Factors Contributing to Yeast Prion Generation and Propagation

Prions are self-perpetuating proteinaceous agents that are associated with degenerative neurological disorder such as Creutzfeldt-Jakob disease, kuru, and mad cow diseases. In addition, prions have also been discovered in yeast, which has become an outstanding model to investigate the mechanisms of prion formation and propagation. It is now generally accepted that the same amyloidogenic protein can adopt multiple self-perpetuating conformations, which are referred to as prion strains. Different strains confer various neuropathology and incubation periods in mammalian prion diseases, and cause different cellular phenotypes in [PRION+] yeast cells. Interestingly, this strain feature has also been recently correlated to the variation in pathology of neurodegenerative diseases caused by “prion-like” proteins, such as different tauopathies and synucleinopathies caused by distinct Tau and α-synuclein strains, respectively. However, the generation of prion or prion-like amyloid, and the formation of various strains are not fully understood. Here, I use yeast prion [PSI+] as a model to explore the protein sequence requirement for de novo prion generation and prion propagation, as well as investigate how such requirements differ between distinct strains. The protein determinant of [PSI+] is Sup35, which is a translation termination factor. In [psi-] cells, Sup35 is soluble and functions in translation termination. By contrast, Sup35 forms an amyloid structure in [PSI+] cells, resulting in a loss-of-function and nonsense suppression cellular phenotype. The “N” domain, composed of amino acids 1-123 of Sup35, is required to induce and maintain [PSI+], and thus was classified as the prion-forming domain (PFD). However, by generating recombinant Sup35 truncations with different lengths of the M domain (amino acids 124-253), I found that most of the M domain is required for de novo [PSI+] formation, and a small region of the M domain is also required for [PSI+] propagation. In addition, I found that the requirement for both [PSI+] formation and propagation are strain dependent, and I further identified the regions required for various [PSI+] strains. These data show that the M domain of Sup35, and thus the protein sequence flanking the PFD of amyloidogenic protein, is crucial in prion formation and propagation, as well as in dictating prion strain polymorphism. I further characterized a critical region of amino acids 129-148 in the M domain, which was identified as a binding site for a chaperone protein required for [PSI+] propagation, called Hsp104. I showed the protein sequence of this region greatly affect the propagation of weak [PSI+], and to a less extent on strong [PSI+]. Deletion of this region also revealed strain specificity of nonsense suppression level upon the overexpression of the other termination factor, Sup45. The results also suggest that there are other Hsp104 binding sites outside of amino acids 129-148, and such interactions can partially maintain [PSI+] propagation. Lastly, I found the Sup35 sequence requirement for [PSI+] formation is also dependent on the strain of another yeast prion, [RNQ+]. In addition, the [PSI+] strains formed were also [RNQ+] strain-dependent. This provides further evidences for a previous hypothesis that aggregated Rnq1 in [RNQ+] cells directly interact with soluble Sup35, by cross-seeding Sup35 monomers to induce [PSI+] aggregates. The cross-seeding hypothesis is attractive and will provide significant insight in understanding the mechanism of neurodegenerative diseases that involve multiple amyloidogenic proteins that might cross seed each other. Taken together, this dissertation highlights the factors crucial for de novo prion generation and prion propagation, and expands the understanding of prion strain polymorphisms. It thus has broad implications in understanding the mechanism of amyloidogenic protein misfolding in neurodegenerative diseases, as well as the variety of the disease pathology