The Golgi complex: in vitro veritas?

Modulation of the Golgi apparatus in Saccharomyces cerevisiae sec7 mutants as seen by three-dimensional electron microscopy

The three-dimensional configuration of the Golgi apparatus has been examined with the electron microscope in thick Golgi sections of Saccharomyces cerevisiae prepared from a wild-type strain and from sec7 mutants maintained for various periods of time at the nonpermissive temperature of 37 degrees C and then returned to the permissive temperature of 24 degrees C. Reduced osmium postfixation of glutaraldehyde fixed specimens stained intensely the content of Golgi elements and thus facilitated their three-dimensional characterization. In wild-type S. cerevisiae, the Golgi elements usually appeared as isolated networks of membranous tubules dispersed throughout the cytoplasm. Along such networks, distensions filled with stained material were similar in size to nearby secretory granules, suggesting that the latter formed by fragmentation of the Golgi elements. In sec7 mutants maintained at 37 degrees C in low (0.1%) glucose medium, secretion granules progressively decreased in number and soon disappeared. Concomitantly the networks of Golgi tubules increased in size and complexity, lost their distensions, and then transformed into flattened saccules forming stacks of up to seven or eight saccules that were similar to the Golgi stacks seen in mammalian cells. However in contrast to the latter, connections between the saccules were evident and Golgi-associated small vesicles were generally absent. Following return to the permissive temperature (24 degrees C), secretion granules reappeared, the Golgi stacks progressively decreased in size, and resumed their initial state of separated small tubular networks. Thus in sec7 mutant, grown at 37 degrees C in low glucose medium, segregation of secretory granules is blocked. As a result, Golgi membranes accumulate to form a continuous system of stacked and interconnected saccules.

Molecular cloning and subcellular localization of three GTP-binding proteins of the rab subfamily

Small GTPases of the rab subfamily are involved in regulation of intracellular membrane transport events. We recently used a PCR approach to isolate short cDNA fragments of a number of novel rab sequences. These PCR fragments have not been used with cDNA library screening and PCR-based techniques to clone the cDNAs encoding three of these proteins, rab12, rab22, and rab24. By northern blot analysis, the messages were found to be present in a wide variety of mouse tissues. However, quantitative differences in the mRNA levels between the tissues were detected. We determined the subcellular localization of the GTPases by expressing the c-myc epitope-tagged proteins with the Semliki Forest virus and the vaccinia T7 vector systems. Transiently expressed rab12 was localized to the Golgi complex. This localization was confirmed using a polyclonal anti-peptide antibody detecting the endogenous protein in BHK cells. rab22 expressed from the cDNA was localized to endosomal compartments and to the plasma membrane. After longer periods of expression, the protein was found on abnormally large perinuclear endosomal structures, suggesting that it is a potent regulator of events in the endocytic pathway. Finally, rab24 was found in the endoplasmic reticulum/cis-Golgi region and on late endosomal structures. The localization of rab24 may indicate its involvement in autophagy-related processes.

Regulation of pathways within cultured epithelial cells for the transcytosis of a basal membrane-bound peroxidase-polylysine conjugate

A conjugate of horseradish peroxidase (HRP) to poly(L-lysine) (PLL) was used as a non-specific adsorptive probe to study transcytosis in MDCK strain I and Caco-2 epithelial cells. As we have shown previously, HRP-PLL transcytosis proceeds via an intracellular, non-lysosomal proteolytic compartment in MDCK cells; yet, this compartment is utilized for transcytosis only in the basal-to-apical direction (Taub, M. E. and Shen, W.-C. J. Cell. Physiol., 150, 283–290, 1992). Using size exclusion chromatography, we demonstrate that the PLL moiety of the conjugate is effectively cleaved during transcytosis, thus releasing free HRP from the apical surface of the cells. Pulse-chase studies indicate that approximately 6% of basolateral surface-associated HRP-PLL conjugate in Transwell-grown cell monolayers enters the basal-to-apical transcytotic pathway. Brief (1 hour) treatment with 160 nM phorbol ester (PMA), a protein kinase C stimulator, elicits a 2-fold increase in the rate and amount of HRP-PLL transcytosis following a 1 hour lag time. Treatment with 1.6 micrograms/ml brefeldin A (BFA) inhibits HRP-PLL transcytosis by approximately 30%; additionally, BFA is able to abolish completely the PMA stimulatory effect. Removal of BFA causes a re-establishment of the normal rate of transcytosis within 2 hours, demonstrating the reversibility of BFA inhibition with respect to HRP-PLL transcytosis. Thus, PMA most likely elicits an increase in the amount of basally internalized conjugate delivered to BFA-sensitive transcytotic compartments.(ABSTRACT TRUNCATED AT 250 WORDS)

Dynamic retention of TGN membrane proteins in Saccharomyces cerevisiae

In a secretory pathway organelle like the Golgi complex, resident proteins are retained in the face of substantial protein flux to subsequent destinations. Recently, molecular genetic strategies have been used to study membrane protein retention in a compartment of the yeast Saccharomyces cerevisiae that is analogous to the trans Golgi network (TGN) of mammalian cells. These studies have defined retention signals containing aromatic amino acids in the TGN proteins' cytoplasmic domains. The identification of mutants that fail to retain TGN proteins has offered the first glimpse into the components involved in retention. The phenotypes of these mutants suggest that retention involves retrieval of TGN proteins from an endosomal compartment.

