HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Linstedt, A. D. Articles by Sztul, E. Search for Related Content PubMed PubMed Citation Articles by Linstedt, A. D. Articles by Sztul, E. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 14, 10196-10201, April 7, 2000

Binding Relationships of Membrane Tethering Components
THE GIANTIN N TERMINUS AND THE GM130 N TERMINUS COMPETE FOR BINDING TO THE p115 C TERMINUS^*

Adam D. Linstedt^§, Stephen A. Jesch, Amy Mehta, Tina H. Lee, Rafael Garcia-Mata^¶, David S. Nelson^¶, and Elizabeth Sztul^¶

From the  Department of Biological Sciences, Carnegie Mellon University, Pittsburgh, Pennsylvania 15213 and the ^¶ Department of Cell Biology, University of Alabama Medical School, Birmingham, Alabama 35294

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

By forming a molecular tether between two membranes, p115, giantin, and GM130 may mediate multiple Golgi-related processes including vesicle transport, cisternae formation, and cisternal stacking. The tether is proposed to involve the simultaneous binding of p115 to giantin on one membrane and to GM130 on another membrane. To explore this model, we tested for the presence of the putative giantin-p115-GM130 ternary complex. We first mapped p115-binding site in giantin to a 70-amino acid coiled-coil domain at the extreme N terminus, a position that may exist up to 400 nm away from the Golgi membrane. We then generated glutathione S-transferase (GST) fusion proteins containing either giantin's or GM130's p115 binding site and tested whether such proteins could bind p115 and GM130 or bind p115 and giantin, respectively. Unexpectedly, GST fusions containing either the giantin or the GM130 p115 binding site efficiently bound p115, but the p115 bound to GST-giantin did not bind GM130, and the p115 bound to GST-GM130 did not bind giantin. To explain this result, we mapped the giantin binding site in p115 and found that it is located at the C-terminal acidic domain, the same domain involved in binding GM130. The presence of a single binding site in p115 for giantin and GM130 was confirmed by demonstration that giantin and GM130 compete for binding to p115. These results question a simple tethering model involving a ternary giantin-p115-GM130 complex and suggest that p115-giantin and p115-GM130 interactions might mediate independent membrane tethering events.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Selective docking prior to membrane fusion ensures that transport intermediates only fuse efficiently at appropriate destinations. Arguably, the importance of vesicle docking is reflected by the number and conservation of distinct protein complexes involved in this process (1-3). These complexes probably participate in coupled sequential interactions between vesicle and target components. The first of these sequential interactions may be the formation of a tether. Fibers, up to 100 nm in length, that appear to connect 100-nm diameter coat protein I (COPI)¹-coated vesicles to the Golgi are evident in electron micrographs of isolated Golgi after incubation under transport conditions (4), suggesting that vesicles might be tethered to the Golgi by interacting with fibrous membrane proteins. As vesicles form, tethers might anchor them to the site of vesicle formation, thus limiting their diffusion away from the Golgi stack and promoting their interaction with nearby cisternae. Alternatively, tethers could trap vesicles formed at distant sites once they have migrated to the Golgi region and anchor them to the site of vesicle docking and fusion. In both cases, tether formation would have important consequences for vesicle transport to or through the Golgi stack in either the anterograde or the retrograde direction (5).

The membrane transport factor p115 has been proposed to mediate tethering of COPI vesicles to the Golgi membrane (6). p115 is a peripheral Golgi membrane protein (7) associated primarily with cis-Golgi elements including the intermediate compartment (8, 9). p115 is required for endoplasmic reticulum to Golgi traffic in semi-intact cells (10) and for in vitro assays measuring intra-Golgi transport (7), binding of transcytotic vesicles to acceptor membranes (11), regrowth of Golgi cisternae from post-mitotic fragments (12, 13), cisternal stacking (14), and binding of COPI vesicles to Golgi membranes (6). Sequence analysis indicates that p115 is homologous to Uso1p, a protein required for endoplasmic reticulum to Golgi transport in Saccharomyces cerevisiae (15, 16). During endoplasmic reticulum to Golgi transport in vitro, Uso1p mediates binding of endoplasmic reticulum-derived coat protein II vesicles to Golgi membranes (17). Both p115 and Uso1p have an analogous three-domain structure: an N-terminal globular domain, a coiled-coil dimerization domain, and a C-terminal acidic domain (11, 18, 19). Both proteins form a myosin II-like parallel homodimer with two globular heads and an extended kinked tail (18, 19).

