JBC Advanced Glycation Endproducts

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M407813200 on October 8, 2004

J. Biol. Chem., Vol. 279, Issue 53, 55545-55555, December 31, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
279/53/55545    most recent
M407813200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Paulhe, F.
Right arrow Articles by Wehrle-Haller, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Paulhe, F.
Right arrow Articles by Wehrle-Haller, B.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

A Specific Endoplasmic Reticulum Export Signal Drives Transport of Stem Cell Factor (Kitl) to the Cell Surface*{boxs}

Frédérique Paulhe, Beat A. Imhof, and Bernhard Wehrle-Haller{ddagger}

From the Department of Pathology and Immunology, Centre Medical Universitaire, 1 rue Michel Servet, 1211 Geneva 4, Switzerland

Received for publication, July 12, 2004 , and in revised form, September 15, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Stem cell factor, also known as Kit ligand (Kitl), belongs to the family of dimeric transmembrane growth factors. Efficient cell surface presentation of Kitl is essential for the migration, proliferation, and survival of melanocytes, germ cells, hemopoietic stem cells, and mastocytes. Here we demonstrate that intracellular transport of Kitl to the cell surface is driven by a motif in the cytoplasmic tail that acts independently of the previously described basolateral sorting signal. Transport of Kitl to the cell surface is controlled at the level of the endoplasmic reticulum (ER) and requires a C-terminal valine residue positioned at a distance of 19–36 amino acids from the border between the transmembrane and cytoplasmic domains. Deletion or substitution of the valine with other hydrophobic amino acids results in ER accumulation and reduced cell surface transport of Kitl at physiological expression levels. When these mutant proteins are overexpressed in the ER, they are transported by bulk flow to the cell surface albeit at lower efficiency. A fusion construct between Kitl and the green fluorescent protein-labeled extracellular domain of a temperature-sensitive mutant of vesicular stomatitis virus G protein revealed the valine-dependent recruitment into coat protein complex II-coated ER exit sites and vesicular ER to Golgi transport in living cells. Thus the C-terminal valine defines a specific ER export signal in Kitl. It is responsible for the capture of Kitl at coat protein complex II-coated ER exit sites, leading to subsequent cell surface transport under physiological conditions.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Stem cell factor, also known as Kit ligand (Kitl),1 mast cell growth factor, or Steel factor, belongs to the family of transmembrane-anchored growth factors with highly conserved cytoplasmic domains (1). It is a growth factor required for survival of germ cells, melanocytes, hemopoietic stem cells, and mastocytes. Kitl mRNA is alternatively spliced and forms two transmembrane proteins (2, 3). The larger form (Kitl-M1) contains a major proteolytic cleavage site that generates soluble Kitl protein (S-Kitl). The smaller splice variant (Kitl-M2) lacks this proteolytic site (2). However, Kitl-M2 can be processed with less efficiency at other proteolytic sites (3). Soluble Kitl mediates cell migration, and the membrane form is required for cell survival (4, 5). Soluble Kitl is required for the recruitment of mast cells to the skin, and animals with a deleted proteolytic cleavage site in Kitl lack dermal mast cells (6, 7). In contrast, the membrane-bound form of Kitl delivers a cell survival and proliferation signal to epidermal melanocytes and a homing signal for hemopoietic stem cells in the bone marrow (8, 9). The Kitl signal, however, can only be given when the molecule is correctly presented to the responsive cells. For example, survival of melanocytes or spermatogonia requires basolateral expression of Kitl in skin keratinocytes or Sertoli cells respectively (10, 11). Basolateral sorting of Kitl is determined by a leucine residue and an acidic cluster in its cytoplasmic tail (12). The phylogenetic conservation of this motif illustrates the importance of this Kitl sorting domain (13). Alteration of the cytoplasmic tail of Kitl abrogates not only polarized expression but reduces the number of Kitl molecules on the cell surface (5, 11, 14). Consequently, diminished cell surface expression of Kitl results in reduced numbers of Kitl-dependent mastocytes, germ cells, and melanocytes. This demonstrates that the cytoplasmic tail controls the number of Kitl molecules on the cell surface and the basolateral sorting of Kitl in polarized cells. Polarized intracellular sorting of proteins occurs within the trans-Golgi network, whereas the export from the endoplasmic reticulum (ER) determines the number of molecules destined to be expressed at the cell surface.

The mechanism of protein export from the ER remains controversial. Although some proteins seem to exit the ER by a non-selective process (bulk flow) (15–17), current reports favor the notion that at least some transmembrane proteins are selectively exported from the ER. Three classes of ER export signals have been characterized in the cytoplasmic domain of transmembrane proteins, di-acidic, di-basic, and hydrophobic motifs. A di-acidic DXE-containing motif is required for the efficient ER-to-Golgi transport of VSV G protein in mammalian cells (18–20). Recently, ER export of glycosyltransferases was shown to depend on a di-basic (RK(X)RK) motif (21). Furthermore, a hydrophobic motif consisting of a di-phenylalanine at the C terminus of the protein ERGIC-53 mediates selective protein ER export (22). These examples support the existence and the potential biological importance of ER export signals. Cargo proteins to be delivered from the ER to the Golgi such as VSV G protein and ERGIC-53 have been shown to associate with the activated Sar1GTP·Sec23·Sec24 complex (23). This complex binds to ER membranes and further recruits the Sec13 ·Sec31 complex, which establishes a protein network (coat protein complex II (COPII)) enabling the budding of ER membrane. This forms COPII-coated transport vesicles at predetermined ER exit sites. After uncoating, the vesicles fuse with the VTC compartment ("vesiculo-tubular clusters" also called ERGIC) mediating further protein transport to the cis-Golgi (17, 24). A temperature-sensitive mutant of the G protein of vesicular stomatitis virus (VSV G ts-045) has proven to be an invaluable tool for the analysis of this transport mechanism (25–31). While VSV G ts-045 folds normally at 32 °C, it misfolds at 39.5 °C and is retained in the ER by the quality control system, failing to be transported to the cell surface (32, 33).

To investigate the potential role of the cytoplasmic tail of Kitl in ER export, we developed reporter constructs, consisting of the GFP-tagged ectodomain of the temperature-sensitive form of the VSV G protein fused to wild type and mutant Kitl transmembrane and cytoplasmic sequences. These constructs allowed the identification of the C-terminal valine as a critical and highly selective ER export motif that recruits Kitl to Sec24-positive ER exit sites, resulting in a vesicular transport of Kitl to the cis-Golgi. The specificity of the Kitl ER export signal determines the number of Kitl proteins brought to the cell surface, acting independently of the Kitl basolateral-sorting motif.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Kitl Constructs—Kitl-EGFP-M2/M1 chimeras and mouse KL-S have been described previously (11, 12) KL-S encodes the proteolytically released form of Kitl derived from the Kitl-M1 splice variant (2). A human version of the cytoplasmic tail sequence of mouse Kitl was created by PCR-introducing mutations that result in the replacement of serine by proline and glutamine by glutamic acid at positions 5 and 28 of the cytoplasmic tail of Kitl (Fig. 1A), respectively (S5P, AGAAA-CAGCCAAGTCTTACAAGG and AAGACTTGGCTGTTTCTTCTTCC; Q28E, TGCTGCAAGAGAAAGAGAGAGAA and CTCTTTCTCTTG-CAGCATACTTA). This humanized version of the cytoplasmic tail of Kitl was used for the VSV Gts chimeric proteins.



