Sgf1p, a new component of the Sec34p/Sec35p complex.

Here we report the identification of SGF1 as a high-copy suppressor of the sec35-1 mutant. SGF1 encodes an essential hydrophilic protein of approximately 100 kDa. Using the yeast two-hybrid system and coprecipitation studies, we demonstrate that Sgf1p is a new subunit of the multiprotein Sec34p/Sec35p complex. Reduced levels of Sgf1p lead to the accumulation of a variety of membranes as well as a kinetic block in endoplasmic reticulum to Golgi traffic. Immunofluorescence studies demonstrate that Sec34p is found throughout the Golgi, with a high concentration on early Golgi. Although an earlier study suggested that Sec34p (Grd20p) is not required for protein secretion, we show here that the sec34-2 and sec35-1 mutations lead to a pleiotropic block in the secretion of all proteins into the growth medium.

Distinction between signaling mechanisms in lipid rafts vs. caveolae.

The relative importance of lipid rafts vs. specialized rafts termed caveolae to influence signal transduction is not known. Here we show that in cells lacking caveolae, the dually acylated protein, endothelial nitric oxide synthase (eNOS), localizes to cholesterol-rich lipid raft domains of the plasma membrane. In these cells, expression of caveolin-1 (cav-1) stimulates caveolae biogenesis, promotes the interaction of cav-1 with eNOS, and the inhibition of NO release from cells. Interestingly, in cells where cav-1 does not drive caveolae assembly, despite equal levels of cav-1 and eNOS and localization of both proteins to raft domains of the plasmalemma, the physical interaction of eNOS with cav-1 is dramatically less resulting in less inhibition of NO release. Thus, cav-1 concentrated in caveolae, not in rafts, is in closer proximity to eNOS and is necessary for negative regulation of eNOS function, thereby providing the first clear example of spatial regulation of signaling in this organelle that is distinct from raft domains.

Chronic activation of AMP kinase results in NRF-1 activation and mitochondrial biogenesis.

The underlying mechanism by which skeletal muscle adapts to exercise training or chronic energy deprivation is largely unknown. To examine this question, rats were fed for 9 wk either with or without beta-guanadinopropionic acid (beta-GPA; 1% enriched diet), a creatine analog that is known to induce muscle adaptations similar to those induced by exercise training. Muscle phosphocreatine, ATP, and ATP/AMP ratios were all markedly decreased and led to the activation of AMP-activated protein kinase (AMPK) in the beta-GPA-fed rats compared with control rats. Under these conditions, nuclear respiratory factor-1 (NRF-1) binding activity, measured using a cDNA probe containing a sequence encoding for the promoter of delta-aminolevulinate (ALA) synthase, was increased by about eightfold in the muscle of beta-GPA-fed rats compared with the control group. Concomitantly, muscle ALA synthase mRNA and cytochrome c content were also increased. Mitochondrial density in both extensor digitorum longus and epitrochlearis from beta-GPA-fed rats was also increased by more than twofold compared with the control group. In conclusion, chronic phosphocreatine depletion during beta-GPA supplementation led to the activation of muscle AMPK that was associated with increased NRF-1 binding activity, increased cytochrome c content, and increased muscle mitochondrial density. Our data suggest that AMPK may play an important role in muscle adaptations to chronic energy stress and that it promotes mitochondrial biogenesis and expression of respiratory proteins through activation of NRF-1.

Aux1p/Swa2p is required for cortical endoplasmic reticulum inheritance in Saccharomyces cerevisiae.

In the yeast Saccharomyces cerevisiae, the endoplasmic reticulum (ER) is found at the periphery of the cell and around the nucleus. The segregation of ER through the mother-bud neck may occur by more than one mechanism because perinuclear, but not peripheral ER, requires microtubules for this event. To identify genes whose products are required for cortical ER inheritance, we have used a Tn3-based transposon library to mutagenize cells expressing a green fluorescent protein-tagged ER marker protein (Hmg1p). This approach has revealed that AUX1/SWA2 plays a role in ER inheritance. The COOH terminus of Aux1p/Swa2p contains a J-domain that is highly related to the J-domain of auxilin, which stimulates the uncoating of clathrin-coated vesicles. Deletion of the J-domain of Aux1p/Swa2p leads to vacuole fragmentation and membrane accumulation but does not affect the migration of peripheral ER into daughter cells. These findings suggest that Aux1p/Swa2p may be a bifunctional protein with roles in membrane traffic and cortical ER inheritance. In support of this hypothesis, we find that Aux1p/Swa2p localizes to ER membranes.