p115 interacts with two Golgi proteins, GM130 and giantin (6, 9, 20). GM130 is a tightly associated cytoplasmically oriented peripheral component of the Golgi, predicted to form a segmented coiled-coil dimer (21, 22). Giantin is a C-terminally anchored dimeric integral component of the Golgi membrane with a large cytoplasmic domain also predicted to form a segmented coiled-coil (23-25). Because giantin, but not GM130, is recovered in COPI vesicles generated from Golgi fractions in the presence of GTPS, and p115 promotes binding of these COPI vesicles to Golgi membranes, it was proposed that p115 mediates vesicle tethering by simultaneously binding giantin in vesicles and GM130 on the Golgi (6). Indeed, anti-giantin and anti-GM130 antibody treatments that reduce p115 membrane binding also block vesicle docking (6). Similarly, N-ethylmaleimide-sensitive fusion protein-dependent cisternae formation as well as Golgi reassembly stacking protein (GRASP) 65-dependent cisternal stacking also require p115 and are inhibited by anti-giantin or anti-GM130 antibodies (14). Together, the data suggest that multiple processes involving close apposition of Golgi membranes might utilize p115, giantin, and GM130 for tether formation. However, a key prediction of the proposed model, that membrane tethering is mediated by a simultaneous binding of p115 to giantin and GM130, has not been documented.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Plasmid Constructions-- Constructs used for the generation of glutathione S-transferase (GST) fusion proteins were in pGEX (Amersham Pharmacia Biotech), a vector encoding the GST gene followed by a multiple cloning site followed by stop codons in all three frames. In each case, fragments of giantin, GM130, or p115 were cloned in frame at the multiple cloning site. For GST-giantin I-V, GST-giantin X, and GST-giantin I and II, the restriction fragments from the full-length giantin cDNA (24) used were BamHI-XhoI (28 to 2086), EcoRV-KpnI (4535-5630), and BamHI-HindIII (28 to 778), respectively. For the nucleotide positions indicated here, 1 equals the A in giantin's start codon, and nucleotides upstream of the start codon were derived from the pBluescript-II-KS polylinker (Stratagene, La Jolla, CA). For GST-giantin I and GST-giantin II, the giantin fragments were generated with the polymerase chain reaction using the primer pairs 5'-CGGGATCCAATTTAATAATACTACA with 5'-CGGGATCCAGAACAGTCCCTCCTTG (122-367) and 5'-CGGGATCCAGTCAGAGGAGCAACTT with 5'-CGGGATCCGGGCATCTTTCTCTCG (379-675), respectively. For GST-p115, the restriction fragments used were EcoRI-NotI (for GST-p115-(710-959)) or EcoRI-XhoI (for GST-p115-(710-934)). Constructs used for generation of in vitro translated p115 protein fragments were in pBluescript-II-KS (Stratagene), a vector containing the T3 promoter site for transcription. In each case, a p115 fragment from the full-length cDNA (11) was cloned into the plasmid multicloning site. Constructs used for production of p115 and GM130 in bacteria were previously described (9). All constructs used were confirmed by restriction mapping.

Protein Expression and Purification-- Individual colonies of DH5aF' E. coli transformed with a fusion protein encoding the pGEX vector were grown to OD 0.6 in 1 liter of LB at 37 °C, induced with 0.5 mM isopropyl-1-thio--D-galactopyranoside for 5 h at 30 °C, and collected as a cell pellet. After washing with 150 mM NaCl, the cell pellet was resuspended in lysis buffer (50 mM Tris, pH 8.0, 5 mM EDTA, 150 mM NaCl, 10% glycerol, 5 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 10 µg/ml leupeptin, and 10 µg/ml pepstatin A), incubated on ice for 30 min, adjusted to 0.5% Triton X-100, sonicated four times for 30 s each, and centrifuged at 100,000 rpm for 30 min in a TLA 100.3 rotor (Beckman Instruments, Palo Alto, CA). The cleared lysate was collected and incubated with 0.5 ml of glutathione-agarose beads for 1 h at 4 °C. The beads were washed with 40 ml of PD buffer (phosphate-buffered saline, 1 mM dithiothreitol) containing 0.1% Triton X-100, washed with 10 ml of PD buffer, and then eluted in PD buffer containing glutathione. The purity of the eluted fractions was monitored by Coomassie staining after SDS-PAGE. Following dialysis against phosphate-buffered saline at 4 °C, the proteins were aliquoted and frozen. In some cases, the washed column beads were used directly in binding experiments. Preparation of bacterial lysates containing GM130 was as described previously (9). In vitro transcription and translation reactions for p115 were carried out using the TNT system (Promega, Madison, WI) according to the manufacturer's specifications. The reactions were aliquoted and frozen at 80 °C before use. All frozen samples were thawed and centrifuged at 100,000 × g immediately before use in binding assays.

Cell Extract Preparation-- For the preparation of cell extracts, normal rat kidney (NRK) cells were rinsed three times and collected by scraping in ice-cold phosphate-buffered saline. After centrifugation, the cell pellet was resuspended with 3 volumes of 2× HKT (20 mM Hepes, pH 7.4, 200 mM KCl, 1% Triton X-100, 2 mM EDTA, 2 mM dithiothreitol, 2 mM phenylmethylsulfonyl fluoride, 20 µg/ml leupeptin, and 20 µg/ml pepstatin A), passed through a 25-gauge needle, adjusted with 3 volumes of H₂O, and centrifuged for 30 min at 100,000 × g. For the preparation of membrane and cytosol extracts, rinsed NRK cells were homogenized by 10 passes through a 25-gauge needle in 250 mM sucrose, 10 mM Hepes, 1 mM EDTA containing the protease inhibitors. The postnuclear supernatant was collected after a 2-min spin at 1000 × g, underlayered with 10 µl of 50% sucrose, and centrifuged for 15 min at 14,000 × g. The resulting supernatant was further centrifuged for 30 min at 100,000 × g to yield the cytosol fraction. The membrane pellet on its sucrose cushion was lysed with 5 volumes of 2× HKT containing the protease inhibitors, passed through a needle, adjusted to 1× HKT with H₂O, and clarified by centrifugation for 30 min at 100,000 × g.

