| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biomembrane and Biofunctional Chemistry, Graduate School of Pharmaceutical Sciences, Hokkaido University, Japan
1Correspondence: Nishi 6, Kita 12, Kita-ku, Sapporo 060-0812 Japan. E-mail: yigarash{at}pharm.hokudai.ac.jp
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: Edg-1 • G-protein-coupled receptor • glycosylation • sphingosine 1-phosphate • membrane microdomain
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 2)
. Edg-1 was originally cloned as one of the immediate-early genes induced during in vitro differentiation of human endothelial cells (3)
and was recently identified as a receptor molecule for Sph-1-P. Edg-1 belongs to the receptor family for Sph-1-P and lysophosphatidic acid (LPA) (4)
, which consists of Edg-2/Vzg-1 (5)
, Edg-3 (6)
, Edg-4 (7)
, Edg-5/Agr16/H218 (8
, 9)
, Edg-6 (10)
, Edg-7 (11)
, and Edg-8/Nrg-1 (12
, 13)
. Recent studies have shown that Edg-1 regulates diverse signal transduction pathways implicated in cell motility (14
, 15)
, cell differentiation (4)
, and cell growth and angiogenesis (16)
through intracellular signaling molecules, such as MAP kinase and calcium, via heterotrimeric G-proteins (4
, 17
, 18)
and rho family small-G-proteins (4
, 16)
.
We first postulated that like most other GPCRs (19)
, Edg-1 would be post-translationally modified. One possible modification is the phosphorylation by a family of G-protein-coupled receptor kinases, which might enable its internalization through binding to another family of proteins called arrestins (20)
. Another possible modification of Edg-1 would be N-glycosylation in its amino-terminal extracellular portion (21)
, since many GPCRs possess one or more putative glycosylation sites this region, although the role of such N-glycosylation of GPCR is not clear at present.
In this study, we found for the first time that Edg-1 is indeed glycosylated in its amino-terminal extracellular portion. We further identified the specific glycosylation site as asparagine 30 by creating a nonglycosylated mutant of Edg-1 (N30D-Edg-1) and transfecting it into mammalian cell lines. The nonglycosylated receptor was normally expressed in the plasma membrane and exhibited normal ligand binding ability. This glycosylation seems to be important for Sph-1-P-induced receptor internalization. We observed that the lack of N-glycosylation in Edg-1 shows a poor tendency to accumulate in caveolin-enriched fractions based on results with sucrose density gradient centrifugation. Thus, this study has implied possible roles for the N-glycosylation of Edg-1 in regulating its ligand-induced dynamics in the ligand-stimulated cells.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-isoform) from Upstate Biotechnology (Lake Placid, NY). Alexa 488 goat anti-rabbit IgG (H+L) conjugate was from Molecular Probes (Eugene, OR) and anti-rabbit Ig Texas red-linked whole antibody, anti-rabbit IgG antibodies conjugated to horseradish peroxidase, and enhanced chemiluminescence (ECL or ECL-plus) detection kits were from Amersham Pharmacia Biotech (Piscataway, NJ).
DNA constructs
A FLAG epitope tag was introduced into the HindIII-BamHI sites of a mammalian expression vector pcDNA3 (Invitrogen, Palo Alto, CA) by ligation with the oligonucleotides 5'-AGCTTGCCACCATGGATTACAAGGATGACGACGATAAGG-3' and 5'-GATCCCTTATCGTCGTCATCCTTGTAATCCATGGTGGCA-3' yielding pcDNA3-FLAG1. An HA epitope tag was introduced into the BamHI-EcoRI sites of pcDNA3-FLAG1 by ligation with the oligonucleotides 5'-GATCCCTACCCATACGACGTCCCAGATTACGCTTG-3' and 5'-AATTCAAGCGTAATCTGGGACGTCGTATGGGTAGG-3' yielding pcDNA3-FLAG1-HA4. Murine Edg-1 cDNA was amplified from mRNA of NIH3T3 cells by reverse transcriptase-polymerase chain reaction (RT-PCR) using primers 5'-GTTCATTCTCATCTGCTGCT-3' and 5'-TATAGTGCTTGTGGTAGAGC-3'. The amplified Edg-1 cDNA was introduced into BamHI sites just before the start codon and after the stop codon by PCR using primers 5'-CGGGATCCATGGTGTCCACTAGCATCCC-3' and 5'-CGGGATCCGAAGAAGAATTGACGTTTCC-3'. The cDNA was subcloned into the BamHI site of an expression vector pcDNA3-FLAG1-HA4.
