A peripheral protein associated with the cis-Golgi network redistributes in the intermediate compartment upon brefeldin A treatment

A Peripheral Protein Associated with the cis-Golgi Network Redistributes in the Intermediate Compartment upon Brefeldin A Treatment Rosa M. Rios,** Anne-Marie Tassin,* Claude Celati* Claude Antony,§ Marie-Christophe Boissier, II Jean-Claude Homberg,~ and Michel Bornens* * Centre de Genetique Moleculaire, Centre National de la Recherche Scientifique, 91198-Gif/Yvette, France; * Departamento de Microbiologfa, Facultad de Biologia, Universidad de Sevilla, 41080-SeviUa, Spain; § Institut Jacques Monod, Centre National de la Recherche Scientifique-Universit6 Pads VII, F-75005, Paris, France; II D6partement de Rhumathologie, Hopital Avicenne, 93000 Bobigny, France; and qlD6partement d'Immunologie, Facult6 de M&lecine Saint-Antoine, 75571, Paris, France Abstract. Human autoantibodies offer unique tools for the study of cellular constituents since they usually recognize highly conserved components, the most difficult to detect due to their low immunogenicity. The serum from a patient with Sj6gren's syndrome (RM serum) showing a very high reactivity to the Golgi complex has been shown to immunoprecipitate and to immunodetect by Western blotting experiments a protein of mol wt 210,000 (p210) that was shown to be peripheral and cytoplasmically disposed. A close examination of the p210 labeling revealed some differences with Golgi markers: RM serum staining was slightly more extensive than several Golgi markers and showed a discontinuous or granular appearance. Nocodazole induced a specific and early segregation of many p210-associated vesicles or tubules from Golgi apparatus. Upon brefeldin A treatment, p210 did not redistribute in the ER as did other Golgi proteins. In contrast, it exhibited a vesicular pattern reminiscent to that displayed by proteins residing in the intermediate compartment. Double staining immunofluorescence using the RM serum and the marker of the intermediate compartment, p58, revealed segregation of both proteins in control conditions but colocalization in BFA-treated cells. We have further demonstrated by combining different drug treatments that p210containing elements in brefeldin A-treated cells belong indeed to the intermediate compartment. Experiments on brefeldin A recovery suggested that these p210 elements might play a role in reformation and repositioning of the Golgi apparatus. Ultrastructural localization performed by immunoperoxidase staining allowed us to establish that p210 interacted with the external side of an abundant tubulo-vesicular system on the cis side of the Golgi complex which extended to connecting structures and vesicles between saccules or stacks of cisternae, p210 appears to be a novel protein residing in the cis-Golgi network that may cycle between the Golgi apparatus and the intermediate compartment. T HE Golgi apparatus (GA) t is a dynamic membranous network which plays a key role in processing, maturation, and sorting of the newly synthesized proteins destined to specific membrane domains, or to secretion, and in recycling of receptors involved in endocytosis. Whereas ER is usually extended throughout the cytoplasm, the Golgi complex appears centrally located in close proximity to the microtubule-organizing center (Farquhar and Palade, 1981; Address all correspondence to Rosa M. Rios, Departamento de Microbiologfa, Facultad de Biologfa, Universidad de Sevilla, Apdo. 1095, 41080Sevilla, Spain. 1. Abbreviations used in this paper: ARF, ADP-ribosylation factor; CGN, cis-Golgi network; TGN, trans-Golgi network; BFA, brefeldin A; GA, Golgi apparatus; GalTf, antigalactosyl transferase; SLO, streptolysin O. Kupfer et al., 1982). This pericentrosomal organization that is generated and maintained by microtubules (Wehland and Willingham, 1983; Wehland et al., 1983; Rogalsld and Singer, 1984; Sandoval et al., 1984) enables the Golgi complex to receive and to direct both exocytic and endocytic material that traffic through it. Functional and morphological studies have shown a polarized organization of the Golgi complex. A three compartment subdivision of the Golgi complex is thought to be required to accomplish all of its functions: the cis-Golgi network (CGN) next to the ER, where newly synthesized material is received and sorted, the medial Golgi (containing the cis, medial, and tmns cisternae) where glycosylation and trimming of carbohydrate moieties takes place, and the trans-Golgi network (TGN) that mediates the sorting and exit of material from the GA before a final vesicular trans- © The Rockefeller University Press, 0021-9525/94/06/997/17 $2.00 The Journal of Cell Biology, Volume 125, Number 5, June 1994 997-1013 997 on July 26, 2017jcb.rupress.orgDownloaded from port step to the terminal destination (Mellman and Simons, 1992). Transport events through the GA are thought to occur by movement of carrier vesicles (Palade, 1975). Nonclathrincoated vesicles have been proposed to mediate transport from ER to the Golgi complex and between the different Golgi compartments (Orci et al., 1986; Maihotra et al., 1989; Serafini et al., 1991a). The coat of these vesicles comprises a set of four proteins named COPs (or, /~, 3' and t$-COPs) and probably two additional subunits of 20 and 36 kD (Serafini et ai., 1991a). All of these proteins exist in a cytosolic complex, the coatomer, that may represent an unassembled precursor of the coat (Waters et al., 1991). In addition ADP-ribosylation factor (ARF), a small GTP-binding protein, has been identified as a component of the Golgiderived coated vesicles (Serafini et al., 1991b). Both the coatomer and ARF associate reversibly to Golgi membranes (Donaldson et ai., 1991a; Serafini et al., 1991b) and dynamics of this cycle is now well established (Donaldson et al., 1992a). Recently, a Golgi membrane enzyme has been identified which might determine the localization of membrane-bound ARF-GTP and hence the sites for vesicle budding (Donaldson et al., 1992b; Helms and Rothman, 1992). Other small GTP-binding proteins of the rab family have also been found in association with specific vesicles and organelles and are thought to have a crucial role in the targeting of vesicles to, and their fusion with, the appropriate acceptor organelles (Bourne, 1988; Goud and McCaffrey, 1991; Pfeffer, 1992). For many of them however, more conclusive information is still required to decide whether their localization is restricted to a single pair of donor-acceptor compartments (see Antony et al., 1992). Heterotrimeric G proteins have also been reported to be part of the transport machinery (Ercolani et al., 1990; Barr et al., 1991; Donaldson et al., 1991b; Stow et ai., 1991; Ktistakis et ai., 1992). Various drugs proved to be useful to elucidate the dynamic equilibrium on which Golgi membrane organization is based. The fungal metabolite brefeldin A (BFA) had been shown to inhibit protein secretion and dramatically disintegrates the Golgi apparatus, resulting in the distribution of Golgi membranes (lipids and proteins) in the ER by a mechanism that requires energy and microtubules. Uncoated tubules emanating from the Golgi complex are observed rapidly after addition of BFA to ceils suggesting that they are the structural intermediates that