Role of the central lysine cluster and scrapie templating in the transmissibility of synthetic prion protein aggregates.

Mammalian prion structures and replication mechanisms are poorly understood. Most synthetic recombinant prion protein (rPrP) amyloids prepared without cofactors are non-infectious or much less infectious than bona fide tissue-derived PrPSc. This effect has been associated with differences in folding of the aggregates, manifested in part by reduced solvent exclusion and protease-resistance in rPrP amyloids, especially within residues ~90-160. Substitution of 4 lysines within residues 101-110 of rPrP (central lysine cluster) with alanines (K4A) or asparagines (K4N) allows formation of aggregates with extended proteinase K (PK) resistant cores reminiscent of PrPSc, particularly when seeded with PrPSc. Here we have compared the infectivity of rPrP aggregates made with K4N, K4A or wild-type (WT) rPrP, after seeding with scrapie brain homogenate (ScBH) or normal brain homogenate (NBH). None of these preparations caused clinical disease on first passage into rodents. However, the ScBH-seeded fibrils (only) led to a subclinical pathogenesis as indicated by increases in prion seeding activity, neuropathology, and abnormal PrP in the brain. Seeding activities usually accumulated to much higher levels in animals inoculated with ScBH-seeded fibrils made with the K4N, rather than WT, rPrP molecules. Brain homogenates from subclinical animals induced clinical disease on second passage into "hamsterized" Tg7 mice, with shorter incubation times in animals inoculated with ScBH-seeded K4N rPrP fibrils. On second passage from animals inoculated with ScBH-seeded WT fibrils, we detected an additional PK resistant PrP fragment that was similar to that of bona fide PrPSc. Together these data indicate that both the central lysine cluster and scrapie seeding of rPrP aggregates influence the induction of PrP misfolding, neuropathology and clinical manifestations upon passage in vivo. We confirm that some rPrP aggregates can initiate further aggregation without typical pathogenesis in vivo. We also provide evidence that there is little, if any, biohazard associated with routine RT-QuIC assays