View larger version (67K):
[in this window]
[in a new window]
  		FIG. 1.
C-terminal deletions of Kitl result in ER accumulation. A, scheme representing the alignment of wild-type (wt) and C-terminal deletion mutants of mouse Kitl cytoplasmic tail. The name of the constructs represents the site of amino acid deletions. B, fluorescence images of live COS-7 cells 48 h after transient transfection with wild-type (wt), C-terminal deletion mutants of Kitl-EGFP-M1 (del36, del22, del12, and del12–28), soluble form of Kitl (KLS), or two specific markers of the ER and Golgi compartments, ER-GFP and Gal T-GFP, respectively. In addition to cell surface staining, a typical ER localization was observed after deletion of the last 25 (del12), 15 (del22), or the very last C-terminal amino acid (del36). The soluble form of Kitl lacking the transmembrane and cytoplasmic domains was located in the ER but not at the cell surface (KLS). An internal deletion of the cytoplasmic domain leaving the C terminus intact resulted in a wild-type staining pattern (del12–28). C, COS-7 cells were transiently transfected with HA-tagged versions of wild-type (wt) or del36 mutant Kitl. 24 and 48 h after transfection, cell surfaces were biotinylated for 15 min, washed, blocked in DMEM containing 10% FCS, and lysed with detergent. Control and Endo-H-treated cell lysates (lys) or avidin-precipitated fractions (av) were analyzed on 7.5% SDS-PAGE minigels followed by Western blotting (WB) with polyclonal anti-Kitl antibody. Mature (m), immature (im), and deglycosylated (de) forms of Kitl are indicated by arrows. Scale bar = 10 µm.

 
A HA-tagged form of Kitl-M2 was generated by replacing the EGFP sequence (between ClaI and PinAI sites (12)) in Kitl-EGFP-M2 chimeras with the complementary annealed primer pair (CGATATAC-CCATACGATGTTCCTGATTACGCTAGCCTCA and CCGGTGAGGCT-AGCGTAATCAGGAACATCGTATGGGTATAT) coding for the HA epitope (YPYDVPDYASL).

To generate the temperature-sensitive VSV Gts-EGFP-Kitl-M2 and VSV Gts-HA-Kitl-M2 chimeras, we amplified the DNA sequence that encodes the extracellular N-terminal domain of the temperature-sensitive (ts-O45) mutant of VSV G with the respective forward and reverse primer containing a HindIII and ClaI restriction site (TATAAGCTT-GACACCATGAAGTG and TATATCGATGAGCTCTTCCAACTACTG). The cleaved PCR fragment was cloned into the Kitl-EGFP-M2 and Kitl-HA-M2 constructs in such a way that the nucleotide sequence coding for the extracellular Kitl domain was replaced with that of the temperature-sensitive ts-045 mutant of VSV G (VSV Gts). All constructs were verified by dideoxy sequencing.

VSV Gts-YFP and GalT-GFP were kindly provided by Dr. Paccaud. ER-GFP was from Clontech (San Jose, CA).

Cell Culture, Transfection, and Live Fluorescence Microscopy— COS-7 cells were obtained from ATCC, HeLa cells were provided by Dr. Paccaud, MDCK II cells were given by Dr. Matter and cultured in 10% fetal calf serum (PAA Laboratories, Linz, Austria) in Dulbecco's modified Eagle's medium (DMEM) (Invitrogen, Paisley, Scotland). Cells were fed three times weekly and passaged by treatment with 0.05% trypsin and 0.53 mM EDTA. Cultures were maintained in a humidified incubator with a 10% CO2/90% air mixture at 37 °C.

Transient transfections were performed on glass coverslips with FuGENE 6 (Roche Applied Science) according to the manufacturer's recommendations. Wild-type or mutant VSV Gts-GFP-M2-transfected cells were kept at 39.5 °C for 36 h to accumulate chimeric proteins in the ER. The coverslips were then transferred to a 32 °C incubator for the indicated periods. Living or fixed cells (5 min in 2% paraformaldehyde, 0.05% glutaraldehyde) were observed with either an inverted confocal or epifluorescent microscope (LSM510; Zeiss-Axiovert 100, Zurich, Switzerland), using a PlanNeofluar x63 NA 1.4 oil immersion objective (Zeiss), equipped with an omega GFP filter set (XF100, Chroma, Brattelboro, VT) and an incubation chamber with the temperature and CO2 set at 39.5 or 32 °C and 10%, respectively. Pictures were acquired with a Hamamatsu C4742-95-10 digital CCD camera (Hamamatsu Photonics) controlled by the Openlab software (Improvision, Coventry, UK).

Immunofluorescence—HeLa cells in DMEM, 10% FCS were plated on human fibronectin (10 µg/ml)-coated glass coverslips and incubated for 24 h at 37 °C. Cells were transfected as described above. For calreticulin immunostaining, cells were fixed for 5 min in 4% paraformaldehyde and permeabilized for 15 min in PBS containing 1% Triton X-100 and 1% BSA. For immunostaining of TGN46, cells were fixed for 20 min in 4% paraformaldehyde and permeabilized in PBS containing 0.05% saponin for 5 min at room temperature. Fixed cells were incubated with monoclonal anti-calreticulin (BD Transduction Laboratories; catalog number 612136) or sheep anti-human TGN46 (Serotec, AHP500G) for 1 h at room temperature. Sec24C was revealed after fixation with ice-cold methanol for 2 min followed by two washes in PBS by a polyclonal anti-Sec24C antibody from Dr. Paccaud (34) suspended in PBS, 1% BSA for 1 h. After three washes in PBS, 1% BSA, cells were incubated with Texas Red-conjugated goat anti-rabbit (Jackson ImmunoResearch Laboratories; catalog number 111-075-144), donkey anti-sheep (Jackson ImmunoResearch Laboratories), or goat anti-mouse (Jackson ImmunoResearch Laboratories; catalog number 115-075-166) antibody for 1 h at room temperature, respectively, and cells were washed three times and mounted in PBS for microscopic analysis.

Fluorescence-activated Cell Sorting—After treatment with 0.0 trypsin and 0.53 mM EDTA, 120,000 MDCK II cells, wild-type or stably expressing wild-type or del36 mutant Kitl-EGFP protein, were seeded in 6-well plates and cultured for 1 day at 37 °C. For fluorescenceactivated cell sorter analysis, cells were removed from the dish by a treatment with Cell Dissociation Medium (Sigma; catalog number C-5789) for 20 min at 37 °C, blocked with 10% FCS containing RPMI, and suspended in PBS, 0.5% BSA (Axonlab) at 4 °C. As indicated, cells were incubated with monoclonal anti-EGFP (Clontech, Basel, Switzerland; catalog number 8363-2) antibody in PBS, 0.5% BSA on ice for 15 min. After washing with PBS, 0.5% BSA, cells were further incubated with biotinylated goat anti-mouse antibody (Vector Laboratories, Burlingame, CA; catalog number BA-9200) on ice for 15 min and revealed after washing with PBS, 0.5% BSA using cychrome-5-coupled streptavidin (Southern Biotechnology; catalog number 7100-15). After washing with PBS, 0.5% BSA, flow cytometry was performed on a FACSCalibur (BD Biosciences). Control stainings were performed with non-transfected MDCK II cells or omitting the primary EGFP antibody on Kitl-EGFP expressing cells. Dead cells were excluded from the analysis by staining for propidium iodide (Sigma; catalog number P-4864). Results are expressed as a plot of frequency versus log fluorescence.

