Nutrient supply dictates cell signaling changes, which in turn regulate membrane protein trafficking. To better exploit nutrients, cells relocalize membrane transporters via selective protein trafficking. Key in this reshuffling are the α-arrestins, selective protein trafficking adaptors conserved from yeast to man. α-Arrestins bind membrane proteins, controlling the ubiquitination and endocytosis of many transporters. To prevent the spurious removal of membrane proteins, α-arrestin-mediated endocytosis is kept in check through phospho-inhibition. This phospho-regulation is complex, with up to 87 phospho-sites on a single α-arrestin and many kinases/phosphatases targeting α-arrestins. To better define the signaling pathways controlling paralogous α-arrestins, Aly1 and Aly2, we screened the kinase and phosphatase deletion (KinDel) library, which is an array of all non-essential kinase and phosphatase yeast deletion strains, for modifiers of Aly-mediated phenotypes. We identified many Aly regulators, but focused our studies on the TORC1 kinase, a master regulator of nutrient signaling across eukaryotes. We found that TORC1 and its signaling effectors, the Sit4 protein phosphatase and Npr1 kinase, regulate the phosphorylation and stability of Alys. When Sit4 is lost, Alys are hyperphosphorylated and destabilized in an Npr1-dependent manner. These findings add new dimensions to our understanding of TORC1 regulation of α-arrestins and have important ramifications for cellular metabolism.

Background information: Phosphatidylinositol (PI) is an essential phospholipid, critical to membrane bilayers. The complete deacylation of PI by B-type phospholipases produces intracellular and extracellular glycerophosphoinositol (GPI). Extracellular GPI is transported into the cell via Git1, a member of the Major Facilitator Superfamily of transporters at the yeast plasma membrane. Internalized GPI is degraded to produce inositol, phosphate and glycerol, thereby contributing to these pools. GIT1 gene expression is controlled by nutrient balance, with phosphate or inositol starvation increasing GIT1 expression to stimulate GPI uptake. However, less is known about control of Git1 protein levels or localization. Results: We find that the α-arrestins, an important class of protein trafficking adaptor, regulate Git1 localization and this is dependent upon their interaction with the ubiquitin ligase Rsp5. Specifically, α-arrestin Aly2 stimulates Git1 trafficking to the vacuole under basal conditions, but in response to GPI-treatment, either Aly1 or Aly2 promote Git1 vacuole trafficking. Cell surface retention of Git1, as occurs in aly1∆ aly2∆ cells, is linked to impaired growth in the presence of exogenous GPI and results in increased uptake of radiolabeled GPI, suggesting that accumulation of GPI somehow causes cellular toxicity. Regulation of αarrestin Aly1 by the protein phosphatase calcineurin improves steady-state and substrate-induced trafficking of Git1, however, calcineurin plays a larger role in Git1 trafficking beyond regulation of α-arrestins. Interestingly, loss of Aly1 and Aly2 increased phosphatidylinositol-3-phosphate on the limiting membrane of the vacuole, and this was further exacerbated by GPI addition, suggesting that the effect is partially linked to Git1. Loss of Aly1 and Aly2 leads to increased incorporation of inositol label from [ 3 H]-inositol-labelled GPI into PI, confirming that internalized GPI influences PI balance and indicating a role for the a-arrestins in this regulation. Conclusions: The α-arrestins Aly1 and Aly2 are novel regulators of Git1 trafficking with previously unanticipated roles in controlling phospholipid distribution and balance.

Phosphatidylinositol (PI) is an essential phospholipid and critical component of membrane bilayers. The complete deacylation of PI by phospholipases of the B-type leads to the production of intracellular and extracellular glycerophosphoinositol (GPI), a water-soluble glycerophosphodiester. Extracellular GPI is transported into the cell via Git1, a member of the Major Facilitator Superfamily of transporters that resides at the plasma membrane in yeast. Once internalized, GPI can be degraded to produce inositol, phosphate and glycerol, thereby contributing to reserves of these building blocks. Not surprisingly, GIT1 gene expression is controlled by nutrient balance, with limitation for phosphate or inositol each increasing GIT1 expression to facilitate GPI uptake. Less is known about how Git1 protein levels or localization are controlled. Here we show that the α-arrestins, an important class of protein trafficking adaptor, regulate the localization of Git1 in a manner dependent upon their association with the ubiquitin ligase Rsp5. Specifically, α-arrestin Aly2 is needed for effective Git1 internalization from the plasma membrane under basal conditions. However, in response to GPI-treatment of cells, either Aly1 or Aly2 can promote Git1 trafficking to the vacuole. Retention of Git1 at the cell surface, as occurs in aly1∆ aly2∆ cells, results in impaired growth in the presences of excess exogenous GPI and results in increased uptake of radiolabeled GPI, suggesting that accumulation of this metabolite or its downstream products leads to cellular toxicity. We further show that regulation of α-arrestin Aly1 by the protein phosphatase calcineurin improves both steady-state and ligand-induced trafficking of Git1 when a mutant allele of Aly1 that mimics the dephosphorylated state at calcineurin-regulated residues is employed. Thus, calcineurin regulation of Aly1 is important for the GPI-ligand induced trafficking of Git1 by this α-arrestin, however, the role of calcineurin in regulating Git1 trafficking is much broader than can simply be explained by regulation of the α-arrestins. Finally, we find that loss of Aly1 and Aly2 leads to an increase in phosphatidylinositol-3-phosphate on the limiting membrane of the vacuole and this alteration is further exacerbated by addition of GPI, suggesting that the effect is at least partially linked to Git1 function. Indeed, loss of Aly1 and Aly2 leads to increased incorporation of inositol label from 3H-inositol-labelled GPI into PI, confirming that internalized GPI influences PI synthesis and indicating a role for the α-arrestins in regulating the process.