Mutagenesis
Point mutations were introduced into pcDNA3-FLAG-Edg-1-HA by site-directed mutagenesis. PCR was performed using 2 ng of the DNA template and primers enclosing the entire codon region (N21D; sense, 5'-AGCTCAGTCTCTGACTATGGGGACTATGA-3', antisense, 5'-GCGGAGAGCTTTAACCTCCGGGATGCT-3', N30D; sense, 5'-AAGTTGAACATCGGGGCGGAGAAGGACCA-3', antisense, 5'-GCCTGTGTAGTCGTAATGCCGGACTATGA-3', N36D; sense, 5'-AAGTTGGACATCGGGGCGGAGAAGGACCA-3', antisense, 5'-GCCTGTGTAGTTGTAATGCCGGACTATGA-3'). The antisense primers were treated with T4 polynucleotide kinase (Takara, Tokyo, Japan). The PCR was carried out in the presence of 0.2 mM dNTPs, 1 mM MgCl2, and 0.3 µM polynucleotide kinase-treated primers for 30 cycles. The PCR products were ligated by Ligation High (TOYOBO, Tokyo, Japan). All genes were completely sequenced after mutagenesis.
Cell culture and transfections
NIH3T3 cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum (Sigma). Chinese hamster ovary (CHO) cells were cultured in Ham’s F-12 medium (Sigma) with 10% fetal bovine serum (complete media). These cells were cultured at 37°C in a humidified 5% CO2 atmosphere. Cells were transfected using the LipofectAMINE Plus kit (Life Technologies, Gaithersburg, MD) according to the manufacturer’s instructions. Stably transfected clones expressing epitope-tagged Edg-1 were selected with 600 µg/mL Geneticin (G-418 sulfate; Life Technologies).
Electrophoresis and immunoblotting
Reducing SDS-PAGE was performed on 10% Laemmli gels. After electrophoresis, the proteins were transferred to a PVDF membrane (Millipore, Bedford, MA). Blocking was performed with 5% skim milk in TBS-T (20 mM Tris-HCl (pH 7.5), 137 mM NaCl, 0.05% Tween-20) for 1 h at room temperature. This was followed by incubation overnight at 4°C with the primary antibody. The blot was washed with TBS-T and incubated for 1 h at room temperature with the secondary antibody. The blot was washed again with TBS-T and detection was performed with ECL or ECL-plus and quantitated by densitometry with the NIH Image program.
Endoglycosidase digestion
Cells were washed twice with ice-cold PBS and lysed with extraction buffer (20 mM Tris-HCl (pH 7.5), 0.25 mM EDTA, 12 mM ß-glycerophosphate, 1 mM Na4P2O7, 5 mM NaF, 5 mM sodium orthovanadate, 5% glycerol, 1 mM phenylmethylsulfonyl fluoride, 2 mM dithiothreitol, and 1.5 µg/mL aprotinin). After incubation for 30 min at 0°C, membrane pellets were separated from extracts by centrifugation at 100,000 g at 2°C for 1 h. The pellets were resuspended in denaturing buffer (containing 0.5% SDS and 1% 2-mercaptoethanol) and boiled for 10 min. Endo H or PNGase F digestion was performed on the lysates at 37°C for 2 h according to the manufacturer’s recommended procedure. The control samples were incubated in parallel. The reactions were stopped by adding 4x Laemmli sample buffer (250 mM Tris-HCl (pH 6.8), 8% SDS, 40% glycerol, and 20% 2-mercaptoethanol). These samples were analyzed by Western blot.