Surface Biotinylation, Avidin Precipitation, and Western Blotting— Transient or stable transfected cells were washed three times with pre-warmed biotinylation buffer (140 mM NaCl, 5 mM KCl, 0.5 mM MgSO4, 1.5 mM CaCl2, 0.05% NaHCO3, and 0.1% glucose at pH 7.5). The cells were biotinylated for 15 min at the appropriate temperature in a solution of 0.5 mg of sulfo-N-hydroxysuccinimide-biotin (Pierce)/ml of biotinylation buffer. After biotinylation, cells were rinsed twice with ice-cold biotinylation buffer and subsequently blocked with DMEM, 10% FCS for 10 min. Biotinylated cells were washed with ice-cold biotinylation buffer and then extracted with 0.5 ml of lysis buffer for 10 min on ice (120 mM NaCl, 50 mM Tris-HCl, pH 8.0, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS, 1 mM phenylmethylsulfonyl fluoride, and 1 µg/ml chymostatin, leupeptin, antipain, and pepstatin, each). Biotinylated proteins were precipitated with 50 µl of a 50% suspension of avidin-agarose beads (Pierce) for 1 h at 4°C. Beads were washed once with 0.5 ml of lysis buffer and twice with 1 ml of TBST (10 mM Tris-HCl, pH 7.5, 150 mM NaCl; 0.1% Tween 20). Beads were boiled for 5 min in 50 µl of 1x SDS-PAGE buffer containing {beta}-mercaptoethanol. Proteins were separated on 7.5% SDS-PAGE and transferred to nitrocellulose according to standard protocols. Blots were blocked in 1% BSA in TBST and incubated with either a polyclonal anti-Kitl antibody (35) or a monoclonal anti-EGFP antibody (Clontech, Basel, Switzerland; catalog number 8363-2) in blocking buffer. Nitrocellulose blots were washed with TBST, incubated with anti-rabbit or rat anti-mouse antibodies conjugated with peroxidase (Jackson ImmunoResearch Laboratories; catalog numbers 111-035-144 and 415-035-100) and revealed with ECL reagents (Amersham Biosciences).

As an internal control for transfection efficiency and respective synthesis of Kitl proteins, we compared identical fractions of total cell lysates for Kitl or EGFP immunoreactivity on Western blots. Quantification of bands was performed from digitally scanned blots using the Openlab software (Improvision).

Endoglycosidase Digestion— endoglycosidase-H (Endo-H) digestion was performed on biotinylated cells (see above), which were lysed for 15 min in 500 µl of Endo-H buffer. A fraction of 100 µl of cell lysate was heated at 95 °C for 5 min. The samples were cooled and divided into two equally sized aliquots, and 3 milliunits/ml endoglycosidase-H (New England Biolabs) was added to 1 aliquot. Digestion at 37 °C was terminated after 24 h by heating for 5 min at 95 °C in SDS-PAGE sample buffer. Digested samples were analyzed under reducing conditions by 10% SDS-PAGE.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
C-terminal Valine Is Required for Export of Kitl from the ER—The highly conserved cytoplasmic domain of Kitl directs the molecule to the basolateral cell surface of polarized epithelial cells (5, 11, 14). In addition, substitution of the cytoplasmic domain of Kitl with an irrelevant sequence reduces cell surface expression of Kitl (5, 11, 14). Here we examined the nature of Kitl transport to the cell surface. To monitor intracellular traffic, we recently established GFP-tagged Kitl constructs. However, due to the delay in fluorophore formation of newly synthesized GFP-proteins, the early steps of protein transport of the GFP-Kitl chimera cannot be observed (36, 37). Although this property seems to be a disadvantage, it can be exploited as a built in timer to determine the speed of intracellular transport. First, COS-7 cells were transiently transfected with the wild-type form of GFP-tagged Kitl leading to homogenous cell surface staining. A residual intracellular vesicular staining was identical to that obtained with a Golgi specific marker construct (Gal T-GFP), while the ER compartment was not visible (Fig. 1B). This expression pattern was compared with that of various Kitl mutant proteins carrying deletions in the cytoplasmic tail (Fig. 1A). In addition to the cell surface staining, all mutant proteins exhibiting C-terminal deletions were observed in an intracellular reticular network. The removal of the single C-terminal valine was sufficient to induce this reticular location. (Fig. 1B, del36). Soluble Kitl that comprise a deletion of the transmembrane and cytoplasmic tail also showed a reticular pattern (Fig. 1B, KLS). This pattern was identified as ER, since cells transfected with a commercial ER specific reporter construct showed identical staining (Fig. 1B, ER-GFP). In contrast, internal deletions of amino acids 12–28 of the cytoplasmic tail leaving the C terminus intact resulted in a wild-type expression pattern (Fig. 1B, del12–28). These observations demonstrate that deletion of the C-terminal valine is sufficient to induce accumulation of Kitl in the ER. The delayed visibility of nascent GFP combined with rapid passage through the ER compartment of wild-type Kitl could explain the absence of visible wild-type GFP-Kitl in the ER. In contrast, C-terminally mutated Kitl proteins remain longer in the ER than the time required for GFP-fluorophore formation, allowing their detection in this compartment.

ER Accumulation of C-terminal Mutant Kitl Protein Leads to Reduced Cell Surface Expression—Based on these results we suspected that delayed intracellular transport of Kitl mutant proteins reduced their cell surface expression. This may be due to retention of Kitl mutant protein in the ER, while bulk flow out of the ER may account for residual cell surface expression. To test this hypothesis, we determined ER to Golgi passage and cell surface expression levels of Kitl by Endo-H sensitivity and cell surface biotinylation. Cell lysates and cell surface proteins were analyzed 12, 24, 36, and 48 h following transfection of Kitl into COS-7 cells (Fig. 1C). After 12 h, Kitl expression was not detected (data not shown). A single, Endo-H-resistant band was detected after 24, 36, and 48 h in cells transfected with wild-type Kitl. This band corresponded to the size of biotinylated wild-type Kitl protein found at the cell surface. After 24 h, the majority (70%) of C-terminal, valine-deleted Kitl (del36) appeared as an Endo-H-sensitive, faster migrating band on SDS gels. Compared with wild-type, only a small quantity of del36 mutant protein was expressed at the cell surface (22%). At 36 and 48 h after Kitl transfection, the ratio between mature (Endo-H-resistant) and immature (Endo-H-sensitive) del36 mutant protein increased (43 and 49%, respectively). This resulted in a delayed cell surface expression of del36 Kitl, eventually reaching wild-type levels (Fig. 1C; 84% at 36 h and 110% at 48 h). It suggests that the Kitl del36 protein first accumulated in the ER compartment and was subsequently transported to the cell surface by bulk flow requiring high protein concentrations in the ER. The absence of apparent Endo-H-sensitive wild-type Kitl (only 1% at 24 h) shows rapid passage through the ER and suggests efficient recognition by the ER export machinery. Due to the continuous proteolytic shedding of wild-type and mutant cell surface Kitl, similar steady state surface expression levels are eventually reached in transiently transfected COS-7 cells. To approach physiological levels of Kitl expression, we established polarized epithelial cells (MDCK) stably expressing wild-type and mutant Kitl protein. Similar to the findings described above at early time points after transfection of COS-7 cells, MDCK cells showed reduced levels of del36, valine-deleted Kitl on the cell surface when compared with wild-type (Fig. 2, B and D). The del36 mutant protein was also detected in a reticular staining pattern indicating ER retention (Fig. 2, G and H). This confirms a delay in ER export found with cells transiently transfected with Kitl and demonstrates that ER retention results in reduced cell surface expression under steady state conditions.