Binding Assays-- Unless otherwise indicated, each assay was carried out with a 10-µl volume of packed glutathione-agarose beads (Sigma) coated with approximately 2 µg of GST fusion protein. The beads were prewashed three times with 0.5 ml of 1× HKT, rotated with the lysate (reticulocyte lysate, NRK lysate, or bacterial lysate) for 60 min at 4 °C at a final volume between 50 and 200 µl, washed four times with 0.5 ml of HKT for each wash, and eluted with either sample buffer or 0.5 ml of 0.6 M KCl. Fractions corresponding to total, unbound, each wash, and the salt eluate were concentrated by precipitation with trichloroacetic acid. The samples were analyzed on 7% SDS-polyacrylamide gels. The entire eluate was analyzed, whereas the amount of total analyzed varied and is indicated in the figure legends. The gels were processed for silver staining, fluorography, or immunoblotting. Antibodies and their dilutions used for immunoblotting were anti-p115 (9) at 1:5000, anti-GM130 (9) at 1:5000, and anti-giantin (23) at 1:200. For binding to antibody-coated beads, each assay was carried out with a 10-µl volume of packed protein A-Sepharose (Amersham Pharmacia Biotech) coated with anti-p115 (0.5 µl). The binding assays were otherwise carried out as described for the glutathione-agarose beads. For two-stage assays, the washed beads from the first stage, prior to elution, were carried through a second round of binding and washes exactly as described for the first stage.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Strategy to Detect a Ternary Giantin-p115-GM130 Complex-- Ternary complex formation was not analyzed in previous reports, and the recovery of giantin and GM130 on beads containing attached p115 (6) could be explained by the recovery of distinct p115-giantin and p115-GM130 dimers. Our approach differed in that we tested directly the formation of a ternary giantin-p115-GM130 complex by incubating beads containing bound giantin or beads containing bound GM130 with cell extracts containing p115, GM130, and giantin. p115-dependent recovery of GM130 on giantin beads would illustrate that p115 can mediate the hypothesized bridge formation by linking giantin to GM130. Similarly, the bridge would also be illustrated by p115-dependent giantin recovery on GM130 beads.

Mapping p115 Binding Site in Giantin: the N-terminal Coil I-- To test the model, we needed to generate GST fusion proteins containing the p115 binding site of giantin or GM130. To identify the p115 binding site in giantin, different fragments of giantin were fused in frame to the C terminus of GST (Fig. 1A). The constructs were expressed in bacteria, purified and immobilized on glutathione-agarose beads, and tested for their ability to bind radiolabeled full-length rat p115 generated by in vitro transcription/translation. The cytoplasmic domain (3238 amino acids) of giantin contains 13 discrete regions strongly predicted to form coiled-coil structures (coils I-XIII; Fig. 1A). As shown in Fig. 1B, an N-terminal fragment of giantin containing coils I-V bound p115 (lane 2), as did an N-terminal fragment containing coils I and II (lane 4) and an N-terminal fragment containing only coil I (lane 6). A fragment containing part of coil X (lane 3), a fragment containing coil II (lane 5), and GST alone (lane 1) did not bind p115. These results indicate that the most N-terminal piece of giantin tested, coil I (amino acids 47-117), was sufficient for p115 binding. Identical results were obtained when the GST-giantin fragments were tested for binding to cellular p115 (see below; Fig. 2A); i.e. all giantin fragments that contained coil I bound endogenous p115, and all fragments that lacked coil I did not bind. As shown in Fig. 1C, the concentration of GST-giantin for half-maximal p115 binding in vitro was ~260 nM.

View larger version (25K): [in this window] [in a new window]   Fig. 1.   Giantin's N-terminal coil I contains the p115 binding site. A, schematic diagram of giantin structure and used constructs. The N-terminal coil I is shaded. B, radiolabeled in vitro transcribed/translated full-length p115 was incubated with the indicated GST-giantin fusion protein bound to glutathione-agarose beads. After washing, the recovery of p115 was determined by SDS-PAGE and fluorography. Only constructs that contain giantin coil I bound p115. C, radiolabeled in vitro transcribed/translated full-length p115 was incubated with the indicated concentrations of GST-giantin I-V bound to glutathione-agarose beads. After washing, the recovery of p115 was determined by SDS-PAGE and fluorography. The total (T) represents 100% of the total p115 present in the assay. D, COS-7 cells were transfected with either HA tagged wild-type (wt) or coil I-minus (N) versions of giantin. Detergent lysates of the transfected cells were incubated with GST-p115-(710-959) bound to glutathione agarose beads. The GST-p115-(710-959) construct is described in the legend to Fig. 5. After washing, the recovery of transfected giantin was determined by immunoblotting with antibodies against HA. The total (T) represents 10% of the total HA-tagged giantin present in the binding assay. Only full-length giantin bound GST-p115.

View larger version (46K): [in this window] [in a new window]   Fig. 2.   Giantin bound p115 does not bind GM130. A, indicated GST-giantin fusion proteins bound to glutathione-agarose beads were incubated with NRK detergent extracts. The amount of cellular GM130 and p115 recovered on the beads was determined by immunoblotting with anti-GM130 antibodies (GM130 panel) or anti-p115 antibodies (p115 panel). The total (T) represents 10% of the total GM130 and p115 present in the binding assays. Only giantin fragments containing coil I bound p115 (lanes 3, 4, and 6). Despite ample recovery of p115, GM130 was not recovered. B, GST-giantin I-V bound to glutathione-agarose beads (lanes 1 and 2) or anti-p115 antibodies bound to protein A-Sepharose beads (lanes 3 and 4) were used in a two-stage incubation. In stage 1, the beads were incubated with NRK detergent lysates. After washing the beads, stage 2 reactions were carried out by incubating the beads with buffer alone (lanes 1 and 3) or with buffer containing bacterially produced full-length GM130 (lanes 2 and 4). The amount of endogenous p115 and of endogenous and bacterially produced GM130 recovered on the beads was determined by immunoblotting with a mixture of anti-GM130 and anti-p115 antibodies. Equivalent amounts of endogenous p115 were recovered with GST-giantin and anti-p115 antibodies (lanes 1-4). However, endogenous GM130 was recovered only with immunoprecipitated p115 (lane 3) but not with GST-giantin precipitated p115 (lane 1). Similarly, exogenous GM130 bound to immunoprecipitated p115 (lane 4) but not to GST-giantin precipitated p115 (lane 2).