MAPK activation
Stably transfected CHO cells expressing Edg-1 or a mutated form were grown in 6-well plates for 24 h in complete medium, followed by serum starvation for 16 h. Cells were incubated with serum-free medium containing various concentrations of Sph-1-P. After the indicated times, these cells were washed twice with ice-cold PBS and solubilized with Laemmli sample buffer. Proteins were separated by SDS-PAGE and analyzed by Western blotting as described above. MAPK activation was determined by Western blot analysis with anti-phospho-MAPK and anti-MAPK antibodies and quantified by densitometry with the NIH Image program.
Immunofluorescence microscopy
Stably transfected CHO cells expressing WT- or N30D-Edg-1 were grown on glass coverslips (Matsunami, Kyoto, Japan). The cells were washed twice with PBS, fixed for 15 min in 3.7% formaldehyde in PBS, washed three times with PBS, and permeabilized in 0.5% Triton X-100 in PBS for 5 min. Blocking was performed with PBS containing 10% BSA for 1 h at room temperature. All immunostaining was performed using a 1:100 dilution of the anti-HA polyclonal antibody and detected using a 1:50 dilution of the anti-rabbit-Ig Texas red conjugated antibody or a 1:300 dilution of the Alexa 488 anti-rabbit IgG conjugate. These primary and secondary antibodies were incubated for 1 h at room temperature. The cells were washed three times with PBS after incubation. The coverslips were rinsed in water and mounted onto glass slides using Mowiol 4–88 (Calbiochem, San Diego, CA). Cell Images were digitally captured from a Zeiss Axiophot 2 (Carl Zeiss, Thornwood, NJ) microscope using CCD camera (ProgRes3008, Kontron Electronik, Everett, MA) and Metamorph software (Universal Imaging Corp., Downingtown, PA).
Immunostaining of internalized Edg-1
Stably transfected CHO cells expressing WT- or N30D-Edg-1 were grown on glass coverslips for 16 h in complete medium, followed by serum starvation for 5 h. The cells were incubated with serum-free medium containing the indicated concentrations of Sph-1-P. After the indicated times, cells were fixed and immunostained as described above. To calculate the receptor internalization, 
100 cells were selected in random fields using a Zeiss Axiophot 2 microscope with a x63 objective and the number of cells showing internalized receptors was counted.
Receptor recovery
Stably transfected CHO cells expressing WT- or N30D-Edg-1 were grown on glass coverslips for 16 h in complete medium, followed by serum starvation for 5 h in the presence of 10 µg/mL cycloheximide. The cells were incubated for 30 min with serum-free medium containing 100 nM Sph-1-P and 10 µg/mL cycloheximide. After removal of Sph-1-P, cells were immunostained as described above. To calculate the receptor recovery, 100 cells were selected in random fields and the number of cells exhibiting receptor recovery to the cell surface was counted.
Proteinase K digestion
Stably transfected CHO cells expressing WT- or N30D-Edg-1 were grown in 6-well plates for 24 h in complete medium, followed by changing medium containing 0.2% FBS for 16 h. The cells were incubated with the indicated concentrations of Sph-1-P. Cells were washed twice with cold PBS, and scraped with 1 mM EDTA-PBS. The cell suspensions were treated with 0.1 mg/mL proteinase K with or without 1% Triton X-100 at 4°C for 1 h. Reactions were stopped by adding an equal volume of 10% trichloroacetic acid (TCA), mixed thoroughly, and incubated at 4°C for 20 min. The TCA precipitates were washed once with 5% TCA and twice with cold acetone. The precipitates were lysed with Laemmli sample buffer and analyzed by Western blot.
