In many cell types, lateral diffusion barriers compartmentalize the plasma membrane and, at least in budding yeast, the endoplasmic reticulum (ER). However, the molecular nature of these barriers, their mode of action and their cellular functions are unclear. Here, we show that misfolded proteins of the ER remain confined into the mother compartment of budding yeast cells. Confinement required the formation of a lateral diffusion barrier in the form of a distinct domain of the ER-membrane at the bud neck, in a septin-, Bud1 GTPase- and sphingolipid-dependent manner. The sphingolipids, but not Bud1, also contributed to barrier formation in the outer membrane of the dividing nucleus. Barrier-dependent confinement of ER stress into the mother cell promoted aging. Together, our data clarify the physical nature of lateral diffusion barriers in the ER and establish the role of such barriers in the asymmetric segregation of proteotoxic misfolded proteins during cell division and aging.
Bik1p is the budding yeast counterpart of the CLIP-170 family of microtubule plus-end tracking proteins, which are required for dynein localization at plus ends and dynein-dependent spindle positioning. CLIP-170 proteins make up a CAP-Gly microtubule-binding domain, which sustains their microtubule plus-end tracking behaviour. However, in yeast, Bik1p travels towards plus ends as a cargo of the plus-end-directed kinesin Kip2p. Additionally, Kip2p behaves as a plus-end-tracking protein; hence, it has been proposed that Bik1p might track plus ends principally as a cargo of Kip2p. Here, we examined Bik1p localization in yeast strains expressing mutant tubulin lacking the C-terminal amino acid (Glu tubulin; lacking Phe), the interaction of which with Bik1p is severely impaired compared with wild type. In Glu-tubulin strains, despite the presence of robust Kip2p comets at microtubule plus ends, Bik1p failed to track plus ends. Despite Bik1p depletion at plus ends, dynein positioning at the same plus ends was unperturbed. Video microscopy and genetic evidence indicated that dynein was transported at plus ends in a Kip2p-Bik1p-dependent manner, and was then capable of tracking Bik1p-depleted plus ends. These results indicate that Bik1p interactions with tubulin are important for Bik1p plus-end tracking, and suggest alternative pathways for Bik1p-Kip2p-dependent dynein localization at plus ends.
In budding yeast, the phosphatase Cdc14 orchestrates progress through anaphase and mitotic exit, thereby resetting the cell cycle for a new round of cell division. Two consecutive pathways, Cdc fourteen early anaphase release (FEAR) and mitotic exit network (MEN), contribute to the progressive activation of Cdc14 by regulating its release from the nucleolus, where it is kept inactive by Cfi1. In this study, we show that Cdc14 activation requires the polo-like kinase Cdc5 together with either Clb–cyclin-dependent kinase (Cdk) or the MEN kinase Dbf2. Once active, Cdc14 triggers a negative feedback loop that, in the presence of stable levels of mitotic cyclins, generates periodic cycles of Cdc14 release and sequestration. Similar phenotypes have been described for yeast bud formation and centrosome duplication. A common theme emerges where events that must happen only once per cycle, although intrinsically capable of oscillations, are limited to one occurrence by the cyclin–Cdk cell cycle engine.