Probing natural amyloid fibril assembly by protein display

The structure and substance of living things are made from self-assembling molecules, primarily proteins. Protein aggregates have become notorious for being associated with some diseases of old age such as Alzheimer's and Parkinson's Diseases. In these disorders, proteins mis-fold and aggregate into large rope-like deposits known as amyloid fibrils. It turns out that most if not all proteins can fold to form these amyloid filaments. The fibrils have a very regular structure and some remarkable physical properties. It turns out that Nature has evolved some proteins to be in this state deliberately, to make fibrils on bacterial cell surfaces, to help bacteria stick to other cells. These are some of the most stable fibrils of their type. The regular and stable structure of amyloid fibrils makes them a good scaffold on which to hang other, functional proteins, and that is this basis of this proposal. We will use the proteins that form the natural amyloid fibrils on bacterial cells and fuse them, at the level of their genes, to other proteins and enzymes. To do this, we need to understand more about how individual proteins, which are about 2 nanometres in all dimensions, assemble to form fibrils that are 10 nanometres wide and 10 micrometres long. We can alter the protein's sequence and observe how the fibrils change. Once we have established more about how the building blocks are assembled, we will fuse other proteins to the building blocks so that their function becomes displayed on the fibrils. To start with we will display proteins that carry electrons around in cells. This way we can make a conducting amyloid fibril. We will also display on the fibrils enzymes that destroy antibiotics. This can be used to protect the cell against an antibiotic, giving the cell an advantage. This property can be used to evolve the fibrils to be more useful technologically. Once we have established the best type of fibril and protein display methods, we can then make fibrils at will and use them to build up networks of fibrils on the nanometre scale. This is a new approach to building molecular devices for electronics and diagnostics, out of completely self-assembling molecules. So while amyloid is a serious problem in many disease states, it may also be remarkably useful for making molecular circuits.

This proposal is to optimise the use of the E. coli curli fibres as a display vehicle for folded proteins and enzymes. Once established and under control, this protein system will be used as a cassette for generating a family of functional fusion proteins for controlled deposition into catalytic, 2D and 3D structures. E.coli and Salmonella species with the curli phenotype express highly aggregated, extracellular fibrils from the csg operon. By every definition these are amyloid fibrils, constructed from folded proteins. A key feature of Csg fibres is the production of two distinct proteins responsible separately for nucleation and precipitation. We will use this two-component fibril forming system to display functional proteins on amyloid fibrils by making fusions to the Csg proteins. We will create the genes for the fusion of cytochrome b562 and the TEM-1 beta-lactamase to the Csg fibril forming units. The cytochrome-decorated fibrils will be studied for their electronic and conductive properties. The beta-lactamase decorated fibres will be used as an in vivo selection tool for evolving the Csg proteins for optimal enzyme display. We will explore the structural mechanisms of fibril assembly using protein engineering and a variety of biophysical methods. We will then use this information to develop methods for the controlled formation of fibres at defined locations on surfaces. The result will be a useful fibre scaffold which can be used for display of a variety of functional proteins.