To test whether coil I was sufficient and necessary for the interaction between giantin and p115, we generated an expression construct to be used for transfection in COS-7 cells, in which the sequences corresponding to coil I were deleted and sequences corresponding to the hemagglutinin (HA) epitope were added at the N terminus. As a control, a construct corresponding to an HA-tagged version of full-length giantin was also generated. Extracts from transfected cells were incubated with beads containing p115, and after washing, the recovery of the HA-tagged protein on the beads was determined by immunoblotting with anti-HA antibodies. Fig. 1D illustrates that only the full-length giantin bound to bead-attached p115 (lane 2), while the giantin fragment lacking coil I did not bind (lane 4). Both HA-tagged proteins were abundantly expressed (lanes 1 and 3) and localized to the Golgi in transfected cells (data not shown). Therefore, giantin's N-terminal coil I, a domain that may extend up to 400 nm away from the membrane, is necessary and sufficient to mediate giantin's interaction with p115.

Binary p115-Giantin and p115-GM130 but Not Ternary Giantin-p115-GM130 Complexes Can Be Detected-- To test whether ternary complexes could form on GST-giantin, different GST-giantin fragments were immobilized on glutathione-agarose beads and incubated with detergent extracts of NRK cells. Bound proteins were eluted, and the presence of p115 and GM130 was analyzed by immunoblotting. As expected, cellular p115 bound to all GST-giantin constructs containing giantin's coil I (Fig. 2A, p115 panel, lanes 3, 4, and 6) but not to constructs lacking coil I (lanes 2 and 5). However, despite the abundant recovery of p115, GM130 was not recovered on the beads (GM130 panel). Therefore, endogenous p115 did not mediate the indirect association of endogenous GM130 with giantin. Because a significant proportion of cellular p115 is bound to GM130 (9, 20), this suggested that the p115-GM130 complex did not bind to giantin. Instead, only uncomplexed p115 bound giantin, and when this occurred the giantin-bound p115 no longer could interact with GM130.

To ensure that p115 remained associated with GM130 in our NRK extracts, we performed parallel precipitations using either anti-p115 antibodies or GST-giantin. As shown in Fig. 2B, p115 was recovered together with GM130 following immunoprecipitation (lane 3), indicating that stable p115-GM130 complexes were present in the cell lysate. However, such complexes were not recovered during precipitation with GST-giantin (lane 1). Only uncomplexed p115 bound to giantin. These results provide strong evidence that both free p115 and p115-GM130 complexes were present in the NRK extracts, that only the free p115 bound giantin, and that once bound to giantin p115 no longer could interact with GM130.

We next tested whether p115 bound to GST-giantin was competent to bind exogenously added bacterially produced GM130. As a control, an analogous incubation was carried out with p115 isolated on anti-p115 antibodies bound to protein A-Sepharose. The exogenous GM130 did not bind to the GST-giantin-precipitated p115 (Fig. 2B, lane 2). Under the conditions used in these experiments for binding of p115 to GST-giantin, the p115 was at saturating levels. Therefore, the lack of GM130 binding is unlikely to be due to GST-giantin occupying two sites (one of which might normally be available for GM130 binding) on a single molecule of p115. The results indicate that p115 bound to GST-giantin is unable to bind GM130.

Mapping and Testing the p115 Binding Site in GM130: the N-terminal Basic Domain Binds p115, but Ternary GM130-p115-Giantin Complexes Are Not Detected-- To confirm that our results were not due to some peculiarity in the GST-giantin proteins used, we also tested for ternary complex formation using beads containing GM130. GM130 is known to bind p115 via its N terminus, which contains a stretch of basic residues (20). To confirm that the GM130 N-terminal domain is sufficient for p115 binding, the N terminus of GM130 (amino acids 1-74) was fused in frame to the C terminus of GST (Fig. 3A). The resulting GST-GM130 fusion protein was expressed in bacteria, purified, and immobilized on glutathione agarose beads. This matrix was incubated with detergent-solubilized NRK membranes to determine whether the GM130 fragment was capable of isolating endogenous p115. GST-giantin was used as a positive control, and GST alone served as a negative control. As shown in Fig. 3B, beads containing GST-GM130 specifically bound cellular p115 (p115 panel, lane 4). Therefore, the GM130 N terminus is sufficient for binding to p115. Nevertheless, similar to the result for GST-giantin, no evidence for formation of a ternary GM130-p115-giantin complex was obtained; i.e. despite abundant recovery of p115 on GST-GM130, no giantin was recovered (Fig. 3B, Gtn panel).