View larger version (55K):
[in this window]
[in a new window]
  		FIG. 2.
C-terminal deletions of Kitl result in reduced cell surface expression. A–D, determination of total (A, C) and cell surface expression (B, D) of wild-type and del36 mutant EGFP-Kitl proteins in stably transfected MDCK II cells by fluorescence-activated cell sorter. In B and D, the control staining without anti-EGFP antibody is shown as a dotted line. Representative results from three independent experiments are shown. E–H, fluorescence images of live confluent cultures of MDCK II cells that stably express wild-type (E) or del36 mutant (G) forms of Kitl-EGFP-M2 proteins. Magnified views of the cell surface expression and ER-accumulation of wild-type and del36 mutant forms of EGFP-Kitl-M2 proteins are shown in F and H, respectively. Scale bar = 10 µm.

 
Characterization of the ER Export Signal Using a VSV Gts-Kitl Chimera—Since transport of wild-type Kitl-EGFP from ER to Golgi occurs too rapidly to be observed in living cells, a strategy was applied that allowed controlled ER retention or release from the ER. In addition, we wanted to distinguish between signal-dependent and bulk flow transport of Kitl from the ER to the Golgi. Therefore, we fused the temperature-sensitive mutant of the VSV G protein (ts-045 VSV G) to the EGFP-tagged transmembrane and cytoplasmic domains of Kitl (VSV Gts-EGFP-Kitl, Fig. 3A). This construct showed a behavior comparable with the original VSV Gts mutant protein, i.e. it was retained in the ER at the non-permissive temperature (39.5 °C) and was rapidly transported from ER to Golgi at the permissive temperature (32 °C) (Fig. 3, B and E, respectively). These respective locations were confirmed by immunohistochemistry with markers for ER (calreticulin; Fig. 3C) and Golgi (TGN46; Fig. 3F).



View larger version (39K):
[in this window]
[in a new window]
  		FIG. 3.
VSV Gts-EGFP-Kitl-M2 protein accumulates in the ER at 39.5 °C and is exported to the Golgi at 32 °C. A, a schematic representation of wild-type and chimeric forms of Kitl used in the current study. The non-cleavable forms of wild-type Kitl (Kitl-M2) and the EGFP-tagged chimeric constructs between Kitl and VSV Gts are shown. SP, signal peptide; RBD, Kitl receptor binding domain; MD, transmembrane domain. B–D, HeLa cells transiently transfected with VSV Gts-EGFP-Kitl-M2 were incubated for 36 h at 39.5 °C prior shifting to 32 °C for 1 h (E–G). The localization of Kitl protein was determined by GFP fluorescence (B, E); ER and Golgi localization were determined by staining with anti-calreticulin (C) or anti-TGN46 (F) antibodies, respectively. The merged images are indicated in the right panels (D, G). Scale bars = 10 µm.

 
The importance of the C-terminal valine of Kitl in ER export was further studied by shifting the cells from the non-permissive to the permissive temperature. Observation at 10, 30, and 60 min after the temperature shift showed that the del36 mutant Kitl protein remained in the ER, whereas wild-type Kitl was efficiently accumulated in the Golgi area (Fig. 4A). We quantified this transport by counting the cells displaying Kitl in the Golgi (Fig. 4B). Wild-type Kitl was detectable after 10 min in 25% and after 30 min in 46% of the cells. In contrast, only 8% of cells transfected with the del36 mutant showed Golgi staining. For comparison, full-length VSV G ts-045 protein was detected in the Golgi of 55 and 80% of the cells after shifting the temperature for 10 or 30 min, respectively (Fig. 4B). Cell surface expression of the VSV Gts-Kitl chimeras was analyzed after 120 min. Full-length VSV Gts and wild-type VSV Gts-Kitl proteins, but not the del36 mutant proteins, were detected at the cell surface (Fig. 4C). These results demonstrate that the VSV Gts-Kitl chimera behaves similarly to the original Kitl protein in terms of ER export and transport to the cell surface. Thus it represents a valuable tool to study ER export signals in the cytoplasmic tail of Kitl and other transmembrane proteins (38). In addition, these experiments proved that the C-terminal valine residue of Kitl is essential for rapid and efficient ER export and may represent a bona fide ER export signal.



View larger version (45K):
[in this window]
[in a new window]
  		FIG. 4.
The C-terminal valine determines Kitl export from ER. A, fluorescence images of live COS-7 cells transiently transfected with wild-type (left column) and del36 mutant (middle column) forms of VSV Gts-EGFP-Kitl-M2 or with VSV Gts-YFP (right column). The cells were incubated for 36 h at 39.5 °C prior to shifting to 32 °C for 10, 30, or 60 min, as indicated. Scale bar = 10 µm. B, 36 h after transfection with VSV Gts-YFP, wild-type or del36 mutant forms of VSV Gts-EGFP-Kitl-M2 protein; COS-7 cells were shifted from 39.5 to 32 °C for the indicated periods of time. The percentage of cells exhibiting Golgi staining was then determined. Error bars correspond to the mean ± S.E. of at least three experiments. C, cell surface transport of wild-type and del36 mutant forms of the VSV Gts-Kitl chimeras were analyzed in transfected COS-7 cells, which were shifted to 32 °C for 2 h before cell surface biotinylation. Total cell lysates (Lys) or avidin-precipitated fractions (av) were analyzed by 7.5% SDS-PAGE and detected after Western blotting with monoclonal anti-EGFP antibody.

 
The Transport Signal of Kitl at Position 36 Is Most Efficiently Mediated by C-terminal Valine—We next explored whether other amino acids can substitute the C-terminal valine residue in ER export. We generated Kitl mutant proteins in which the C-terminal valine was replaced by amino acids with hydrophobic or aromatic side chains. Surprisingly, the most closely related hydrophobic amino acids, alanine, leucine, or isoleucine, were not able to restore efficient transport. Substitution by the aromatic phenylalanine, tryptophan, tyrosine, or the hydrophobic methionine also showed dramatic reduction of ER export. However, partial restoration of ER to Golgi transport (50% compared with wild-type Kitl) was obtained with substitution of valine with proline (Fig. 5A). Moreover, systematic substitution of the 8 amino acids upstream of the valine with alanine did not modify wild-type Kitl transport (data not shown). Moreover, the addition of amino acids C-terminal of the valine completely abrogated the transport signal (data not shown). This suggests that the C-terminal amino acid valine is essential for ER export of Kitl.



View larger version (30K):
[in this window]
[in a new window]
  		FIG. 5.
Quantification of wild-type and mutant Kitl protein export from the ER. A, COS-7 cells were transfected with wild-type or valine 36-substituted forms of VSV Gts-EGFP-Kitl-M2. 36 h after transfection, cells were shifted from 39.5 to 32 °C for 30 min. Then, the percentage of GFP fluorescent cells exhibiting a Golgi staining was evaluated. Results are expressed in respect to the Golgi staining in wild-type Kitl. B, COS-7 cells were transfected for 36 h with wild-type or mutant forms of VSV Gts-EGFP-Kitl-M2 protein at 39.5 °C. Golgi staining of wild-type, an internal cytoplasmic deletion (del12–28; see Fig. 1), and a C-terminal deletion exposing an new Ala-Val motif (del12) were compared with the original VSV Gts construct after 30 min at 32 °C. Then, the percentage of GFP fluorescent cells exhibiting a Golgi staining was evaluated. Error bars in A and B correspond to the mean ± S.E. of at least three experiments.