carry the Golgi membranes to the ER (for review see Klausner et ai., 1992). The immediate effect of BFA is to prevent the assembly of COPscoated vesicles (Orci et al., 1991) probably by interfering with the association of ARF and consequently of the coatomer complex with Golgi membranes (Donaldson et al., 1992a). Thus BFA would block anterograde traffic between ER and GA without affecting retrograde transport from GA to ER. Morphology and functioning of other organelles of the secretory pathway such as TGN or the endosomal system are also perturbed by BFA (Klausner et al., 1992). BFA-like phenotypes have been described in two different situations: overexpression of a human ERD-2-1ike protein (Hsu et al., 1992) and in a mutant CHO cell line (Zuber et al. 1991). All of these data indicate that the Golgi apparatus exists as a steady state structure and that the maintenance of its integrity requires a rigorous control of the balance between anterograde and retrograde traffic from ER up to terminal GA. In response to BFA treatment, three different behaviors of Golgi proteins have been reported. Most of the Golgi proteins such as mannosidase II and IA (Lippincott-Schwartz et al., 1989), galactosyl-transferase (Lippincott-Schwartz et al., 1990), the proteins p58, p54, and p86 (Donaldson et al., 1990), and GMPc.t.2, MG 160, and GMPtq.2, (Alcaide et al., 1992) redistribute to ER. ~COP (Donaldson et al., 1990), ARF (Donaldson et al., 1991a), and a p200 (Narula et al., 1992) dissociate from Golgi membranes to cytosol. Finally p58 (Saraste and Svensson, 1991) and p53 (Lippincott-Schwartz et ai., 1990), two proteins which reside in the intermediate compartment, do not modify their localization since intermediate compartment preserves its identity in the presence of BFA. Despite this major progress in understanding the function and organization of the Golgi complex, many Golgi membrane proteins remain to be discovered or characterized. Recently, a new immunological approach has been used in order to identify highly conserved Golgi components which have shown to be the most difficult to detect using classical strategies, namely the screening of sera libraries from patients with systemic autoimmune diseases (Kooy et al., 1992). Human autoantibodies offer unique tools for the study of cellular components since they recognize highly conserved and functionally important molecules. In addition, the autoantibodies usually react with the active sites of the antigen and therefore they are often capable of inhibiting its functional activity (Tan, 1991). By using this approach, we have selected a serum from a patient with Sjtgren's syndrome containing high titer autoantibodies to the GA (Rios, R. M., M. C. Boissier, J. C. Homberg, and M. Bornens, manuscript submitted for publication). It recognizes a novel Golgi protein of 210 kD in a wide variety of cells and organisms. Biochemical and EM investigations indicate that p210 is a peripheral protein whose localization is restricted to the cis side of the Golgi complex and seems to correspond to the CGN. Interestingly, p210 displays an unusual behavior, compared to bona fide Golgi markers, when cells are treated with nocodazole or with BFA. In experiments using BFA, unlike most Golgi proteins, p210 does not redistribute to ER but to the intermediate compartment. These results suggest that certain Golgi proteins might be excluded from the pathway that redistributes Golgi membrane proteins to the ER and that the time course of Golgi reconstitution after BFA treatment occurs by sequential events, probably involving a subset of specific proteins. Materials and Methods Cell Culture The KE37 cell line of T lymphoblastic origin was grown in RPMI 1640 medium containing 7% fetal calf serum. HeLa cells were maintained in DME supplemented with 10% fetal calf serum. Normal muscles were obtained from patients undergoing surgical operations. The biopsy specimens were collected in a sterile container containing culture medium and stored at 4°C. The specimens were dissociated as previously described (Tassin et al., 1985a). 2 mM Glutamine, 50 U/ml penicillin, and 50 #g/m streptomycin were included in all culture media. Antibodies Serum of the patient R.M. suffering from Sjtgren's syndrome (see Rodriguez el al., 1982 and Blascheck et al., 1988) was aliquoted, sodium azide added, and then stored at -70"C. IgG fraction (10 mg/ml final concentration) was purified from whole serum on protein A-Sepharose columns and The Journal of Cell Biology, Volume 125, 1994 998 on July 26, 2017jcb.rupress.orgDownloaded from stored in 50% glycerol at -70°C. CTR 433 was obtained from a library of monoclonal antibodies against centrosomes isolated from human lymphoblasts and has been previously characterized as a marker of the med/al compartment of the GA (Jasmin et al., 1989). Affinity-purified antigalactosyl transferase (GalTf) antibody was kindly provided by Dr. Eric Berger (Institute of Physiology, Zurich, Switzerland). Affinity-purified anti-p58 polyclonal antibody (Saraste et al., 1987) was obtained from Dr. Jaako Saraste (Ludwig Institute of Cancer Research, Stockholm, Sweden). A specific antiendoplasmic reticulum antibody (Louvard et al., 1982) was the kind gift of Dr. D. Louvard (Institut Pasteur, Paris, France). Anti-/~-tubulin antibody was purchased from Amersham Corp. (Arlington Heights, IL). Specific antibodies from RM serum were affinity purified on nitrocellulose strips using immunoreactive proteins from different fractions, according to Krolme et al. (1982). Immunofluorescence Microscopy Ceils were grown on culture-treated slides for 2 d before an experiment. Cells were rinsed twice with PBS and incubated in methanol at -20°C for 6 min to simultaneously fix and permeabilize the cells. After methanol treatment, cells were rinsed with PBS containing 0.1% Tween 20 (PBST). Primary antibodies diluted in PBST containing 3 % BSA were added for 1 h at room temperature. The cells were then rinsed with PBST three times to wash away unbound primary antibodies. The same incubations and washing procedures were used for fluoresceinor rhodamine-labeled secondary antibodies. The cells were finally dehydrated, mounted in citifluor, examined, and photographed with a Zeiss Axiophot microscope. For the visualization of p58, the cells were fixed for 15 min at room temperature with 3% paraformaldehyde in 0.1% phosphate buffer, pH 7.4, and thereafter permeabilized for 15 min in PBS containing 0.1% saponin. After a rinse, the cells were incubated with 50 mM NI-I4C1 to inactivate free aldehyde groups and then processed as described above. Drug Treatments BFA was purchased from Epicentre Technologies (Madison, WI) or obtained from Sandoz Ltd. (Basel, Switzerland). It was stored at -200C as a stock solution of 5 rng/ml in methanol. Nocodazole was obtained from Sigma Chem. Co. (St. Louis, MO) and stored as a 5-mM stock solution in DMSO at -200C. Immediately prior to use, solutions were prepared in culture medium. Ceils were incubated at 37°C with 10/tM nocodazole or 1-5 t~g/ml BFA for various periods up to 2 h. The calcium iunophore A23187 (Sigma Chem. Co.) was administered to cells in culture medium at 5/~M final concentration. Cells were incubated with A23187 for 4 h at 37°C before fixation. Metabolic Labeling and lmmunoprecipitation Subconfluent