The small GTP-binding protein rab6p is redistributed in the cytosol by brefeldin A

Rab6 protein belongs to the Sec4/Ypt/rab subfamily of small GTP-binding proteins involved in intracellular membrane trafficking in yeast and mammalian cells. Its localization both in medial and trans-Golgi network prompted us to study the effects of brefeldin A (BFA) on rab6p redistribution. By two techniques, indirect immunofluorescence and cell fractionation, we investigated the fate of rab6p and compared it to other Golgi or trans-Golgi network markers in BHK-21 and NIH-3T3 cells. BFA, at 5 micrograms/ml, induced redistribution of rab6p according to a biphasic process: during the first 10–15 minutes, tubulo-vesicular structures--colabelled with a bona fide medial Golgi marker called CTR 433--were observed; these structures were then replaced by punctate diffuse staining, which was stable for up to 3 hours. The 110 kDa peripheral membrane protein beta-COP was released much more rapidly from the Golgi membranes, whereas the trans-Golgi network marker TGN 38 relocated to the microtubule organizing center. The kinetics of reversion of BFA action on these antigens was also followed by immunofluorescence. Consistent with these results, rab6 antigen, originally found as 40% in the cytosolic versus 60% in the particulate (P 150,000 g) fraction, became almost entirely cytosolic; moreover, it partitioned in the aqueous phase of Triton X-114 whereas the membrane fraction was detergent-soluble. Rab6p did not become part of the coatomers after its BFA-induced release from Golgi structures. Three requirements seemed to be necessary for such a release: integrity of the microtubules, presence of energy, and a hypothetical trimeric G protein, as revealed by the respective roles of nocodazole, ATP depletion, and sensitivity to aluminium fluoride. Finally, we have shown that BFA does not prevent attachment of newly synthesized rab6p to membranes.

Structure of the Golgi apparatus in stimulated and nonstimulated acinar cells of mammary glands of the rat

The structural features of the Golgi apparatus of acinar cells of mammary glands were examined with the electron microscope in 3 groups of rats: (1) in lactating female animals at 8 days postpartum, which served as controls; (2) in female rats sacrificed at various intervals from 2 to 30 hours following separation from their 8-day old pups; and (3) in females separated from their 8-day-old pups for a period of 12 hours and returned to their litters for durations of 1, 2, 4, and 8 hours. In animals of group 2, the Golgi stacks remained identical to that of controls between 2 and 8 hours. At 12 hours and later, the Golgi stacks decreased progressively in size, but the number of elements composing the stacks remained similar to that of lactating females and all contained casein submicelles. At 24 and 30 hours, typical secretory granules containing casein micelles disappeared from the trans aspect of the stacks. The earliest and most striking changes observed in the Golgi apparatus of the rats of group 2 took place at 12 hours. At this time, the prosecretory and secretory granules decreased considerably in volume and lost most of their electron-lucent content. This indicated that the delivery of small molecules, i.e., lactose and H2O, to these structures was soon altered following arrest of the sucking stimulus. In animals of group 3, the size of prosecretory and secretory granules and the amount of their electron-lucent content reverted to normal at 4 hours. Thus the influx of lactose and H2O into these structures appears to be rapidly restored after returning the pups to their mothers. The decrease in size of the Golgi stacks noted at 12, 18, and 24 hours following arrest of lactation (group 2), was accompanied by an increase in number of small vesicles that formed clusters next to the Golgi stacks and in "wells." Thus in these regressing Golgi stacks, many of the associated small vesicles appear to arise by vesiculation of the saccules.

Immunoisolation of Kex2p-containing organelles from yeast demonstrates colocalisation of three processing proteinases to a single Golgi compartment

One of the Golgi compartments of Saccharomyces cerevisiae is defined by the presence of a specific endoproteinase, Kex2p, which cleaves precursor polypeptides at pairs of basic residues. We have used antibodies directed against the cytoplasmically disposed C-terminal domain of Kex2p to develop an immuno-affinity procedure for the isolation of Kex2p-containing organelles. The method gives a high yield of sealed organelles that are essentially free of contamination from other secretory pathway organelles while being significantly enriched for two other late Golgi enzymes, dipeptidylaminopeptidase A and the Kex1 carboxypeptidase. Our findings provide clear evidence for a single yeast Golgi compartment containing all three late-processing enzymes, which is likely to be the functional equivalent in yeast of the mammalian trans-Golgi network.

Brefeldin A affects West Nile virus replication in Vero cells but not C6/36 cells

A fungal metabolite brefeldin A (BFA) was used to study virus-host interaction in glycoprotein processing in West Nile virus-infected Vero and C6/36 cells. The results indicated that as little as 1 microgram/ml of BFA resulted in complete breakdown in the Golgi organelle in infected Vero cells. This led to modifications of the glycoproteins which could not be efficiently used in infectious virion formation. In contrast, as much as 10 micrograms/ml of BFA in culture medium did not affect either glycoprotein formation or production of infectious particles in C6/36 cells. The results showed that in Vero cells, the transport of glycoproteins to the Golgi apparatus is important in West Nile virus infection. It also showed that BFA could be used as a tool to understand further the trafficking of glycoprotein from the ER to Golgi in flavivirus infection in Vero cells.

The actin-binding protein comitin (p24) is a component of the Golgi apparatus

Comitin (p24) was first identified in Dictyostelium discoideum as a membrane-associated protein which binds in gel overlay assays to G and F actin. To analyze its actin-binding properties we used purified, bacterially expressed comitin and found that it binds to F actin in spin down experiments and increases the viscosity of F actin solutions even under high-salt conditions. Immunofluorescence studies, cell fractionation experiments and EM studies of vesicles precipitated with comitin-specific monoclonal antibodies showed that comitin was present in D. discoideum on: (a) a perinuclear structure with tubular or fibrillary extensions; and (b) on vesicles distributed throughout the cell. In immunofluorescence experiments using comitin antibodies NIH 3T3 fibroblasts showed a similar staining pattern as D. discoideum cells. Using bona fide Golgi markers the perinuclear structure was identified as the Golgi apparatus. The results were supported by an electron microscopic study using cryosections. Based on these data we propose that also in Dictyostelium the stained perinuclear structure is the Golgi apparatus. In vivo the perinuclear structure was found to be attached to the actin and the microtubule network. Alteration of the actin network or depolymerization of the microtubules led to its dispersal into vesicles distributed throughout the cell. These results suggest that the Golgi apparatus in D. discoideum is connected to the actin network by comitin. This protein seems also to be present in mammalian cells.