Preparation of caveolae-enriched membrane fractions
Stably transfected CHO cells confluently grown in 150 mm dishes were scraped into 2 ml of 500 mM sodium carbonate (pH 11), homogenized (5 strokes in a TeflonTM homogenizer), and sonicated (6x10 s). The homogenate was adjusted to 45% sucrose by the addition of 2 ml of 90% sucrose in MBS (25 mM MES, pH 6.5, 150 mM NaCl) and placed at the bottom of an ultracentrifuge tube. A 5–30% discontinuous sucrose gradient was formed above (4 ml of 5% sucrose, 4 ml of 30% sucrose; both in MBS). After centrifugation for 16–18 h at 100,000 g in an SW41 rotor (Beckman Instruments, Fullerton, CA), 1 ml fractions were collected and analyzed by Western blot.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A revealed that some modification had occurred in Edg-1 (lane 2), since the molecular mass of the stained band is higher than the predicted one from the cDNA sequence (42 kDa) (3)
. In light of previous reports showing that some GPCRs are glycosylated in their amino-terminal portion (19)
, we considered the possibility that transfected Edg-1 is glycosylated in the NIH3T3 cells,. We performed an endoglycosidase digestion of Edg-1 expressed in NIH3T3 cells, followed by SDS-PAGE and Western blotting with anti-tag antibodies (lane 3, 4). As predicted, with endoglycosidase treatments, the shift in the mobility was observed both in anti-FLAG (amino-terminal) and anti-HA (carboxyl-terminal) blots, clearly indicating the presence of oligosaccharide. Moreover, glycosylation of Edg-1 by Endo H-sensitive oligosaccharides modified in ER (high mannose-type) were observed with anti-FLAG antibody, whereas PNGase F-sensitive glycosylation form modified in Golgi (complex or hybrid type) of Edg-1 were observed only with anti HA-antibody. There were two or more bands of Edg-1 in an anti-HA antibody blotting, the reasons for which are not clear. Similar phenomena were observed when the same cDNA was transfected into CHO, COS-7 and HEK293 cells or when using another construction of the Edg-1, which has an HA tag at the amino terminus and a FLAG tag at the carboxyl terminus (data not shown). These results together suggest that the amino-terminal portion of Edg-1 was partially cleaved at some stage of its transport process through Golgi or after its expression on the cell surface (unpublished results).
|
As shown in Fig. 1B
, the predicted amino acid sequence of Edg-1 reveals that Asn-30 and Thr-32 embody the consensus sequence (-NXS/T-) for the N-glycosylation site. To determine the exact site of glycosylation, we prepared three Edg-1 mutants in which each of three asparagine residues (Asn-21, Asn-30, and Asn-36) was replaced by aspartic acid; epitope-tagged forms of these mutants (N21D, N30D, N36D) or WT-Edg-1 were transfected into NIH3T3 cells. Western blots clearly showed that N30D-Edg-1 was not glycosylated, whereas N21D- or N36D-Edg-1 was (Fig. 1C
). We concluded that Asn-30 is the major glycosylation site of Edg-1.
Stably expressed N30D-Edg-1 is predominantly located in the plasma membrane
In place of transiently transfected NIH3T3 cells, we established CHO clones stably expressing either WT- or N30D-mutant Edg-1 in order to further investigate the importance of the glycosylation in Edg-1. To confirm the glycosylation states of WT- or N30D-Edg-1 in CHO cells, we treated the extracts of those clones with endoglycosidase under similar conditions as in the NIH3T3 cells (Fig. 1)
. The results in Fig. 2
A indicate that the stably transfected WT-Edg-1 is also glycosylated on Asn-30 in CHO cells (lane 2), since nonglycosylated N30D-Edg-1 (lane 4) and endoglycosidase-treated WT-Edg-1 (lane 3) appeared identical, similar to results observed with the transient transfected NIH3T3 cells (Fig. 1)
. Again, stably expressing WT- or N30D-Edg-1 in CHO cells yielded two or more bands as well as transiently expressing experiments (Fig. 1)
.
|
To observe the localization of WT- or N30D-Edg-1 in CHO cells, these cells were immunostained with an anti-HA antibody in the steady state (Fig. 2B
). The results show that expression of the WT- or N30D-Edg-1 localizes predominantly in plasma membrane, with a small amount of perinuclear distribution (e.g., Golgi). To examine the ligand binding sensitivity of the N-glycosylation-deficient mutant of Edg-1, we measured the specific binding of ligand in WT- or N30D-Edg-1-expressing CHO cells; Scatchard analysis for WT-or N30D-Edg-1-expressing cells indicated a dissociation constant (Kd) of 25.7 nM or 26.4 nM, respectively. These results show there is no apparent difference in Kd values between WT- and N-30D-Edg-1, indicating that the N-glycans of Edg-1 do not affect the binding ability of the receptor for Sph-1-P.
