Golgi-resident Protein Research Articles

The length of cargo-protein transmembrane segments drives secretory transport by facilitating cargo concentration in export domains

The cellular destination of secretory proteins is determined by interactions of their targeting motifs with coat-protein complexes. The transmembrane domain (TMD) of secretory proteins also plays a central role in their transport and targeting. However, a comprehensive model that considers both TMD- and targeting-sequence-mediated transport has never been advanced. We focused on the secretory transport of two fluorescently tagged membrane proteins: vesicular stomatitis virus G tsO45 (VSVG), which is a cargo protein that is a thermoreversible mutant, and the Golgi-resident protein GalT-CFP. A quantitative approach was applied to analyze, in living cells, secretory transport dynamics, as well as cargo concentration of YFP-tagged VSVG mutants with one, three, five, seven, eight or nine amino acids deleted from their TMD, as well as two or four amino acids added to their TMD. Changes in TMD length affected secretory transport dynamics and the extent of cargo concentration in the ER exit sites, demonstrating that the capacity of the transport machinery to concentrate cargo depends on the length of the TMD of the cargo protein.

The Arabidopsis thaliana putative sialyltransferase resides in the Golgi apparatus but lacks the ability to transfer sialic acid

A common feature of the animal sialyltransferases (STs) is the presence of four conserved motifs, namely large (L), small (S), very small (VS) and motif III. Although sialic acid (SA) has not been detected in plants, three orthologues containing sequences similar to the ST motifs have been identified in the Arabidopsis thaliana L. database. In this study, we report that the At3g48820 gene (Gene ID: 824043) codes for a Golgi resident protein lacking the ability to transfer SA to asialofetuin or Galbeta1,3GalNAc and Galbeta1,4GlcNAc oligosaccharide acceptors. Restoration of deteriorated motifs S, VS and motif III by constructing chimeric proteins consisting of the 28-308 amino acid region of the A. thalianaAt3g48820 ST-like protein and the 264-393 amino acid region of the Oryza sativa L. AK107493 ST-like protein, or of the 28-240 amino acid region of the At3g48820 protein and the 204-350 amino acid region of the Homo sapiens L. alpha2,3-ST (NP_008858) was not able to recover sialyltransferase activity. Altering the appropriate amino acid regions of the A. thalianaAt3g48820 ST-like protein to those typical for the mammalian motif III (HHYWE) and VS motif (HDADFE) also did not have any effect. Our data, together with previous results, indicate that A. thaliana in particular, and plants in general, do not have transferases for SA. Substrates for the plant ST-like proteins might be compounds involved in secondary metabolism.

Ion‐binding properties of Calnuc, Ca2+ versus Mg2+– Calnuc adopts additional and unusual Ca2+‐binding sites upon interaction with G‐protein

Calnuc is a novel, highly modular, EF-hand containing, Ca(2+)-binding, Golgi resident protein whose functions are not clear. Using amino acid sequences, we demonstrate that Calnuc is a highly conserved protein among various organisms, from Ciona intestinalis to humans. Maximum homology among all sequences is found in the region that binds to G-proteins. In humans, it is known to be expressed in a variety of tissues, and it interacts with several important protein partners. Among other proteins, Calnuc is known to interact with heterotrimeric G-proteins, specifically with the alpha-subunit. Herein, we report the structural implications of Ca(2+) and Mg(2+) binding, and illustrate that Calnuc functions as a downstream effector for G-protein alpha-subunit. Our results show that Ca(2+) binds with an affinity of 7 mum and causes structural changes. Although Mg(2+) binds to Calnuc with very weak affinity, the structural changes that it causes are further enhanced by Ca(2+) binding. Furthermore, isothermal titration calorimetry results show that Calnuc and the G-protein bind with an affinity of 13 nm. We also predict a probable function for Calnuc, that of maintaining Ca(2+) homeostasis in the cell. Using Stains-all and terbium as Ca(2+) mimic probes, we demonstrate that the Ca(2+)-binding ability of Calnuc is governed by the activity-based conformational state of the G-protein. We propose that Calnuc adopts structural sites similar to the ones seen in proteins such as annexins, c2 domains or chromogrannin A, and therefore binds more calcium ions upon binding to Gialpha. With the number of organelle-targeted G-protein-coupled receptors increasing, intracellular communication mediated by G-proteins could become a new paradigm. In this regard, we propose that Calnuc could be involved in the downstream signaling of G-proteins.

