Many neurodegenerative diseases are caused by the accumulation of intracellular or extracellular proteins that form toxic amyloid deposits. The cell has developed a complex protein quality control system involving molecular chaperones to protect against the accumulation of these toxic amyloid aggregates. Yeast is an ideal system to study the mechanisms the cell has developed to eliminate amyloids. Amyloids occur in yeast in the form of prion proteins and more than 10 prions have been identified. Prions have two conformations: an amyloid conformation, which is infectious, and a normal folded conformation. The prion in its amyloid conformation occurs as cytosolic seeds that propagates by transmission of the seeds to the daughter cells. To cure yeast of the misfolded amyloid form of the prion, the prion seeds must be eliminated, which can occur by several different mechanisms: (1) inhibition of severing (2) asymmetric segregation of the seeds between mother and daughter cells and (3) dissolution of the seeds. We have investigated the curing under different conditions of two well-characterized yeast prions, PSI+ and URE3 prions by expressing GFP-labeled prion proteins. As yeast were cured, cells were imaged and plated to measure the extent of curing using a standard plating assay. Cells with amyloid prion conformation have fluorescent foci, whereas cured cells show diffuse fluorescence. Using these techniques, we have examined curing by many agents to elucidate the different mechanism used by yeast to eliminate the infectious conformation. The molecular chaperone, Hsp104, severs the prions seeds so inhibition of its severing activity cures prions by diluting out the seeds by cell division. This mechanism of curing predicts a progressive loss of foci per cell until the yeast are cured and the percent cells with foci should be equivalent to the percent cells with the amyloid prion conformation. As predicted, these results we obtained when URE3 was cured by inactivating Hsp104 activity with guanidine. However, when PSI+ was cured by inactivating Hsp104 with guanidine, the foci were not detectable in cells that were still PSI+ based on the plating assay. Foci became detectable in the non-cured cells when the yeast were stressed. These results led us to proposed that in addition to severing activity, Hsp104 has another activity that we called trimming. Trimming activity reduces the size of the seeds by removing monomers from the filaments, but unlike severing activity, it does not increase the number of seeds. Our imaging studies established that the infectious amyloid conformation is also cured by dissolution of the prion seeds. For more than two decades, it has been recognized that in addition to inactivation of Hsp104, overexpression of Hsp104 also cures the PSI+ prion, but not other prions. Over the years, several different models have been proposed to explain this unusual property of PSI+. We investigated how Hsp104 overexpression cures by imaging the prion seeds during the curing process. These studies showed that the mechanism of curing is by dissolution of the prion seeds and trimming of the prion seeds by Hsp104 is an integral part of this curing process. Hsp104 overexpression caused a loss of detectable foci in cells prior to curing, whereas overexpression of Hsp104 mutants that lacked trimming activity did not cure PSI+ yeast. Furthermore, we observed that the rate of trimming was different depending on the PSI+ variants. Variants that cure at a faster rate by Hsp104 overexpression showed faster trimming and were trimmed at low levels of Hsp104 overexpression and by Hsp104 orthologs from S. pombe and C. albicans. On the hand, variants that cure at a slower rate by Hsp104 overexpression showed slower trimming, were only partially trimmed at low levels of Hsp104 overexpression and were not trimmed by Hsp104 orthologs from S. pombe and C. albicans. The trimming of the prion seeds is also affected by the Hsp70 chaperone, Ssa1. Overexpression of Ssa1 inhibits the trimming of the prion seeds and antagonizes the curing of PSI+ by Hsp104 overexpression. Conversely, overexpression of a dominant negative mutant of Ssa1 increased the rate of trimming of the prion seeds and markedly increased the rate of PSI+ curing by Hsp104 overexpression. In fact, we found that overexpression of the dominant negative Ssa1 mutant cures some variants of PSI+ at endogenous levels of Hsp104. Here again, we observed trimming of the prion seeds, which occurred at a faster rate than the curing. These results indicate that trimming activity of Hsp104 causes dissolution of the prion seeds in PSI+ yeast. Dissolution of the prion seeds as a mechanism of curing can also be achieved by inhibiting the growth of the amyloid. This is the mechanism proposed for the curing of PSI+ and URE3 prions by overexpression of human DnaJb6 based on in vitro data showing that DnaJb6 blocks the growth of amyloids including yeast prions, huntingtin fragments, and amyloid-beta fragments. Consistent with the in vitro data, we observed a loss of detectable foci in cells that were not cured, but unlike curing by Hsp104 overexpression, the foci do not become detectable after stressing the cells. It appears that overexpression of DnaJb6 inhibits seed growth, but not the severing of the seeds by Hsp104. This continues to reduce the size of the seeds until they reach a point where the seeds are no longer stable. Curing of the PSI+ and URE3 prions also occurs by asymmetric segregation due to aggregation of the seeds, which impedes their transmission to the daughter cells. Imaging studies showed that overexpression of Hsp42, Btn2, Cur1, or Ydj1 cure URE3 by inducing the formation of cytosolic aggregates. Interestingly, these proteins all aggregate the prion seeds even though they have different cellular functions and localizations. The large prion aggregates colocalized with Btn2 and Hsp42, but not Ydj1 or Cur1. In fact, Cur1 localizes predominantly to the nucleus, whereas Ydj1, a member of the Hsp40 family, is cytosolic. We also found that overexpression of Ydj1 and Cur1 cures PSI+ prion, but not overexpression of Hsp42 and Btn2. Like URE3, overexpression of Ydj1 and Cur1 induced aggregation of the prion seeds in PSI+ cells, indicating curing by asymmetric segregation. The curing of PSI+ and URE3 by overexpression of Cur1 and Ydj1 appears to stem from a reduction in the cytosolic pool of the Sis1-Ssa1 complex. Sis1, a member of the Hsp40 family is recruited into the nucleus by Cur1, whereas Ydj1 competes with Sis1 for the molecular chaperone, Ssa1. In contrast to other laboratories that have proposed that these proteins cure prions by inhibiting the severing of the seeds, our data support a model of curing by asymmetric segregation. This model is further supported by our findings that overexpression of a dominant negative Sis1 mutant cured both the PSI+ and URE3 prions by aggregating the prion seeds, whereas overexpression of wild-type Sis1 inhibited the aggregation of the prion seeds. Our imaging studies of the PSI+ and URE3 prions show different mechanisms that the yeast utilize to eliminate the prion seeds, ultimately curing these prions. We are now extending our studies to examine factors affecting amyloid formation of huntingtin fragments in yeast. It has been reported that prions have an essential role in huntingtin amyloid formation in yeast. Since in mammalian cells, these huntingtin fragments always form amyloid, we are trying to understand what keeps these fragments soluble in the absence of prion and whether prion is necessary only for nucleation or for nucleation and maintenance of the huntingtin amyloid.
Showing the most recent 10 out of 19 publications