monolayers of HeLa cells were labeled with 100-200 tzCi/ml of [35Slmethionine/cysteine (Trans-Label; ICN Biomedicals, Inc., Costa Mesa, CA) for 4 h in DMEM supplemented with 5% FCS. Cells were washed with ice-cold PBS and incubated with 1 ml of iysis buffer (50 mM Tris-HCi, pH 8.0, 150 mM NaC1, 1% Triton X-100) containing 1/1,000 CLAP (10/~g/ml in DMSO of each chymostatin, leucopeptin, antipaln, and pepstatin), 1 mM PMSE and 1 ~g/rrd aprotinin for 30 min in ice with rocking. The cell lysate was collected and clarified by centrifugation at top speed in a microfuge for 5 rain at 4°C. 50 #1 of a 50% solution of protein A-Sepharose beads in lysis buffer were incubated with 1 or 2/~1 of either autoimmune serum or normal human serum for 2 h in ice with rocking, washed, collected by centrifugatiun, and added to 50/~1 of labeled samples. The mixture was rotated for 2 h at 4°C, then the protein A-Sepharose beads were collected and washed 5 times in lysis buffer followed by two washes in 50 mM Tris-HC1, pH 8.0. Proteins were released from the beads by boiling in SDS-PAGE sample buffer. Subcellular Fractionation Fractions enriched in Golgi vesicles were isolated from KE37 cells by flotation in a sucrose gradient as described in Rios et al. (1992). Prior to further processing, Golgi-enriched fractions were diluted by addition of several volumes of 10 mM Tris-HC1, pH 7.4, and protease inhibitors were added. The suspensions were centrifuged at 200,000 g for 2 h at 40C and pellets were used for subsequent experiments. Electrophoresis and lmmunoblot Analyses Proteins were separated by SDS-PAGE on 8% acrylamide gels (Laemmli, 1970) and stained with Coomassie blue. Two-dimensional gel electrophoresis of Golgi-enriched fractions was performed according to O'Farrell (1975) and stained with silver nitrate. Oneor two-dimensional gels were electrophoretically transferred to nitrocellulose filters according to Towbin et al. (1979). Nitrocellulose filters were blocked for 1 h at 37°C in TBST (10 mM Tris-HCl, pH 7.4, 150 mM NaCI, 0.1% Tween 20) containing 5% nonfat dry milk. Then filters were incubated for 1-2 h at 37°C in the primary antibody diluted in TBST, washed in the same buffer, and incubated for 45 min at 37°C with secondary anti-rabbit or anti-human IgG antibodies conjugated with either alkaline phosphatase (Promnga Corp., Madison, WI) or peroxidase (Catalg, San Francisco, CA). After washes, alkaline phosphatase activity was revealed with 0.1 M Tris-HCl, pH 9.5, 100 mM NaCi, 5 mM MgCI2 containing both nitroblue tetrazolium and 5-bromo-4-chioroindolyl phosphate. Peroxidase activity was revealed using the ECL system (Amersham Corp.). Membrane Extraction and Washing Procedures Golgi vesicles (50 tzg protein) were extracted with Triton X-114 as described by Bordier (1981). The detergent phase was taken up directly in SDS sample buffer; the aqueous phase was first concentrated by precipitation with 10% (wt/vol) TCA. Gulgi vesicles were also washed with 1 M NaC1, 0.2 M sodium carbonate, pH 11.0, or 6 M urea. After incubation for 30 rain on ice, the samples were centrifuged for 1 h at 200,000 g. The supernatants (precipitated with 10 % TCA) and pellets were processed for SDS-PAGE followed by immunoblotting with either autoimmune serum or anti-CralTf antibody. PermeabiUzation Experiments HeLa cells grown on glass slides at subconfluent densities were washed with PBS and buffer A (25 mM Hepes-KOH, 125 mM potassium acetate, and 2.5 mM magnesium acetate) and then incubated for different times with Streptolysin O (SLO; Burroughs Wellcome, Research Triangle Park, NC) prepared at varying concentrations in buffer A supplemented with 1 mM dithiothreitol. Cold buffer A was added to SLO-treated cells to stop the permeabilization process and cells were washed two times with buffer A containing 1% BSA. Permeabilized HeLa ceils were incubated with the autoimmune serum diluted 1/100 in buffer A containing 1% BSA. After that, cells were washed, fixed in methanol at -20°C, and incubated with anti-GalTf. As a control, an analogous experiment was performed in which permeabilized cells were incubated with the anti-GalTf antibody before fixation and thereaRer with the autoimmune serum. Permeabilization using 0.5 U/ml SLO for 7 rain at 37°C gave the best results and was routinely used. Electron Microscopy Staining of myotubes by immunoperoxidase and processing for electron microscopy was carried out according to Saraste et al. (1987). Briefly, cells were fixed in paraformaldehyde-lysine-periodate fixative for 4 h at room temperature, permeabilized with 0.05 % saponin, and incubated with antip210 IgG (1/100) overnight. Cells were then incubated with peroxidaseconjugated Fab fragments of sheep anti-human IgG for 60 min, followed by reaction with diaminobenzidine and Epon embedding. Results The Auto-immune RM Serum Decorates the Golgi Apparatus in a Veide Variety of Cells The serum from patient R.M., hereafter termed "RM serum; was assayed by indirect immunofluorescence in a variety of mammalian cells (HeLa, KE37, human myoblasts, NRK, and CHO). In all cases, RM serum showed a very high reactivity to the GA. A weaker nuclear staining could also be observed but disappeared at higher dilution of the serum. An example is presented in Fig. 1 where double staining of HeLa cells with RM serum and with a monospecific antibody directed against the trans-Golgi marker GalTf is shown (Fig. 1, A and B). Distributions of both antigens were virtually identical. In human myoblasts decorated with RM serum and with the monoclonal antibody CTR433, a medialRios et al. A Novel 210-kD cis-Golgi Network-associated Protein 999 on July 26, 2017jcb.rupress.orgDownloaded from Figure 1. Immunofluorescence microscopy of methanol-fixed HeLa cells (A and B), human myoblasts (C and D) stained with RM serum (A and C), and the Golgi markers anti-GalTf (B) and CTR 433 (D). In both cell types, RM serum gave a strong fluorescent signal in the Golgi region that colocalized with those shown by the Golgi markers. No staining of the ER or the cytoplasm was detected. A more detailed observation (A'-D') revealed that the staining patterns displayed by RM serum (A' and C') and GaiTf (B') or CTR 433 (D') were not identical (compare A' and B' and C' and D'). The Golgi staining of both Golgi markers was thinner and more uniform than that of RM serum which appeared patchy or vesicular and sometimes larger over the Golgi area. Bar, 5 /~m. Golgi marker (Jasmin et al., 1989), the colocalization of both antigens was also apparently perfect (Fig. 1, C and D). Despite the low resolution of immunofluorescence microscopy, a close examination of the staining patterns revealed some differences between Golgi markers and RM serum. Notably, the latter showed a rather discontinuous or granular appearance (see Fig. 1, A' and C'), and was often slightly larger than the staining obtained with Golgi markers. This was even more obvious for myoblasts in which the GA is more extended (Fig. 1, C' and D'). These observations suggested that RM serum could decorate structures tightly associated to those defined by reference Golgi markers. Structures Decorated by RM Serum Are Particularly Sensitive to MT-disruption Segregation between the staining observed with RM serum and that obtained with Golgi markers became evident when cells were