Intracellular Trafficking Pathway of Yeast Long-chain Base Kinase Lcb4, from Its Synthesis to Its Degradation*

Sphingoid long-chain base 1-phosphates act as bioactive lipid molecules in eukaryotic cells. In budding yeast, long-chain base 1-phosphates are synthesized mainly by the long-chain base kinase Lcb4. We recently reported that, soon after yeast cells enter into the stationary phase, Lcb4 is rapidly degraded by being delivered to the vacuole in a palmitoylation- and phosphorylation-dependent manner. In this study, we investigated the complete trafficking pathway of Lcb4, from its synthesis to its degradation. After membrane anchoring by palmitoylation at the Golgi apparatus, Lcb4 is delivered to the plasma membrane (PM) through the late Sec pathway and then to the endoplasmic reticulum (ER). The yeast ER consists of a cortical network juxtaposed to the PM (cortical ER) with tubular connections to the nuclear envelope (nuclear ER). Remarkably, the localization of Lcb4 is restricted to the cortical ER. As the cells reach the stationary phase, G1 cell cycle arrest initiates Lcb4 degradation and its delivery to the vacuole via the Golgi apparatus. The protein transport pathway from the PM to the ER found in this study has not been previously reported. We speculate that this novel pathway is mediated by the PM-ER contact.

Sphingoid long-chain base 1-phosphates act as bioactive lipid molecules in eukaryotic cells. In budding yeast, long-chain base 1-phosphates are synthesized mainly by the long-chain base kinase Lcb4. We recently reported that, soon after yeast cells enter into the stationary phase, Lcb4 is rapidly degraded by being delivered to the vacuole in a palmitoylation-and phosphorylation-dependent manner. In this study, we investigated the complete trafficking pathway of Lcb4, from its synthesis to its degradation. After membrane anchoring by palmitoylation at the Golgi apparatus, Lcb4 is delivered to the plasma membrane (PM) through the late Sec pathway and then to the endoplasmic reticulum (ER). The yeast ER consists of a cortical network juxtaposed to the PM (cortical ER) with tubular connections to the nuclear envelope (nuclear ER). Remarkably, the localization of Lcb4 is restricted to the cortical ER. As the cells reach the stationary phase, G 1 cell cycle arrest initiates Lcb4 degradation and its delivery to the vacuole via the Golgi apparatus. The protein transport pathway from the PM to the ER found in this study has not been previously reported. We speculate that this novel pathway is mediated by the PM-ER contact.
Sphingolipids are major constituents of eukaryotic plasma membranes. Ceramide, the structural backbone of sphingolipids, is generated from long-chain base (LCB) 3 and an amidelinked fatty acid. In addition, phosphorylation of the major mammalian LCB, sphingosine, by sphingosine kinase yields sphingosine 1-phosphate (S1P). By binding to receptors in the S1P/Edg family of G-protein-coupled receptors, S1P regulates various cellular processes including proliferation, differentia-tion, motility, and adherens junction assembly (1)(2)(3)(4). In addition to its extracellular effects, S1P is thought to function as a second messenger of various stimuli, such as growth factors and cytokines, and to elicit cellular responses including Ca 2ϩ mobilization and apoptosis inhibition (1,2,4).
The yeast Saccharomyces cerevisiae does not contain S1P but instead has two other long-chain base 1-phosphates (LCBPs), dihydrosphingosine 1-phosphate and phytosphingosine 1-phosphate. The synthesis and metabolic pathways of LCBPs are highly conserved between mammalian and yeast cells, and yeast LCBPs are produced by LCB kinases that correspond to mammalian sphingosine kinases. Yeast have two LCB kinases, Lcb4 and Lcb5, although Lcb4 is responsible for most cellular activity. Yeast cells, however, do not have cell surface receptors for LCBPs, so these compounds function only intracellularly. Similar to the intracellular functions of S1P in mammals, LCBPs in yeast are involved in diverse cellular processes including heat stress resistance and Ca 2ϩ mobilization (5)(6)(7)(8).
