When yeast cells growing on a poor nitrogen source are supplied with NH4+ ions, several nitrogen permeases including the general amino acid permease (Gap1p) are rapidly and completely inactivated. This report shows that inactivation by NH4+ of the Gap1 permease is accompanied by its degradation. A functional NPl1 gene product is required for both inactivation and degradation of Gap1p. Molecular analysis of the NPl1 gene showed that it is identical to RSP5. The RSP5 product is a ubiquitin-protein ligase (E3 enzyme) whose physiological function was, however, unknown. Its C-terminal region is very similar to that of other members of the E6-AP-like family of ubiquitin-protein ligases. Its N-terminal region contains a single C2 domain that may be a Ca(2+)-dependent phospholipid interaction motif, followed by several copies of a recently identified domain called WW(P). The Npi1/Rsp5 protein has a homologue both in humans and in mice, the latter being involved in brain development. Stress-induced degradation of the uracil permease (Fur4p), a process in which ubiquitin is probably involved, was also found to require a functional NPl1/RSP5 product. Chromosomal deletion of NPl1/RSP5 showed that this gene is essential for cell viability. In the viable npi1/rsp5 strain, expression of NPl1/RSP5 is reduced as a result of insertion of a Ty1 element in its 5' region. Our results show that the Npi1/Rsp5 ubiquitin-protein ligase participates in induced degradation of at least two permeases, Gap1p and Fur4p, and probably also other proteins.

Uracil uptake by Saccharomyces cerevisiae is mediated by the FUR4-encoded uracil permease. This permease undergoes endocytosis and subsequent degradation in cells subjected to adverse conditions. The data presented here show that uracil permease also undergoes basal turnover under normal growth conditions. Both basal and induced turnover depend on the essential Npi1p/Rsp5p ubiquitin-protein ligase. Epitope-tagged ubiquitin variants have been used to show that uracil permease is ubiquitinated in vivo. The ubiquitin-permease conjugates that are readily demonstrated in wild type cells were barely detectable in npi1 mutant cells, indicating that uracil permease may be a physiological substrate of the Npi1p ubiquitin ligase. The lack of ubiquitination of the permease in npi1 cells resulted in an increase in active, i.e. plasma membrane-located, permease, suggesting that there is a direct relationship between ubiquitination and removal of the permease from the plasma membrane. The accumulation of ubiquitin-permease conjugates in thermosensitive act1 mutant cells, deficient in the internalization step of endocytosis is consistent with this idea. On the other hand, the degradation of uracil permease does not require a functional proteasome since the permease was not stabilized in either pre1 pre2 or cim3 and cim5 mutant cells that have impaired catalytic (pre) or regulatory (cim) proteasome subunits. In contrast, both basal and stress-stimulated turnover rates were greatly reduced in pep4 mutant cells having defective vacuolar protease activities. We therefore propose that ubiquitination of uracil permease acts as a signal for endocytosis of the protein that is subsequently degraded in the vacuole.

The yeast uracil permease, Fur4p, is downregulated by uracil, which is toxic to cells with high permease activity. Uracil promotes cell surface Rsp5p-dependent ubiquitylation of the permease, signaling its endocytosis and further vacuolar degradation. We show here that uracil also triggers the direct routing of its cognate permease from the Golgi apparatus to the endosomal system for degradation, without passage via the plasma membrane. This early sorting was not observed for a variant permease with a much lower affinity for uracil, suggesting that uracil binding is the signal for the diverted pathway. The FUI1-encoded uridine permease is similarly sorted for early vacuolar degradation in cells exposed to a toxic level of uridine uptake. Membrane proteins destined for vacuolar degradation require sorting at the endosome level to the intraluminal vesicles of the multivesicular bodies. In cells with low levels of Rsp5p, Fur4p can be still diverted from the Golgi apparatus but does not reach the vacuolar lumen, being instead missorted to the vacuolar membrane. Correct luminal delivery is restored by the biosynthetic addition of a single ubiquitin, suggesting that the ubiquitylation of Fur4p serves as a specific signal for sorting to the luminal vesicles of the multivesicular bodies. A fused ubiquitin is also able to sort some Fur4p from the Golgi to the degradative pathway in the absence of added uracil but the low efficiency of this sorting indicates that ubiquitin does not itself act as a dominant signal for Golgi-to-endosome trafficking. Our results are consistent with a model in which the binding of intracellular uracil to the permease signals its sorting from the Golgi apparatus and subsequent ubiquitylation ensures its delivery to the vacuolar lumen.