Tissue-specific overexpression of lipoprotein lipase causes tissue-specific insulin resistance.

Insulin resistance in skeletal muscle and liver may play a primary role in the development of type 2 diabetes mellitus, and the mechanism by which insulin resistance occurs may be related to alterations in fat metabolism. Transgenic mice with muscle- and liver-specific overexpression of lipoprotein lipase were studied during a 2-h hyperinsulinemic-euglycemic clamp to determine the effect of tissue-specific increase in fat on insulin action and signaling. Muscle-lipoprotein lipase mice had a 3-fold increase in muscle triglyceride content and were insulin resistant because of decreases in insulin-stimulated glucose uptake in skeletal muscle and insulin activation of insulin receptor substrate-1-associated phosphatidylinositol 3-kinase activity. In contrast, liver-lipoprotein lipase mice had a 2-fold increase in liver triglyceride content and were insulin resistant because of impaired ability of insulin to suppress endogenous glucose production associated with defects in insulin activation of insulin receptor substrate-2-associated phosphatidylinositol 3-kinase activity. These defects in insulin action and signaling were associated with increases in intracellular fatty acid-derived metabolites (i.e., diacylglycerol, fatty acyl CoA, ceramides). Our findings suggest a direct and causative relationship between the accumulation of intracellular fatty acid-derived metabolites and insulin resistance mediated via alterations in the insulin signaling pathway, independent of circulating adipocyte-derived hormones.

TRAPP I implicated in the specificity of tethering in ER-to-Golgi transport.

TRAPP is a conserved protein complex required early in the secretory pathway. Here, we report two forms of TRAPP, TRAPP I and TRAPP II, that mediate different transport events. Using chemically pure TRAPP I and COPII vesicles, we have reconstituted vesicle targeting in vitro. The binding of COPII vesicles to TRAPP I is specific, blocked by GTPgammaS, and, surprisingly, does not require other tethering factors. Our findings imply that TRAPP I is the receptor on the Golgi for COPII vesicles. Once the vesicle binds to TRAPP I, the small GTP binding protein Ypt1p is activated and other tethering factors are recruited.

Distribution and function of AP-1 clathrin adaptor complexes in polarized epithelial cells.

Expression of the epithelial cell-specific heterotetrameric adaptor complex AP-1B is required for the polarized distribution of many membrane proteins to the basolateral surface of LLC-PK1 kidney cells. AP-1B is distinguished from the ubiquitously expressed AP-1A by exchange of its single 50-kD mu subunit, mu1A, being replaced by the closely related mu1B. Here we show that this substitution is sufficient to couple basolateral plasma membrane proteins, such as a low-density lipoprotein receptor (LDLR), to the AP-1B complex and to clathrin. The interaction between LDLR and AP-1B is likely to occur in the trans-Golgi network (TGN), as was suggested by the localization of functional, epitope-tagged mu1 by immunofluorescence and immunoelectron microscopy. Tagged AP-1A and AP-1B complexes were found in the perinuclear region close to the Golgi complex and recycling endosomes, often in clathrin-coated buds and vesicles. Yet, AP-1A and AP-1B localized to different subdomains of the TGN, with only AP-1A colocalizing with furin, a membrane protein that uses AP-1 to recycle between the TGN and endosomes. We conclude that AP-1B functions by interacting with its cargo molecules and clathrin in the TGN, where it acts to sort basolateral proteins from proteins destined for the apical surface and from those selected by AP-1A for transport to endosomes and lysosomes.

A snc1 endocytosis mutant: phenotypic analysis and suppression by overproduction of dihydrosphingosine phosphate lyase.

The v-SNARE proteins Snc1p and Snc2p are required for fusion of secretory vesicles with the plasma membrane in yeast. Mutation of a methionine-based sorting signal in the cytoplasmic domain of either Sncp inhibits Sncp endocytosis and prevents recycling of Sncp to the Golgi after exocytosis. snc1-M43A mutant yeast have reduced growth and secretion rates and accumulate post-Golgi secretory vesicles and fragmented vacuoles. However, cells continue to grow and secrete for several hours after de novo Snc2-M42A synthesis is repressed. DPL1, the structural gene for dihydrosphingosine phosphate lyase, was selected as a high copy number snc1-M43A suppressor. Because DPL1 also partially suppresses the growth and secretion phenotypes of a snc deletion, we propose that enhanced degradation of dihydrosphingosine-1-phosphate allows an alternative protein to replace Sncp as the secretory vesicle v-SNARE.