Recently, we found that the stability and localization of Lcb4 were regulated by its post-translational modifications, including palmitoylation, phosphorylation, and ubiquitination (9,10). Palmitoylation of Lcb4 is catalyzed by the palmitoyltransferase Akr1, whereas Pho85 cyclin-dependent kinase is responsible for its phosphorylation. Interestingly, the phosphorylation of Lcb4 is greatly reduced in ⌬akr1 cells (10), suggesting that the palmitoylation is a prerequisite for the phosphorylation. Palmitoylation is also involved in the membrane anchoring of Lcb4, which otherwise would be a soluble protein. Although Lcb4 is relatively stable in the log phase of cell growth, it is destabilized upon the cell reaching the stationary phase. Both palmitoylation and phosphorylation are required during the stationary phase for the degradation of Lcb4, which is delivered to the vacuole via the multivesicular body (MVB), then degraded (9,10).
In a previous report, we demonstrated that in log phase, Lcb4 is localized primarily in the endoplasmic reticulum (ER), although it is also found in the plasma membrane (PM) (11). However, it was unclear how Lcb4 reaches the ER or PM. In the present study, we investigated the complete trafficking pathway of Lcb4, from its synthesis to its degradation. We found that localization of Lcb4 in the ER was unusually restricted to the cortical ER. In addition, our results strongly support the exist-ence of a novel PM-to-ER transport pathway. Transport of Lcb4 via the PM-ER contact site is an intriguing possibility that would also explain its unique, cortical ER-specific localization.

EXPERIMENTAL PROCEDURES
Yeast Strains and Media-S. cerevisiae strains used in this study are listed in Table 1. For construction of the ⌬pep12::LEU2 cells, the 0.56-kb BglII-KpnI region of the PEP12 gene was replaced with the LEU2 marker. TS335 cells were prepared by backcrossing HMSF1 (sec1-1) cells with cells of the SEY6210 background strain four times.
The plasmid pUG34, a yeast expression vector encoding a fusion protein with an N-terminal enhanced green fluorescent protein (EGFP) under the control of the MET15 promoter, was a gift from Dr. J. H. Hegemann (Heinrich-Heine University, Düsseldorf, Germany). The pWK63 (MET15 promoter, EGFP-LCB4, CEN, HIS3 marker) plasmid was described previously (9).
Microscopic EGFP fluorescence was performed as described previously (9). Cells were analyzed by fluorescence microscopy using an Axioskop 2 PLUS microscope (Carl Zeiss, Oberkochen, Germany).
Pulse-Chase and Immunoprecipitation-Pulse-Chase experiments and subsequent immunoprecipitation were performed  (14), and Lcb4 was then immunoprecipitated using anti-Lcb4 antiserum (9) and protein A-Sepharose beads (GE Healthcare). Samples were separated by SDS-PAGE, and the gels were fixed, treated with the Amplify fluorographic reagent TM (GE Healthcare), dried, and exposed to x-ray film at Ϫ80°C.

RESULTS
Lcb4 Is Localized in the Cortical ER-We first examined the localization of Lcb4 by indirect immunofluorescence microscopy using anti-Lcb4 antibodies. A weak signal was detected near the cell surface in wild-type cells but was not observed in ⌬lcb4 cells (Fig. 1A). In cells in which Lcb4 was produced from a 2 plasmid, cell surface staining became more prominent (Fig. 1A), agreeing with our previous result (10).
We also investigated the localization of Lcb4 by sucrose density gradient. As reported previously (11), endogenous Lcb4 was localized mainly in the ER fractions but partly in the PM fractions (Fig. 1B). When Lcb4 was expressed from a CEN plasmid, its subcellular localization was almost the same as that observed in wild-type cells (Fig. 1C).
In fluorescence microscopy, yeast ERs typically appear as two ring structures, the nuclear ER and the cortical ER, as demonstrated by the staining patterns of the ER protein Kar2 (Fig. 1A). However, the cortical ER is closely adjacent to the PM, identified by its marker Pma1 (Fig. 1A), so immunofluorescence cannot distinguish between the two. Sucrose density gradient separation revealed that most of the endogenous Lcb4 was localized in the ER (Fig. 1B), although no nuclear ER staining was observed (Fig. 1A). These results suggest that the ER localization of Lcb4 is restricted to the cortical ER and that the cell surface staining observed corresponds to both the PM and the cortical ER.
To confirm the cortical ER localization of Lcb4, we investigated endogenous and overproduced Lcb4 by immunoelectron microscopy. As presented in Fig. 2A, the ER structures were distinguishable as characteristic transparent structures (19) that included the cortical ER (closed arrows) and the nuclear ER (open arrows). Overproduced Lcb4 was indeed detected in the cortical ER (closed arrowheads) but not in the nuclear ER (Fig.  2B). Plasma membrane localization was also observed in some cells (data not shown). Similarly, endogenous Lcb4 was observed in the cortical ER ( Fig. 2C) but not in the nuclear ER.