 
The Valine Signal Is Independent of the Kitl Basolateral Targeting Motif—We now asked whether the ER export signal was independent of the previously identified basolateral targeting signal of Kitl (12). To test this, we created the internal deletion mutant del12–28 of Kitl protein, which lacked the cytoplasmic domain required for basolateral sorting. This Kitl mutant protein was similarly transported from ER to Golgi as wild-type Kitl (Fig. 5B see also Fig. 1). In addition, this Kitl mutant protein revealed that the C-terminal valine when spaced 19 amino acids from the border between the transmembrane and cytoplasmic domains is as efficient as in the wild-type position. In conclusion, the cytoplasmic tail of Kitl exhibits two independent intracellular transport motifs, one for basolateral sorting and the other for ER export.

The Valine Signal Requires Minimal Spacing from the Border between the Transmembrane and Cytoplasmic Domains—We wanted to know whether further reduction of the distance from the border between the transmembrane and cytoplasmic domains would affect the efficiency of the valine signal. Therefore, an internal valine at position 11 of the cytoplasmic tail was exposed by C-terminal deletion (del12). This mutant protein, however, was no longer exported from the ER (Fig. 5B). This inhibition of ER export was not due to a change in the environment proximal to the valine, since a cytoplasmic tail of similar length, but consisting of the wild-type C-terminal Kitl sequence (KKKEREFQEV), was not exported either (data not shown). This suggests that the C-terminal valine in Kitl requires a minimal distance from the border between the transmembrane and cytoplasmic domains to be functional as an ER export signal.

The C-terminal Valine of Kitl Is a Key Determinant for COPII-mediated ER Export—Recent data showed that COPII-positive vesicles are involved in ER export (34, 39, 40). Since Kitl does not exhibit one of the known ER export signals, we now inquired whether the C-terminal valine docks Kitl to the COPII ER export mechanism. However, the transport of Kitl from ER to Golgi was too rapid to be analyzed in detail. Therefore we studied ER export of Kitl at low temperature as described previously for VSV Gts protein (30, 41). Incubation of the cells at 10 °C arrests ER export of VSV Gts at the COPII-positive exit sites, whereas 15 °C allows ER to Golgi transport but at reduced velocity. Along this line, wild-type and mutant VSV Gts-EGFP-Kitl proteins were accumulated in the ER at the non-permissive temperature for 36 h. The Kitl export from the ER was then analyzed at 10 °C or 15 °C respectively (Fig. 6). After 30 min at 10 °C, wild type VSV Gts-EGFP-Kitl appeared as punctiform structures associated with the ER. This behavior was comparable with the original VSV Gts protein. The punctiform staining was not found in cells expressing the del36 Kitl mutant protein (Figs. 6 and 7). The dynamic analysis of the dot like structures revealed a large pool of stationary structures and may represent immature COPII-coated ER exit sites (Fig. 6, A and C, and see videos 1 and 3 in the supplementary material). This notion was confirmed after fixation, showing that the Kitl dots co-localized with the COPII marker Sec24C (Fig. 7) (34). Occasionally, a small fraction of Kitl-positive dots detached from the ER and moved toward the cell center (data not shown). At 15 °C ER-associated dots containing wild-type Kitl formed already after 10 min. These structures were initially immobile and then detached from the ER and moved as vesicles toward the Golgi. After 1 h of incubation at 15 °C, tubular, Kitl-positive structures appeared in addition to the dots (Fig. 6D and see video 4 in the supplemental materials). These tubules were apparently formed by elongation of larger, immobile dots, suggesting that they emerge from Kitl-positive vesicles. They moved toward the center of the cell, occasionally capturing and transporting further Kitl positive vesicles. Subsequently tubules started to shuttle between the periphery and the center of the cell. The same type and timing of intracellular transport was observed with the VSV Gts control protein (Fig. 6F and see video 6 in the supplemental material). In contrast, the del36 mutant of Kitl protein, lacking the ER export signal, was not found in transport vesicles and tubules at 10 or 15 °C (Fig. 6, B and E, and see videos 2 and 5 in the supplemental material).



View larger version (144K):
[in this window]
[in a new window]
  		FIG. 6.
Dynamic analysis of ER export of Kitl at 10 or 15 °C. Transiently transfected COS-7 cells were kept 36 h at 39.5 °C before shifting to 10 °C (A–C) or 15 °C(D–F) for 1 h. Time-lapse sequences were then recorded from cells transfected with wild-type (A, D) and del36 mutant forms (B, E) of VSV Gts-EGFP-Kitl-M2 and compared with cells expressing the original VSV Gts-YFP construct (C, F). Magnified sequences of the boxed areas in A, C, D, and F (A', C', D', and F') are shown below the respective panels. Scale bar = 25 µm.

 



View larger version (70K):
[in this window]
[in a new window]
  		FIG. 7.
COPII-mediated ER export of VSV Gts-EGFP-Kitl and EGFP-Kitl protein. COS-7 cells were transiently transfected with wild-type or del36 mutant forms of VSV Gts-EGFP-Kitl-M2 (A–C and D–F, respectively) and EGFP-Kitl-M2 (G–I and J–L, respectively). The cells were subsequently incubated for 36 h at 39.5 °C. After shifting the temperature to 10 °C for 2 h, the localization of Kitl protein and COPII-coated ER exit sites were determined by confocal fluorescence microscopy for EGFP (A, D, G, J) and staining for Sec24C (B, E, H, K). The merged images are indicated in the right panel (C, F, I, L), and magnified views of A–F and G–L are shown below the respective series (A'–L'). Scale bars = 10 µm.

 
We excluded an artifactual role of the temperature-sensitive VSV Gts protein in the recruitment of the chimeric Kitl to COPII-coated vesicles at 10 °C by showing that non-tagged EGFP-Kitl protein could also accumulate in Sec24C-positive ER exit sites during a 2-h incubation period (Fig. 7G). In contrast, del36 mutant Kitl protein was excluded from Sec24C-positive sites (Fig. 7J). These experiments proved that wild-type Kitl is efficiently exported from the ER by recruitment into COPII-coated vesicles. In contrast, the deletion of the C-terminal valine abrogated the recruitment of Kitl into the COPII coated vesicular export system resulting in delayed forward transport and reduced cell surface expression.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Characterization of the C-terminal Valine of Kitl as an ER Export Signal—In this study, we demonstrate that efficient export of Kitl from the ER relies on the cytoplasmic C-terminal valine. This is the first report describing the physiological role of this valine signal for the ER export of a functional oligomeric cell surface protein under homeostatic conditions. We also show that this signal is essential for the capture of Kitl into ER exit sites allowing further transport to the cell surface.