Uracil permease is a multispanning protein of the Saccharomyces cerevisiae plasma membrane which is encoded by the FUR4 gene and produced in limited amounts. It has a long N-terminal hydrophilic segment, which is followed by 10 to 12 putative transmembrane segments, and a hydrophilic C terminus. The protein carries seven potential N-linked glycosylation sites, three of which are in its N-terminal segment. Overexpression of this permease and specific antibodies were used to show that uracil permease undergoes neither N-linked glycosylation nor proteolytic processing. Uracil permease N-terminal segments of increasing lengths were fused to a reporter glycoprotein, acid phosphatase. The in vitro and in vivo fates of the resulting hybrid proteins were analyzed to identify the first signal anchor sequence of the permease and demonstrate the cytosolic orientation of its N-terminal hydrophilic sequence. In vivo insertion of the hybrid protein bearing the first signal anchor sequence of uracil permease into the endoplasmic reticulum membrane was severely blocked in sec6l and sec62 translocation mutants.Many specific yeast permeases have been subjects of genetic and physiological studies (8), while more recent reports have concentrated on their molecular biology (1,5,6,13,21,30,43,51). Protein structure prediction programs have been used to propose preliminary structural models based on the corresponding DNA sequences (21, 42, 51). The sequencing data indicate that several yeast sugar transporters are homologous with bacterial and mammalian sugar transporters (5, 6, 42), such as several human glucose transporters (27). The members of this sugar transporter superfamily all have the same general structure, consisting of 12 potential membrane-spanning segments connected by hydrophilic loops of similar relative lengths (42). The structures of these transporters differ mainly in the lengths of their N termini and to a lesser extent in the lengths of their C-terminal segments. The human glucose transporters have very short N termini (12 amino acids long) (27), as do the bacterial transporters (24), while the yeast members of this family have N termini of 66 (42) to 97 (5) amino acids. All of the other yeast permeases sequenced so far (1,21,30,43,51) have similarly long N-terminal hydrophilic extensions. The orientation of these segments with respect to the plasma membrane has not been demonstrated experimentally for these yeast proteins. A number of yeast permeases carry potential N glycosylation sites, but there is as yet no clear indication as to whether any of these permeases is indeed a glycoprotein. These proteins have not yet been analyzed biochemically and no yeast permease has been purified, whereas several mammalian glucose transporters have been purified (19). Purification and extensive biochemical studies have been performed on the Escherichia coli lactose permease (11, 48), for which overexpression of active protein could be obtained (44). In contrast, several attempts at overexpression of active yeast transporters ...

In yeast, membrane proteins from the biosynthetic and endocytic pathways must be ubiquitylated for sorting to inward-budding vesicles in late endosomes, which give rise to multivesicular bodies. A conserved protein complex containing the yeast Vps23p or its mammalian counterpart Tsg101 may act as the ubiquitin receptor.

In Saccharomyces cerevisiae theFUR4-encoded uracil permease catalyzes the first step of the pyrimidine salvage pathway. The availability of uracil has a negative regulatory effect upon its own transport. Uracil causes a decrease in the level of uracil permease, partly by decreasing theFUR4 mRNA level in a promoter-independent fashion, probably by increasing its instability. Uracil entry also triggers more rapid degradation of the existing permease by promoting high efficiency of ubiquitination of the permease that signals its internalization. A direct binding of intracellular uracil to the permease is possibly involved in this feedback regulation, as the behavior of the permease is similar in mutant cells unable to convert intracellular uracil into UMP. We used cells impaired in the ubiquitination step to show that the addition of uracil produces rapid inhibition of uracil transport. This may be the first response prior to the removal of the permease from the plasma membrane. Similar down-regulation of uracil uptake, involving several processes, was observed under adverse conditions mainly corresponding to a decrease in the cellular content of ribosomes. These results suggest that uracil of exogenous or catabolic origin down-regulates the cognate permease to prevent buildup of excess intracellular uracil-derived nucleotides.

No abstract

We have screened the EUROFAN (European Functional Analysis Network) deletion strain collection for yeast mutants defective in secretory/vacuolar pathways and/or associated biochemical modifications. We used systematic Western immunoblotting to analyse the electrophoretic pattern of several markers of the secretory/vacuolar pathways, the soluble a-factor, the periplasmic glycoprotein invertase, the plasma membrane GPIanchored protein Gas1p, and two vacuolar proteins, the soluble carboxypeptidase Y and the membrane-bound alkaline phosphatase, which are targeted to the vacuole by different pathways. We also used colony immunoblotting to monitor the secretion of carboxypeptidase Y into the medium, to identify disruptants impaired in vacuolar targeting. We identified 25 mutants among the 631 deletion strains. Nine of these mutants were disrupted in genes identified in recent years on the basis of their involvement in trafficking (VPS53, VAC7, VAM6, APM3, SYS1), or glycosylation (ALG12, ALG9, OST4, ROT2). Three of these genes were identified on the basis of trafficking defects by ourselves and others within the EUROFAN project (TLG2, RCY1, MON2). The deletion of ERV29, which encodes a COPII vesicle protein, impaired carboxypeptidase Y trafficking from the endoplasmic reticulum to the Golgi apparatus. We also identified eight unknown ORFs, the deletion of which reduced Golgi glycosylation or impaired the Golgi to vacuole trafficking of carboxypeptidase Y. YJR044c, which we identified as a new VPS gene, encodes a protein with numerous homologues of unknown function in sequence databases.