These results indicate that Lcb4 is indeed localized specifically in the cortical ER.
Lcb4 Is Transported to the PM via the Golgi Apparatus-Recently, we demonstrated that Lcb4 is anchored to the membrane through its palmitoylation by the palmitoyltransferase Akr1 (10). Considering that Akr1 is predominantly localized in the Golgi apparatus (20), we speculated that Lcb4 is transported to the PM/ER after being palmitoylated at the Golgi apparatus. To test this theory, we prepared an EGFP-Lcb4 construct under the control of the inducible MET15 promoter, which enabled us to monitor only the newly synthesized EGFP-Lcb4. We introduced this construct into wild-type or sec1-1 temperaturesensitive mutants, in which the late Sec pathway involved in vesicular transport from the Golgi apparatus to the PM is abolished at restrictive temperatures (37°C). In wild-type cells grown at either 23 or 37°C, EGFP-Lcb4 was observed at the cell periphery. In contrast, in the sec1-1 mutants, EGFP-Lcb4 was distributed in punctate structures characteristic of the Golgi apparatus, at the restrictive temperature (37°C) but not at the permissive temperature (23°C) (Fig. 3A). This result indicates that Golgi-to-PM transport via the late Sec pathway is essential not only for the PM localization but also for the cortical ER localization of Lcb4. Thus, it is highly likely that Lcb4 is transported to the ER after reaching the PM.
To determine the site of Lcb4 phosphorylation, we next examined the phosphorylation in sec1-1 mutants over time using a pulse-chase experiment. Lcb4 was gradually converted to its phosphorylated form in wild-type cells at either temperature and in the sec1-1 mutants at the permissive temperature (Fig. 3B). On the other hand, almost no phosphorylation was observed in sec1-1 mutants at the restrictive temperature. These results indicate that Lcb4 is phosphorylated after reaching the PM.
Lcb4 Is Transported from the ER to the Vacuole via the Golgi/ MVB in the Stationary Phase-Previously, we demonstrated that upon reaching the stationary phase, Lcb4 is rapidly delivered via the MVB to the vacuole for degradation (9). However, it is unclear how the Lcb4, localized in the ER during the log phase, is transported to the MVB/vacuole. One possibility is that Lcb4 is transported from the ER to the Golgi apparatus via vesicular transport (early Sec pathway) and then to the MVB/ vacuole, or perhaps, Lcb4 returns to the PM and is transported to the MVB/vacuole via endocytosis. To distinguish between these possibilities, we investigated the degradation of Lcb4 in early sec and endocytosis mutants. The temperature-sensitive mutants sec12-1 and sec23-1 carry mutations in the component of the coat protein complex II (COPII) complex that is involved in the transport from the ER to the Golgi apparatus. Cells were cultured until just before entering the stationary phase, and then they were shifted to the restrictive temperature. Although it was degraded in a time-dependent manner in wild-type cells, Lcb4 was not degraded in either the sec12-1 or the sec23-1 mutants (Fig. 4A), indicating that Lcb4 is indeed transported from the ER to the Golgi via the early Sec pathway. On the other hand, mutations affecting endocytosis (⌬sla1, ⌬chc1, and ⌬clc1) did not affect the stability of Lcb4, confirming that it is not transported to the vacuole via the endocytosis pathway (Fig. 4B). These results also support the existence of a PM-to-ER transport pathway since the PM-localized Lcb4 must be delivered to the ER prior to its degradation.
Next, we examined the stability of Lcb4 using deletion mutants with defective post-Golgi transport. As shown in Fig. 4C, the stability of Lcb4 during the stationary phase was increased in mutants carrying deletions (⌬vps45 and ⌬pep12) that affect the soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNARE) complex, which is involved in transport between the Golgi apparatus and the MVB. We also examined the localization of EGFP-Lcb4 in log and stationary phases in these mutants. In wild-type cells, EGFP-Lcb4 was observed near the cell surface in the log phase, whereas vacuolar staining was evident in the stationary phase, reflecting the accumulation of the stable EGFP moiety (Fig. 4D). On the other hand, EGFP-Lcb4 was observed in intracellular organelles in both ⌬vps45 and ⌬pep12 cells at the stationary phase (Fig. 4D). These results indicate that Lcb4 is transported from the ER to the vacuole via the Golgi/MVB at the stationary phase.