treated with nocodazole (Fig. 2). After short treatments of nocodazole (up to 15-30 min), the morphology of the GA visualized with the anti-GalTf antibody was not significantly modified (Fig. 2 B). By contrast, numerous small vesicles, or tubules, decorated by RM serum appeared around the GA but also outside the Golgi region (Fig. 2 A). By 2-3 h of nocodazole treatment, the GA had completely fragmented and the Golgi elements had dispersed throughout the cytoplasm. In these conditions, staining patterns observed with RM serum and with anti-GalTf were only partially coincident, RM serum recognizing a number of elements that did not contain GalTf (Fig. 2, compare C and D). Furthermore, a detailed observation revealed that even in elements containing both antigens, the staining patterns were not identical either, RM serum decorating distinct domains often at the periphery of the GalTf-containing domains (see Fig. 2, E and F, arrows). From this set of experiments, we concluded that RM serum additionally decorates distinct structures closely asThe Journal of Cell Biology, Volume 125, 1994 I000 on July 26, 2017jcb.rupress.orgDownloaded from Figure 2. Effect of nocodazole on the distribution of RM serum autoantigen. HeLa cells were treated with 10 #M nocodazole for 15 min (A and B) and 180 min (C and D), fixed and double stained with autoimmune RM serum (A and C) and the anti-GalTf antibody (B and D). After 15 min of nocodazole treatment numerous vesicles and tubules staining with autoimmune RM serum but negative for GalTf began to distribute throughout the cytoplasm (A, arrows; this figure has been slightly underexposed to show the small tubules). After 3 h nocodazole, a number of discrete Golgi dements were observed with the anti-GalTf antibody, all of which (E, large arrows) were also labeled with autoimmune RM serum. Other membranous elements uniquely stained by RM serum were also apparent (E, circles with arrows). Note that labeling patterns of Golgi elements displayed by both antibodies were not identical, that of RM serum being again more extensive than the pattern produced by the anti-GalTf antibody. Bar, 5 t~m. sociated with the GA and highly dependent on microtubules for both their integrity and their association with the GA. RM Serum Reacts with a Golgi-enriched 210-kD Protein In order to characterize the autoantigen recognized by RM serum, the Triton-soluble fraction of metabolically labeled HeLa cells (see Materials and Methods) was used for immunoprecipitation. Immune complexes were analyzed by SDSPAGE and fluorography. Three major bands of 210, 130, and 45 kD were immunoprecipitated by RM serum (Fig. 3 A, lane /). Several other bands were also occasionally observed. In control experiments using normal human serum these bands were not detected (Fig. 3 A, lane 2). A complex Rios et al. A Novel 210-kD cis-Golgi Network-associated Protein 1001 on July 26, 2017jcb.rupress.orgDownloaded from Figure 3. Autoimmune serum recognizes a 210-kD Golgi protein. Immunoprecipitation and Western blotting were used to identify the Golgi antigen of the autoimmune serum. (A) HeLa cells were metabolically labeled with 35S-Trans-Label and detergent extracts of cells were used for immunoprecipitation. Autoimmune RM serum (AS) immunoprecipitated three major proteins of 210, 130, and 45 kD that did not appear in immunoprecipitates using normal human serum (NHS). (B) KE37 cells were fractionated and a Golgienriched fraction prepared by flotation in a sucrose gradient. The enrichment of different subcellular fractions in Golgi membranes was tested by Western blotting using an anti-GalTf antibody. Proteins from total homogenate (H), nuclear pellet (N), post-nuclear supernatant (PNS), and the Golgi-rich fraction were separated by SDS-PAGE, transferred to nitrocellulose filters, and assayed for their content in GalTf. (C) A blot with identical samples to those in (B) was incubated with autoimmune RM serum and bound antibody was detected by chemiluminescence. Proteins of 200, 130, and 45 kD were the major proteins observed in homogenate, nuclear pellet, or postnuclear supernatant. In Golgi-enriched fractions only a 210-kD band was recognized by autoimmune RM serum; in some experiments a lighter band with a slightly faster mobility was also detected. (D) The pellet obtained by high-speed centrifugation of the postnuclear supernatant, which was highly enriched in proteins of 210 and 130 kD, as judged by Western blotting with RM serum and devoid of any 45-kD signal (lane/), was used to affinity-purify RM serum on individual bands. When assayed on Golgi-enriched fraction (lanes 2 and 3), p210 affinity-purified immunoglobulins reacted only with the 210-kD band (lane 2) whereas reactive pattern was indeed expected with an autoimmune serum, and in order to ascertain which of these bands corresponded to the Golgi autoantigen, we turned to human lymphoblasts from the KE37 cell line, which provided us, in a reproducible manner, with a Golgi-enriched fraction prepared by flotation on a sucrose gradient (see Rios et al., 1992). Total homogenate, nuclear pellet, postnuclear supernatant (75 #g), and Golgi-enriched fraction (50 #g) were independently examined for their content in Golgi membranes by Western blotting, using an afffinity-purified anti-GalTf antibody (Fig. 3 B): only the Golgi-enriched fraction contained detectable amounts of GalTf. An identical nitrocellulose filter was incubated with RM serum (Fig. 3 C). Reactive bands with apparent molecular masses of 200, 130, and 45 kD were the major components detected in the nuclear pellet or in the postnuclear supernatant of KE37 cells. However, a different pattern was observed in Golgi-enriched fractions: a 210-kD protein, that sometimes appeared as a doublet, was highly reactive (Fig. 3 C). These results strongly suggested that the 210-kD band was the Golgi autoantigen. Definitive demonstration of this was obtained by affinity purifying RM serum on individual reactive bands after electrophoretic separation and transfer onto nitrocellulose filters (Fig. 3 D). As a source of antigens, we used a membrane-enriched fraction obtained by high-speed centrifugation of the postnuclear supernatant: the 210and the 130-kD bands were highly enriched in this pellet and were the only reactive bands with RM serum (Fig. 3 D, lane/), the 45-kD band turning out to be soluble (not shown). Antibodies retro-eluted from each band were further tested on a Golgi-enriched fraction: the 210-kD band was revealed by p210 affinity-purified antibodies (Fig. 3 D, lane 2) whereas the p130 affinity-purified antibodies did not reveal any component in this fraction (Fig. 3 D, lane 3) although it revealed the p130 in the membraneenriched postnuclear fraction (not shown). Moreover, when assayed by immunofluorescence, p210 affinity-purified antibodies produced a strong Golgi staining pattern on HeLa cells (Fig. 3 E), whereas p130 affinity-purified antibodies did not give any Golgi staining nor any obvious other membrane compartment (not shown). Finally, proteins from the Golgi-enriched fraction were resolved by two-dimensional electrophoresis and were either stained with silver nitrate or transferred to nitrocellulose filters and incubated with the autoimmune serum (Fig. 4). The serum recognized one spot of molecular mass >200 kD and pI 5.6 (Fig. 4 A) which corresponded to an abundant protein in silver-stained gels (Fig. 4 B). The 210-kD Protein Is a Peripheral Membrane Protein The membrane localization of the 210-kD protein was investigated by using different extraction procedures. Isolated Golgi membranes (50 #g) were extracted with 2 % Triton X-114 (the most commonly employed method for separating hydrophilic and hydrophobic proteins) or washed with 6 M the pl30 affinity-purified immunoglobulins did not reveal any band (lane 3). The MW markers were stained by Ponceau red S. (E) p2 l0 affinity-purified immunoglobulins revealed the GA when used for immunofluorescence on HeLa cells after methanol fixation. Bar, 10 #m. The Journal of Cell Biology, Volume 125, 1994 1002 on July 26, 2017jcb.rupress.orgDownloaded from Figure 4. The Golgi-enriched fraction was resolved by twodimensional electrophoresis. Two-dimensional gels were either transferred to nitrocellulose filters and incubated with autoimmune RM serum (.4) or stained with silver nitrate (B). Autoimmune RM serum recognized a spot of molecular mass 210 kD and pI 5.6. urea, 1 M NaC1, and 0.2 M sodium carbonate, pH 11.0. Supernatants and pellets of different treatments and aqueous and detergent phases of Triton X-114 extraction were analyzed by SDS-PAGE and transferred to nitrocellulose filters. The upper halves of filters were processed for immunolabeling with RM serum (Fig. 5 A). The lower halves of Western blots were immunolabeled with anti-GalTf antibody to monitor the different extraction procedures (Fig. 5 B). GalTf is an integral membrane protein disposed at the luminal face of the trans cisternae. As expected, GalTf remained associated to membranes after high salt, high pH, and urea treatments and partitioned in the detergent phase of Triton X-114 extraction. In contrast, the 210-kD protein entirely partitioned in the aqueous phase of Triton X-114 extraction and became soluble upon all other treatments, indicating that Figure 5. p210 is an external and peripheral protein in Golgi membranes. Golgi membranes were treated with 2% Triton X-I14 (TX114) as described by Bordier et ai. (1981) and both the aqueous (Aqu) and detergent (Det) phases were separated by SDS-PAGE and transferred to nitrocellulose filters. Other Golgi membranes were treated with 6 M urea, 1 M NaCI, and 200 mM sodium carbonate (pH 11.0), and after centrifugation, the resulting supernatants (S) and pellets (P) were processed as described above. The upper half of the filter was assayed by immunoblotting to detect p210 (A) and the lower half was incubated with anti-GalTf antibody as a control for a Golgi integral membrane protein (B). (C-F) HeLa cells were permeabilized using 0.5 U/m1 SLO for 7 min at 37°C and then incubated for I h with either autoimmune RM serum (C) or the antiGalTf antibody (E). After washes, permeabilized cells were fixed and stained for GalTf (D) or p210 (F), respectively. No staining of the GA was observed in permeabilized cells incubated with antiGalTf indicating that Golgi membranes remained intact after permeabilization treatment. Bar, 5 gm. Rios et al. A Novel 210-kD cis-Golgi Network-associated Protein 1003 on July 26, 2017jcb.rupress.orgDownloaded from p210 protein is a peripheral cytoplasmically disposed membrane protein. A different approach was also undertaken to confirm the localization of p210 in Golgi membranes. Semi-intact Hela cells were prepared using the bacterial toxin SLO to selectively permeabilize the plasma membrane to allow the access of antibodies to cytoplasm while maintaining the morphological integrity of the GA. In addition, in cells perforated with SLO, the GA and most intracellular organelles were shown to be well preserved at light and electron microscopic level (Miller and Moore, 1991). HeLa cells were incubated with 0.2-2 U/ml SLO for 5-10 min at 37°C, washed, and incubated with RM serum for 1 h. Then, cells were washed, fixed, and stained for GalTf. Antibodies bound before and after fixation were visualized using fluorescein and rhodamine secondary antibodies, respectively (Fig. 5, C and D). Treatment with 0.5 U/ml SLO for 7 min at 37°C gave the best results: more than 80% of the cells were permeabilized and showed Golgi staining after incubation with RM serum for 1 h (Fig. 5 C). By contrast, no GA staining was observed in permeabilized cells incubated with anti-GalTf antibody before fixation (Fig. 5 E), indicating that the GA itself was not perforated by SLO and remained impermeant to antibody molecules. p210 Is Concentrated at the Cis Side of the GA In an attempt to localize p210 at the ultrastructural level, we made use principally of primary cultures obtained from human muscle biopsies that are capable of producing in vitro myogenesis. Although this system was not tractable for immunoelectron cryomicroscopy, therefore imposing localization by immuno-peroxidase technique, it had two significant advantages: (a) it provided us with mononucleated cells displaying a highly extended GA; it was in these cells that RM serum decorated the GA with the most granular aspect (see Fig. 1, C and C'), suggesting that one could hopefully identify the reactive structures within the GA; and (b) the GA in differentiating myotubes displays a characteristic perinuclear distribution, in which the cis-trans polarity is easily identifiable from the nuclear periphery towards the plasma membrane (Tassin et al., 1985b). In mononucleated cells p210 demonstrated a clear asymmetrical distribution on one side of the GA (Fig. 6), the labeling being externally associated with budding or tubular processes in zones which, from their aspect, could correspond to the so-called "non-compact zones" described by Rambourg and Clermont (1990) (Fig. 6, A and B, curved arrows). Another constant feature was that face views of Golgi membranes demonstrated a discontinuous appearance of the labeling, corresponding to more or less regularly spaced small cavities (Fig. 6 A, straight arrows), which could correspond to the granular aspect observed at the optical level. Higher magnification suggested that these cavities corresponded to large perforations of the saccules of the mid compartment described as "wells" by Rambourg and Clermont (1990) in the so-called "compact zones" of the GA. A face and a side view from the same Golgi area are indicated in Fig. 6 B (straight arrows). By immunofluorescence microscopy, young myotubes were decorated by RM serum in a manner strikingly similar to that obtained with any Golgi marker, i.e., essentially as a perinuclear ring (Fig. 7 A; for comparison see Tassin et al., 1985b). One could, however, observe a rather punctuated aspect of the perinuclear GA when decorated with RM serum (Fig. 7A, arrows). Ultrastructural localization demonstrated that p210 was associated with tubulovesicular structures at the cis side of the GA, and next to the nuclear envelope, which could penetrate deeply between saccules, the precipitate being detected between the lateral sides of the Golgi stacks. In these perforations, the labeling appeared abundant at the edge of the cisternae (Fig. 7 C, arrows). Overall, despite the difficulty of precisely deciding the extent of the peroxidase labeling, the observation of numerous sections led us to conclude that p210 could interact with the external side of an abundant