Activity, Splice Variants, Conserved Peptide Motifs, and Phylogeny of Two New α1,3-Fucosyltransferase Families (FUT10 and FUT11)

We report the cloning of three splice variants of the FUT10 gene, encoding for active alpha-l-fucosyltransferase-isoforms of 391, 419, and 479 amino acids, and two splice variants of the FUT11 gene, encoding for two related alpha-l-fucosyltransferases of 476 and 492 amino acids. The FUT10 and FUT11 appeared 830 million years ago, whereas the other alpha1,3-fucosyltransferases emerged 450 million years ago. FUT10-391 and FUT10-419 were expressed in human embryos, whereas FUT10-479 was cloned from adult brain and was not found in embryos. Recombinant FUT10-419 and FUT10-479 have a type II trans-membrane topology and are retained in the endoplasmic reticulum (ER) by a membrane retention signal at their NH(2) termini. The FUT10-479 has, in addition, a COOH-ER membrane retention signal. The FUT10-391 is a soluble protein without a trans-membrane domain or ER retention signal that transiently localizes to the Golgi and then is routed to the lysosome. After transfection in COS7 cells, the three FUT10s and at least one FUT11, link alpha-l-fucose onto conalbumin glycopeptides and biantennary N-glycan acceptors but not onto short lactosaminyl acceptor substrates as do classical monoexonic alpha1,3-fucosyltransferases. Modifications of the innermost core GlcNAc of the N-glycan, by substitution with ManNAc or with an opened GlcNAc ring or by the addition of an alpha1,6-fucose, suggest that the FUT10 transfer is performed on the innermost GlcNAc of the core chitobiose. We can exclude alpha1,3-fucosylation of the two peripheral GlcNAcs linked to the trimannosyl core of the acceptor, because the FUT10 fucosylated biantennary N-glycan product loses both terminal GlcNAc residues after digestion with human placenta alpha-N-acetylglucosaminidase.

GPHR is a novel anion channel critical for acidification and functions of the Golgi apparatus

The organelles within secretory and endocytotic pathways in mammalian cells have acidified lumens, and regulation of their acidic pH is critical for the trafficking, processing and glycosylation of cargo proteins and lipids, as well as the morphological integrity of the organelles. How organelle lumen acidification is regulated, and how luminal pH elevation disturbs these fundamental cellular processes, is largely unknown. Here, we describe a novel molecule involved in Golgi acidification. First, mutant cells defective in Golgi acidification were established that exhibited delayed protein transport, impaired glycosylation and Golgi disorganization. Using expression cloning, a novel Golgi-resident multi-transmembrane protein, named Golgi pH regulator (GPHR), was identified as being responsible for the mutant cells. After reconstitution in planar lipid bilayers, GPHR exhibited a voltage-dependent anion-channel activity that may function in counterion conductance. Thus, GPHR modulates Golgi functions through regulation of acidification.

Charged single α‐helix: A versatile protein structural motif

A few highly charged natural peptide sequences were recently suggested to form stable alpha-helical structures in water. In this article we show that these sequences represent a novel structural motif called "charged single alpha-helix" (CSAH). To obtain reliable candidate CSAH motifs, we developed two conceptually different computational methods capable of scanning large databases: SCAN4CSAH is based on sequence features characteristic for salt bridge stabilized single alpha-helices, whereas FT_CHARGE applies Fourier transformation to charges along sequences. Using the consensus of the two approaches, a remarkable number of proteins were found to contain putative CSAH domains. Recombinant fragments (50-60 residues) corresponding to selected hits obtained by both methods (myosin 6, Golgi resident protein GCP60, and M4K4 protein kinase) were produced and shown by circular dichroism spectroscopy to adopt largely alpha-helical structure in water. CSAH segments differ substantially both from coiled-coil and intrinsically disordered proteins, despite the fact that current prediction methods recognize them as either or both. Analysis of the proteins containing CSAH motif revealed possible functional roles of the corresponding segments. The suggested main functional features include the formation of relatively rigid spacer/connector segments between functional domains as in caldesmon, extension of the lever arm in myosin motors and mediation of transient interactions by promoting dimerization in a range of proteins.