G 1 Cell Cycle Arrest Accelerates the Degradation of Lcb4-Yeast cells in glucose-containing medium proliferate by fermenting glucose, but the exhaustion of glucose causes a conversion in metabolism from fermentation to respiration, which utilizes ethanol. This conversion is called the diauxic shift. After yeast cells reach the diauxic shift, they undergo G 1 cell cycle arrest and enter into the stationary phase. To test whether the degradation of Lcb4 at the stationary phase is related to the G 1 cell cycle arrest, we examined the stability of Lcb4 in cultures treated with drugs that synchronize the cells at the G 1 phase. Rapamycin inhibits Tor kinase and arrests the cell cycle at the G 1 phase (21,22). Treatment with 200 ng/ml rapamycin resulted in the stimulation of Lcb4 degradation (Fig. 5A). The mating pheromone ␣-factor similarly arrests the cell cycle in the late G 1 phase. Since exogenous ␣-factor is degraded rapidly by the secreted protease Bar1 (23), we used ⌬bar1 cells. Treatment with 6 g/ml ␣-factor resulted in enhanced Lcb4 degradation (Fig. 5B). In contrast, treatment with 15 g/ml nocodazole, which inhibits the polymerization of tubulin molecules and arrests cells at the M phase, or with 100 mM hydroxyurea, which inhibits ribonucleotide reductase and arrests cells at the S phase, did not affect the stability of Lcb4 (Fig. 5A). These results indicate that G 1 cell cycle arrest causes stimulation of Lcb4 degradation.

DISCUSSION
There have been discrepancies among studies as to the intracellular localization of Lcb4. Using indirect immunofluorescence microscopy, we detected non-tagged Lcb4 near the cell surface, regardless of its expression level (Fig. 1A) (10). However, two other groups obtained different results using C-terminally HA-tagged Lcb4 (Lcb4-HA) expressed at the endogenous level. One group detected Lcb4-HA as a punctate pattern and diffuse patches, which overlapped with the Golgi and endosome markers (24). The other group observed punctate staining as well but also a ring-like staining, which co-localized with an ER marker (25). We considered that these differences might be caused by the C-terminal tag. In fact, we observed punctate structures when examining the localization of overproduced Lcb4-HA (data not shown), confirming that the C-terminal tag does indeed affect the intracellular localization of Lcb4. In contrast, the observed ER localization is quite reasonable since the substrates of Lcb4, LCBs, are synthesized in the ER. PM localization has also been expected since exogenously added LCBs are converted in the PM to LCBPs to be utilized as sphingolipid precursors (26 -28).
The ER of the budding yeast S. cerevisiae appears as two simple rings, the cortical ER juxtaposed with the PM and the nuclear ER equivalent to the nuclear envelope. The cortical ER is a network of highly dynamic tubules that undergo ring closure and tubule-branching movements, similar to the mammalian peripheral ER (29). The cortical-specific ER localization we observed for Lcb4 is unusual (Figs. 1 and 2); however, similar localization has been reported for other ER proteins such as Rtn1 and Rtn2 (30).