tubulovesicular system on the cis side of the Golgi, and that this system could encompass, or extend to, connecting structures or vesicles between individual saccules and dictyosomes. p210 Redistributes in the Intermediate Compartment upon BFA Treatment BFA treatment of cells results in the rapid loss of the Golgi as a distinct organelle due to the redistribution of Golgi content and membranes into the ER. Double immunofluorescence staining with RM serum and with the trans-Golgi marker GalTf was performed on BFA-treated HeLa cells to determine the fate of p210. Within 5 min of BFA treatment (1 pg/ml), tubular processes extending out of enlarged Golgi structures were visible with the anti-GalTf antibody (Fig. 8 B). RM serum showed a similar pattern although the tubular extensions were less evident (Fig. 8 A). After 30 rain of BFA treatment, in most cells GalTf was redistributed in a fine reticular pattern characteristic of ER labeling (Fig. 8 D). By contrast, p210 was present in numerous small vesicles scattered throughout the cytoplasm (Fig. 8 C), a pattern that remained unchanged upon longer incubations with BFA. Under these conditions the staining patterns displayed by GalTf and p210 were completely distinct whereas they were congruent in control cells. The effect of BFA on p210 was entirely reversible. HeLa cells treated with BFA for 1 h were allowed to recover during 10 or 30 min in BFA-free medium. 10 min after removal of the drug, p210-containing vesicles appeared larger and tubular structures extending towards a perinuclear region could be observed (Fig. 8 E). After 30 min of recovery, the spotty staining was greatly reduced and p210 was concentrated in a compact structure close to the nucleus (Fig. 8 G). The relocalization of GalTf was significantly slower although some vesicles could be observed 10 rain after removal of the drug (Fig. 8 F). These vesicles always colocalized with p210-containing elements, suggesting that movement of proteins out of the ER to the GA occurred through these elements. After 30 min of BFA recovery, both proteins colocalized again (Fig. 8 H). The time course of the relocalization of both proteins strongly suggested that p210 elements were involved in the repositioning of the Golgi complex after removal of BFA, a possibility which would fit with the localization of p210 in the CGN. Until now, three proteins have been described to exhibit a similar staining pattern after BFA treatment: p58 (Saraste and Svensson, 1991), p53 (Lippincott-Schwartz et al., 1990), and p23 (Luen Tang et al., 1993), all of which reside in the intermediate compartment. To determine the nature of the spotty staining displayed by p210, we compared the localizaThe Journal of Cell Biology, Volume 125, 1994 1004 on July 26, 2017jcb.rupress.orgDownloaded from Figure 6. Ultrastructural localization of p210 in mononucleated cells from human myogenic cultures, using indirect immunoperoxidase techniques. At low magnification (A), the (3A displays a large spatial extension on one side of the nucleus (the nuclear envelope is indicated in A, NE) and surrounds the centrosome (the two centrioles are indicated in A, C). Side views of the Golgi elements demonstrate that p210 is associated with one side of the GA, where particularly tubular membrane extensions, or vesicles, appear labeled (curved arrows). Face views demonstrate numerous and rather even-spaced spots of regular size (straight arrows) which look like cavities. On higher magnification (B), and although ultrathin sections do not allow one to easily appreciate the overall architecture of the GA, one can recognize such cavities on face view (horizontal straight arrow) and tentatively identify them on side views (vertical straight arrow). They apparently correspond to perforations in register, or wells, which often interrupt the mid saccules (Rambourg and Clermont, 1990) and their decoration by RM serum could be responsible for the granular aspect of immunofluorescent staining at the light level (see Fig. 1). Curved arrows in B point to labeled budding elements. on July 26, 2017jcb.rupress.orgDownloaded from preparation) and in those which mediate the reconstruction of the GA after BFA treatment (this work). In addition, p210 appears to be associated with the CGN, a tubular-cisternal network, and presumably with tubules connecting different Golgi stacks. From these results there emerges a relation of this protein with the formation of microtubule-dependent membrane tubules or with processes mediated by them. Since human autoantibodies are usually capable of inhibiting the functional activity of the antigens, the permeabilization experiments used in this work (Fig. 5) represent a first step to directly examine the function of p210 in the cell. Preliminary data from this type of functional study using SLO-permeabilized cells have revealed that RM immunoglobulins block nocodazole-induced fragmentation of the GA, possibly by stabilizing membrane tubules connecting different Golgi stacks. In addition, p210 was found to be enriched in preparations of Taxol-induced microtubules from KE37 and HeLa cells (Rios, R. M., and M. Bornens, unpublished resuits). The possibility of a direct interaction of p210 with microtubules is now being investigated. This would well agree with the fact that this protein is peripheral. Moreover, together with the behavior of p210 during BFA treatment, this fits nicely with the known involvement of the intermediate compartment and microtubules in the biogenesis of the Golgi complex after BFA treatment. We have shown that p210 cycles between the CGN and the intermediate compartment by effect of BFA. Does p210 cycle similarly under normal conditions? Although more experiments are necessary, our preliminary data showing localization of p210 in tubules that emanate from the Golgi complex at 16°C would suggest that it does. In that case, it must be rapidly recycled to the CGN since we have been unable to detect it in the intermediate compartment. Therefore, the intermediate compartment could serve to recycle not only proteins moving out from ER but also from the GA to their original locations. Three other autoantigens have been previously reported to be associated with the Sj6gren's syndrome: a component of ribonucleoprotein particles of mol wt 60,000, a 48-kD protein that associates to nascent transcript (Tan, 1991), and a p230 localized in the Golgi complex (Kooy et al., 1992). This p230 Golgi protein exhibits some characteristics similar to our p210 autoantigen, but also some marked differences: first, the p230 is associated with the trans cisternae of the GA, and second, BFA induces the dissociation of p230 from the Golgi complex to cytosol. A 200-kD Golgi protein has also been reported by Narula et al. (1992). This protein is not cisternae specific but accumulates to dilated rims of cisternae and on vesicles scattered in the Golgi region. Like p230, p200 redistributes in the cytoplasm upon BFA action, but both proteins differ in the kinetics of BFA action. In conclusion, p210 appears to be a novel Golgi protein which might be a useful marker in deciphering the dynamics of the ER-Golgi pathway. Morphological criteria have demonstrated that p53 and p58 cycle constitutively between the ER, the intermediate compartment, and the CGN. This dynamic behavior