: any correlation? Kenig M, Jenko-Kokalj S, Tusek-Znidaric M, Pompe-Novak M, Guncar G, Turk D, Waltho JP, Staniforth RA, Avbelj F, Zerovnik E. Proteins 2006 Mar 1;62(4):918-27.

Staniforth,R.A., Giannini,S., Higgins,L.D., Conroy,M.J., Hounslow,A.M., Jerala,R., Craven,C.J. and Waltho,J.P. (2001) Three-dimensional domain swapping in the folded and molten-globule states of cystatins, an amyloid-forming structural superfamily. EMBO J. 20, 4774-4781.

Staniforth,R.A., Dean,J.L.E., Zhong,Q., Clarke,A.R., Zerovnik,E. and Waltho,J.P. (2000) The major transition state in protein folding need not involve the immobilizsation of side chains. Proc Nat. Acad. Sci. USA 97, 5790-5795.