We also identified the complete intracellular trafficking pathways of Lcb4, from synthesis to degradation, as illustrated in Fig. 6. Since Lcb4 has no signal sequence or transmembrane domain, it must be synthesized in the cytosol and anchored to the Golgi apparatus through palmitoylation by Akr1. Lcb4 is then delivered to the PM via the late Sec pathway and then to FIGURE 3. Lcb4 is transported from the Golgi apparatus to the PM via the late Sec pathway. A, SEY6210 (wild-type) cells and the late Sec pathwaydefective TS335 (sec1-1) cells harboring pUG34 (EGFP-vector) or pWK63 (EGFP-LCB4) were grown to log phase at 23°C in SC medium lacking histidine. To induce the expression of EGFP-Lcb4, the culture medium was replaced with SC medium lacking both histidine and methionine, and the cells were immediately transferred to 37°C or kept at 23°C for 2 h. Cells were fixed with formaldehyde and observed under a fluorescence microscope. B, SEY6210 (wild-type) and TS335 (sec1-1) cells, grown at 23°C to log phase, were pulselabeled with [ 35 S]methionine/cysteine for 15 min at 23 or 37°C and then chased with excess unlabeled methionine (final concentration of 0.5 mg/ml) and cysteine (final concentration of 0.1 mg/ml) for 0, 1, 2, and 3 h at their respective temperatures. At each time point, lysates were prepared and subjected to immunoprecipitation using anti-Lcb4 antibodies. Immunoprecipitated proteins were separated by SDS-PAGE and detected by autoradiography. p-Lcb4, phosphorylated Lcb4. were harvested and lysed as indicated. Total cell lysates (10 g of protein) were separated by SDS-PAGE followed by immunoblotting with anti-Lcb4 antibodies. p-Lcb4, phosphorylated Lcb4. A, RSY255 (wild-type) cells and the early Sec pathway-deficient RSY263 (sec12-4) and RSY281 (sec23-1) cells were grown at 23°C until just prior to entering the stationary phase. At time 0, the cultures were shifted to 37°C, and an aliquot of cells was collected at each indicated time point. B, BY4741 (wild-type) cells and cells with deletions in genes affecting endocytosis, including 3033 (⌬sla1), 4572 (⌬chc1), and 4797 (⌬clc1) cells, were grown at 30°C and then harvested in early logphase (E) or 16 h after reaching stationary phase (S). C, BY4741 (wild-type) cells and cells with deletions in genes affecting the late Golgi-to-MVB pathway, including SIY259 (⌬pep12) and 4462 (⌬vps45) cells, were grown at 30°C and harvested in early log-phase or 16 h after reaching stationary phase. D, BY4741 (wild-type) cells harboring pUG34 (EGFP-vector), and BY4741 (wild-type), 4462 (⌬vps45), and SIY259 (⌬pep12) cells, each harboring pWK63 (EGFP-LCB4) were grown to log phase at 30°C in SC medium lacking histidine. The expression of EGFP and EGFP-Lcb4 was then induced by changing the medium to SC medium lacking both histidine and methionine. The subcellular localization of EGFP-Lcb4 was examined under a fluorescence microscope at 3 h after induction (log phase) or 24 h after induction (stationary phase).
the ER. Lcb4 is degraded slowly during the log phase, but its degradation is accelerated at the stationary phase. For degradation, Lcb4 is delivered from the ER to the Golgi via the early Sec pathway and then finally to the vacuole through the MVB.
Transport from the PM to the ER has been reported for lipid molecules such as sterols and fatty acids (31,32), although the molecular mechanism is mostly unknown. Recently, however, a role for the ATP binding cassette transporters Aus1 and Pdr11 in ergosterol transport from the PM to the ER was reported (33). In contrast, PM-to-ER transport has not been established for proteins. At present, the mechanism responsible for transporting Lcb4 from the PM to the ER is unclear. However, since recent studies revealed that the ER is in contact with various organelles, such as the mitochondria, the PM, the trans-Golgi network, the endosome, and the vacuole (34 -37), we speculate that the PM-ER contact may be responsible for the Lcb4 transport. Yeast cortical ER has been shown to be associated with the PM at a site called the PM-associated membrane (PAM) (36,38,39). The unique, cortical ER-specific localization of Lcb4 can be nicely explained by its hypothesized transport via the PAM. Since the PM contacts the cortical ER only at the PAM, Lcb4 is supplied to the cortical ER but not to the nuclear ER in this model. The cortical ER is connected with the nuclear ER via the tubular structure (29), so diffusion of Lcb4 to the nuclear ER via the tubules may be possible. However, considering the much larger area of the cortical ER, as compared with that of the tubules, it is highly likely that diffusion within the cortical ER would surpass that to the nuclear ER.
Reportedly, the PAM is enriched in several lipid biosynthetic enzymes (31,40), implying that it contains a characteristic lipid composition (40). We recently reported that the transport of Lcb4 to the ER is influenced by the lipid environment (11). The localization of Lcb4 in the PM is greatly reduced in ergosterol biosynthetic mutants (11), suggesting that the PM-to-ER transport is accelerated in these mutants.
Until now, the components of the contact sites between the ER and other organelles have remained almost unknown. However, recent studies in yeast have revealed that the contact site between the nuclei and the vacuole, which is called a nucleus-vacuole junction, is formed by the interaction of the nuclear ER protein Nvj1 with the vacuolar protein Vac8 (31,(35)(36)(37). This suggests that a protein-protein interaction is responsible for the formation of contact sites between the ER and other organelles.
