Preview

PDF, English - main document Download (3MB) | Terms of use

Abstract

Increasing evidence suggests that the accumulation of misfolded protein species into specific spatially separated deposition sites is a cytoprotective response of the cell. Yeast has at least three different protein quality control sites for the deposition of aggregated proteins. The JUxtaNuclear Quality control (JUNQ)/IntraNuclear Quality control site (INQ) and the Cyto-Q harbours unstructured, amorphously misfolded proteins, while the perivacuolar Insoluble PrOtein Deposit (IPOD) is regarded as a specialized deposition site for highly ordered amyloid aggregates. Recently, it was found that targeting of amyloid aggregates to the IPOD depends on proteins that function either in actin cable-based transport processes (Myo2, Tpm1/2) or in vesicular transport and vesicle fusion events (Sec18, Sec14, Sec21, Vps1). Knockdown/deletion of either of the above-mentioned factors resulted reversibly in multiple small aggregates of the model amyloid PrD-GFP dispersed throughout the cytoplasm instead of its proper accumulation at the IPOD. These multiple aggregates, also interpreted as transport intermediates co-localized with the Atg9 vesicle marker, Atg9, and the CVT (Cytoplasm-to-Vacuole Targeting) pathway substrate preApe1. Based on these findings, it was hypothesized that the recruitment machinery for amyloid aggregates to the IPOD overlaps with that for preApe1 to the neighbouring PAS (Phagophore Assembly Site) and involved Atg9 or related vesicles that are transported along actin cables to the IPOD. In the current study, we falsified this hypothesis by evaluating the effects of in vivo gene knockout studies of ATG9 and other key components of this pathway on the recruitment machinery.

In order to identify the key molecular factors and narrow down the vesicular pathways involved in the recruitment machinery for amyloid aggregates to the IPOD, we performed an unbiased mass spectrometry approach to isolate the transport intermediates of PrD-GFP generated in a VPS1 null mutant or a SEC18 knockdown strain. Candidates tested in follow-up experiments using in vivo knockdown/knock-out, and co-localization techniques confirmed our initial hypothesis that vesicular transport is involved in amyloid recruitment to the IPOD. Furthermore, we found that proper recruitment of PrD-GFP to the IPOD is disrupted upon depletion/deletion of components involved in Golgi to endosome targeting and intra-Golgi transport processes (Mon2, Dop1, Cop1) as well as candidates mediating homotypic membrane fusion events as well as endosome to vacuole transport (Vps33, Vps45). Using fluorescence microscopy, we observed that Vps33, a Sec1/Munc18 family (SM) protein and core component of HOPS/CORVET multisubunit tethering complexes co-localized with the multiple PrD-GFP aggregates generated in MON2 null mutants. Based on these findings, it was proposed that PrD-GFP aggregates are recruited to the IPOD via endosomes/MVB (Multivesicular bodies), which are known for their role in delivering substrates to the vacuole as a part of endosomal vacuolar transport.