α‐Arrestins, a novel class of protein trafficking adaptors, help cells survive environmental changes by controlling membrane protein trafficking. One hurdle to understanding α‐arrestins is that few α‐arrestin‐cargo pairs have been identified. It is technically challenging to identify membrane cargos trafficked by a‐arrestins due to their transient associations and their biochemical nature. To identify α‐arrestin‐regulated cargos, we used Evolutionary Rate Covariation (ERC), which employs sequence‐based signatures to identify genes with similar evolutionary histories. We compared ERC values for α‐arrestins with proteins across 18 yeast species. Among the top co‐evolving proteins were those previously defined as α‐arrestin cargos. We are determining if membrane proteins, not yet identified as α‐arrestin cargos, with the highest ERC values (>0.5), interact with α‐arrestins by assessing their localization and relative protein abundances in wild‐type cells versus those lacking the specific α‐arrestins. Fluorescence intensity was used to quantify the abundance and/or subcellular distribution of GFP tagged proteins. Statistically significant changes in GFP abundance or localization between wild‐type cells and those lacking α‐arrestins demonstrate dependence on these protein trafficking adaptors. This makes them good candidates as new α‐arrestin‐dependent cargos, which we are confirming by co‐localization, bimolecular fluorescence complementation, and biochemical approaches. Using ERC and our fluorescence imaging pipeline, we have quantitatively confirmed that 36 integral membrane proteins, previously unassociated with α‐arrestins, display α‐arrestin‐dependent localization changes. This greatly expands the repertoire of α‐arrestin cargo within cells, and raises important functional implications for this family. In conclusion, the ERC approach is a powerful new tool that is able to define protein trafficking regulatory networks, which will undoubtedly be of interest to the cell biology community.

Glucose is the preferred carbon source for most eukaryotes, and the first step in its metabolism is phosphorylation to glucose-6-phosphate. This reaction is catalyzed by a family of enzymes called either hexokinases or glucokinases depending on their substrate specificity. The yeast Saccharomyces cerevisiae encodes three such enzymes, Hxk1, Hxk2 and Glk1. In yeast and mammals, some isoforms of this enzyme are found in the nucleus, suggesting a possible moonlighting function beyond glucose phosphorylation. In contrast to mammalian hexokinases, the yeast Hxk2 enzyme has been proposed to shuttle into the nucleus in glucose replete conditions where it reportedly moonlights as part of a glucose-repressive transcriptional complex. To achieve this role in glucose repression, Hxk2 reportedly binds the Mig1 transcriptional repressor, is dephosphorylated at serine 15 in its N-terminus, and requires an Nterminal nuclear localization sequence (NLS). In this study, we use high-resolution, quantitative, fluorescent microscopy of live cells to determine the conditions, residues, and regulatory proteins required for Hxk2 nuclear localization. In direct contradiction to previous yeast studies, our quantitative imaging demonstrates that Hxk2 is largely excluded from the nucleus under glucose replete conditions but is retained in the nucleus under glucose limiting conditions. Our data show that the Hxk2 N-terminus does not contain an NLS but instead comprises sequences necessary for nuclear exclusion and multimerization regulation. Amino acid substitutions of the phosphorylated residue, serine 15, disrupt Hxk2 dimerization but have no effect on its glucose-regulated nuclear localization. Substitution of alanine at the nearby residue, lysine 13, affects both dimerization and maintenance of nuclear exclusion under glucose replete conditions. Modeling and simulation provide insight into the molecular mechanisms of this regulation. In marked contrast to earlier studies, we find that the transcriptional repressor Mig1 and the protein kinase Snf1 have little effect on Hxk2 localization. Instead, the protein kinase Tda1 is a key regulator of Hxk2 localization. Finally, RNAseq analyses of the yeast transcriptome further dispel the idea that Hxk2 moonlights as a transcriptional repressor, demonstrating that Hxk2 has a negligible role in transcriptional regulation in both glucose replete and limiting conditions. Taken together, our studies provide a paradigm shift for the conditions, residues, and regulators controlling Hxk2 dimerization and nuclear localization. Based on our data, the nuclear translocation of Hxk2 in yeast occurs in glucose starvation conditions, a finding that aligns well with the nuclear regulation of mammalian orthologs of this enzyme. Our findings lay the foundation for future studies of Hxk2 nuclear activity.