View larger version (33K): [in this window] [in a new window]   Fig. 3.   GM130-bound p115 does not bind giantin. A, diagram of GM130 and the GST-GM130-(1-74) construct used. B, aliquots of an NRK membrane detergent extract were incubated with ~10 µg of GST, GST-giantin I and II, or GST-GM130-(1-74) bound to glutathione-agarose beads. After washing, the recovery of cellular giantin and p115 was determined by immunoblotting with anti-giantin antibodies (Gtn panel) or anti-p115 antibodies (p115 panel). The total (T) represents 5% of the total giantin and p115 present in the binding assay. GST-giantin (lane 3) and GST-GM130 (lane 4), but not GST (lane 2), bound cellular p115. Giantin was not recovered on GST-GM130 beads containing p115 (lane 4). Note that, compared with the experiment in Fig. 2, this experiment used five times the amount of binding protein, and half the amount of total was loaded. This accounts for the apparent increase in p115 recovery.

Together, the results indicate that, under the conditions tested, although p115 can bind either giantin or GM130, once it is bound to one it loses its capacity for binding the other.

Mapping the Giantin Binding Site in p115: The Same C-terminal Acidic Domain Required for GM130 Binding-- p115 binding to GM130 appears to be mediated by a charge interaction between the C-terminal acidic tail of p115 and a stretch of basic residues in the N terminus of GM130 (9, 20). Interestingly, giantin's coil I also has a marked cluster of basic residues. For comparison, the sequences of GM130 N terminus, giantin's N-terminal coil I, and p115 C-terminal acidic domain are shown in Fig. 4.

View larger version (14K): [in this window] [in a new window]   Fig. 4.   Sequence comparison of GM130 N terminus, giantin N terminus, and p115 C terminus. GM130 and giantin contain clusters of positively charged residues, while p115 contains a relatively long linear stretch of negatively charged residues. Serines present in GM130 and p115 that are known to be phosphorylated are shadowed.

A simple explanation for the lack of ternary complex formation would be that giantin and GM130 bind to the same site on p115. To test whether giantin binds to the same region of p115 that binds GM130, various fragments of p115 cDNA were generated (shown schematically in Fig. 5A), and the corresponding radiolabeled proteins were produced by in vitro transcription/translation. Binding of these proteins to GST-giantin immobilized on beads was then determined. As shown in Fig. 5B, p115 constructs containing the C-terminal acidic domain (i.e. full-length p115-(1-959) (lanes 1 and 2) and p115-(283-959) (lanes 5 and 6)) bound giantin. Constructs lacking the acidic domain (i.e. p115-(1-324) (lanes 3 and 4), p115-(1-934) (lanes 7 and 8), and p115-(1-766) (lanes 9 and 10)) did not bind giantin. These results suggested that the acidic domain at the extreme C terminus of p115 is required for binding.

View larger version (20K): [in this window] [in a new window]   Fig. 5.   The C-terminal acidic tail of p115 binds giantin. A, schematic diagram of used p115 constructs. The C-terminal acidic domain is shaded. B, GST-giantin I-V immobilized on glutathione-agarose beads was incubated with the indicated radiolabeled in vitro transcribed/translated p115 fragment. Recovery of bound (B) and unbound (U) p115 fragment is shown after SDS-PAGE and fluorography. Only p115 constructs containing the C-terminal acidic domain (lanes 2 and 6) bound to giantin. Constructs lacking the C-terminal region (lanes 4, 8, and 10) did not bind. The position of molecular mass markers (in kDa) is indicated. C, GST-p115-(710-959) or GST-p115-(710-934) bound to glutathione beads were incubated with NRK detergent extracts. The amount of endogenous giantin and GM130 recovered on each matrix was determined by immunoblotting with anti-giantin and anti-GM130 antibodies. Only the p115 construct containing the C-terminal acidic domain bound cellular giantin and GM130.

To test whether the region containing the acidic domain is sufficient for binding giantin, the C terminus of p115 including amino acids 710-959 was fused in frame to the C terminus of GST. As a control, the same region but lacking the most C-terminal 25 amino acids (p115-(710-934)) was fused to GST. The resulting fusion proteins were expressed in bacteria and were purified and immobilized on glutathione-agarose beads. Each matrix was then incubated with NRK lysates to determine whether the p115 fragment is capable of isolating endogenous giantin. As shown in Fig. 5C, GST-p115-(710-959) bound giantin (lane 1), whereas a fusion protein lacking the acidic tail, GST-p115-(710-934) did not bind giantin (lane 2). Similarly, GM130 was recovered with the GST-p115 containing the acidic C terminus but not when the acidic terminus was removed (Fig. 5, GM130 panel). These data are analogous to those showing endogenous giantin and GM130 recovery on beads containing full-length p115 (6).

Giantin and GM130 Compete for Binding to p115-- To test directly whether giantin and GM130 interactions with p115 are mutually exclusive, we measured the binding of GST-giantin to immobilized p115 in the presence of increasing concentrations of GST-GM130. The addition of increasing concentrations of GST was used as a control. As expected, GST-giantin bound to beads containing immunoprecipitated cytosolic p115 (Fig. 6, GST-Gtn panel, lanes 2 and 6) but not to control beads without p115 (lane 1). In contrast, GST-giantin binding was significantly blocked if the binding reaction was carried out in the presence of increasing concentrations of GST-GM130 (lanes 3-5). When GST was added in place of GST-GM130, GST-giantin binding to p115 was not altered (lanes 7-9). These findings suggest that giantin and GM130 compete for binding to p115. Indeed, the added GST-GM130 that prevented GST-giantin binding was recovered bound to p115 (Fig. 6, GST-GM130 panel, lanes 3-5). GST did not interact with the immunoprecipitated p115 (Fig. 6, GST panel). The differences in recovery of GST-giantin in the presence or absence of GST-GM130 were not due to differences in p115 levels in the immunoprecipitates (Fig. 6, p115 panel, lanes 2-9). Together, the data indicate that p115 independently, but not simultaneously, interacts with either giantin or GM130.