DHHC2 Affects Palmitoylation, Stability, and Functions of Tetraspanins CD9 and CD151

Although palmitoylation markedly affects tetraspanin protein biochemistry and functions, relevant palmitoylating enzymes were not known. There are 23 mammalian "DHHC" (Asp-His-His-Cys) proteins, which presumably palmitoylate different sets of protein substrates. Among DHHC proteins tested, DHHC2 best stimulated palmitoylation of tetraspanins CD9 and CD151, whereas inactive DHHC2 (containing DH-->AA or C-->S mutations within the DHHC motif) failed to promote palmitoylation. Furthermore, DHHC2 associated with CD9 and CD151, but not other cell surface proteins, and DHHC2 knockdown diminished CD9 and CD151 palmitoylation. Knockdown of six other Golgi-resident DHHC proteins (DHHC3, -4, -8, -17, -18, and -21) had no effect on CD9 or CD151. DHHC2 selectively affected tetraspanin palmitoylation, but not the palmitoylations of integrin beta4 subunit and bulk proteins visible in [(3)H]palmitate-labeled whole cell lysates. DHHC2-dependent palmitoylation also had multiple functional effects. First, it promoted physical associations between CD9 and CD151, and between alpha3 integrin and other proteins. Second, it protected CD151 and CD9 from lysosomal degradation. Third, the presence of DHHC2, but not other DHHC proteins, shifted cells away from a dispersed state and toward increased cell-cell contacts.

DGRASP-Mediated Noncanonical Integrin Secretion Is Required for Drosophila Epithelial Remodeling

Integral plasma membrane proteins are typically transported in the secretory pathway from the endoplasmic reticulum and the Golgi complex. Here we show that at specific stages of Drosophila development corresponding to morphological changes in epithelia, apposed basolateral membranes separate slightly, allowing new plasma membrane contacts with basal extracellular matrix. At these sites, newly synthesized integrin alpha subunits are deposited via a mechanism that appears to bypass the Golgi. We show that the Drosophila Golgi resident protein dGRASP localizes to these membrane domains and that, in the absence of dGRASP, the integrin subunit is retained intracellularly in both follicular and wing epithelia that are found disrupted. We propose that this dGRASP-mediated noncanonical secretion route allows for developmental regulation of integrin function upon epithelial remodeling. We speculate that this mechanism might be used during development as a means of targeting a specific subset of transmembrane proteins to the plasma membrane.

Loss of PL6 protein expression in renal clear cell carcinomas and other VHL‐deficient tumours

Mutations in the von Hippel-Lindau tumour suppressor gene (VHL) cause the VHL hereditary cancer syndrome and occur in most sporadic clear cell renal cell cancers (CC-RCCs). The mechanisms by which VHL loss of function promotes tumour development in the kidney are not fully elucidated. Here, we analyse expression of PL6, one of the potential tumour suppressor genes from the critical 3p21.3 region involved in multiple common cancers. We classify PL6 as a Golgi-resident protein based on its perinuclear co-localization with GPP130 in all cells and tissues analysed. We show that PL6 RNA and protein expression is completely or partially lost in all analysed CC-RCCs and other VHL-deficient tumours studied, including the early precancerous lesions in VHL disease. The restoration of VHL function in vitro in the VHL-deficient CC-RCC cell lines was found to reinstate PL6 expression, thus establishing a direct link between VHL and PL6. Insensitivity of PL6 to hypoxia suggested that PL6 is regulated by VHL via a HIF-1-independent pathway. We ruled out mutations and promoter methylation as possible causes of PL6 down-regulation in CC-RCC. We hypothesize that loss of a putative PL6 secretory function due to VHL deficiency is an early and important event that may promote tumour initiation and growth.

Targeting the expression of functional murine CMP-sialic acid transporter to the E. coli inner membrane

The heterologous expression of functional mammalian integral membrane proteins still represents a significant hurdle towards their crystallization and structure elucidation. We have therefore explored the use of the OmpA signal sequence to deliberately target the expression of the murine CMP-sialic acid transporter, a Golgi-resident protein with 10 putative transmembrane domains, to the Escherichia coli inner membrane. Here, we show that the expression of an OmpA signal sequence-FLAG-CMP-sialic acid transporter fusion protein in E. coli results in the targeting and insertion of recombinant protein within the inner membrane. Significantly, functionality was confirmed by the ability of spheroplasted E. coli and mixed phosphatidylcholine– E. coli inner membrane proteoliposomes incorporating recombinant CMP-sialic acid transporter to accumulate CMP-sialic acid in vitro.