N-glycosylation of Edg-1 does not affect ligand-induced MAP kinase activation
It is well known that p42-mitogen-activated protein kinase (p42-MAP kinase, also known as ERK2) is activated by Sph-1-P through Edg-1 (4
, 17
, 18)
. Therefore, we examined whether the ligand-induced MAP kinase activation through Edg-1 was affected by its amino-terminal glycosylation. We analyzed the time course of MAP kinase activation by Sph-1-P through WT- or N30D-Edg-1. CHO cells stably expressing WT- or N30D-Edg-1 were treated with 10 nM Sph-1-P. After the indicated times, cells were extracted and Western blotting was performed. MAP kinase activation was analyzed by Western blot using an anti-phospho MAP kinase antibody and an anti-MAP kinase antibody. As shown in Fig. 3
, there was no apparent difference in a time-dependent manner of MAPK activation between WT- and N30D-Edg-1-expressing CHO cells. Similar results were obtained in a dose-response manner (data not shown). These results suggest that amino-terminal glycosylation is not necessary for ligand-induced intracellular signal transductions through heterotrimeric G-proteins, as shown here in MAP kinase activation. However, since receptor-induced signaling is known to be amplified through cascade reactions, it can be cautiously said that measuring the MAPK activity (a downstream event in a cascade) alone might be insufficient to evaluate to the full extent the role of N-glycosylation.
|
Effect of N-glycosylation of Edg-1 on the receptor internalization observed with immunofluorescence microscopy
Internalization of ligand-induced G-protein-coupled receptors is required for resensitization and recycling to the cell surface, as well as for lysosomal degradation. Recent reports have shown that Edg-1 undergoes internalization after treatment with its ligand Sph-1-P from the cell surface to intracellular vesicles (22)
. This process has been proposed to be crucial for the regulation of receptor signaling and cellular responsiveness (23)
. We examined whether the N-glycan chains of the Edg-1 may affect its ligand-induced receptor internalization. WT- and N30D-Edg-1 were immunostained using the anti-HA antibody, followed by treatment with 10 to 50 nM of Sph-1-P for 10 min (Fig. 4
A). Receptor internalized cells were counted as described in Materials and Methods. The results in Fig. 4B
indicate that WT-Edg-1-expressing CHO cells were rapidly internalized with a 10 min treatment of 
10 nM of Sph-1-P whereas N30D-Edg-1 needed 50 nM ligand for its complete internalization. The rate of ligand-induced receptor internalization did not increase in N30D-Edg-1-expressing cells after long-term treatment (
60 min) with 10 nM Sph-1-P (Fig. 4C, D
). These results suggest that WT-Edg-1-expressing CHO cells were rapidly internalized by the treatment of low, physiological concentrations of Sph-1-P, whereas N30D-Edg-1, the mutant without a glycosylation site, was not internalized effectively under these conditions; a much higher concentrations of the ligand were needed for the internalization.
|
N-glycosylation of Edg-1 does not affect receptor recycling from intracellular vesicles to the plasma membrane
Internalized Edg-1-GFP expressed in HEK293 cells has been shown to be recycled to the plasma membrane within 2 h (22)
. We examined the difference of receptor recycling between WT- and N30D-Edg-1-expressing CHO cells. Cells were preincubated for 0.5 h with 100 nM Sph-1-P and 10 µg/ml cycloheximide in order to internalize all the ligand bound receptors. Sph-1-P was removed by washing, and the cells were incubated for the indicated times with the same concentration of cycloheximide to block the de-novo synthesis of Edg-1. More than half the expressed WT-Edg-1 was recovered to the plasma membrane for 120 min after Sph-1-P washout (Fig. 5
A). N30D-Edg-1 was also recycled, and at the same rate as WT-Edg-1 (Fig. 5B
). These results strongly suggest that in contrast to its effect on the internalization of the receptor, the presence of N-glycosylation on Edg-1, does not affect its recycling from intracellular vesicles to the plasma membrane.