together with the lack of a marker for the CGN has hindered the comprehension of structural and functional relationships between these three membranebound compartments. Recently, a novel protein, p63, that resides in the intermediate compartment and does not cycle between ER and the CGN has been identified (Schweizer et al., 1993). Here, we have identified a protein residing in the CGN that cycles between this structure and the intermediate compartment, at least in the presence of BFA. We think that both proteins, p63 and p210, represent useful tools to clarify the intimate relationships between the intermediate compartment and the GA. In addition, p210 might be an important protein for the microtubule-dependent organization of the GA. We thank Dr. J. C. Courvalin (Institut Jacques Monod, Paris, France) for his help at the beginning of this work, and Dr. S. Brown (Institut de Sciences Vegetales, Gif/Yvette, France) as well as Drs. E. Coudrier and B. Goud (Institut Pasteur, Paris, France) for stimulating discussions and critical reading of the manuscript. We are in debt to Drs. J. Saraste, D. Louvard, and E. Berger for the gift of antibodies against p58, endoplasmic reticulum, and galactosYl transferase, respectively. We wish to thank M. Fardeau for providing muscle biopsies. We thank D. Meur and I. Gaspar for photographic work and N. Bordes for her expert assistance. We are also grateful to Dr. C. Fedriani for her material support, stimulating discussions, and critical reading of the manuscript. This work has been supported by Centre National de la Recherche Scientifique and by a grant from the Ministerio de Sanidad y Consumo of Spain (FIS 93/0824) to R. M. Rios and one from Association Fran~aise contre les Myopathies (954817) to A. M. Tassin. Received for publication 3 September 1993 and in revised form 22 February 1994. References Alcalde, J., P. Bonay, A. Roa, S. Vilaro, and I. V. Sandoval. 1992. Assembly and disassembly of the Golgi complex: two processes arranged in a cis-trans direction. J. Cell Biol. 116:69-83. Antony, C., C. Cibert, G. Geraud, A. Santa Maria, B. Maro, V. Mayau, and B. Goud. 1992. The small GTP-binding protein rab6p is distributed from medial Golgi to the trans-Golgi network as determined by a confocal microscopy approach. J. Cell Sci. 103:785-796. Barr, F. A., A. Leyte, S. MoUner, T. Pfeuffer, S. A. Tooze, and W. B. Huttner. 1991. Trimeric G-proteins of the trans-Golgi network are involved in the formarion of constitutive secretory vesicles and immature secretory granules. FEBS (Fed. Fur. Biol. Soc.) Left. 294:239-243. Blaschek, M. A., Y. L. Pennec, A. M. Simitzis, P. Le Goff, A. Lamour, G. Kerdraon, J. Jouquan, and P. Youinou. 1988. Anti-Golgi complex autoantibodies in patients with primary Sj6gren's syndrome. Scand. J. Rheumatol. 17:291-296. Bordier, C. 1981. Phase separation of integral membrane proteins in Triton X-l14 solutions. J. Biol. Chem. 256:1604-1607. Booth, C., and G. L. E. Koch. 1989. Perturbation of cellular calcium induces secretion of luminal ER proteins. Cell. 59:729-737. Bourne, H. R. 1988. Do GTPases direct membrane traflSc in secretion? Cell. 53:669-671. Cooper, M. S., A. H. Cornell-Bell, A. Chernjavsky, J. W. Dani, and S. J. Smith. 1990. Tubulovesicular processes emerge from trans-Golgi cisternae, extend along microtubules and interlink adjacent trans-Golgi elements into a reticulum. Cell. 61:135-145. Donaldson, J. G., J. Lippincot-Schwartz, G. S. Blom, T. E. Kreis, and R. D. Klansner. 1990. Dissociation ofa 110-kD peripheral membrane protein from the Golgi apparatus is an early event in brefeldin A action. J. Cell Biol. 111:2295-2306. Donaldson, J. G., R. A. Kahn, J. Lippineott-Schwartz, and R. D. Klausner. 199 la. Binding of ARF and B-COP to Golgi membranes: possible regulation by a trimeric G protein. Science (Wash. DC.). 254:1197-1199. Donaldson, J. G., J. Lippincott-Schwartz, and R. D. Klausner. 1991b. Guanine nucleotides modulate the effects of brefeldin A in semipermeable cells: regulation of the association of a 110-kD peripheral membrane protein with the Golgi apparatus. J. Cell Biol. 112:579-588. Donaldson, J. G., D. Cassel, R. A. Kahn, and R. D. Klausner. 1992a. ADPribosylation factor, a small GTP-binding protein, is required for binding of the coatomer protein B-COP to Golgi membranes. Proc. Natl. Acad. Sci. USA. 89:6408-6412. Donaldson, J. G., D. Finazzi, and R. D. Klansner. 1992b. Brefeldin A inhibits Golgi membrane-catalysed exchange of guanine nucleotide onto ARF protein. Nature (Lond.). 360:350--352. Ercolani, L., J. L. Stow, J. F. Boyle, E. J. Holtzmann, H. Lin, J. R. Grove, and D. A. Ausiello. 1990. Membrane localization of the pertussis toxinsensitive G-protein subunit cd-2 and cd-3 and expression of a metallothionein-txi-2 fusion gene in LLC-PK1 cells. Proc. Natl. Acad. Sci. USA. 87: 4637--4639. The Journal of Cell Biology, Volume 125, 1994 1012 on July 26, 2017jcb.rupress.orgDownloaded from Farquhar, M. G., and G. E. Palade. 1981. The Golgi apparatus (complex)- (1954-1981)-from artifact to center stage. J. Cell Biol. 91:77-103. Fritzler, M. J., J. Etherington, C. Sokoluk, T. D. KinseUa, and D. W. Valanca. 1984. Antibodies from patients with autoimmune diseases react with a cytoplasmic antigen in the Golgi apparatus. J. lmmunol. 132:2904-2908. Gaspar, M. L., M. A. R. Marcos, C. Gutierrez, M. J. Martin, J. S. Bonifacino, and I. V. Sandoval. 1988. Presence of an autoantibody against a Golgi cisternal membrane protein in the serum and cerebrospinal fluid from a patient with idiopathic late onset cerebellar ataxia. J. Neuroimmunol. 17:287-299. Goud, B., and M. McCaffrey. 1991. Small GTP-binding proteins and their role in transport. Curr. Opin. Cell Biol. 3:626-633. Hauri, H. P., and A. Schweizer. 1992. The endoplasmic reticulum-Golgi intermediate compartment. Curr. Opin. Cell Biol. 4:600-608. Helms, J. B., and J. E. Rothman. 1992. Inhibition by Brefeldin A of a Golgi membrane enzyme that catalyses exchange of guanine nucleotide bound to ARF. Nature (Lond.). 360:352-354. Hsu, W. H., N. Shah, and R. D. Klansner. t992. A brefeldin-like phenotype is induced by the overexpression of a human ERD-2 protein, ELP-1. Cell. 69:625-635. Jasmin, B. J., J. Cartaud, M. Bornens, and J. P. Changeux. 1989. Golgi apparatus in chick skeletal muscle: changes in its distribution during end plate development and after denervation. Proc. Natl. Acad. Sci. USA. 86:72187222. Klansner, R. D., J. G. Donaldson, andJ. Lippincott-Schwartz. 1992. Brefeldin A: insights into the control of membrane traffic and organdie structure. J. Cell Biol. 116:1071-1080. Kooy, J., B. H. Toh, J. M. Pettitt, R. Erlicb, and P. A. Gleeson. 1992. Human autoantibodies as reagents to conserved Golgi components. Characterization of a peripheral, 230-kDa compartment-specific Golgi protein. J. Biol. Chem. 267:20255-20263. Krohne, G., R. Stick, J. A. Kleinschmidt, R. Moll, W. W. Franke, and P. Hansen. 1982. Immunolocalization of a major karyoskeletal protein in nucleoli of oocytes and somatic cells of Xenopus laevis. J. Cell Biol. 94:749-754. Ktistakis, N. T., M. E. Linder, M. G. Roth. 1992. Action of brefeldin A blocked by activation of a pertussis-toxin-sensitive G protein. Nature (Lond.). 356:344-346. Kupfer, A., D. Louvard, and J. Singer. 