Identification of a Redox-sensitive Cysteine in GCP60 That Regulates Its Interaction with Golgin-160

Golgin-160 is ubiquitously expressed in vertebrates. It localizes to the cytoplasmic side of the Golgi and has a large C-terminal coiled-coil domain. The noncoiled-coil N-terminal head domain contains Golgi targeting information, a cryptic nuclear localization signal, and three caspase cleavage sites. Caspase cleavage of the golgin-160 head domain generates different fragments that can translocate to the nucleus by exposing the nuclear localization signal. We have previously shown that GCP60, a Golgi resident protein, interacts weakly with the golgin-160 head domain but has a strong interaction with one of the caspase-generated golgin-160 fragments (residues 140-311). This preferential interaction increases the Golgi retention of the golgin-160 fragment in cells overexpressing GCP60. Here we studied the interaction of golgin-160-(140-311) with GCP60 and identified a single cysteine residue in GCP60 (Cys-463) that is critical for the interaction of the two proteins. Mutation of the cysteine blocked the interaction in vitro and disrupted the ability to retain the golgin-160 fragment at the Golgi in cells. We also found that Cys-463 is redox-sensitive; in its reduced form, interaction with golgin-160 was diminished or abolished, whereas oxidation of the Cys-463 by hydrogen peroxide restored the interaction. In addition, incubation with a nitric oxide donor promoted this interaction in vitro. These findings suggest that nuclear translocation of golgin-160-(140-311) is a highly coordinated event regulated not only by cleavage of the golgin-160 head but also by the oxidation state of GCP60.

Disruption of Golgi processing by 2-phenyl benzimidazole analogs blocks cell proliferation and slows tumor growth

Cancer chemotherapy continues to be challenged by the emergence of resistant tumors, and one organelle entwined in the development of drug resistance is the Golgi apparatus. Recently, we discovered a group of 2-(substituted phenyl)-benzimidazole (2-PB) compounds that displace resident Golgi proteins from the juxtanuclear region resulting in their degradation. These compounds are also potent anti-proliferative agents, which together with their action on the Golgi made a compelling case for testing them against cancer. The anti-tumor activity of a group of 2-PB compounds was examined both in vitro and in vivo. The role of the Golgi in the anti-proliferative effect was assessed by comparing the proliferation of individual cell lines with the distribution and total cellular expression of selected resident Golgi proteins. The anti-proliferative activity of 2-PB compounds is partially reversible (time- and concentration-dependent), non-cell-cycle-specific, and translates to tumor growth inhibition in vivo. While 2-PB compounds displace resident Golgi proteins from the juxtanuclear region in all cells, those that are resistant to the anti-proliferative effects differ from sensitive cells in that they have the capacity to protect these Golgi proteins from degradation. These results illustrate the utility of targeting the Golgi for cancer drug development. They also reveal a cellular strategy for resisting 2-PB drug effects through protection of displaced Golgi proteins from degradation thus allowing their continued function.

Regulation of Heparan Sulfate 6-O-Sulfation by β-Secretase Activity

The enzymes involved in glycosaminoglycan chain biosynthesis are mostly Golgi resident proteins, but some are secreted extracellularly. For example, the activities of heparan sulfate 6-O-sulfotransferase (HS6ST) and heparan sulfate 3-O-sulfotransferase are detected in the serum as well in the medium of cell lines. However, the biological significance of this is largely unknown. Here we have investigated by means of monitoring green fluorescent protein (GFP) fluorescence how C-terminally GFP-tagged HS6STs that are stably expressed in CHO-K1 cell lines are secreted/shed. Brefeldin A and monensin treatments revealed that the N-terminal hydrophobic domain of HS6ST3 is processed in the endoplasmic reticulum or cis/medial Golgi. Treatment of HS6ST3-GFP-expressing cells with various protease inhibitors revealed that the cell-permeable beta-secretase inhibitor N-benzyloxycarbonyl-Val-Leu-leucinal (Z-VLL-CHO) specifically inhibits HS6ST secretion, although this effect was specific for HS6ST3 but not for HS6ST1 and HS6ST2. However, Z-VLL-CHO treatment did not increase the molecular size of the HS6ST3-GFP that accumulated in the cell. Z-VLL-CHO treatment also induced the intracellular accumulation of SP-HS6ST3(-TMD)-GFP, a modified secretory form of HS6ST3 that has the preprotrypsin leader sequence as its N-terminal hydrophobic domain. Diminishment of beta-secretase activity by coexpressing the amyloid precursor protein of a Swedish mutant, a potent beta-secretase substrate, also induced intracellular HS6ST3-GFP accumulation. Moreover, Z-VLL-CHO treatment increased the 6-O-sulfate (6S) levels of HS, especially in the disaccharide unit of hexuronic acid-GlcNS(6S). Thus, the HS6ST3 enzyme in the Golgi apparatus and therefore the 6-O sulfation of heparan sulfates in the cell are at least partly regulated by beta-secretase via an indirect mechanism.