|
Proteinase K digestion in surface remaining Edg-1 with Sph-1-P stimulated cells
As shown in Fig. 4
, observed with immunofluorescence microscopy, the amino-terminal glycosylation seems to be essential for efficient ligand-induced internalization of Edg-1. However, this observation alone is not sufficient to unambiguously decide whether Edg-1 was really internalized from the cell surface or formed clusters in the cell surface, not the internalized vesicles, on Sph-1-P stimulation. To distinguish surface receptors from internalized ones, we performed a proteinase K digestion of Edg-1 in which proteinase K can attack only cell surface receptors but not internalized ones (Fig. 6
A). WT- or N30D-Edg-1-expressing CHO cells were treated with 0.1 mg/mL proteinase K under nonpermeabilized Triton X-100 (-) conditions, followed by stimulation with Sph-1-P for 3 min. Proteinase K-resistant, internalized Edg-1 was detected by Western blot using an anti-HA antibody. The quantitative calculation of these results (Fig. 6B
) was consistent with the immunofluorescence microscopic analyses (Fig. 4)
. In contrast, proteinase K digestion on the cell surface WT- or N30D-Edg-1 in the absence of Sph-1-P under nonpermeabilized conditions showed no apparent differences, suggesting that WT- and N30D-Edg-1 distributed equally to the cell surface (Fig. 6A
, lane 3).
|
To be more precise, the apparent decrease in proteinase K-resistant receptors (shown in Fig. 6
) may be due to less internalization or more degradation after internalization. Therefore, to exclude the latter possibility, we examined the receptor degradation after Sph-1-P-induced receptor internalization of WT- and N30D-Edg-1-expressing CHO cells (Fig. 7
). These cells were treated with the Sph-1-P for 10 min after serum starvation for 16 h. Whole cell extracts were analyzed by Western blot using an anti-HA antibody. These results clearly revealed that WT- and N30D-Edg-1 are not degraded on Sph-1-P stimulation for 10 min.
|
Based on these three criteria—1) observation with immunofluorescence microscopy, 2) proteinase K resistance, 3) distinction between internalization and possible after degradation of the receptors—we concluded that nonglycosylated form of Edg-1 is much less responsive for ligand-induced internalization.
Distribution of WT- or N30D-Edg-1 in sucrose density gradient fractions from CHO cells
We tried to rationalize the different sensitivity for the internalization between nonglycosylated and WT-Edg-1. One possibility is that the orientation of WT-Edg-1 and N30D-Edg-1 in the plasma membrane, especially in the so-called microdomain or raft, may differ and may result in the change in the rate of receptor internalization as observed in Figs. 4
and 6
. We performed a sucrose density gradient centrifugation to separate the microdomain fractions from CHO cells stably expressing WT- or N30D-Edg-1. As shown in Fig. 8
, most of the WT-Edg-1 was clearly localized in the caveolae-enriched fractions whereas N30D-Edg-1 was dispersed across the broader fractions under the same conditions (fractions 5–8), although most of the caveolin accumulated in the fifth fraction. These results suggest that the presence of N-glycans of Edg-1 may play a pivotal role in their orientation to the caveolae-enriched fractions in the cells and possibly affect the rate of ligand-induced receptor internalization, although the precise mechanism of internalization through the involvement of microdomain is not determined.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. During these experiments, it was also noticed that Edg-1 might be cleaved at the amino-terminal portion, as described in Results. Moreover, the existence of the broad 40–55 kDa band observed with anti-HA antibody strongly suggested possible modification such as lipidation or phosphorylation after cleavage at the amino-terminal portion. So far, the roles or meaning of these post-translational modifications (cleavage and other processing) for the receptor functions are not known.