View larger version (61K): [in this window] [in a new window]   Fig. 6.   Giantin and GM130 compete for binding to p115. Control (lane 1) or anti-p115 (lanes 2-9) antibodies were bound to protein A-Sepharose beads, and each matrix was incubated with NRK cytosol in 1× HKT buffer. After washing, the immunoprecipitates were incubated with 1 µg of GST-giantin and the indicated amounts of GST-GM130 (lanes 3-5) or GST (lanes 7-9). After further washing, recovery of GST-giantin (GST-Gtn panel), GST-GM130 (GST-GM130 panel), GST (GST panel), and cellular p115 (p115 panel) was determined by immunoblotting. The total (T) represents 0.16 µg of GST-giantin, 0.08 µg of GST-GM130, and 0.08 µg of GST. Similar amounts of endogenous p115 were recovered in the anti-p115 immunoprecipitates (lanes 2-9). GST-giantin bound to the p115 beads when buffer alone (lanes 2 and 6) or buffer with increasing concentrations of GST (lanes 7-9) was added. The addition of GST-GM130 (lanes 3-5) resulted in GST-GM130 binding to the p115 beads and prevented GST-giantin binding. Background bands (*) in this experiment were variable and not present in assays where higher amounts of the competing proteins were added. They were probably due to minor amounts of antibodies that eluted from the beads.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Dissection of molecular events that mediate target recognition by vesicles is key to our understanding of trafficking and subcellular compartmentalization. One of the earliest events in the sequence of interactions that mediate vesicle docking is likely to involve stable tethering of vesicles to the target membrane. Fibrous material visualized by electron microscopy appears to attach COPI vesicles to Golgi cisternae (4), suggesting that long extended proteins might mediate the tethering process. The transport factor p115, which binds to two extended Golgi proteins, giantin and GM130 (6, 9, 20), was recently proposed to physically link COPI vesicles to the Golgi by simultaneously binding giantin present in the vesicles and GM130 present on the Golgi membranes (6). The same mechanism was proposed to mediate cisternae formation and cisternal stacking (14).

To test this model, we asked whether giantin could bind GM130 in a p115-dependent manner and whether p115 could bind both giantin and GM130 simultaneously. To generate exogenous immobilized giantin capable of binding p115, we needed to map giantin's p115 binding site. Interestingly, a 70-amino acid giantin domain that is predicted to form a coiled-coil segment at the extreme N terminus of the protein was required and sufficient for p115 binding. The length of giantin's cytoplasmic domain is estimated to be 250 nm (calculated from measured sedimentation and diffusion coefficients (24)) and could be as long as 430 nm if the entire protein were to form an extended coiled-coil. Because giantin is anchored to the membrane through a stretch of hydrophobic amino acids at its extreme C terminus, these estimates indicate that p115 binding to giantin may take place at a relatively great distance from the membrane. GM130 also has an N-terminal p115 binding site, and GM130 is predicted to form an elongated (up to 130-nm) coiled-coil anchored to the Golgi membrane via its C-terminal binding to the myristoylated membrane-anchored protein GRASP65 (26, 27). p115 itself is 50 nm long (16). Therefore, end to end binding of giantin-p115-GM130-GRASP65 into a complex would generate an almost absurdly long membrane-membrane linkage. Of course, the actual length of such a tether would depend on the conformation of the complex. If the ~100-nm filaments visualized connecting COPI vesicles to the Golgi were a giantin-p115-GM130-GRASP65 tether complex, then such complex would have to fold to form a shorter tether. However, under a variety of tested conditions p115 did not bind giantin and GM130 simultaneously. Specifically, giantin bound p115 but not complexes of p115-GM130, and preformed complexes of giantin-p115 did not bind GM130. Thus, although dimeric complexes of giantin-p115 and GM130-p115 formed, such complexes did not interact with the third possible binding partner. An explanation was provided by the observation that giantin competed with GM130 for binding to p115, suggesting that p115 contains a single site that binds either giantin or GM130. This was confirmed by mapping p115's giantin binding site to its acidic C terminus, the same site used by GM130 (9). These observations are not consistent with a straightforward model in which p115 tethers COPI vesicles to Golgi cisternal elements by physically linking giantin in one membrane to GM130 in the other membrane.

What, then, is the role of giantin, p115, and GM130 in vesicle transport, cisternae formation, and cisternal stacking? Our observations do not preclude the expected docking roles for each of these proteins. Vesicle transport, cisternae formation, and cisternal stacking each involve membrane docking and would be facilitated by long distance stable linkages. What could be the mechanism underlying the p115-mediated docking mechanism? A number of possibilities could be envisioned. p115 could act catalytically, perhaps to facilitate the bridging of giantin to GM130 by a yet to be identified protein. Alternatively, our results would not rule out a p115 bridge between giantin and GM130 if p115 existed as oligomers. Because the p115 dimer appears to contain a single binding site for either molecule, tetramers or higher order oligomers of p115 could theoretically bind both giantin and GM130 to form tethers. If this were the case, an activity not present in our extracts must be required to drive p115 oligomer formation. Finally, p115 may physically link membranes but in a manner distinct from the proposed model. One possibility consistent with a single p115 site for binding either GM130 or giantin is that the tethering reactions of p115, giantin, and GM130 are mediated by giantin-p115-X and GM130-p115-Y complexes, where X and Y are novel p115 receptors. In this regard, it is interesting that p115 binding to the intermediate compartment may be mediated by an unknown component, since it appears to be independent of giantin and GM130 (9). This reasoning raises the intriguing possibility that p115-giantin and p115-GM130 interactions might mediate tethering in spatially or temporally distinct events.