Retinoic-acid-induced augmentation of molecular species carrying sialosyl Lewis(a) antigen in colorectal-carcinoma cell cultures

In order to study the effects of vitamin-A metabolites on long-term carcinoma-antigen secretion, colorectal-carcinoma cells SW1116 were cultured on membrane filters in totally synthetic media with 0 to 2.6 microM retinoic acid (RA). RA altered cell division, cell size and soluble-sialosyl Le(a) (S-Le(a) secretion and S-Le(a) accumulation within cells and apical-membrane domains. Cultures treated with RA for 10-12 days grew to lower cell densities (60% of controls) and contained more protein per cell (140% of controls). RA treated cells also had 5-fold higher levels of S-Le(a) in cells and secreted 9-fold more S-Le(a) into culture media assayed per 24 hr by (ELISA) 19-9 monoclonal antibody binding. As total media S-Le(a) increased, polarity of non-lipid S-Le(a) antigen secretion increased toward the interior (apical) media. High-performance thin-layer immunobinding showed that ganglioside S-Le(a) was higher in RA-fed cells, but could not be detected in apical media of RA-fed or control cells after 24 hr. Western blots indicated that non-lipid sialosyl Lewis(a) was bound to 150- to 180-kDA molecular species principally in cells, but 210- to 300-kDa molecular species appeared in the non-lipid extract of media. Thus, the above RA alterations, monitored by 3 immunochemical techniques, include up to 9-fold stimulation of "constitutive" 150- to 300-kDa sialosyl-Lewis(a) secretion, but ganglioside Lewis(a) is sorted differently and retained by apical membranes.

Relationship between the Golgi complex and microtubules enriched in detyrosinated or acetylated alpha-tubulin: studies on cells recovering from nocodazole and cells in the terminal phase of cytokinesis

Double immunofluorescence microscopy was used to study the relationship between the Golgi complex and microtubules enriched in posttranslationally modified tubulins in cultured mouse L929 fibroblasts. In interphase cells, the elements of the Golgi complex were grouped around the microtubule-organizing center. From here, tyrosinated microtubules extended to the periphery of the cells, whereas the distribution of detyrosinated and acetylated microtubules largely overlapped with that of the Golgi complex. Treatment of cells with 10 microM nocodazole led to the disruption of all microtubules and dispersion of the Golgi elements. Following withdrawal of the drug, tyrosinated microtubules reformed first, followed by acetylated and then detyrosinated microtubules. In parallel, the Golgi elements moved back toward the juxtanuclear region and reestablished a close spatial relationship first with the acetylated and later also with the detyrosinated microtubules. Long-term recovery in the presence of 0.15 or 0.3 microM nocodazole allowed partial reformation of tyrosinated and acetylated microtubules, whereas no or only a few detyrosinated microtubules were detected. At the same time, the Golgi elements were grouped closer together around or on one side of the nucleus in close relation to acetylated microtubules. In synchronized cells released from a mitotic block, a radiating array of tyrosinated microtubules was first formed, followed by acetylated and detyrosinated microtubules. The Golgi elements initially came together in a few groups and thereafter took an overall morphology similar to that in interphase cells. During this reunification, they showed a close spatial relationship to acetylated microtubules, whereas detyrosinated microtubules appeared only later. Microtubules enriched in acetylated and/or detyrosinated tubulin thus appear to take part in establishing and maintaining the organization of the Golgi elements within an interconnected supraorganellar system. Whether the acetylation and detyrosination of tubulin are directly involved in this process or merely represent two modifications within this subpopulation of microtubules remains unknown.

A cytosolic complex of p62 and rab6 associates with TGN38/41 and is involved in budding of exocytic vesicles from the trans-Golgi network

TGN38/41, an integral membrane protein predominantly localized to the trans-Golgi network, has been shown to cycle to the plasma membrane and return to the TGN within 30 min. (Ladinsky, M. S., and K. E. Howell. 1992. Eur. J. Cell Biol. 59:92-105). In characterizing the proteins which associate with TGN38/41, a peripheral 62-kD protein, two forms of rab6 and two other small GTP-binding proteins were identified by coimmunoprecipitation. However, approximately 90% of the 62-kD protein is cytosolic and is associated with the same subset of small GTP-binding proteins. Both the membrane and cytoplasmic complexes were characterized by sizing column fractionation and velocity sedimentation. The membrane complex was approximately 250 kD (11.6 S) consisting of the cytosolic complex and a heterodimer of TGN38/41 (160 kD). The cytosolic complex was approximately 86 kD (6.1 S) consisting of p62 and one small GTP-binding protein. Preliminary evidence indicates that phosphorylation of the p62 molecule regulates the dissociation of the cytosolic complex from TGN38/41. Functionally the cytosolic p62 complex must bind to TGN38/41 for the budding of exocytic transport vesicles from the TGN as assayed in a cell-free system (Salamero, J., E. S. Sztul, and K. E. Howell. 1990. Proc. Natl. Acad. Sci. USA. 87:7717-7721). Interference with p62, rab6 or TGN38, and TGN41 cytoplasmic domains by immunodepletion or competing peptides completely inhibited the budding of exocytic transport vesicles. These results support an essential role for interaction of the cytosolic p62/rab6 complex with TGN38/41 in budding of exocytic vesicles from the TGN.

Targeting and retention of Golgi membrane proteins

Recent cloning of genes encoding membrane proteins of the Golgi complex has allowed investigation of protein targeting to this organelle. Targeting signals have been identified in three glycosyltransferases, a viral envelope protein and several proteins of the trans-Golgi network. Interestingly, the targeting signals for membrane proteins of the Golgi stacks seem to be contained in transmembrane domains. Information in the cytoplasmic tails is required for the targeting of trans-Golgi network proteins. Mechanisms involving both retention and retrieval have been invoked.

