The yeast arrestin-related protein Bul1 is a novel actor of glucose-induced endocytosis

Abstract

Yeast cells have a remarkable ability to adapt to nutritional changes in their environment. During adaptation, nutrient-signaling pathways drive the selective endocytosis of nutrient transporters present at the cell surface. A current challenge is to understand the mechanistic basis of this regulation. Transporter endocytosis is triggered by their ubiquitylation, which involves the ubiquitin ligase Rsp5 and its adaptors of the arrestin-related family (ART). This step is highly regulated by nutrient availability. For instance, the monocarboxylate transporter Jen1 is ubiquitylated, endocytosed, and degraded upon exposure to glucose. The ART protein Rod1 is required for this overall process; yet Rod1 rather controls Jen1 trafficking later in the endocytic pathway and is almost dispensable for Jen1 internalization. Thus, how glucose triggers Jen1 internalization remains unclear. We report that another ART named Bul1, but not its paralogue Bul2, contributes to Jen1 internalization. Bul1 responds to glucose availability, and preferentially acts at the plasma membrane for Jen1 internalization. Thus, multiple ARTs can act sequentially along the endocytic pathway to control transporter homeostasis. Moreover, Bul1 is in charge of Jen1 endocytosis after cycloheximide treatment, suggesting that the functional redundancy of ARTs may be explained by their ability to interact with multiple cargoes in various conditions.

FIGURE 1:

Bul1 assists Rod1 in the glucose-induced endocytosis of Jen1, and acts primarily at the plasma membrane. (A) Localization of Jen1-GFP in WT, rod1∆, rod1∆ bul1∆, and rod1∆ bul2∆ cells after 4 h growth in lactate medium (Jen1 induction) and at various times after the addition of glucose. (B) From the experiment presented in A, quantification of the ratio of fluorescence at the cell periphery over total fluorescence at 60 min after glucose treatment (see Materials and Methods). (C) Localization of Jen1-GFP in WT, rod1∆, bul1∆, and rod1∆ bul1∆ cells after 4 h growth in lactate medium and after glucose treatment. (D) Model of Jen1 trafficking after internalization. The deletion of VPS52 abrogates the vacuolar targeting of Jen1. (E) Localization of Jen1-GFP in vps52∆, vps52∆ rod1∆, and vps52∆ rod1∆ bul1∆ cells grown in the indicated conditions. LatA: latrunculin A. (F) From the experiment presented in E, quantification of the ratio of fluorescence at the cell periphery over total fluorescence at 30 min after glucose treatment.

FIGURE 2:

Bul1 is responsible for the residual ubiquitylation and degradation of Jen1 observed in the rod1∆ strain. (A) Degradation of Jen1-GFP over time after glucose treatment in WT, rod1∆, bul1∆, and rod1∆ bul1∆ cells. Total protein extracts were analyzed by SDS–PAGE and Western blotting using the indicated antibodies. PGK: phosphoglycerate kinase (loading control). (B) Quantification of the degradation of Jen1-GFP for each strain (ratio of free GFP over total GFP signal at the 60 min time point; see Materials and Methods). (C) Line scan of the Western blot presented in A showing pixel intensity (relative to the max intensity for each lane) over 100 pixels. (D) Jen1-GFP was immunopurified in denaturing conditions from the indicated strains, 10 min after glucose treatment. Immunoprecipitates were blotted with anti-GFP and anti-ubiquitin antibodies.

FIGURE 3:

Expression of a TGN-localized version of Rod1 leads to an increased dependency toward the function of Bul1 for Jen1 degradation. (A) Localization of Rod1-GFP and Rod1-PH^FAPP1-GFP in wild-type yeast cells expressing the TGN marker, Sec7-mCherry, in the indicated culture conditions. Bottom, false-color images of the 10 min Glc time point showing increased TGN association of Rod1-PH^FAPP1-GFP over Rod1-GFP (generated using the “Fire” lookup table in ImageJ). (B) Degradation of Jen1-GFP over time after glucose treatment in rod1∆ or rod1∆ bul1∆ cells, carrying an empty plasmid or plasmids encoding Rod1-PH^FAPP1-GFP or Rod1-GFP. Note that both Rod1/Rod1-PH^FAPP1 and Jen1 are tagged with GFP; however, Rod1/Rod1-PH^FAPP1 are expressed at a much lower level (8- to 16-fold less than Jen1, respectively; see Supplemental Figure S2) and do not interfere with the detection of Jen1-GFP signals. (C) Quantification of Jen1-GFP degradation for each strain (60 min time point; see Materials and Methods).

FIGURE 4:

Glucose regulates Bul1 phosphorylation. (A) Western blot from total protein extracts prepared from cells expressing ether Bul1-Flag or Bul1(PYm)-Flag and grown as indicated. (B) Line scan of the Western blot presented in A showing pixel intensity (relative to the max intensity for each lane) over 100 pixels. (C) Western blot on total protein extracts prepared from cells expressing ether Bul1-Flag or Bul1(PYm)-Flag and grown as indicated, and treated or not with calf intestinal phosphatase (CIP). (D) Line scan of the Western blot presented in C.

FIGURE 5:

The glucose-induced dephosphorylation of Bul1 requires Sit4. (A) Western blot from total protein extracts prepared from WT or reg1∆ cells expressing either Bul1-Flag or Bul1(PYm)-Flag and grown as indicated. (B) Western blot from total protein extracts prepared from WT or sit4∆ cells expressing Bul1-Flag grown as indicated and treated or not with calf intestinal phosphatase (CIP). (C) Line scan of the Western blot presented in B. (D) Localization of a galactose-inducible Jen1-GFP in WT, sit4∆, rod1∆, and rod1∆ sit4∆ cells after galactose induction or at various times after glucose treatment. (E) From the experiment presented in D, quantification of the ratio of fluorescence at the cell periphery over total fluorescence. (F) Localization of Jen1-GFP in WT, rod1∆, bul1∆, and rod1∆ bul1∆ cells after lactate induction or various times after CHX treatment. (G) From the experiment presented in H, quantification of the ratio of fluorescence at the cell periphery over total fluorescence.