Role of molecular chaperones in protein folding diseases

许多神经退行性疾病是由形成淀粉样沉积的细胞内或细胞外蛋白质的积累引起的。一般来说，淀粉样蛋白沉积可以穿过细胞膜，将有毒的淀粉样蛋白扩散到邻近的细胞。由于这种传播方式类似于感染性朊病毒疾病中的播种，朊病毒样传播似乎在许多神经退行性疾病中很常见。为了防止有毒淀粉样蛋白聚集体的积累，分子伴侣的作用是分解它们，但是当它们不能保持质量控制时，蛋白质聚集体就形成了。我的研究处于这两个领域的交叉点；研究酵母和哺乳动物细胞中蛋白质聚集体的形成和繁殖及其分子伴侣的分解。

Many neurodegenerative diseases are caused by the accumulation of intracellular or extracellular proteins that form amyloid deposits. In general, amyloid deposits can cross cell membranes to spread the toxic amyloids to neighboring cells. Since this mode of propagation is similar to the seeding that occurs in infectious prion disease, prion-like transmission appears to be common to many neurodegenerative diseases. To protect against the accumulation of toxic amyloid aggregates, molecular chaperones function to disaggregate them, but when they do not maintain quality control, protein aggregates are formed. My research is at the crossroads of these two areas; examining the formation and propagation of protein aggregates and their disaggregation by molecular chaperones in yeast and mammalian cells. The main focus of my laboratory has been on prion propagation in both mammalian and yeast cells. Prions are infectious proteins that self-propagate by changing from their properly folded conformation to a misfolded amyloid conformation. The conversion to the amyloid conformation can occur either spontaneously or by seeding with the misfolded amyloid protein. Once conversion occurs, the amyloid prion protein continues to propagate from mother to daughter cells. In mammalian cells, the properly folded prion protein (PrPc) is a GPI-anchored protein that resides primarily on the plasma membrane. The conversion of PrPc to the amyloid scrapie conformation (PrPsc) occurs as PrPc traffics along the endosomal pathway, but the compartment where conversion takes place has been controversial. By inhibiting the cellular trafficking of PrPsc at various endosomes, we found that the major intracellular site of prion conversion is the multivesicular body (MVB). This observation is of particular interest because the MVB has an unusual membrane topology that might enable trans-interaction between PrPc and PrPsc molecules within the cell as well as at the plasma membrane. We are now perturbing other trafficking pathways both to confirm our results and to understand how blocking different pathways for the internalization of cargo affects the propagation of PrPsc. Knocking down caveolin, and flotillin had no significant effect on PrPc and PrPsc levels. On the other hand, knocking down clathrin and AP2 caused an increase in PrPsc levels, but did not significantly affect PrPc levels. These results suggest that both PrPc and PrPsc are internalized in lipid rafts when clathrin-mediated endocytosis is blocked, which in turn increases the conversion of PrPc to PrPsc. We have also been studying prions in yeast, which is an excellent system for understanding the role of molecular chaperones in prion propagation. There are more than a dozen known prions in budding yeast and their propagation is dependent on the molecular chaperone Hsp104. Hsp104 severs the prion seeds so inhibiting its activity prevents new seed production. This in turn cures the yeast of the prion when the seeds are diluted out by cell division. In addition to severing activity, we found a new activity for Hsp104 that we called trimming. Unlike severing activity, trimming activity makes the seeds smaller, but does not increase the seed number. Paradoxically, one of the yeast prions, called PSI+ prion, is not only cured by inactivation of Hsp104, but also by overexpression of Hsp104. To understand the mechanism by which Hsp104 overexpression cures PSI+, the curing process was examined using live cell imaging of yeast expressing GFP-labeled Sup35 in combination with standard plating assays. We determined that overexpression of Hsp104 cured PSI+ by dissolution of the prion seeds in a process that was dependent on the trimming activity of Hsp104. Mutants of Hsp104 that were defective in trimming activity did not cure PSI+ when overexpressed. We determined that the molecular chaperone Ssa1, a member of the Hsp70 family, inhibits the trimming activity of Hsp104, which in turn inhibits the curing of PSI+ by Hsp104 overexpression. Conversely, a dominant negative Ssa1 mutant increases the trimming of the prion seeds, resulting in faster curing of PSI+. Therefore, Ssa1 concentration plays a critical role in determining whether overexpression of Hsp104 cures PSI+ prion by regulating the trimming activity of Hsp104. We also examined the curing of PSI+ by overexpression of different fungal homologs of Hsp104. The rate of curing was examined in weak and strong PSI+ variants, which differ in that weak variants have fewer seeds than strong PSI+ variants. Overexpression of Hsp104 from Saccharomyces cerevisiae, Candida albicans or Schizosaccharomyces pombe cured the weak PSI+ variants, but only S. cerevisiae Hsp104 cured the strong PSI+ variants. We determined that Hsp104 from C. albicans and S. pombe had reduced trimming activity, which prevented their curing the strong PSI+ variants by Hsp104 overexpression. Finally, we examined the role of the highly acidic C-terminal extension, which is present among members of the fungal Hsp104 family, on the propagation of PSI+ and the curing of PSI+ by dissolution of the prion seeds. Without the C-terminal acidic extension, the different fungal homologs did not propagate PSI+. When these Hsp104 fragments were overexpressed in PSI+ cells, the S. cerevisiae Hsp104 fragment acted as a dominant negative mutant, whereas the C. albicans and S. pombe Hsp104 fragments caused aggregation of the prion seeds, which in turn cured PSI+ curing by asymmetric segregation of the prion seeds. To test whether the C-terminal extension of Hsp104 was essential for propagating and for curing PSI+ by dissolution of the prion seeds when overexpressed, we used a 444B chimera. This chimera contains all the domains of S. cerevisiae Hsp104 except for the C-terminal domain of Hsp104, which was swapped with the homologous domain from ClpB, the E.coli paralog of Hsp104. Unlike the fungal Hsp104 homologs, the C-terminal domain of ClpB does not have the acidic extension. We found that 444B propagated PSI+ and cured PSI+ by dissolution of the prion seeds when overexpressed. Therefore, the function of the C-terminal domain has been conserved among members of the Hsp104/ClpB family, but there has not been conservation of the primary sequence.