Beta-COP is essential for biosynthetic membrane transport from the endoplasmic reticulum to the Golgi complex in vivo

Microinjection of antibodies against a synthetic peptide of a non-clathrin-coated vesicle-associated coat protein, beta-COP, blocks transport of a temperature-sensitive vesicular stomatitis virus glycoprotein (ts-O45-G) to the cell surface. Transport is inhibited upon release of the viral glycoprotein from temperature blocks at 39.5 degrees C (endoplasmic reticulum [ER]) and 15 degrees C (intermediate compartment), but not at 20 degrees C (trans-Golgi network). Ts-O45-G is arrested in tubular membrane structures containing p53 at the interface of the ER and the Golgi stack. This is consistent with inhibition of acquisition of endoglycosidase H resistance of ts-O45-G in injected cells. Secretion of endogenous proteins and maturation of cathepsin D are also inhibited. These data provide in vivo evidence that beta-COP has an important function in biosynthetic membrane traffic in mammalian cells.

Molecular characterization of two human autoantigens: unique cDNAs encoding 95- and 160-kD proteins of a putative family in the Golgi complex

Serum autoantibodies from a patient with autoantibodies directed against the Golgi complex were used to screen clones from a HepG2 lambda Zap cDNA library. Three related clones, designated SY2, SY10, and SY11, encoding two distinct polypeptides were purified for further analysis. Antibodies affinity purified by adsorption to the lambda Zap-cloned recombinant proteins and antibodies from NZW rabbits immunized with purified recombinant proteins reproduced Golgi staining and bound two different proteins, 95 and 160 kD, from whole cell extracts. The SY11 protein was provisionally named golgin-95 and the SY2/SY10 protein was named golgin-160. The deduced amino acid sequence of the cDNA clone of SY2 and SY11 represented 58.7- and 70-kD proteins of 568 and 620 amino acids. The in vitro translation products of SY2 and SY11 cDNAs migrated in SDS-PAGE at 65 and 95 kD, respectively. The in vitro translated proteins were immunoprecipitated by human anti-Golgi serum or immune rabbit serum, but not by normal human serum or preimmune rabbit serum. Features of the cDNA suggested that SY11 was a full-length clone encoding golgin-95 but SY2 and SY10 together encoded a partial sequence of golgin-160. Analysis of the SY11 recombinant protein identified a leucine zipper spanning positions 419-455, a glutamic acid-rich tract spanning positions 322-333, and a proline-rich tract spanning positions 67-73. A search of the SwissProt data bank indicated sequence similarity of SY11 to human restin, the heavy chain of kinesin, and the heavy chain of myosin. SY2 shared sequence similarity with the heavy chain of myosin, the USO1 transport protein from yeast, and the 150-kD cytoplasmic dynein-associated polypeptide. Sequence analysis demonstrated that golgin-95 and golgin-160 share 43% sequence similarity and, therefore, may be functionally related proteins.

Retention of a cis Golgi protein requires polar residues on one face of a predicted alpha-helix in the transmembrane domain

The first membrane-spanning domain (m1) of the model cis Golgi protein M (formerly called E1) from the avian coronavirus infectious bronchitis virus is required for targeting to the Golgi complex. When inserted in place of the membrane-spanning domain of a plasma membrane protein (vesicular stomatitis virus G protein), the chimeric protein ("Gm1") is retained in the Golgi complex of transfected cells. To determine the precise features of the m1 domain responsible for Golgi targeting, we produced single amino acid substitutions in m1 and analyzed their effects on localization of Gm1. Expression at the plasma membrane was used as the criterion for loss of Golgi retention. Rates of oligosaccharide processing were used as a measure of rate and efficiency of transport through the Golgi complex. We identified four uncharged polar residues that are critical for Golgi retention of Gm1 (Asn465, Thr469, Thr476, and Gln480). These residues line one face of a predicted alpha-helix. Interestingly, when the m1 domain of the homologous M protein from mouse hepatitis virus is inserted into the G protein reporter, the chimeric protein is not efficiently retained in the Golgi complex, but transported to the cell surface. Although it possesses three of the four residues we identified as important in the avian m1 sequence, other residues in the membrane-spanning domain from the mouse protein must prevent efficient recognition of the polar face within the lipid bilayer of the cis Golgi.

Double immunofluorescent staining of alpha 2,6 sialyltransferase and beta 1,4 galactosyltransferase in monensin-treated cells: evidence for different Golgi compartments?

Beta 1,4 galactosyl- and alpha 2,6 sialyltransferase (gal-T EC 2.4.1.22 and sialyl-T EC 2.4.99.1) sequentially elongate and terminate complex N-glycan chains of glycoproteins. Both enzymes reside in trans Golgi cisternae; their ultrastructural relationship, however, is unknown. To delineate their respective Golgi compartment(s) we conducted a double label immunofluorescent study by conventional and confocal laser scanning microscopy in HepG2, HeLa, and other cells in presence of Golgi-disturbing agents. Polyclonal, peptide-specific antibodies to human sialyl-T expressed as a beta-galactosidase-sialyl-T fusion protein in E. coli were developed and applied together with mABs to human milk gal-T. In untreated HepG2 and HeLa cells Golgi morphology identified by immunofluorescent labeling of sialyl-T and gal-T, respectively, was nearly identical. Treatment of cells with brefeldin A (BFA) led to rapid and coordinated disappearance of immunostaining of both enzymes; after BFA washout, vesicular structures reappeared which first stained for gal-T followed by sialyl-T; in the reassembled Golgi apparatus sialyl-T and gal-T were co-localized again. In contrast, monensin treatment produced a reversible swelling and scattering of gal-T positive Golgi elements while sialyl-T positive structures showed little change. Treatment with nocodazole led to dispersal of Golgi elements in which gal-T and sialyl-T remained co-localized. Treatment with chloroquine affected Golgi structures less than monensin and led to condensation of gal-T positive and to slight enlargement of sialyl-T positive structures. Sequential recovery from BFA of gal-T and sialyl-T and their segregation by monensin suggest that these enzymes are targeted to different Golgi subcompartments.