Matrix proteins can generate the higher order architecture of the Golgi apparatus.

The Golgi apparatus in animal cells comprises a reticulum of linked stacks in the pericentriolar and often in the juxtanuclear regions of the cell. The unique architecture of this organelle is thought to depend on the cytoskeleton and cytoplasmic matrix proteins--the best characterized being the golgin family of fibrous, coiled-coil proteins and the GRASP family of stacking proteins. Here we show that these matrix proteins can be separated from oligosaccharide-modifying enzymes in the Golgi stack without affecting their ability to form a ribbon-like reticulum in the correct location near to the nucleus. Our data suggest that the Golgi is a structural scaffold that can exist independently of, but is normally populated by, the enzyme-containing membranes that modify transiting cargo. This new concept of the Golgi further indicates that the Golgi may be an autonomous organelle rather than one that is in simple dynamic equilibrium with the endoplasmic reticulum.

Transport through the yeast endocytic pathway occurs through morphologically distinct compartments and requires an active secretory pathway and Sec18p/N-ethylmaleimide-sensitive fusion protein.

Molecules travel through the yeast endocytic pathway from the cell surface to the lysosome-like vacuole by passing through two sequential intermediates. Immunofluorescent detection of an endocytosed pheromone receptor was used to morphologically identify these intermediates, the early and late endosomes. The early endosome is a peripheral organelle that is heterogeneous in appearance, whereas the late endosome is a large perivacuolar compartment that corresponds to the prevacuolar compartment previously shown to be an endocytic intermediate. We demonstrate that inhibiting transport through the early secretory pathway in sec mutants quickly impedes transport from the early endosome. Treatment of sensitive cells with brefeldin A also blocks transport from this compartment. We provide evidence that Sec18p/N-ethylmaleimide-sensitive fusion protein, a protein required for membrane fusion, is directly required in vivo for forward transport early in the endocytic pathway. Inhibiting protein synthesis does not affect transport from the early endosome but causes endocytosed proteins to accumulate in the late endosome. As newly synthesized proteins and the late steps of secretion are not required for early to late endosome transport, but endoplasmic reticulum through Golgi traffic is, we propose that efficient forward transport in the early endocytic pathway requires delivery of lipid from secretory organelles to endosomes.

In vitro import of proteins into mitochondria of Trypanosoma brucei and Leishmania tarentolae.

In eukaryotic evolution, the earliest branch of organisms to have mitochondria are the trypanosomatids. Their mitochondrial biogenesis not only includes import of most proteins, but also, unlike in other organisms, import of the whole set of tRNAs. In order to investigate these processes, we devised novel procedures for the isolation of mitochondria from two trypanosomatid species: Trypanosoma brucei and Leishmania tarentolae. Isotonic cell lysis followed by equilibrium density centrifugation in Nycodenz gradients yielded mitochondrial fractions exhibiting a membrane potential. Furthermore, we have used these fractions to reconstitute import of mitochondrial matrix proteins in vitro. Energy-dependent uptake of an artificial precursor protein, containing a trypanosomal presequence attached to mouse dihydrofolate reductase and of yeast mitochondrial alcohol dehydrogenase could be demonstrated. The presequences of both proteins were processed in T. brucei whereas only the trypanosomal one was cleaved in L. tarentolae. Trypsin pretreatment abolished the ability of the mitochondria to import proteins, indicating the involvement of proteinaceous components at the surface of mitochondria.

The Golgi stack reassembles during telophase before arrival of proteins transported from the endoplasmic reticulum.

HeLa cells arrested in prometaphase were pulse-labeled with [35S]methionine and chased in the absence of nocodazole to allow passage through mitosis and into G1. Transport of histocompatibility antigen (HLA) molecules to the medial- and trans-Golgi cisternae was measured by monitoring the resistance to endoglycosidase H and the acquisition of sialic acid residues, respectively. Transport to the plasma membrane was measured using neuraminidase to remove sialic acid residues on surface HLA molecules. The half-time for transport to each of these compartments was about 65-min longer in cells progressing out of mitosis than in G1 cells. This delay was only 5-min longer than the half-time for the fall in histone H1 kinase activity suggesting that inactivation of the mitotic kinase triggers the resumption of protein transport. The half-time for reassembly of the Golgi stack, measured using stereological procedures, was also 65 min, suggesting that both transport and reassembly are triggered at the same time. However, since reassembly was complete within 5 min, whereas HLA took 25 min to reach the medial-cisterna, we can conclude that the Golgi stack has reassembled by the time HLA reaches it.