	FOOTNOTES

^* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ To whom correspondence should be addressed: Dept. of Biological Sciences, Carnegie Mellon University, 4400 5th Ave., Pittsburgh, PA 15213. Tel.: 412-268-8274; Fax: 412-268-7129; E-mail: linstedt@andrew.cmu.edu.

	ABBREVIATIONS

The abbreviations used are: COPI, coat protein I; GTPS, guanosine 5'-3-O-(thio)triphosphate; GRASP, Golgi reassembly stacking protein; GST, glutathione S-transferase; HA, hemagglutinin; NRK, normal rat kidney; PAGE, polyacrylamide gel electrophoresis.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

1.	Ferro-Novick, S., and Jahn, R. (1994) Nature 370, 191-193[CrossRef][Medline] [Order article via Infotrieve]
2.	Pfeffer, S. R. (1999) Nat. Cell Biol. 1, E17-E22[CrossRef][Medline] [Order article via Infotrieve]
3.	Rothman, J. E. (1994) Nature 372, 55-63[CrossRef][Medline] [Order article via Infotrieve]
4.	Orci, L., Perrelet, A., and Rothman, J. E. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 2279-2283[Abstract/Free Full Text]
5.	Orci, L., Stamnes, M., Ravazzola, M., Amherdt, M., Perrelet, A., Sollner, T., and Rothman, J. E. (1997) Cell 90, 335-349[CrossRef][Medline] [Order article via Infotrieve]
6.	Sönnichsen, B., Lowe, M., Levine, T., Jamsa, E., Dirac-Svejstrup, B., and Warren, G. (1998) J. Cell Biol. 140, 1013-1021[Abstract/Free Full Text]
7.	Waters, M. G., Clary, D. O., and Rothman, J. E. (1992) J. Cell Biol. 118, 1015-1026[Abstract/Free Full Text]
8.	Bannykh, S. I., Nisihmura, N., and Balch, W. (1998) Trends Cell Biol. 8, 21-25[CrossRef][Medline] [Order article via Infotrieve]
9.	Nelson, D. S., Alvarez, C., Gao, Y.-s., Garcia-Mata, R., Fialkowski, E., and Sztul, E. (1998) J. Cell Biol. 143, 319-332[Abstract/Free Full Text]
10.	Alvarez, C., Fujita, H., Hubbard, A., and Sztul, E. (1999) J. Cell Biol. 147, 1205-1221[Abstract/Free Full Text]
11.	Barroso, M., Nelson, D. S., and Sztul, E. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 527-531[Abstract/Free Full Text]
12.	Rabouille, C., Misteli, T. R., Watson, R., and Warren, G. (1995) J. Cell Biol. 129, 605-618[Abstract/Free Full Text]
13.	Rabouille, C., Kondo, H., Newman, R., Hui, N., Freemont, P., and Warren, G. (1998) Cell 92, 603-610[CrossRef][Medline] [Order article via Infotrieve]
14.	Shorter, J., and Warren, G. (1999) J. Cell Biol. 146, 57-70[Abstract/Free Full Text]
15.	Nakajima, H., Hirata, A., Ogawa, Y., Yonehara, T., Yoda, K., and Yamasaki, M. (1991) J. Cell Biol. 113, 245-260[Abstract/Free Full Text]
16.	Sapperstein, S. K., Lupashin, V. V., Schmitt, H. D., and Waters, M. G. (1996) J. Cell Biol. 132, 755-767[Abstract/Free Full Text]
17.	Cao, X., N., Ballew, and Barlowe, C. (1998) EMBO J. 17, 2156-2165[CrossRef][Medline] [Order article via Infotrieve]
18.	Sapperstein, S. K., Walter, D. M., Grosvenor, A. R., Heuser, J. E., and Waters, M. G. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 522-526[Abstract/Free Full Text]
19.	Yamakawa, H., Seog, D.-H., Yoda, K., Yamasaki, M., and Wakabayashi, T. (1996) J. Struct. Biol. 116, 356-365[CrossRef][Medline] [Order article via Infotrieve]
20.	Nakamura, N., Lowe, M., Levine, T. P., Rabouille, C., and Warren, G. (1997) Cell 89, 445-455[CrossRef][Medline] [Order article via Infotrieve]
21.	Nakamura, N., Rabouille, C., Watson, R., Nilsson, T., Hui, N., Slusarewicz, P., Kreis, T. E., and Warren, G. (1995) J. Cell Biol. 131, 1715-1726[Abstract/Free Full Text]
22.	Slusarewicz, P., Nillson, T., Hui, N., Watson, R., and Warren, G. (1994) J. Cell Biol. 124, 405-413[Abstract/Free Full Text]
23.	Linstedt, A. D., and Hauri, H.-P. (1993) Mol. Biol. Cell 4, 679-693[Abstract]
24.	Linstedt, A. D., Foguet, M., Renz, M., Seelig, H. P., Glick, B. S., and Hauri, H.-P. (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 5102-5105[Abstract/Free Full Text]
25.	Seelig, H. P., Schranz, P., Schrter, H., Wiemann, C., Griffiths, G., and Renz, M. (1994) Mol. Cell. Biol. 14, 2564-2576[Abstract/Free Full Text]
26.	Barr, F. A., Puype, M., Vandekerckhove, J., and Warren, G. (1997) Cell 91, 253-262[CrossRef][Medline] [Order article via Infotrieve]
27.	Barr, F. A., Nakamura, N., and Warren, G. (1998) EMBO J. 17, 3258-3268[CrossRef][Medline] [Order article via Infotrieve]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				Y. Guo, V. Punj, D. Sengupta, and A. D. Linstedt Coat-Tether Interaction in Golgi Organization Mol. Biol. Cell, July 1, 2008; 19(7): 2830 - 2843. [Abstract] [Full Text] [PDF]