Confocal Microscopy-based Linescan Methodologies for Intra-Golgi Localization of Proteins

Localization of resident Golgi proteins to earlier (cis) or later (trans) Golgi compartments has traditionally required quantitative immunocytochemistry and electron microscopy, which are inaccessible to many researchers. For this reason, light microscopy has often been used, initially for localization of Golgi glycotransferases and, more recently, for other Golgi proteins (e.g., Arf1, GBF1, Rab6). Quantitation of light microscopic intra-Golgi localization can be problematic. We describe here a novel quantitative light microscopic methodology using linescans crossing the Golgi ribbon. Our method determines a localization for the unknown protein in a one-dimensional coordinate system in which 0.0 corresponds to localization of a cis marker and 1.0 to localization of a trans marker. We also describe a variant of this methodology in which Golgi morphology is simplified by nocodazole-induced dispersal into ministacks, allowing a fully automated analysis. In our assay, beta1,4-galactosyltransferase-YFP and Golgin97 localize similarly to trans markers, whereas p115, GBF1, and p58-YFP are similarly near other cis markers. The medial Golgi protein alpha1,3-1,6-mannosidase II gives an intermediate localization in this assay. These methodologies may prove useful in instances where electron microscopy is technically difficult as well as when rapid analysis of large numbers of samples is required.

Activation of NF-κB by the Human T Cell Leukemia Virus Type I Tax Oncoprotein Is Associated with Ubiquitin-dependent Relocalization of IκB Kinase

Activation of NF-κB by the Human T Cell Leukemia Virus Type I Tax Oncoprotein Is Associated with Ubiquitin-dependent Relocalization of IκB Kinase

Involvement of a Golgi-resident GPI-anchored protein in maintenance of the Golgi structure.

The Golgi apparatus consists of a series of flattened cisternal membranes that are aligned in parallel to form stacks. Cytosolic-oriented Golgi-associated proteins have been identified that may coordinate or maintain the Golgi architecture. Here, we describe a novel GPI-anchored protein, Golgi-resident GPI-anchored protein (GREG) that has a brefeldin A-sensitive Golgi localization. GREG resides in the Golgi lumen as a cis-oriented homodimer, due to strong interactions between coiled-coil regions in the C termini. Dimerization of GREG as well as its Golgi localization depends on a unique tandem repeat sequence within the coiled-coil region. RNA-mediated interference of GREG expression or expression of GREG mutants reveals an essential role for GREG in maintenance of the Golgi integrity. Under these conditions, secretion of the vesicular stomatitis virus glycoprotein protein as a marker for protein transport along the secretory pathway is inhibited, suggesting a loss of Golgi function as well. These results imply the involvement of a luminal protein in Golgi structure and function.

Identification and characterization of COPIa- and COPIb-type vesicle classes associated with plant and algal Golgi