Disruption of the ICE2 gene, which encodes a novel ER membrane protein, reportedly results in a defect in the delivery of ER into daughter cells (41). In addition, the cortical ER of ⌬ice2 cells is less continuous in the mother cell as compared with that in wild-type cells (41). However, our sucrose density gradient fractionation studies demonstrated the subcellular localization of Lcb4 in ⌬ice2 cells to be almost the same as in wild-type cells. 4 Furthermore, deletion of the ICE2 gene had no apparent effect on the LCB kinase activity. 4 Thus, in ⌬ice2 cells, Lcb4 is delivered to the ER. The cortical ER does still exist in ⌬ice2 cells (41), and this may account for the ineffectiveness of the ICE2 deletion on Lcb4 localization and activity. 4 S. Iwaki, A. Kihara, and Y. Igarashi, unpublished data. FIGURE 5. G 1 cell cycle arrest triggers degradation of Lcb4. A and B, cells were grown at 30°C in YPD medium to a density of 2.0 A 600 and then treated accordingly. At the indicated times, an aliquot of cells (2 ϫ 10 7 ) was collected, and the cells were lysed. Total cell lysates (10 g of protein) were separated by SDS-PAGE followed by immunoblotting with anti-Lcb4 antibodies. A, BY4741 (wild-type) cells treated with dimethyl sulfoxide (control) or a cell cycle inhibitor, rapamycin (200 ng/ml; G 1 phase), nocodazole (15 g/ml; M phase) or hydroxyurea (100 mM; S phase). B, 1408 (⌬bar1) cells treated with dimethyl sulfoxide (control) or ␣-factor (6 g/ml), a G 1 cell cycle inhibitor. FIGURE 6. A model for the transport of Lcb4 in log phase and stationary phase. Lcb4 is synthesized in the cytosol, palmitoylated at the Golgi apparatus (pathway 1), and then transported to the PM via the late Sec pathway (pathway 2). Most of the Lcb4 is delivered further to the cortical ER (cER, pathway 3), which is closely adjacent to the PM. Upon reaching the stationary phase, the ER-localized Lcb4 is transported to the Golgi apparatus via the early Sec pathway (pathway 4), and finally, to the vacuole, via the MVB pathway (pathway 5 and 6). nER, nuclear ER.
In the study presented here, G 1 arrest by inhibitors of cell cycle progression (rapamycin and ␣-factor) promoted the degradation of Lcb4 (Fig. 5). The exact molecular mechanism by which Lcb4 was degraded, however, remains unclear. We speculate that modification of Lcb4 is involved. We previously reported (9) that Lcb4 is phosphorylated by the cyclin-dependent kinase Pho85 complexed with the G 1 cyclin Pcl1/Pcl2, which are activated at the G 1 phase (42,43). This phosphorylation is indeed required, but is not sufficient, for the degradation of Lcb4. We also demonstrated that Lcb4 is ubiquitinated and that the phosphorylation functions to increase the efficiency of the ubiquitination (9). It is possible that the ubiquitin ligase acting on Lcb4 is activated under G 1 cell cycle-arrested conditions.
The down-regulation of Lcb4 in the G 1 phase is considered to be important in regulating the cellular levels of LCBs and LCBPs. Interestingly, LCBs are required for transient cell cycle arrest induced by heat shock, and cell cycle recovery from this transient arrest is repressed in ⌬lcb4 ⌬lcb5 cells (44). Thus, it is possible that LCBPs are involved in G 1 -to-S phase progression. A similar function has also been proposed for the mammalian LCBP, S1P. Increasing intracellular S1P levels following the overexpression of SPHK1, a homologue of Lcb4, result in a decreased number of cells in G 1 /G 0 -phase and an increase in those in S-phase, under normal and low serum conditions (45). Therefore, down-regulation of Lcb4 in G 1 phase may be important for maintenance of G 1 cell cycle arrest.
Our results suggest the existence of a novel trafficking pathway from the PM to the ER. Recently, it was reported that certain proteins, such as the GTPase Ras2 and the eight-transmembrane protein Ist2, traffic from the ER to the PM via a Sec-independent pathway (46,47). Although the molecular mechanisms behind the transport of these proteins are unknown, the transport of Ras2 and Ist2 via the PAM is an intriguing possibility, as in the case of Lcb4. To test this possibility, identification of proteins involved in the formation of the PAM will be necessary.