Refolding models

The resultant data have led to the proposal of numerous molecular models of amyloid fibril structure (Makin and Serpell, 2005). These models can be separated into three general classes (Fig. 2) (1) the Refolding models,... 

Fig. 2. Classes of structural models of amyloid-like fibrils. The Refolding models propose that a native protein (circle) partially or completely unfolds to attain a new fold (rectangle) in the fibril (stack of rectangles). In contrast, the Gain-of-Interaction models propose that only part of the native protein changes and takes on a new structure in the fibril. The remainder of the protein (partial circle) retains its native structure. The Natively Disordered models begin with disordered proteins or protein fragments, and these become ordered in the fibril. PolyQ refers to polyglutamine.

Refolding models propose that the fibril-forming protein exists in two distinct states the native state and the fibrillar state (Fig. 2). In converting from one to the other, the protein must unfold, then refold. As the fibrillar state is common to proteins with dissimilar sequences, Fandrich et al. (2001) have suggested that refolding into fibrils is dominated by backbone interactions, which are available to all protein sequences. 

In this section, we discuss the Refolding models proposed for insulin, SH3 domain, myoglobin, and prion protein (PrP). 

A prime example of a Refolding model is that of the insulin protofilament (Jimenez et al., 2002). Insulin is a polypeptide hormone composed of two peptide chains of mainly o -helical secondary structure (Fig. 3A Adams et al., 1969). Its chains (21- and 30-amino acids long) are held together by 3 disulfide bonds, 2 interchain and 1 intrachain (Sanger, 1959). These bonds remain intact in the insulin amyloid fibrils of patients with injection amyloidosis (Dische et al., 1988). Fourier transform infrared (FTIR) and circular dichroic (CD) spectroscopy indicate that a conversion to jS-structure accompanies insulin fibril formation (Bouchard et al., 2000). The fibrils also give a cross-jS diffraction pattern (Burke and Rougvie, 1972). 

Fig. 3. Refolding model of insulin protofilaments, from Jimenez et al. (2002). (A) Ribbon diagram of the crystal structure of porcine insulin (PDB ID code 3INS), generated with Pymol (DeLano, 2002). The two chains are shown as dark and light gray with N- and C-termini indicated. The dotted lines represent the three disulfide bonds 1 is the intrachain and 2 and 3 are the interchain bonds. (B) Cartoon representation of the structure of monomeric insulin in the fibril, as proposed by Jimenez et al. (2002). The thick, arrowed lines represent /1-strands, and thinner lines show the remaining sequence. The disulfide bonds are as represented in panel A, and N- and C-termini are indicated. (Components of this panel are not to scale.) (C) Cartoon representation of an insulin protofilament, showing a monomer inside. The monomers are proposed to stack with a slight twist to form two continuous /(-sheets. (Components of this panel, including the protofilament twist, are not to scale.) In the fibril cross sections presented byjimenez et al. (2002), two, four, or six protofilaments are proposed to associate to form the amyloid-like fibrils.

Many amyloid fibrils seem to be made up of smaller protofilaments. Although the number of protofilaments per fibril varies, the protofilaments have a fairly consistent diameter of 30 A (Serpell et al., 2000b Shirahama and Cohen, 1967 Shirahama et al., 1973). For some proteins, for example TTR, the protofilament diameter matches that of the native protein, suggesting that a Gain-of-Interaction model is plausible (Serpell et al., 1995). For other proteins, for example the SH3 domain, the protofilament seems too small to accommodate the native protein structure, suggesting that a Refolding model is plausible (Jimenez et al., 1999). 

The extreme stability of amyloid and amyloid-like fibrils is difficult to understand in terms of the three classes of fibril models. For the Refolding models, it has been suggested that the amyloid conformation is a default conformation for a polypeptide chain (Dobson, 1999). However, these models do not give a clear indication of what types of interactions differ in the amyloid conformation versus the native conformation, and so it is unclear why the amyloid conformation should be more stable. Also, it seems that the elevated protein concentrations associated with fibril formation might disproportionately favor nonspecific aggregation of the destabilized intermediate over amyloid fibril formation. 

---