Coat protein I (COPI) vesicles arise from Golgi cisternae and mediate the recycling of proteins from the Golgi back to the endoplasmic reticulum (ER) and the transport of Golgi resident proteins between cisternae. In vitro studies have produced evidence for two distinct types of COPI vesicles, but the in vivo sites of operation of these vesicles remain to be established. We have used a combination of electron tomography and immunolabeling techniques to examine Golgi stacks and associated vesicles in the cells of the scale-producing alga Scherffelia dubia and Arabidopsis preserved by high-pressure freezing/freeze-substitution methods. Five structurally distinct types of vesicles were distinguished. In Arabidopsis, COPI and COPII vesicle coat proteins as well as vesicle cargo molecules (mannosidase I and sialyltransferase-yellow fluorescent protein) were identified by immunogold labeling. In both organisms, the COPI-type vesicles were further characterized by a combination of six structural criteria: coat architecture, coat thickness, membrane structure, cargo staining, cisternal origin, and spatial distribution. Using this multiparameter structural approach, we can distinguish two types of COPI vesicles, COPIa and COPIb. COPIa vesicles bud exclusively from cis cisternae and occupy the space between cis cisternae and ER export sites, whereas the COPIb vesicles bud exclusively from medial- and trans-Golgi cisternae and are confined to the space around these latter cisternae. We conclude that COPIa vesicle-mediated recycling to the ER occurs only from cis cisternae, that retrograde transport of Golgi resident proteins by COPIb vesicles is limited to medial and trans cisternae, and that diffusion of periGolgi vesicles is restricted.

Quantitative Proteomics Analysis of the Secretory Pathway

We report more than 1400 proteins of the secretory-pathway proteome and provide spatial information on the relative presence of each protein in the rough and smooth ER Golgi cisternae and Golgi-derived COPI vesicles. The data support a role for COPI vesicles in recycling and cisternal maturation, showing that Golgi-resident proteins are present at a higher concentration than secretory cargo. Of the 1400 proteins, 345 were identified as previously uncharacterized. Of these, 230 had their subcellular location deduced by proteomics. This study provides a comprehensive catalog of the ER and Golgi proteomes with insight into their identity and function.

Golgi twins in late mitosis revealed by genetically encoded tags for live cell imaging and correlated electron microscopy

Combinations of molecular tags visible in light and electron microscopes become particularly advantageous in the analysis of dynamic cellular components like the Golgi apparatus. This organelle disassembles at the onset of mitosis and, after a sequence of poorly understood events, reassembles after cytokinesis. The precise location of Golgi membranes and resident proteins during mitosis remains unclear, partly due to limitations of molecular markers and the resolution of light microscopy. We generated a fusion consisting of the first 117 residues of alpha-mannosidase II tagged with a fluorescent protein and a tetracysteine motif. The mannosidase component guarantees docking into the Golgi membrane, with the tags exposed in the lumen. The fluorescent protein is optically visible without further treatment, whereas the tetracysteine tag can be reduced acutely with a membrane-permeant phosphine, labeled with ReAsH, monitored in the light microscope, and used to trigger the photoconversion of diaminobenzidine, allowing 4D optical recording on live cells and correlated ultrastructural analysis by electron microscopy. These methods reveal that Golgi reassembly is preceded by the formation of four colinear clusters at telophase, two per daughter cell. Within each daughter, the smaller cluster near the midbody gradually migrates to rejoin the major cluster on the far side of the nucleus and asymmetrically reconstitutes a single Golgi apparatus, first in one daughter cell and then in the other. Our studies provide previously undescribed insights into Golgi disassociation and reassembly during mitosis and offer a powerful approach to follow recombinant protein distribution in 4D imaging and correlated high-resolution analysis.

IntraGolgi distribution of the Conserved Oligomeric Golgi (COG) complex

The Conserved Oligomeric Golgi (COG) complex is an eight-subunit (Cog1–8) peripheral Golgi protein involved in membrane trafficking and glycoconjugate synthesis. COG appears to participate in retrograde vesicular transport and is required to maintain normal Golgi structure and function. COG mutations interfere with normal transport, distribution, and/or stability of Golgi proteins associated with glycoconjugate synthesis and trafficking, and lead to failure of spermatogenesis in Drosophila melanogaster, misdirected migration of gonadal distal tip cells in Caenorhabditis elegans, and type II congenital disorders of glycosylation in humans. The mechanism by which COG influences Golgi structure and function is unclear. Immunogold electron microscopy was used to visualize the intraGolgi distribution of a functional, hemagglutinin epitope-labeled COG subunit, Cog1-HA, that complements the Cog1-deficiency in Cog1-null Chinese hamster ovary cells. COG was found to be localized primarily on or in close proximity to the tips and rims of the Golgi's cisternae and their associated vesicles and on vesicles and vesiculo-tubular structures seen on both the cis and trans-Golgi Network faces of the cisternal stacks, in some cases on COPI containing vesicles. These findings support the proposal that COG is directly involved in controlling vesicular retrograde transport of Golgi resident proteins throughout the Golgi apparatus.