Anterograde transport from the ER to the Golgi is mediated by vesicles coated with COPII proteins (42–44). Interestingly di-hydrophobic and di-acidic motifs interact with COPII components, and in some cases, accessory factors are required to direct transmembrane cargo into COPII vesicles (44–46). The COPII coat is formed by a membrane proximal component consisting of the Sar1GTP ·Sec23 ·Sec24 complex further cross-linked by the Sec13 ·Sec31 heterodimer (47). The C-terminal valine of Kitl requires a minimal distance from the border between the transmembrane and cytoplasmic domains of 12–20 amino acids to be recognized by the ER export machinery. This distance and the high selectivity for a C-terminal valine residue distinguishes the Kitl signal from previously described hydrophobic C-terminal ER export motifs (18, 19, 22, 48). The specificity of the C-terminal valine signal is in agreement with recent studies demonstrating distinct and independent binding sites for different cargo molecules on the Sec24 subunit of the COPII complex (49, 50). Whether Kitl uses the same structural component of the COPII coat awaits further analysis. Replacement of valine with other apolar amino acids was non-functional, showing that it was not the hydrophobic nature of valine that leads to efficient ER export. Therefore, the Kitl ER export signal is different from the well studied, dihydrophobic ER export signal in ERGIC-53, which can tolerate various combinations of C-terminal hydrophobic residues (48). Partial restoration of Kitl ER export was only observed with a C-terminal proline, an amino acid with a volume similar to valine, which may fit into a putative binding pocket of the COPII, ER export complex. The importance of valine-specific binding pockets other than within the COPII complex has been described in protein-protein interactions occurring at the plasma membrane. Three different PDZ domain recognition motifs (classes I–III) depending on C-terminal valine play a role in protein transport from the trans-Golgi network to domains of the plasma membrane (51). In addition to the C-terminal valine, these domains require a critical amino acid at position –2. They consist of Ser or Thr in the class I motif, hydrophobic amino acids in class II, or acidic residues in class III (52). In the case of Kitl, however, the C-terminal valine is sufficient and the glutamine at position –2 is not essential for ER export (data not shown). It is therefore unlikely that the C-terminal valine of Kitl is recognized during ER export by a protein containing a PDZ domain. However, we cannot exclude the possibility that a PDZ containing adaptor protein recognizes the C-terminal valine during the passage through the Golgi or later phases of the transport (53, 54).

Comparison with Other Hydrophobic ER Export Signals— Different ER to Golgi transport signals have been identified in other transmembrane proteins (19, 21, 55). A structurally Kitl-related ER export signal is formed by two consecutive C-terminal phenylalanines in the lectin ERGIC, a protein involved in ER to Golgi transport of glycoproteins (56). These residues can be mutated into other hydrophobic amino acids conserving ER export of ERGIC, which was not the case in Kitl. Moreover, the ER export motif in ERGIC is located at a distance of 10 amino acids away from the border between the transmembrane and cytoplasmic domains. In contrast, the C-terminal valine in Kitl is located at position 36. In the transmembrane growth factor TGF{alpha} the cytoplasmic tail is similar in length to that of Kitl, but two C-terminal valines form the ER export signal (57, 58). Whether TGF{alpha} uses the Kitl or the ERGIC ER export system is not clear, since there are homologies to the di-hydrophobic signal in ERGIC and the monomeric valine in Kitl. The issue is further complicated, since TGF{alpha} has an isoprenylation site juxtaposed to the di-valine. This modification creates a novel membrane anchor within the cytoplasmic tail that brings the C-terminal di-valine closer to the membrane, converting a membrane distal into a membrane proximal signal. These data demonstrate that the ER export machinery, which recognizes the C-terminal valine in Kitl, seems to be different from the transport strategy taken by ERGIC and TGF{alpha}.

COPII-assisted Recruitment of Kitl in ER Exit Sites—The temperature-sensitive mutant of the G protein of vesicular stomatitis virus represents an invaluable tool to study ER export and visualization of cargo transport through the secretory pathway of living cells (25–31). While the G protein folds normally at 32 °C, it is misfolded at 39.5 °C and is retained in the ER by the quality control system (32, 33). The potential of this system for the study of cargo recruitment into ER exit sites became apparent due to the possibility of blocking ER export at 10 °C (41). This temperature block separates the process of recruitment of oligomeric (trimeric) VSV Gts protein into COPII-coated vesicles from that of the transport to the Golgi. Using this technique, we demonstrated that the C-terminal valine of Kitl is required for the specific recruitment of Kitl or a VSV Gts-Kitl chimera into COPII-coated ER exit sites. The retarded appearance of the VSV Gts-tagged Kitl in ER exit sites at 10 °C is likely due to a slow down of the (oligomerization) trimerization process of the extracellular VSV Gts domain required for ER export (25, 59). In the absence of the C-terminal valine in the VSV Gts Kitl chimera, recruitment to ER exit sites and transport to the cell surface is completely blocked (Fig. 8). Whether this lack of transport is due to aborted formation of the quaternary structure of the VSV Gts ectodomain (trimerization) or the absence of alternative ER export mechanisms (bulk flow) is currently not known (19, 60). In contrast, despite the absence of valine ER export signal in Kitl, the molecule is still transported to the cell surface but with lower efficiency (Fig. 8). In the case of Kitl, bulk flow may take place due to extremely high concentrations of Kitl in the ER of Kitl del36-expressing COS cells (Fig. 8). In accordance, we found that in these cells the number of molecules transported to the cell surface was proportional to the number of molecules in the ER. In cells expressing wild-type Kitl, however, transport is efficient already when a few Kitl molecules are present in the ER (Fig. 8). In summary, our results suggest that the valine signal facilitates recruitment of Kitl near the ER exit sites, inducing capture by the COPII coat immediately after biosynthesis. This concept of valine-dependent COPII recruitment and ER export is illustrated in Fig. 8, B and C. Apparently, this mechanism is critical for proteins that are expressed at low levels such as growth factors to get efficiently and rapidly transported to the cell surface.



View larger version (45K):
[in this window]
[in a new window]
  		FIG. 8.
Model of C-terminal valine-dependent Kitl recruitment to ER exit sites. A, graphic view of protein synthesis, ER accumulation, and ER export of wild-type and del36 mutant forms of EGFP-Kitl and VSV Gts-EGFP-Kitl chimeras. The graph on the left-hand side indicates specific ER export of wild-type Kitl protein at low concentrations in the ER. In contrast, the transport of the C-terminal valine-deleted Kitl mutant protein is achieved only when high Kitl protein concentrations are reached in the ER. Despite the high concentration of the del36 mutant form of the VSV Gts-EGFP-Kitl protein in the ER at 39.5 °C, this construct is not able to exit the ER by bulk flow at the permissive temperature of 32 °C (right-hand side). B, model of signal or bulk flow dependent recruitment of Kitl into ER exit sites. The valine-dependent ER export signal is able to capture and concentrate Kitl at ER exit sites. In the absence of the valine signal, ER export can only be achieved at high Kitl concentrations that are required to induce spontaneous localization into ER export sites (B, right-hand side). C, two-step model of ER exit site recruitment of the temperature-sensitive VSV Gts-EGFP-Kitl chimera. ER export of this construct at permissive temperatures requires an additional step of refolding and trimerization of the VSV Gts-ectodomain prior to ER exit site recruitment. Once the VSV Gts-EGFP-Kitl is released from the chaperone-mediated retention, the functional trimer is recruited to ER exit sites by a C-terminal valine-dependent mechanism. In the absence of the valine signal, spontaneous ER exit site localization and subsequent bulk flow is blocked at permissive temperatures (C, right-hand side). It is currently not known whether the cause of this inhibition of transport is the absence of VSV Gts refolding or lack of valine-dependent recruitment to ER exit sites.