Giantin, a novel conserved Golgi membrane protein containing a cytoplasmic domain of at least 350 kDa

The Golgi complex consists of a series of stacked cisternae in most eukaryotes. Morphological studies indicate the existence of intercisternal cross-bridge structures that may mediate stacking, but their identity is unknown. We have identified a 400-kDa protein, giantin, that is localized to the Golgi complex because its staining in double immunofluorescence experiments was coincident with that of galactosyltransferase, both in untreated cells and in cells treated with agents that disrupt Golgi structure. A monoclonal antibody against giantin yielded Golgi staining in one avian and all mammalian cell types tested, indicating that giantin is a conserved protein. Giantin exhibited reduced mobility on nonreducing sodium dodecyl sulfate-polyacrylamide gel electrophoresis, was recovered in membrane fractions after differential centrifugation or sucrose flotation, and was not released from membranes by carbonate extraction. Thus, giantin appears to be an integral component of the Golgi membrane with a disulfide-linked lumenal domain. Strikingly, the majority of the polypeptide chain is cytoplasmically disposed, because large (up to 350 kDa) proteolytic fragments of giantin could be released from intact Golgi vesicles. This feature, a large contiguous cytoplasmic domain, is present in the calcium-release channel of muscle that cross-bridges the sarcoplasmic reticulum and transverse tubule membranes. Therefore, giantin's localization, conservation, and physical properties suggest that it may participate in forming the intercisternal cross-bridges of the Golgi complex.

Targeting of viral glycoproteins to the Golgi complex

Certain enveloped viruses are known to assemble on membranes of the Golgi complex. Intracellular budding is facilitated by targeting of the viral glycoproteins to this organelle. It is likely that these viral glycoproteins are retained in the Golgi by the same means as are endogenous Golgi proteins. Consequently, the study of Golgi-specific viral proteins has provided important clues to the nature of Golgi retention signals.

Membrane protein retention in the yeast Golgi apparatus: dipeptidyl aminopeptidase A is retained by a cytoplasmic signal containing aromatic residues

The mechanism by which yeast dipeptidyl aminopeptidase (DPAP) A, type II integral membrane protein, is retained in the late Golgi apparatus has been investigated. Prior work demonstrated that the 118-amino acid cytoplasmic domain is both necessary and sufficient for Golgi retention and that mutant or overexpressed DPAP A no longer retained in the Golgi was delivered directly to the vacuolar membrane (Roberts, C. J., S. F. Nothwehr, and T. H. Stevens. 1992. J. Cell Biol. 119:69-83). Replacement of the DPAP A transmembrane domain with a synthetic hydrophobic sequence did not affect either Golgi retention of DPAP A or vacuolar delivery of the retention-defective form of DPAP A. These results indicate that the DPAP A transmembrane domain is not involved in either Golgi retention or targeting of this membrane protein. A detailed mutational analysis of the cytoplasmic domain of DPAP A indicated that the most important elements for retention were within the eight residue stretch 85-92. A 10-amino acid region from DPAP A (81-90) was sufficient for Golgi retention of alkaline phosphatase, a type II vacuolar membrane protein. Detailed mutational analysis within this 10-amino acid sufficient region demonstrated that a Phe-X-Phe-X-Asp motif was absolutely required for efficient retention. The efficiency of Golgi retention via the DPAP A signal could be diminished by overexpression of wild type but not retention-defective versions of Kex2p, another late Golgi membrane protein, suggesting that multiple Golgi membrane proteins may be retained by a common machinery. These results imply a role for a cytoplasmic signal involving aromatic residues in retention of late Golgi membrane proteins in the yeast Saccharomyces cerevisiae.

Yeast vacuolar proenzymes are sorted in the late Golgi complex and transported to the vacuole via a prevacuolar endosome-like compartment

We are studying intercompartmental protein transport to the yeast lysosome-like vacuole with a reconstitution assay using permeabilized spheroplasts that measures, in an ATP and cytosol dependent reaction, vacuolar delivery and proteolytic maturation of the Golgi-modified precursor forms of vacuolar hydrolases like carboxypeptidase Y (CPY). To identify the potential donor compartment in this assay, we used subcellular fractionation procedures that have uncovered a novel membrane-enclosed prevacuolar transport intermediate. Differential centrifugation was used to separate permeabilized spheroplasts into 15K and 150K g membrane pellets. Centrifugation of these pellets to equilibrium on sucrose density gradients separated vacuolar and Golgi complex marker enzymes into light and dense fractions, respectively. When the Golgi-modified precursor form of CPY (p2CPY) was examined (after a 5-min pulse, 30-s chase), as much as 30-40% fractionated with an intermediate density between both the vacuole and the Golgi complex. Pulse-chase labeling and fractionation of membranes indicated that p2CPY in this gradient region had already passed through the Golgi complex, which kinetically ordered it between the Golgi and the vacuole. A mutant CPY protein that lacks a functional vacuolar sorting signal was detected in Golgi fractions but not in the intermediate compartment indicating that this corresponds to a post-sorting compartment. Based on the low transport efficiency of the mutant CPY protein in vitro (decreased by sevenfold), this intermediate organelle most likely represents the donor compartment in our reconstitution assay. This organelle is not likely to be a transport vesicle intermediate because EM analysis indicates enrichment of 250-400 nm compartments and internalization of surface-bound 35S-alpha-factor at 15 degrees C resulted in its apparent cofractionation with wild-type p2CPY, indicating an endosome-like compartment (Singer, B., and H. Reizman. 1990. J. Cell Biol. 110:1911-1922). Fractionation of p2CPY accumulated in the temperature sensitive vps15 mutant revealed that the vps15 transport block did not occur in the endosome-like compartment but rather in the late Golgi complex, presumably the site of CPY sorting. Therefore, as seen in mammalian cells, yeast CPY is sorted away from secretory proteins in the late Golgi and transits to the vacuole via a distinct endosome-like intermediate.

