Structural Underpinnings of Prion Protein Conversion

Hafner-Bratkovič et al. (1) propose a new structural model for the disease-associated form of prion protein, PrP^Sc, in which all three α-helices of the normal prion protein, PrP^C, are hypothesized to remain intact during the conversion. The model is based largely on observations that certain non-native disulfide cross-links in recombinant PrP (rPrP) abrogate amyloid formation in vitro.

We wish to offer the following comments. Although the structure of PrP^Sc remains enigmatic, much is known about the structure of amyloid fibrils formed from rPrP, and available experimental data are contradictory to the model presented by Hafner-Bratkovič et al. (1). In fact, hydrogen/deuterium exchange experiments indicate that the conversion to amyloid fibrils is associated with the unfolding of helix 1 and major rearrangement of the remaining two helices to highly ordered β-structure (2), and these findings are consistent with the pattern of proteinase K resistance (3). Furthermore, distance measurements by site-directed spin labeling (4) and solid state NMR spectroscopy (5) all indicate that two C-terminal α-helices of PrP^C become converted in amyloid fibrils to an in-register, parallel β-structure.

While the cross-linking data shown in Ref. 1 are very interesting, we find the link between these data and the proposed structural model circumstantial at best, especially since this model is not supported by any biophysical measurements. The problem is compounded by the possibility that the amyloid structures formed by rPrP with non-native disulfide bonds might be different from that formed by the native protein and that the structure of amyloid formed by rPrP might not be equivalent to that of infectious PrP^Sc.