It is generally suggested that glycan portion in a glycoprotein may influence its transportation to the cell surface after its synthesis in ER as well as its stability on the cell surface, since nonglycosylated mutants of glycoproteins are often observed to be accumulated in ER or Golgi apparatus (24)
. Some GPCRs are known to be glycosylated in the NH2-terminal portion, although the roles of glycosylation in these receptors have not been clearly shown. One report indicated that a nonglycosylated GPCR failed to activate downstream of the heterotrimeric G-protein (25)
, but contrary observations have also been reported (26)
. We examined the roles of glycosylation of Edg-1 in its expression on the cell surface and its stability there, using stably transfected CHO cells. Although a part of the nonglycosylated N30D-Edg-1 was detected in Golgi apparatus (Fig. 2B
), the majority was expressed on the plasma membrane. No apparent difference was observed between WT- and N30D-Edg-1 expressed in the cell surface because most of the N30D-Edg-1, like the wild-type one, was digested by proteinase K under nonpermeabilized conditions (Fig. 6)
. Moreover, we think that the oligosaccharide of Edg-1 does not influence the folding of this protein. If the folding of protein were incomplete due to deficiency in oligosaccharide, Edg-1 would assumedly accumulate in endoplasmic reticulum, then be sent to a degrading system (27)
. However, in the cells stably expressing nonglycosylated N30D-Edg-1, this mutant receptor was expressed in the cell surface like wild-type Edg-1 and localization to endoplasmic reticulum was not observed at all (Fig. 2B
). The degradation product was not identified by Western blotting (Fig. 2A
). Based on these results, it is suggested that the presence or absence of oligosaccharide in Edg-1 does not influence folding and stability of this protein.
Edg-1 is known to be internalized into the cell on ligand engagement (22)
. We next examined whether N-glycosylation could affect the rate of internalization of the receptor. As shown in Figs. 4
and 6
, we found a significant difference in the rates of internalization of WT-Edg-1 and N30D-Edg-1 on Sph-1-P stimulation. We examined whether the MAP kinase inhibitors PD98059 and U0126 were involved Edg-1 internalization on Sph-1-P stimulation. However, these inhibitors were unaffected by Edg-1 internalization (data not shown), indicating that the internalization step itself is independent from ligand-induced MAP kinase activation, as clearly observed in WT- and N30D-Edg-1 even at the lower concentration of Sph-1-P.
Many GPCRs undergo ligand-induced receptor internalization via caveolae (through coated pit formation) involving arrestin and other intracellular molecules (28)
. An immediate desensitization, a rapid internalization has been reported to occur in the membrane microdomain or raft. Therefore, we examined whether N-glycosylation of the receptor might be related to its plasma membrane orientation, especially localization in a microdomain structure. Microdomains or raft structures rich in glycosphingolipids or cholesterol are generally considered the center of signal transduction, accumulating various signaling molecules such as receptors and src family members (29)
. In experiments where the caveolae-enriched microdomain was fractionated from WT- or N30D-Edg-1-expressing cells, we showed that WT-Edg-1 accumulated in the caveolin-enriched fractions, whereas the nonglycosylated receptor N30D-Edg-1 was dispersed broadly (Fig. 8)
. These results suggest that N-glycosylation affects ligand-induced receptor internalization partly through influencing its orientation in caveolae-enriched microdomain. Although the precise molecular mechanism of the internalization of the Edg-1 receptor localized in the microdomain remains to be examined, the present study strongly suggests that the N-linked glycans of the Edg-1 receptor may play a regulatory role in receptor dynamics in the ligand-stimulated mammalian cells. It might be interesting to investigate how N-glycan in the receptor is associated with each other or with glycosphingolipids that are enriched in the microdomain (30)
through a carbohydrate-carbohydrate interaction, as previously proposed by Hakomori (31)
, or through putative N-glycan binding proteins (32)
.
Using transiently transfected COS 7 cells, others recently reported that lateral movement of Edg-1 into microdomain was observed in a Sph-1-P dependent manner (33)
. Conversely, in our present study using Edg-1 stably-expressing CHO cells, WT-Edg-1 was observed to accumulate in the caveolae-enriched microdomain regardless of ligand concentration. The reason for this discrepancy is not known, but probably depends on the types of transfection, transiently over-expressed or stably expressed, used in the domain fractionation experiments.
|
ACKNOWLEDGMENTS |
|---|
Received for publication October 5, 2001.
