Structural characterization of recombinant mammalian prions

Zusammenfassung

Prion Diseases or Transmissible Spongiform Encephalopathies, are a group of deadly neurodegenerative disorders that are caused by the pathogenic protein PrPSc. The prion diseases that affect humans are Kuru, Creutzfeldt-Jakob Disease (CJD), Gerstmann-Sträussler- Scheinker Syndrome (GSS) and Fatal Familiar Insomnia (FFI). Prion Diseases are not exclusive to human, other species are very susceptible to develop this fatal neurodegenerative disorder; scrapie that affects sheep and goats, the Transmissible Mink Encephalopathy (TME) that was observed in wild mink as well as in farm animal, the Chronic Waste Disease (CWD) that also affect wild and farm cervids, like elk or deer, the Feline Spongiform Encephalopathy (FSE) that affects domestic cats or zoo animals and the Bovine Spongiform Encephalopathy, also known as mad cow, responsible of the epidemic farm disaster that affected thousands of cattle in the UK and whole Europe during the 80’s and 90’s. The central event of the prion diseases is the conformational conversion of the PrPC into the pathogenic isoform PrPSc. The molecular mechanism of the prion replication, molecular basis of the prion biology, the species barrier and the pathogenesis of the neurodegeneration will not be understood until the structure of the PrPSc is solved. Because of that, elucidating the structure of the PrPSc is one major challenge in the prion research. Nowadays, two proposed models have been highlighted for the structure of the PrPSc. The PIRIBS model suggests that PrPSc amyloid fibers are made by a single protofilament where each monomer is parallel in register stacked along the fibrillar axes and argues that PrPSc monomers are flat, with rPrP amyloid-like β-strand rich cores extending up to position ~90. In contrast, the ß-solenoid model suggest that each single PrPSc is formed by the folded of ß- strands in 4-rungs ß-solenoid architecture. In contrast to the PIRIBS model, the ß-soleonoid model argues that the fibers are made by two intertwined protofilament. The classical PK core of the PrPSc, that span from ~90 to C- terminus, has been reported in plenty of studies. Besides this classical core, the PrPSc has an inner PK-resistant core that span from ~152-230. This inner “super-resistant” core resists partial, reversible unfolding induced by guanidine. Based on that, it is feasible to think that if PrPSc is formed by multi-layer of ß-strands, PK-treatment of partially unfolded PrPSc fibers should necessarily result in their complete disassembly, as the N-terminal “base” of each monomer disintegrates. In contrast, if PrPSc monomers are flat, fibers would persist after such treatment, as seen when rPrP fibers are treated with PK under conditions that preserve their C-terminal ß-strands rich core, which remain stacked. In this work, the brain derived GPI-anchorless PrPSc fibers, were treated with increased concentration of Gn/HCl until partial unfold. After guanidine partial unfolding treatment, fibers were digested with PK to destroy the N-terminal half. After guanidine treatment and PK digestion, the resulted PK resistants PrPSc were analyzed by western blotting and visualized by negative stain TEM. The results obtained show virtually complete disassembly of partially unfolded PrPSc fibers after PK treatment, whereas the rPrP remained its fibrillar architecture. These results support a multi-rung, rather than flat, architecture of the PrPSc monomers. Since the development of a method that allows the generation of prions in vitro, some studies that have been focused on the understanding of molecular conversion mechanism of PrPC to PrPSc, go through the use of synthetic PrPC, in which the introduction of all kind of modifications can help to explain the molecular mechanism that governs the conversion into the PrPSc. In this context, one of the most important events in the prion field was the development of a recPMCA, where the production of the recombinant amyloid generated under sonication cycles recapitulates the nature of wild type prions. The second goal of this work is to generate a recPrPSc that recapitulates the nature of the PrPSc and to use this synthetic prion for structural studies. For that purpose, the recombinant mouse PrP23-230 was bacterially expressed and then purified using a HisTrap Column. During the purification, the immobilized protein was refolded and then eluted using a gradient of imidazole. The refolded recPrP23-230 was collected in different fractions and subsequently, each fraction was submitted to conversion into the misfolded recPrPSc form by several rounds of recPMCA. Due to the fact that the misfolded recPrP shows different infectivity rates (non infectious or infectious), to verify the infectivity of the misfolded forms obtained after recPMCA, a bioassay was performed using a set of transgenic mice. Mice were inoculated with different misfolded PrP23-230. Only one of the misfolded PrP23-230 generated by recPMCA resulted infectious in all the mice inoculated. This recPrPSc was used for structural studies. Substantial evidence suggests that PrPSc is a 4-rung β-solenoid, and that individual PrPSc subunits stack to form amyloid fibers. Recently, limited proteolysis was used to map the β-strands and connecting loops that conform the PrPSc solenoid. Using high resolution SDS-PAGE followed by epitope mapping and mass spectrometry, it was possible to identify positions ∼116/118, 133-134, 141, 152-153, 162, 169 and 179 (murine numbering) as Proteinase K (PK) cleavage sites in PrPSc. Such sites define loops and/or borders of β-strands, and help to define the threading of the β-solenoid. This approach was extended to recombinant PrPSc obtained by recPMCA. Limited proteolysis in combination with the MS analysis, allowed the identification of several PK-resistant peptides. The ESI-TOF analysis identified the peaks with the masses: 9513; 9399 y 8184 correspond to the peptides N152-S230; M153-S230 y Y162-S230. Alternatively, MALDI-TOF analysis identified peaks with the masses 13428, 11859, 11380, 10985, 10514, 9504, 9393, 8179, 6092, 3655, 3048 KDa, that correspond to cleavages of the PK at 116, 134, 138, 141, 145, 152, 153, 162, 179, 201, 206. The results obtained in this study are in agreement with the susceptible sites, previously described in the GPI-anchorless PrPSc, 116, 134, 141, 152, 153, 162 y 179, that are placed within unstructured sequences with the additional nicks at 201, 206, that might describe a new loop located at the end of the C-terminal domain. Furthermore, there were observed additional peaks with the masses; 9510, 8658, 6525, 2482 that correspond to the doubly-truncated fragments 96-178, 103-178, 91-151 y 134-178. In summary, this study shows that the infectious recPrPSc generated in vitro by the recPMCA, exhibits biochemical properties that suggest that the architecture of recPrPSc is similar to that brain-derived PrPSc. These promising results suggest that the recPrPSc is a good model for the brain-derived PrPSc and can be a very good tool for structural studies.