 
Independent Basolateral Sorting and ER Export Signals in Kitl—Our results show that the previously identified basolateral targeting motif of Kitl, consisting of a critical leucine residue and an acidic cluster, is not involved in ER export (12). It also suggests that during intracellular transport of Kitl, recognition of the ER export motif precedes that of the basolateral-targeting determinant. In conclusion, the cytoplasmic tail of Kitl controls the number of molecules transported to the cell surface independently of their polarized sorting. This suggests that a C-terminal deletion of Kitl would result in a dramatic reduction of cell surface expression due to a failure of specific ER export. In this respect, two mouse Kitl mutations are of interest, since they may split between the effects of ER export and basolateral sorting mechanisms. The KitlSl-d mutation results from a genomic deletion that spans the transmembrane and cytoplasmic domains of Kitl (2, 61). This mutation is almost as severe as a full deletion, and its phenotype is likely to be caused by the reduced transport of Kitl to the cell surface due to deficient ER export (see Fig. 1, KLS). In contrast, the KitlSl-17H mutant protein has an altered cytoplasmic domain lacking the basolateral targeting signal but proceeds normally through the ER (11). The phenotypes associated with this Kitl mutation are restricted to polarized tissues such as the keratinocytes in the skin and Sertoli cells in the testis causing loss of pigmentation and sterility (11). Hence, the cytoplasmic tail of Kitl contains two distinct motifs, which do not interfere with each other; the C-terminal valine controls ER export and cell surface expression levels, whereas the leucine residue determines its polarized surface expression in epithelial cells.


   FOOTNOTES
 
* This work was supported in part by a grant from the INSERM (to F. P.) and by grants from the Swiss National Science Foundation (31-49241.96, 31-052727.97, 31-059173.99, 31-64000.00, and 5894.1KTS) (to B. A. I. and B. W.-H.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{boxs} The on-line version of this article (available at http://www.jbc.org) contains videos 1–6. Back

{ddagger} To whom correspondence should be addressed. Tel.: 41-22-379-5764; Fax: 41-22-379-5746; E-mail: Bernhard.Wehrle-Haller{at}medecine.unige.ch.

1 The abbreviations used are: Kitl, Kit ligand; COPII, coat protein complex II; GFP, green fluorescent protein; EGFP, enhanced GFP; YFP, yellow fluorescent protein; ER, endoplasmic reticulum; ERGIC, endoplasmic reticulum Golgi intermediate compartment; PBS, phosphate-buffered saline; TGN, trans-Golgi network; VSV, vesicular stomatitis virus; HA, hemagglutinin; MDCK, Madin-Darby canine kidney; DMEM, Dulbecco's modified Eagle's medium; FCS, fetal calf serum; BSA, bovine serum albumin; Endo-H, endoglycosidase-H; TGF{alpha}, tumor growth factor {alpha}. Back


