Search

Oxidative stress is known to trigger protein misfolding and has been implicated in promoting spontaneous prion formation in both yeast and mammalian cells. Previous studies have suggested that oxidative stress causes protein misfolding and prion formation via oxidation of methionine residues. Methionine residues in proteins are particularly prone to oxidation, forming a racemic mixture of methionine- S-sulfoxide (Met-S-O) and methionine-R-sulfoxide (Met-R-O). In the current study, we directly tested the role of methionine oxidation in prion formation by mutating methionine residues in Sup35 and Rnq1, which are the soluble forms of the [PSI+] and [PIN+] prions, respectively. Substituting methionine residues for alanine residues in Sup35 or Rnq1 significantly reduced the frequency [PSI+] and [PIN+] prion formation consistent with the idea that methionine oxidation in a normally soluble protein underlies its transition to the amyloidogenic prion form. Previous studies have shown that methionine oxidation and prion formation is increased in mutants lacking antioxidant enzymes that act to detoxify reactive oxygen species (ROS) including peroxiredoxins, catalases and superoxide dismutases. Methionine-sulfoxide reductases (MXRs) are ubiquitous antioxidant enzymes that protect against methionine oxidation by directly reducing MetO. The current study looks at two yeast MXRs: Mxr1, which reduces Met-S-O, and Mxr2, which reduces Met-R-O. Mutants deleted for MXR1 or MXR2 are shown to have increased Sup35 methionine oxidation and an elevated frequency of [PSI+] prion formation consistent with the idea that methionine oxidation promotes prion formation. Surprisingly, [PSI+] prion formation is not so significantly increased in a double mutant lacking MXR1 and MXR2. [PSI+] prion formation can be induced by overexpression of Sup35 (NM-GFP) but this occurs at a much-reduced frequency in the double compared with the single mxr mutant strains. Visualisation of the Sup35 aggregates using fluorescence microscopy reveals that larger sized aggregates are formed in the double mxr mutant strain compared with the single and wild-type strains. The Hsp104 chaperone, that is required to generate [PSI+] propagons, is shown to localize to Sup35 aggregates and 15 a model is presented, where Hsp104 is less efficient at disaggregating the large Sup35 aggregates present in the mxr1 mxr2 mutant, which explains the decreased frequency of prion formation. Finally, we examined overall protein aggregation in mxr mutants by isolating aggregates and identifying the aggregated proteins using mass spectrometry. Similarly, high levels of protein aggregation are found in single and double mxr mutants compared with a wild-type strain. Analysis of the biophysical properties of the aggregates revealed that they are enriched for properties which have previously been described to promote aggregation including higher protein abundances, higher translation rates, and higher hydrophobicity. However, these biophysical properties are reduced in the mxr deletion strains compared to the wild-type strain indicating that the threshold for proteins to aggregate may be lower in the mutant strains. Analysis of the chaperones present in the aggregated fractions revealed that Hsp104 disaggregase is present in the aggregated fractions from the mxr2 and mxr1/2 deletion strain. Analysis of Hsp104 activity using a luciferase-GFP re-solubilisation assay shows that disaggregase activity is decreased in the mxr2 and mxr1/2 deletion strains suggesting one possible reason for the increased aggregation detected in these strains.

Abstract: Prions are infectious proteins. In their prion conformation, they catalyze the transformation of non-prion isomers into prions. Another characteristic common to currently identified prions is that they form amyloid aggregates. They occur in several mammals and multiple species of yeast, acting in a variety of biological processes, and their numbers continue to grow. Prions can adopt different infectious conformations known as “strains” in mammals or “variants” in yeast, each with distinct, epigenetically inheritable phenotypes. Mechanisms by which prion variants are determined remain unclear. In this thesis, I describe the characterization of a potential novel prion in Saccharomyces cerevisiae, Riq1p. I show that Riq1p can form SDS/heat-resistant aggregates and is sufficient to maintain a reversibly curable, nonsense-suppression phenotype when it is used to replace Sup35p’s prion determining domain. These findings are significant because, unlike the majority of previously identified prion proteins, Riq1p is richer in glutamine (Q) than asparagine (N) and its prion-like behavior cannot be isolated to its Q/N rich domain. In this way, Riq1p provides an important exception to amino-acid composition trends proposed to govern prion behavior and suggests that such theories require revision. I also report the results of a targeted screen for genes affecting early stages of de novo conversion of the prion protein Sup35p into its [PSI+] conformation. From this screen, I identify several chaperone genes that affect the variant of the S. cerevisiae prion Rnq1p/[PIN+]. I show that the deletion of specific chaperone genes alters [PIN+] variant phenotypes, including [PSI+] induction efficiency, Rnq1p aggregate morphology/size, and variant dominance. Genetic analysis demonstrated that the deletion-induced phenotypic changes are stably inherited in a non-Mendelian manner even after restoration of the deleted gene, confirming that they are due to a bona fide change in the [PIN+] variant. Taken together, these results not only show that it is possible to alter an established prion variant in vivo, but also that molecular chaperones may play an important part in this mechanism for regulating prion variants.