1982. Polarization of the Golgi apparatus and the microtubule-organizing center in cultured fibroblasts at the edge of an experimental wound. Proc. Natl. Acad. Sci. USA. 79:2603-2607. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lond.). 227:680-685. Lippincott-Schwartz, J. 1993. Bidirectional membrane traffic between the endoplasmic reticulum and Golgi apparatus. Trends Cell Biol. 3:81-88. Lippincott-Schwartz, J., J. G. Donaldson, A. Schweizer, E. G. Berger, H. P. Hanri, L. C. Yuan, R. D. Klausner. 1990. Microtubule-dependent retrograde transport of proteins into ER in the presence of brefeldin A suggests an ER recycling pathway. Cell. 60:821-836. Louvard, D., H. Reggio, and G. Warren. 1982. Antibodies to the Golgi complex and the rough endoplasmic reticulum. J. Cell Biol. 92:92-106. Lueng-Tang, B., S. H. Wong, X. L. Qi, S. H. Low, andW. Hong. 1993. Molecular cloning, characterization, subcellular localization, and dynamics of p23, the mammalian KDEL receptor. J. Cell Biol. 120:325-338. Malhotra, V., T. Serafini, L. Orci, J. C. Sheperd, andJ. E. Rothrmm. 1989. Purification of a novel class of coated vesicles mediating biosynthetic protein transport through the Golgi stack. Cell. 58:329-336. Mellman, I., and K. Simons. 1992. The Golgi Complex: in vitro veritas? Cell. 68:824-840. Miller, S. G., and H. P. H. Moore. 1991. Reconstitution of constitutive secretion using semi-intact cells: regulation by GTP but not calcium. J. Cell Biol. 112:39-54. Narula, N., I. McMorrow, G. Plopper, J. Doberty, K. S. Matlin, B. Burke, and J. L. Stow. 1992. Identification ofa 200-kD, brefeldin-sensitive protein on Golgi membranes. J. Cell Biol. 117:27-38. O'Farrell, P. H. 1975. High resolution two dimensional electrophoresis of proteins. J. Biol. Chem. 250:4007-4021. Orci, L., B. S. Glick, and J. E. Rothman. 1986. A new type of coated vesicular carders that appear not to contain clathrin: its possible role in protein transport within the Golgi stack. Cell. 46:171-184. Orci, L., M. Tagaya, M. Amherdt, A. Perrelet, J. G. Donaldson, J. LippincottSchwartz, R. D. Klausner, and J. E. Rorhrnan. 1991. Brefeldin A, a drug that blocks secretion, prevents the assembly of non-clathrin-coated buds on Golgi cisternae. Cell. 64:1183-1195. Palade, G. E. 1975. Intracellular aspects of the process of protein sorting. Science (Wash. DC). 189:347-358. Pavelka, M., and A. Ellinger. 1983. Effect of colchicine on the Golgi complex of rat pancreatic acinar cells. J. Cell Biol. 97:737-748. Pfeffer, S. R. 1992. GTP-binding proteins in intracellular transport. Trends Cell Biol. 2:41-45. Rarnbourg, A., and Y. Clermont. 1990. Three-dimeasional electron microscopic structure of the Golgi apparatus. Eur. J. Cell Biol. 51:189-200. Rios, R. M., C. Celati, S. M. Lohrnann, M. Bornens, and G. Keryer. 1992. Identification of a high affinity binding protein for the regulatory subunit RII~ of cAMP-dependent protein kinase in Golgi enriched membranes of human lympboblasts. EMBO (Fur. Mol. Biol. Organ.) J. 11:1723-1731. Robbins, E., and N. K. Gonatas. 1964. The uitrastructure of a mammalian cell during mitotic cycle. J. Cell Biol. 21:429--463. Rodriguez, J. L., C. Gelpi, M. Thompson, E J. Real, andJ. Fernandez. 1982. Anti-Golgi complex autoantibodies in a patient with Sj6gren's syndrome and lymphoma. Clin. Exp. lmmunol. 49:579-586. Rogalski, A. A., and S. J. Singer. 1984. Association of the Golgi elements with microtubules. J. Cell BioL 99:1092-1100. Rogalski, A. A., J. E. Bergmann, and S. J. Singer. 1984. Effect of microtubule assembly status on the intracellular processing and surface expression of an integral protein of the plasma membrane. J. Cell Biol. 99:1101-1109. Sandoval, I. V., J. S. Bonifacino, R. D. Klausner, M. Henkart, and J. Wehiand. 1984. Role of microtubules in the organization and localization of the Golgi apparatus. J. Cell Biol. 99:l13s-l18s. Saraste, J., and K. Svensson. 1991. Distribution of the intermediate elements operating in ER to Golgi transport. J. Cell Sci. 100:415--430. Saraste, J., G. E. Palade, and M. G. Farquhar. 1987. Antibodies to rat pancreas Golgi subfractions: identification of a 58-kD cis-Golgi protein. Z Cell Biol. 105:2021-2029. Schweizer, A., M. Ericsson, T. Biichi, G. Griffiths, and H. P. Hauri. 1993. Characterization of a novel 63 kDa membrane protein. Implications for the organization of the ER-to-Golgi pathway. J. Cell Sci. 104:671-683. Serafini, T., G. Stenbeck, A. Brecht, E Lottspeich, L. Orci, J. E. Rothman, and E T. Wieland. 1991a. A coat subunit of Golgi-derived non-clathrin-coated vesicles with homology to the clathrin-coated vesicle coat protein ~-adaptin. Nature (Lond.). 349:215-220. Serafini, T., L. Orci, M. Amhert, M. Brurmer, R. A. Kahn, andJ. E. Rothman. 1991b. ADP-ribosylation factor is a subunit of the coat of Golgi-derived COP-coated vesicles: a novel role for a GTP-binding protein. Cell. 67: 239-253. Stow, J. L., J. Bruno de Almeida, N. Narula, E. J. Holtzman, L. Ercolani, and D. A. Ausiello. 1991. A heterotrimeric G protein, God-3, on Golgi membranes regulates the secretion of heparan sulfate proteoglycan in LLC-PK1. J. CellBiol. 114:1113-1124. Tan, E. M. 1991. Autoantibodies in Pathology and Cell Biology. Cell. 67:841-842. Tassin, A. M., B. Maro, and M. Bornens. 1985a. Fate of microtubuleorganizing centers during myogenesis in vitro. J. Cell Biol. 100:35--46. Tassin, A. M., M. Paintrand, E. Berger, and M. Bornens. 1985b. The Golgi apparatus remains associated with microtubule-organizing centers during myogenesis. J. Cell Biol. 101:630-638. Thyberg, J., and S. Moskalewski. 1985. Microtubules and the organization of the Golgi complex. Exp. Cell Res. 159:1-16. Towbin, H., T. Staehlin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gel to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. USA. 76:4350-4354. Turner, J. R., and A. M. Tartakoff. 1989. The response of the Golgi complex to microtubule alterations: the roles of metabolic energy and membrane traffic in Golgi complex organization. J. Cell Biol. 109:2081-2088. Waters, M. G., T. Serafini, and J. E. Rothman. 1991. "Coatomer': a cytosolic protein complex containing subunits of non-clathrin-coated Golgi transport vesicles. Nature (Lond. ). 349:248-251. Wehland, J., and M. C. Willingham. 1983. A rat monoclonal antibody reacting specifically with the tyrosylated form of u-tubulin. IL Effects on cell movement, organization of microtubule and intermediate filaments, and arrangements of Golgi elements. J. Cell Biol. 97:1476-1490. Wehland, J., M. Henkart, R. D. Klausner, and I. V. Sandoval. 1983. Role of microtubules in the distribution of the Golgi apparatus: effect of taxol and microinjected anti-ot-tubulin antibodies. Proc. Natl. Acad. Sci. USA. 80: 4286-4290. Yuang, L., J. G. Barriocanal, J. S. Bonifacino, and I. V. Sandoval. 1987. Two integral membrane proteins located in the cis-middle and trans-part of the Goigi system acquire sialylated N-linked carbohydrates and display different turnovers and sensitivity to cAMP-dependent phosphorylation. J. Cell Biol. 105:215-227. Zuber, C., J. Roth, T. Misteli, A. Nakano, and K. Moremen. 1991. DS28-6, a temperature-sensitive mutant of Chinese hamster ovary cells, expresses key phenotypic changes associated with brefeldin A treatment. Proc. Natl. Acad. Sci. USA. 88:9818-9822. Rios et al. A Novel 210-kD cis-Gol~i Network-associated Protein 1013 on July 26, 2017jcb.rupress.orgDownloaded from