   ACKNOWLEDGMENTS
 
We thank Drs. Jean-Pierre Paccaud and Alessandra Pagano for providing antibodies against Sec24C. We are grateful to Drs. Michel Aurrands-Lions and Caroline Johnson-Léger for helpful discussion and to Drs. Paul Bradfield, Guy Brighouse, and Monique Wehrle-Haller for carefully reading the manuscript. We thank Marie-Claude Jacquier for excellent technical assistance.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Massague, J., and Pandiella, A. (1993) Annu. Rev. Biochem. 62, 515–541[CrossRef][Medline] [Order article via Infotrieve]
  2. Flanagan, J. G., Chan, D. C., and Leder, P. (1991) Cell 64, 1025–1035[CrossRef][Medline] [Order article via Infotrieve]
  3. Huang, E. J., Nocka, K. H., Buck, J., and Besmer, P. (1992) Mol. Biol. Cell 3, 349–362[Abstract]
  4. Wehrle-Haller, B., and Weston, J. A. (1995) Development (Camb.) 121, 731–742[Abstract]
  5. Tajima, Y., Huang, E. J., Vosseller, K., Ono, M., Moore, M. A., and Besmer, P. (1998) J. Exp. Med. 187, 1451–1461[Abstract/Free Full Text]
  6. Tajima, Y., Moore, M. A., Soares, V., Ono, M., Kissel, H., and Besmer, P. (1998) Proc. Natl. Acad. Sci. U. S. A. 95, 11903–11908[Abstract/Free Full Text]
  7. Kunisada, T., Lu, S. Z., Yoshida, H., Nishikawa, S., Mizoguchi, M., Hayashi, S., Tyrrell, L., Williams, D. A., Wang, X., and Longley, B. J. (1998) J. Exp. Med. 187, 1565–1573[Abstract/Free Full Text]
  8. Heissig, B., Hattori, K., Dias, S., Friedrich, M., Ferris, B., Hackett, N. R., Crystal, R. G., Besmer, P., Lyden, D., Moore, M. A., Werb, Z., and Rafii, S. (2002) Cell 109, 625–637[CrossRef][Medline] [Order article via Infotrieve]
  9. Kunisada, T., Yoshida, H., Yamazaki, H., Miyamoto, A., Hemmi, H., Nishimura, E., Shultz, L. D., Nishikawa, S., and Hayashi, S. (1998) Development (Camb.) 125, 2915–2923[Abstract]
  10. Brannan, C. I., Bedell, M. A., Resnick, J. L., Eppig, J. J., Handel, M. A., Williams, D. E., Lyman, S. D., Donovan, P. J., Jenkins, N. A., and Copeland, N. G. (1992) Genes Dev. 6, 1832–1842[Abstract/Free Full Text]
  11. Wehrle-Haller, B., and Weston, J. A. (1999) Dev. Biol. 210, 71–86[CrossRef][Medline] [Order article via Infotrieve]
  12. Wehrle-Haller, B., and Imhof, B. A. (2001) J. Biol. Chem. 276, 12667–12674[Abstract/Free Full Text]
  13. Wehrle-Haller, B. (2003) Pigment Cell Res. 16, 287–296[CrossRef][Medline] [Order article via Infotrieve]
  14. Cheng, H. J., and Flanagan, J. G. (1994) Mol. Biol. Cell 5, 943–953[Abstract]
  15. Pfeffer, S. R., and Rothman, J. E. (1987) Annu. Rev. Biochem. 56, 829–852[CrossRef][Medline] [Order article via Infotrieve]
  16. Wieland, F. T., Gleason, M. L., Serafini, T. A., and Rothman, J. E. (1987) Cell 50, 289–300[CrossRef][Medline] [Order article via Infotrieve]
  17. Martinez-Menarguez, J. A., Geuze, H. J., Slot, J. W., and Klumperman, J. (1999) Cell 98, 81–90[CrossRef][Medline] [Order article via Infotrieve]
  18. Nishimura, N., and Balch, W. E. (1997) Science 277, 556–558[Abstract/Free Full Text]
  19. Sevier, C. S., Weisz, O. A., Davis, M., and Machamer, C. E. (2000) Mol. Biol. Cell 11, 13–22[Abstract/Free Full Text]
  20. Votsmeier, C., and Gallwitz, D. (2001) EMBO J. 20, 6742–6750[CrossRef][Medline] [Order article via Infotrieve]
  21. Giraudo, C. G., and Maccioni, H. J. (2003) Mol. Biol. Cell 14, 3753–3766[Abstract/Free Full Text]
  22. Kappeler, F., Klopfenstein, D. R., Foguet, M., Paccaud, J. P., and Hauri, H. P. (1997) J. Biol. Chem. 272, 31801–31808[Abstract/Free Full Text]
  23. Barlowe, C. (2002) Curr. Opin. Cell Biol. 14, 417–422[CrossRef][Medline] [Order article via Infotrieve]
  24. Alvarez, C., Fujita, H., Hubbard, A., and Sztul, E. (1999) J. Cell Biol. 147, 1205–1222[Abstract/Free Full Text]
  25. Kreis, T. E., and Lodish, H. F. (1986) Cell 46, 929–937[CrossRef][Medline] [Order article via Infotrieve]
  26. Beckers, C. J., Keller, D. S., and Balch, W. E. (1987) Cell 50, 523–534[CrossRef][Medline] [Order article via Infotrieve]
  27. Bergmann, J. E. (1989) Methods Cell Biol. 32, 85–110[Medline] [Order article via Infotrieve]
  28. Balch, W. E., McCaffery, J. M., Plutner, H., and Farquhar, M. G. (1994) Cell 76, 841–852[CrossRef][Medline] [Order article via Infotrieve]
  29. Aridor, M., Bannykh, S. I., Rowe, T., and Balch, W. E. (1995) J. Cell Biol. 131, 875–893[Abstract/Free Full Text]
  30. Presley, J. F., Cole, N. B., Schroer, T. A., Hirschberg, K., Zaal, K. J., and Lippincott-Schwartz, J. (1997) Nature 389, 81–85[CrossRef][Medline] [Order article via Infotrieve]
  31. Scales, S. J., Pepperkok, R., and Kreis, T. E. (1997) Cell 90, 1137–1148[CrossRef][Medline] [Order article via Infotrieve]
  32. Lippincott-Schwartz, J., and Smith, C. L. (1997) Curr. Opin. Neurobiol. 7, 631–639[CrossRef][Medline] [Order article via Infotrieve]
  33. Nehls, S., Snapp, E. L., Cole, N. B., Zaal, K. J., Kenworthy, A. K., Roberts, T. H., Ellenberg, J., Presley, J. F., Siggia, E., and Lippincott-Schwartz, J. (2000) Nat. Cell Biol. 2, 288–295[CrossRef][Medline] [Order article via Infotrieve]
  34. Stephens, D. J., Lin-Marq, N., Pagano, A., Pepperkok, R., and Paccaud, J. P. (2000) J. Cell Sci. 113, 2177–2185[Abstract]
  35. Wehrle-Haller, B., Morrison-Graham, K., and Weston, J. A. (1996) Dev. Biol. 177, 463–474[CrossRef][Medline] [Order article via Infotrieve]
  36. Chalfie, M., and Kain, S. (eds) (1998) Green Fluorescent Protein: Properties, Applications, and Protocols, Wiley-Liss, Inc., New York
  37. Zhang, J., Campbell, R. E., Ting, A. Y., and Tsien, R. Y. (2002) Nat. Rev. Mol. Cell. Biol. 3, 906–918[CrossRef][Medline] [Order article via Infotrieve]
  38. Cole, N. B., Ellenberg, J., Song, J., DiEuliis, D., and Lippincott-Schwartz, J. (1998) J. Cell Biol. 140, 1–15[Abstract/Free Full Text]
  39. Barlowe, C. (1998) Biochim. Biophys. Acta 1404, 67–76[Medline] [Order article via Infotrieve]
  40. Aridor, M., Fish, K. N., Bannykh, S., Weissman, J., Roberts, T. H., Lippincott-Schwartz, J., and Balch, W. E. (2001) J. Cell Biol. 152, 213–229[Abstract/Free Full Text]
  41. Mezzacasa, A., and Helenius, A. (2002) Traffic 3, 833–849[CrossRef][Medline] [Order article via Infotrieve]
  42. Schekman, R., and Orci, L. (1996) Science 271, 1526–1533[Abstract]
  43. Springer, S., Spang, A., and Schekman, R. (1999) Cell 97, 145–148[CrossRef][Medline] [Order article via Infotrieve]
  44. Kuehn, M. J., Herrmann, J. M., and Schekman, R. (1998) Nature 391, 187–190[CrossRef][Medline] [Order article via Infotrieve]
  45. Marzioch, M., Henthorn, D. C., Herrmann, J. M., Wilson, R., Thomas, D. Y., Bergeron, J. J., Solari, R. C., and Rowley, A. (1999) Mol. Biol. Cell 10, 1923–1938[Abstract/Free Full Text]
  46. Powers, J., and Barlowe, C. (2002) Mol. Biol. Cell 13, 880–891[Abstract/Free Full Text]
  47. Matsuoka, K., Schekman, R., Orci, L., and Heuser, J. E. (2001) Proc. Natl. Acad. Sci. U. S. A. 98, 13705–13709[Abstract/Free Full Text]
  48. Nufer, O., Guldbrandsen, S., Degen, M., Kappeler, F., Paccaud, J. P., Tani, K., and Hauri, H. P. (2002) J. Cell Sci. 115, 619–628[Abstract/Free Full Text]
  49. Miller, E. A., Beilharz, T. H., Malkus, P. N., Lee, M. C., Hamamoto, S., Orci, L., and Schekman, R. (2003) Cell 114, 497–509[CrossRef][Medline] [Order article via Infotrieve]
  50. Mossessova, E., Bickford, L. C., and Goldberg, J. (2003) Cell 114, 483–495[CrossRef][Medline] [Order article via Infotrieve]
  51. Fanning, A. S., and Anderson, J. M. (1999) J. Clin. Invest. 103, 767–772[Medline] [Order article via Infotrieve]
  52. Nourry, C., Grant, S. G., and Borg, J. P. (2003) Science's STKE http://stke.sciencemag.org/cgi/content/full/sigtrans;2003/179/re7
  53. Fernandez-Larrea, J., Merlos-Suarez, A., Urena, J. M., Baselga, J., and Arribas, J. (1999) Mol. Cell 3, 423–433[CrossRef][Medline] [Order article via Infotrieve]
  54. Kuo, A., Zhong, C., Lane, W. S., and Derynck, R. (2000) EMBO J. 19, 6427–6439[CrossRef][Medline] [Order article via Infotrieve]
  55. Otte, S., and Barlowe, C. (2002) EMBO J. 21, 6095–6104[CrossRef][Medline] [Order article via Infotrieve]
  56. Zhang, B., Cunningham, M. A., Nichols, W. C., Bernat, J. A., Seligsohn, U., Pipe, S. W., McVey, J. H., Schulte-Overberg, U., de Bosch, N. B., Ruiz-Saez, A., White, G. C., Tuddenham, E. G., Kaufman, R. J., and Ginsburg, D. (2003) Nat. Genet. 34, 220–225[CrossRef][Medline] [Order article via Infotrieve]
  57. Briley, G. P., Hissong, M. A., Chiu, M. L., and Lee, D. C. (1997) Mol. Biol. Cell 8, 1619–1631[Abstract]
  58. Urena, J. M., Merlos-Suarez, A., Baselga, J., and Arribas, J. (1999) J. Cell Sci. 112, 773–784[Abstract]
  59. Doms, R. W., Keller, D. S., Helenius, A., and Balch, W. E. (1987) J. Cell Biol. 105, 1957–1969[Abstract/Free Full Text]
  60. Doms, R. W., Ru