Mitotic Golgi clusters are not tubular endosomes.

HeLa cells were incubated with 15 nm BSA-gold for 1 or 2 hours to mark the endocytic pathway and mitotic cells were then isolated by shake-off. Thin, frozen sections were labelled with antibodies against two resident Golgi markers, beta-(1,4)-galactosyltransferase and N-acetylglucosaminyltransferase I. Detection of the latter was aided by the use of a HeLa cell line stably expressing a myc-tagged version of the endogenous protein. The secondary antibodies were coupled to either 5 or 10 nm gold so that the distribution of each of the three markers could be followed. Qualitative and quantitative studies showed that there were two populations of clusters, those described by us earlier and termed Golgi clusters (Lucocq et al. (1987) J. Cell Biol. 104, 865-874), containing either or both Golgi markers, and clusters of tubular endosomes containing BSA-gold. There was very little overlap showing that Golgi clusters cannot be tubular endosomes as concluded by Tooze and Hollinshead (1992) Eur. J. Cell Biol. 58, 228-242.

Overlapping distribution of two glycosyltransferases in the Golgi apparatus of HeLa cells.

Thin, frozen sections of a HeLa cell line were double labeled with specific antibodies to localize the trans-Golgi enzyme, beta 1,4 galactosyltransferase (GalT) and the medial enzyme, N-acetylglucosaminyltransferase I (NAGT I). The latter was detected by generating a HeLa cell line stably expressing a myc-tagged version of the endogenous protein. GalT was found in the trans-cisterna and trans-Golgi network but, contrary to expectation, NAGT I was found both in the medial- and trans-cisternae, overlapping the distribution of GalT. About one third of the NAGT I and half of the GalT were found in the shared, trans-cisterna. These data show that the differences between cisternae are determined not by different sets of enzymes but by different mixtures.

Mitotic cytosol inhibits invagination of coated pits in broken mitotic cells.

Receptor-mediated endocytosis is inhibited during mitosis in mammalian cells and earlier work on A431 cells suggested that one of the sites inhibited was the invagination of coated pits (Pypaert, M., J. M. Lucocq, and G. Warren. 1987. Eur. J. Cell Biol. 45: 23-29). To explore this inhibition further, we have reproduced it in broken HeLa cells. Mitotic or interphase cells were broken by freeze-thawing in liquid nitrogen and warmed in the presence of mitotic or interphase cytosol. Using a morphological assay, we found invagination to be inhibited only when mitotic cells were incubated in mitotic cytosol. This inhibition was reversed by diluting the cytosol during the incubation. Reversal was sensitive to okadaic acid, a potent phosphatase inhibitor, showing that phosphorylation was involved in the inhibition of invagination. This was confirmed using purified cdc2 kinase which alone could partially substitute for mitotic cytosol.

Formation of coated vesicles from coated pits in broken A431 cells.

Biochemical and morphological techniques were used to demonstrate the early steps in the endocytosis of transferrin in broken A431 cells. After binding 125I-transferrin, the cells were broken by scraping and then warmed. 125I-transferrin became inaccessible to exogenous anti-transferrin antibody providing a measure of the internalization process. Parallel morphological experiments using transferrin coupled to horseradish peroxidase confirmed internalization in broken cells. The process was characterized and compared with endocytosis in intact cells and showed many similar features. The system was used to show that both the appearance of new coated pits and the scission of coated pits to form coated vesicles were dependent on the addition of cytosol and ATP whereas invagination of pits was dependent on neither.

Coated pits in interphase and mitotic A431 cells.

Endocytosis is inhibited during mitosis in A431 cells (Warren et al., 1984) but the site of inhibition is unknown. A quantitative method measuring the extent of budding was used to compare coated pits in interphase and mitotic cells. Every stage of budding found in interphase cells was also found in cells at every stage of mitosis. Flatter coated pits appeared more frequent in mitotic cells but this can be partly, if not entirely, explained by their greater size. We conclude that, if budding is inhibited, inhibition must occur at all stages of the budding process.