Revision received January 8, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Kohno and Y. Igarashi Attenuation of cell motility observed with high doses of sphingosine 1-phosphate or phosphorylated FTY720 involves RGS2 through its interactions with the receptor S1P. Genes Cells, July 1, 2008; 13(7): 747 - 757. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Gohda, J. Kunisawa, F. Miura, Y. Kagiyama, Y. Kurashima, M. Higuchi, I. Ishikawa, I. Ogahara, and H. Kiyono Sphingosine 1-Phosphate Regulates the Egress of IgA Plasmablasts from Peyer's Patches for Intestinal IgA Responses J. Immunol., April 15, 2008; 180(8): 5335 - 5343. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Yokoe, M. Takahashi, M. Asahi, S. H. Lee, W. Li, D. Osumi, E. Miyoshi, and N. Taniguchi The Asn418-Linked N-Glycan of ErbB3 Plays a Crucial Role in Preventing Spontaneous Heterodimerization and Tumor Promotion Cancer Res., March 1, 2007; 67(5): 1935 - 1942. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Deng, J. A. Clemas, G. Chrebet, P. Fischer, J. J. Hale, Z. Li, S. G. Mills, J. Bergstrom, S. Mandala, R. Mosley, et al. Identification of Leu276 of the S1P1 Receptor and Phe263 of the S1P3 Receptor in Interaction with Receptor Specific Agonists by Molecular Modeling, Site-Directed Mutagenesis, and Affinity Studies Mol. Pharmacol., March 1, 2007; 71(3): 724 - 735. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Tani, Y. Igarashi, and M. Ito Involvement of Neutral Ceramidase in Ceramide Metabolism at the Plasma Membrane and in Extracellular Milieu J. Biol. Chem., November 4, 2005; 280(44): 36592 - 36600. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. G. Lo, Y. Xu, R. L. Proia, and J. G. Cyster Cyclical modulation of sphingosine-1-phosphate receptor 1 surface expression during lymphocyte recirculation and relationship to lymphoid organ transit J. Exp. Med., January 18, 2005; 201(2): 291 - 301. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. S. Jolly, M. Bektas, A. Olivera, C. Gonzalez-Espinosa, R. L. Proia, J. Rivera, S. Milstien, and S. Spiegel Transactivation of Sphingosine-1-Phosphate Receptors by Fc{varepsilon}RI Triggering Is Required for Normal Mast Cell Degranulation and Chemotaxis J. Exp. Med., April 5, 2004; 199(7): 959 - 970. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Jeanneteau, J. Diaz, P. Sokoloff, and N. Griffon Interactions of GIPC with Dopamine D2, D3 but not D4 Receptors Define a Novel Mode of Regulation of G Protein-coupled Receptors Mol. Biol. Cell, February 1, 2004; 15(2): 696 - 705. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Kohno and Y. Igarashi Truncation of the N-Terminal Ectodomain Has Implications in the N-Glycosylation and Transport to the Cell Surface of Edg-1/S1P1 Receptor J. Biochem., November 1, 2003; 134(5): 667 - 673. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Gobeil Jr., S. G. Bernier, A. Vazquez-Tello, S. Brault, M. H. Beauchamp, C. Quiniou, A. M. Marrache, D. Checchin, F. Sennlaub, X. Hou, et al. Modulation of Pro-inflammatory Gene Expression by Nuclear Lysophosphatidic Acid Receptor Type-1 J. Biol. Chem., October 3, 2003; 278(40): 38875 - 38883. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Graeler, Y. Kong, J. S. Karliner, and E. J. Goetzl Protein Kinase C {epsilon} Dependence of the Recovery from Down-regulation of S1P1 G Protein-coupled Receptors of T Lymphocytes J. Biol. Chem., July 18, 2003; 278(30): 27737 - 27741. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||