Increased LAMP-2 polylactosamine glycosylation is associated with its slower Golgi transit during establishment of a polarized MDCK epithelial monolayer

An endogenous Madin-Darby canine kidney (MDCK) lysosomal membrane glycoprotein that exhibits a basolateral targeting pathway to the lysosome is shown here to exhibit significant N-terminal amino acid sequence identity to lysosomal associated membrane proteins (LAMP-2) of other species. During establishment of the MDCK monolayer after only 1 d of culture, this canine LAMP-2 has a larger molecular size (110 kDa) than following formation of a confluent monolayer after 3 d of culture (100 kDa) due to the increased presence of N-linked polylactosamine oligosaccharide chains. Neither polylactosamine glycosylation of LAMP-2 in MDCK cells nor truncation of N-linked oligosaccharide chains of LAMP-2 in a ricin-resistant MDCK-RCAR cell line influenced the basolateral polarity of its targeting. However, the rate of basolateral delivery of LAMP-2 in MDCK cells plated for 3 d was significantly faster (t1/2 = 28 min) than in 1-d cells (t1/2 = 40 min); in MDCK-RCAR cells the rate of basolateral delivery at both 1 and 3 d of plating was similar (t1/2 = 40 min). The rate differential in MDCK cells occurred after arrival of LAMP-2 to the Golgi apparatus because the rate of acquisition of endoglycosidase H resistance was the same (t1/2 = 25 min) at both days of plating. The rate of transit of LAMP-2 through the Golgi apparatus to the basolateral domain was therefore far more rapid (approximately 4-fold) in 3 d compared with 1-d MDCK cultures. The increased polylactosamine glycosylation of MDCK LAMP-2 at early times of plating during the establishment of a confluent epithelial monolayer may thus be related to its longer residence time in the Golgi apparatus.

The Golgi apparatus of motor neurons in amyotrophic lateral sclerosis

The Golgi apparatus plays a key role in the posttranslational processing of polypeptides destined for secretion, incorporation into plasma membranes, and fast axoplasmic transport. Dispersion or fragmentation of the Golgi apparatus, experimentally induced by microtubule-disrupting agents, is associated with decreased secretion of immunoglobulins and insulin. The Golgi apparatus is also involved in targeting of lysosomal enzymes and in the endocytosis of certain hormones, receptors, and toxins. There is a paucity of information on this important organelle in human neuropathological conditions. Using an organelle-specific antiserum we have examined by immunocytochemistry the Golgi apparatus of motor neurons in the spinal cord in 4 patients with amyotrophic lateral sclerosis and 1 patient with Werdnig Hoffmann's disease, 1 with infantile neuronal degeneration, 1 with adult-type familial bulbospinal atrophy, 1 with mitochondrial myopathy with cytochrome c oxidase deficiency, 1 with centronuclear myopathy, and 1 with Duchenne's muscular dystrophy, and in 9 age-matched control subjects. In all motor neuronopathies examined and in the patient with mitochondrial myopathy, 20 to 85% of neurons counted had "fragmented" Golgi apparatus. In age-matched control subjects and the other 2 patients with myopathies, 0 to 1.65% of motor neurons had fragmented Golgi apparatus. These findings suggest that the Golgi apparatus of motor neurons is involved in patients with amyotrophic lateral sclerosis and related motor neuron diseases, and perhaps in patients with certain fatal primary myopathies.

Intracellular transport, organelle biogenesis and establishment of Golgi identity: impact of brefeldin A on the activity of lipid synthesizing enzymes

1. The effect of brefeldin A (BFA) on generation of transport vesicles, synthesis of phosphoglycerides, sphingosine and ceramides, and utilization of the sphingolipid precursors in the formation of sphingomyelin and glycosphingolipids in Golgi was investigated. 2. In the presence of 5-10 micrograms/ml BFA, the incorporation of [3H]palmitate into glycerides, phosphoglycerides and sphingolipids decreased 45-60%, and the production of endoplasmic reticulum transport vesicles was reduced 30-50%. 3. In Golgi membranes, the presence of 5-10 micrograms/ml BFA in the mixture, assembled to generate Golgi vesicles, evoked inhibitory effect on the synthesis of sphingomyelin, glycosphingolipids and phosphatidylcholine. On average, the synthesis of the sphingolipids and phosphatidylcholine and production of Golgi transport vesicles declined to 30-40%. 4. Addition of 5-10 micrograms/ml BFA to the assay mixture prepared to measure the activity of cytidylyltransferase, phosphocholine diacylglyceroltransferase, and serine palmitoyltransferase, caused up to 50% inhibition of the enzymes involved in the synthesis of phosphatidylcholine and up to 70% inhibition of the enzyme generating 3-ketosphinganine. 5. The results suggest that BFA inhibits the synthesis of phosphoglycerides and sphingolipids. This, at first, is displayed in reduced production of endoplasmic reticulum and Golgi transport vesicles, while the depletion of sphingolipids abrogates the identity of Golgi membranes.

Molecular cloning and analysis of the protein modules of aggrecans

The large aggregating chondroitin sulfate proteoglycan of cartilage, aggrecan, has served as a prototype of proteoglycan structure. Molecular cloning has elucidated its primary structure and revealed both known and unknown domains. To date the complete structures of chicken, rat and human aggrecans have been deduced, while partial sequences have been reported for bovine aggrecan. A related proteoglycan, human versican, has also been cloned and sequenced. Both aggrecan and versican have two lectin domains, one at the amino-terminus which binds hyaluronic acid and one at the carboxyl-terminus whose physiological ligand is unknown. Both lectins have homologous counterparts in other types of proteins. Within the aggrecans the keratan sulfate domain may be variably present and also has a prominent repeat in some species. The chondroitin sulfate domain has three distinct regions which vary in their prominence in different species. The complex molecular structure of aggrecans is consistent with the concept of exon shuffling and aggrecans serve as suitable prototypes for comprehending the evolution of multi-domain proteins.

