Abstract
Recently, electron cryo-microscopy (cryo-EM) maps of fibrils from the brains of mice and hamsters with five infectious scrapie strains have been published(1-5) and deposited in the electron microscopy data bank (EMDB)(6). This represents long-awaited near-atomic level structural evidence, widely expected to confirm the protein-only prion hypothesis(7,8). Instead, the maps reveal a second component, other than protein. The aim of the present study was to identify the nature of this second component, in the published maps(1-5), using an in silico approach. Extra densities (EDs) containing this component were continuous, straight, axial, at right angles to protein rungs and within hydrogen-bonding distance of protein, consistent with a role as guide and support in fibril construction. EDs co-located with strips of basic residues, notably lysines, and formed a conspicuous cladding over parts of the N-terminal lobe of the protein. In one ED, there was evidence of a Y-shaped polymer forming two antiparallel chains, consistent with replicating RNA. Although the protein-only prion hypothesis(7) is still popular, convincing counter-evidence for an essential role of RNA as a cofactor has amassed in the last 20 years(8). The present findings go beyond this in providing evidence for RNA as the genetic element of scrapie. To reflect the monotonous nature of the protein interface, it is suggested that the RNA may be a tandem repeat. This is against the protein-only prion hypothesis and in favour of a more orthodox agent, more akin to a virus. Fibrils from brains of patients with Alzheimer's disease (AD), Parkinson's disease (PD), amyotrophic lateral sclerosis (ALS) and other neurodegenerations also contain EDs(9) and may be of a similar aetiology.