				A. Diao, L. Frost, Y. Morohashi, and M. Lowe Coordination of Golgin Tethering and SNARE Assembly: GM130 BINDS SYNTAXIN 5 IN A p115-REGULATED MANNER J. Biol. Chem., March 14, 2008; 283(11): 6957 - 6967. [Abstract] [Full Text] [PDF]

				P. Marra, L. Salvatore, A. Mironov Jr, A. Di Campli, G. Di Tullio, A. Trucco, G. Beznoussenko, A. Mironov, and M. A. De Matteis The Biogenesis of the Golgi Ribbon: The Roles of Membrane Input from the ER and of GM130 Mol. Biol. Cell, May 1, 2007; 18(5): 1595 - 1608. [Abstract] [Full Text] [PDF]

				E. Brandon, T. Szul, C. Alvarez, R. Grabski, R. Benjamin, R. Kawai, and E. Sztul On and Off Membrane Dynamics of the Endoplasmic Reticulum-Golgi Tethering Factor p115 In Vivo Mol. Biol. Cell, July 1, 2006; 17(7): 2996 - 3008. [Abstract] [Full Text] [PDF]

				E. Sztul and V. Lupashin Role of tethering factors in secretory membrane traffic Am J Physiol Cell Physiol, January 1, 2006; 290(1): C11 - C26. [Abstract] [Full Text] [PDF]

				M. Beard, A. Satoh, J. Shorter, and G. Warren A Cryptic Rab1-binding Site in the p115 Tethering Protein J. Biol. Chem., July 8, 2005; 280(27): 25840 - 25848. [Abstract] [Full Text] [PDF]

				M. Lowe, J. D. Lane, P. G. Woodman, and V. J. Allan Caspase-mediated cleavage of syntaxin 5 and giantin accompanies inhibition of secretory traffic during apoptosis J. Cell Sci., March 1, 2004; 117(7): 1139 - 1150. [Abstract] [Full Text] [PDF]

				H. Cha, B. L. Smith, K. Gallo, C. E. Machamer, and P. Shapiro Phosphorylation of golgin-160 by mixed lineage kinase 3 J. Cell Sci., February 15, 2004; 117(5): 751 - 760. [Abstract] [Full Text] [PDF]

				M. A. Puthenveedu and A. D. Linstedt Gene replacement reveals that p115/SNARE interactions are essential for Golgi biogenesis PNAS, February 3, 2004; 101(5): 1253 - 1256. [Abstract] [Full Text] [PDF]

				S.-i. Yoshimura, A. Yamamoto, Y. Misumi, M. Sohda, F. A. Barr, G. Fujii, A. Shakoori, H. Ohno, K. Mihara, and N. Nakamura Dynamics of Golgi Matrix Proteins after the Blockage of ER to Golgi Transport J. Biochem., February 1, 2004; 135(2): 201 - 216. [Abstract] [Full Text] [PDF]

				E. Ohta, Y. Misumi, M. Sohda, T. Fujiwara, A. Yano, and Y. Ikehara Identification and Characterization of GCP16, a Novel Acylated Golgi Protein That Interacts with GCP170 J. Biol. Chem., December 19, 2003; 278(51): 51957 - 51967. [Abstract] [Full Text] [PDF]

				R. Chiu, L. Novikov, S. Mukherjee, and D. Shields A caspase cleavage fragment of p115 induces fragmentation of the Golgi apparatus and apoptosis J. Cell Biol., November 25, 2002; 159(4): 637 - 648. [Abstract] [Full Text] [PDF]

				A. K. Gillingham, A. C. Pfeifer, and S. Munro CASP, the Alternatively Spliced Product of the Gene Encoding the CCAAT-Displacement Protein Transcription Factor, Is a Golgi Membrane Protein Related to Giantin Mol. Biol. Cell, November 1, 2002; 13(11): 3761 - 3774. [Abstract] [Full Text] [PDF]

				S. W. Hicks and C. E. Machamer The NH2-terminal Domain of Golgin-160 Contains Both Golgi and Nuclear Targeting Information J. Biol. Chem., September 20, 2002; 277(39): 35833 - 35839. [Abstract] [Full Text] [PDF]

				S.-i. Yoshimura, N. Nakamura, F. A. Barr, Y. Misumi, Y. Ikehara, H. Ohno, M. Sakaguchi, and K. Mihara Direct targeting of cis-Golgi matrix proteins to the Golgi apparatus J. Cell Sci., March 13, 2002; 114(22): 4105 - 4115. [Abstract] [Full Text] [PDF]

				J. R. C. Whyte and S. Munro Vesicle tethering complexes in membrane traffic J. Cell Sci., January 7, 2002; 115(13): 2627 - 2637. [Abstract] [Full Text] [PDF]