Assembly of vaccinia virus: role of the intermediate compartment between the endoplasmic reticulum and the Golgi stacks

Vaccinia virus, the prototype of the Poxviridae, is a large DNA virus which replicates in the cytoplasm of the host cell. The assembly pathway of vaccinia virus displays several unique features, such as the production of two structurally distinct, infectious forms. One of these, termed intracellular naked virus (INV), remains cells associated while the other, termed extracellular enveloped virus (EEV), is released from the cell. In addition, it has long been believed that INVs acquire their lipid envelopes by a unique example of de novo membrane biogenesis. To examine the structure and assembly of vaccinia virus we have used immunoelectron microscopy using antibodies to proteins of different subcellular compartments as well as a phospholipid analysis of purified INV and EEV. Our data are not consistent with the de novo model of viral membrane synthesis but rather argue that the vaccinia virus DNA becomes enwrapped by a membrane cisterna derived from the intermediate compartment between the ER and the Golgi stacks, thus acquiring two membranes in one step. Phospholipid analysis of purified INV supports its derivation from an early biosynthetic compartment. This unique assembly process is repeated once more when the INV becomes enwrapped by an additional membrane cisterna, in agreement with earlier reports. The available data suggest that after fusion between the outer envelope and the plasma membrane, mature EEV is released from the cell.

Protein targeting to dense-core secretory granules

Regulated secretory proteins are stored within specialized vesicles known as secretory granules. It is not known how proteins are sorted into these organelles. Regulated proteins may possess targeting signals which interact with specific sorting receptors in the lumen of the trans-Golgi network (TGN) prior to their aggregation to form the characteristic dense-core of the granule. Alternatively, sorting may occur as the result of specific aggregation of regulated proteins in the TGN. Aggregates may be directed to secretory granules by interaction of a targeting signal on the surface with a sorting receptor. Novel targeting signals which confer on regulated proteins a tendency to aggregate under certain conditions, and in so doing cause them to be incorporated into secretory granules, have been implicated. Specific targeting signals may also play a role in directing membrane proteins to secretory granules.

TGN38 is maintained in the trans-Golgi network by a tyrosine-containing motif in the cytoplasmic domain

Sorting of proteins destined for different plasma membrane domains, lysosomes and secretory pathways takes place in the trans-Golgi network (TGN). TGN38 is an integral membrane protein found in this intracellular compartment. We show that TGN38 contains an autonomous targeting signal within its cytoplasmic domain which determines its intracellular location. Deletion analysis and site-directed mutagenesis of this domain demonstrate that a tyrosine motif homologous to the internalization signal of surface receptors is necessary and sufficient for correct localization. These findings suggest that TGN38 is maintained in the TGN by retrieval from the plasma membrane and employs a different mechanism for retention from that of the transferase enzymes of the trans-Golgi.

Budding from Golgi membranes requires the coatomer complex of non-clathrin coat proteins

Do the coats on vesicles budded from the Golgi apparatus actually cause the budding, or do they simply coat buds (Fig. 1)? One view (the membrane-mediated budding hypothesis) is that budding is an intrinsic property of Golgi membranes not requiring extrinsic coat proteins. Assembly of coats from dispersed subunits is super-imposed upon the intrinsic budding process and is proposed to convert the tips of tubules into vesicles. The alternative view (the coat-mediated budding hypothesis) is that coat formation provides the essential driving force for budding. The membrane-mediated budding hypothesis was inspired by the microtubule-dependent extension of apparently uncoated, 90-nm-diameter membrane tubules from the Golgi apparatus and other organelles in vivo after treatment with brefeldin A, a drug that inhibits the assembly of coat proteins onto Golgi membranes. This hypothesis predicts that tubules will be extended when coat proteins are unavailable to convert tubule-derived membrane into vesicles. Here we use a cell-free system in which coated vesicles are formed from Golgi cisternae to show that, on the contrary, when budding diminishes as a result of immunodepletion of coat protein pools, tubules are not formed at the expense of vesicles. We conclude that coat proteins are required for budding from Golgi membranes.

Protein trafficking along the exocytotic pathway

Proteins of the exocytotic (secretory) pathway are initially targeted to the endoplasmic reticulum (ER) and then translocated across and/or inserted into the membrane of the ER. During their anterograde transport with the bulk of the membrane flow along the exocytotic pathway, some proteins are selectively retained in various intracellular compartments, while others are sorted to different branches of the pathway. The signals or structural motifs that are involved in these selective targeting processes are being revealed and investigations into the mechanistic nature of these processes are actively underway.

Subcellular localization of T-cell receptor complexes containing tyrosine-phosphorylated zeta proteins in immature CD4+CD8+ thymocytes

The T-cell antigen receptor (TCR) is a complex of at least six different proteins (alpha, beta, gamma, delta, epsilon, and zeta) that is assembled in the endoplasmic reticulum (ER) and transported to the cell surface. Unlike mature T cells, most immature CD4+CD8+ thymocytes retain within the ER and degrade greater than 90% of some of the TCR components they synthesize, resulting in low surface expression of TCR complexes. The few surface TCR complexes that most immature CD4+CD8+ thymocytes do express are only marginally capable of transducing signals mobilizing intracellular calcium. The inverse relationship with TCR expression and function suggested that phosphorylated zeta (P-zeta) molecules might function in CD4+CD8+ thymocytes either as an ER retention signal for newly synthesized TCR complexes or as a negative regulatory modification of TCR complexes present on the cell surface. The present study sought to evaluate these two possibilities by determining the subcellular location of TCR complexes containing P-zeta chains. We found that, unlike unmodified zeta chains, all P-zeta chains in CD4+CD8+ thymocytes existed in assembled TCR complexes and that all TCR complexes containing P-zeta molecules had undergone carbohydrate processing events indicative of transit through the Golgi apparatus. These results demonstrate that P-zeta chains are exclusively associated with mature TCR complexes, excluding the possibility that P-zeta serves as an ER retention signal in immature thymocytes. Although we could not directly determine the representation of P-zeta chains among surface TCR complexes, we found that 60-70% of surface TCR complexes on immature CD4+CD8+ thymocytes were associated with tyrosine-phosphorylated protein(s) and that this percentage was inversely correlated with their signaling competence. These results support the concept that tyrosine phosphorylation serves as a negative regulatory modification of certain TCR-associated proteins.

Secretory transport in Plasmodium

The asexual blood stage of the human malaria parasite Plasmodium falciparum resides within the mature erythrocyte - a cell that has no intracellular organelles and few biosynthetic activities. However, Plasmodium, as on actively growing and dividing cell, has numerous requirements for the uptake o f nutrients and expulsion of waste. Hence, the parasite must extensively remodel the erythrocyte to facilitate its survival, not only by exporting numerous proteins, but also by providing the requisite machinery for their .trafficking. In this review, Heidi Elmendorf and Kastun Haldar propose a model for secretion in P. falciparum.

Characterization of a novel 63 kDa membrane protein. Implications for the organization of the ER-to-Golgi pathway

Owing to the lack of appropriate markers the structural organization of the ER-to-Golgi pathway and the dynamics of its membrane elements have been elusive. To elucidate this organization we have taken a monoclonal antibody (mAb) approach. A mAb against a novel 63 kDa membrane protein (p63) was produced that identifies a large tubular network of smooth membranes in the cytoplasm of primate cells. The distribution of p63 overlaps with the ER-Golgi intermediate compartment, defined by a previously described 53 kDa marker protein (here termed ERGIC-53), as visualized by confocal laser scanning immunofluorescence microscopy and immunoelectron microscopy. The p63 compartment mediates protein transport from the ER to Golgi apparatus, as indicated by partial colocalization of p63 and vesicular stomatitis virus G protein in Vero cells cultured at 15 degrees C. Low temperatures and brefeldin A had little effect on the cellular distribution of p63, suggesting that this novel marker is a stably anchored resident protein of these pre-Golgi membranes. p63 and ERGIC-53 were enriched to a similar degree by the same subcellular fractionation procedure. These findings demonstrate an unanticipated complexity of the ER-Golgi interface and suggest that the ER-Golgi intermediate compartment defined by ERGIC-53 may be part of a greater network of smooth membranes.

Localization of TGN38 to the trans-Golgi network: involvement of a cytoplasmic tyrosine-containing sequence

Protein localization to the TGN was investigated by examining the subcellular distribution of chimeric proteins in which the cytoplasmic and/or transmembrane domains of the TGN protein, TGN38, were substituted for the analogous domains of the plasma membrane protein, Tac. Using immunofluorescence and immunoelectron microscopy, the COOH-terminal cytoplasmic domain of TGN38 was found to be sufficient for localization of the chimeric proteins to the TGN. Deletion analysis identified an 11-amino acid segment containing the critical sequence, YQRL, as being sufficient for TGN localization. TGN localization was abrogated by mutation of the tyrosine or leucine residues in this sequence to alanine, or of the arginine residue to aspartate. In addition to specifying TGN localization, the 11-amino acid segment was active as an internalization signal, although the property of internalization alone was insufficient to confer TGN localization. Overexpression of chimeric proteins containing TGN localization determinants resulted in their detection at the plasma membrane and in intracellular vesicles, and abolished detection of endogenous TGN38. These results suggest that discrete cytoplasmic determinants can mediate protein localization to the TGN, and reveal a novel role for tyrosine-based motifs in this process.

Modulation of the Golgi apparatus in stimulated and nonstimulated prolactin cells of female rats

The three-dimensional structure of the Golgi apparatus and its compartments in prolactin cells has been examined in lactating rats in which secretion of prolactin was suppressed by removing the litter or stimulated by allowing the pups to suckle again. As soon as 2 hr after removal of the litter, large irregular progranules and numerous large pale vesicles accumulated in the trans-Golgi area together with vesicular or tubular fragments. The cis-tubular network was no longer recognizable on the cis-face of the Golgi ribbon; the saccules of the midcompartment were partitioned by narrow fissures and also became perforated in register by numerous fenestrations of various sizes and irregular contours. The concomitant appearance of numerous vesicles in the cavities thus formed as well as in the surrounding cytoplasm indicated that they probably arose by the progressive cavitation and fragmentation of saccules of the mid compartment. Such a process, which reached a maximum between 4 and 6 hr after removal of the litter from the mother, was no longer observed at 8 and 12 hr, at which time intervals the Golgi apparatus was reduced in size with no cis-tubular elements and progranules on its trans-aspect and few vesicles in its surroundings. When mothers, separated from their litters for a period of 12 hr, were returned to their pups for 20 min, the cis-tubular network reappeared on the cis-aspect of the Golgi stacks and presumably formed by fusion of vesicles and anastomosed tubules located next to the cisternae of the rough endoplasmic reticulum. In addition, the structure of the midsaccules returned to the stimulated condition, and early progranules were again segregated within the trans-most saccules of the Golgi stack. Hence, the Golgi apparatus of prolactin cells was rapidly and deeply modified in the presence or absence of stimulation.

Transport of casein submicelles and formation of secretion granules in the Golgi apparatus of epithelial cells of the lactating mammary gland of the rat

Lactating mammary glands fixed by perfusion with 5% glutaraldehyde subsequently were postfixed with potassium ferrocyanide reduced osmium or were treated with tannic acid. Stained thin sections were examined with the electron microscope and stereopairs were prepared. The distribution of casein submicelles was analyzed in the various components of the Golgi apparatus. The Golgi stacks were composed of five or six elements, all of which contained casein submicelles 20 nm in diameter. The cis-tubular network or cis-element, as well as the underlying three or four midsaccules, showed these casein submicelles either attached to their membrane or free in the lumen. The trans-most element of the stacks formed distended prosecretory granules in which both isolated or clustered casein submicelles were suspended in an electron-lucent fluid. These micellar aggregates increased in size and became progressively more compact to form spherical dense bodies or casein micelles, in which the individual 20 nm particles could easily be resolved. Casein micelles were seen in secretory granules in addition to a wispy material of low density. The numerous small spherical vesicles (80 nm or larger) seen on the cis, lateral, or trans aspects of the stacks did not appear to contain free casein submicelles. This raises questions regarding the role of these vesicles in the transport of casein macromolecules through the Golgi stacks. It was noticeable that in this Golgi apparatus a trans-Golgi network was limited to a few small residual tubules free from casein submicelles. It thus appears that the greater part of the trans-most Golgi element gives rise to the large prosecretory granules. After leaving the Golgi region and prior to exocytosis, the secretory granules often fuse to form larger granules before exocytosis.

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.