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


     

Genes to Cells (2007) 12, 521-534. doi:10.1111/j.1365-2443.2007.01068.x
© 2007 Blackwell Publishing or its licensors
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Komekado, H.
Right arrow Articles by Kikuchi, A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Komekado, H.
Right arrow Articles by Kikuchi, A.

Glycosylation and palmitoylation of Wnt-3a are coupled to produce an active form of Wnt-3a

 Hideyuki Komekado1,2, Hideki Yamamoto1, Tsutomu Chiba2 and Akira Kikuchi1,*

1 Department of Biochemistry, Graduate School of Biomedical Sciences, Hiroshima University, 1-2-3 Kasumi, Minami-ku, Hiroshima 734-8551, Japan
2 Department of Gastroenterology and Hepatology, Graduate School of Medicine, Kyoto University, 54 Kawara-cho, Shogoin, Sakyo-ku, Kyoto 606-8507, Japan


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Wnt-3a is a representative ligand that activates the ß-catenin-dependent pathway in Wnt signaling and is modified with glycans and palmitate. In this study, we analyzed the relationship between glycosylation and lipidation of Wnt-3a. Secretion of a Wnt-3a mutant that lacks glycosylation (Wnt-3a NQ) was impaired. Wnt-3a C77A, which lacks palmitoylation at Cys77, was secreted with similar efficiency to wild-type Wnt-3a (Wnt-3a WT), but did not induce the internalization of low-density lipoprotein receptor-related protein 6 (LRP6). Furthermore, removal of palmitate from Wnt-3a suppressed the ability to bind to its receptors Frizzled8 and LRP6. Wnt-3a C77A was glycosylated to an extent similar to Wnt-3a WT, while Wnt-3a NQ was not modified with palmitate. Expression of porcupine, which is a putative acyltransferase, enhanced palmitoylation of Wnt-3a WT greatly, but that of Wnt-3a NQ slightly. While Wnt-3a WT was present in both the endoplasmic reticulum (ER) and Golgi, Wnt-3a NQ was located to the ER only. Furthermore, Wnt-3a was not palmitoylated but was glycosylated in the cells treated with Brefeldin A, which inhibits transport of vesicles from the ER to the Golgi. These results indicate that glycosylation of Wnt-3a precedes palmitoylation and that both modifications are necessary for secretion of an active Wnt-3a.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 References
 
Signaling by the Wnt proteins, a large family of cysteine-rich secreted molecules, has a key role in development and diseases (Wodarz & Nusse 1998; Veeman et al. 2003). Wnt proteins regulate many stages of development, including patterning of the embryo, the initiation of axon guidance, and synaptic formation. Defective Wnt signaling plays major roles in diseases such as cancer (Polakis 2000) and osteoporosis (Patel & Karsenty 2002). At least 19 Wnt members have been shown to be present in mammals to date, and each Wnt gene encodes a protein that harbors one or more sites for N-linked glycosylation, and has up to 23 or 24 conserved cysteines.

Wnt-3a has been shown to be expressed in pluripotent ectoderm cells of the primitive streak during gastrulation (Takada et al. 1994). At early somite stages, the Wnt-3a expression domain correlates with a domain of cells in the anterior primitive streak fated to give rise to paraxial mesoderm. Wnt-3a homozygous null mutant embryos lack all but the anterior-most seven to nine somites (Takada et al. 1994). Similar results have been reported for compound mutants in the transcriptional factors lymphoid enhancer factor 1 (Lef1) and T cell factor 1 (Tcf1) (Galceran et al. 1999), indicating that Lef1 and Tcf1 mediate Wnt-3a's effects on paraxial mesoderm development.

These results are consistent with the finding in cell biological studies that Wnt-3a activates the ß-catenin pathway. In the ß-catenin pathway, when Wnt-3a acts on its cell-surface receptors consisting of Frizzled (Fz) and lipoprotein receptor-related protein 5/6 (LRP5/6), ß-catenin escapes from degradation in the Axin complex (Ikeda et al. 1998; Kikuchi 1999; He et al. 2004). The accumulated ß-catenin is translocated to the nucleus, where it binds to Tcf/Lef and thereby stimulates the expression of various genes (Hurlstone & Clevers 2002).

Wnt proteins are also able to regulate the ß-catenin-independent pathways (Veeman et al. 2003). In the pathways, Wnt proteins activate various signaling molecules, including Rho, Rac, calcium/calmodulin-dependent protein kinase II, PKC and phospholipase C, resulting in the coordination of cell motility and polarity. Wnt-3a activates Rac and Rho, which are involved in convergent extension movement during gastrulation of Xenopus embryos and regulation of the neurite extension of PC12 cells (Habas et al. 2003; Kishida et al. 2004). However, it is not fully understood which of 10 Frizzled receptors specifically recognize Wnt-3a, resulting in the activation of distinct downstream pathways.

The cellular responses activated by Wnt-3a have been extensively studied as described above. However, Wnt-3a itself has not been characterized in detail because of the difficulty of purifying Wnt-3a. Although experiments using tunicamycin (an inhibitor of asparagine-linked glycosylation) have shown that Wnt-3a is modified with N-linked glycans (Smolich et al. 1993), the physiological roles of glycosylation of Wnt-3a are not fully understood. In 2003, Wnt-3a was purified for the first time and it was demonstrated that Wnt-3a is modified with palmitate (Willert et al. 2003). However, whether the two different kinds of post-translational modifications (glycosylation and palmitoylation) are linked is not clear. Here we show that glycosylation of Wnt-3a enhances palmitoylation, but that palmitoylation is not required for glycosylation. In addition, we demonstrate that glycosylation and palmitoylation of Wnt-3a are necessary for the secretion and actions of Wnt-3a.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 References
 
Glycosylation of Wnt-3a at Asn87 and Asn298

Since Wnt-3a has been shown to be modified with N-linked glycans (Smolich et al. 1993), first we tried to identify N-glycosylation sites of Wnt-3a. When purified Wnt-3a was treated with Peptide: N-glycosidase F (PNGase F) which cleaves off asparagine-linked carbohydrates, two rapidly migrating bands were observed on an SDS-PAGE gel in a dose-dependent manner (Fig. 1A), suggesting that at least two asparagine residues are glycosylated. There are two potential N-glycosylation sites (Asn87 and Asn298) in Wnt-3a. We generated Wnt-3a mutants in which either Asn87 or Asn298 was mutated to Gln (Wnt-3a N87Q or Wnt-3a N298Q) or both Asn residues were mutated to Gln (Wnt-3a NQ). When Wnt-3a NQ was expressed in HEK-293T cells, a single band was detected and its molecular mass was the same as that of wild-type Wnt-3a (Wnt-3a WT) expressed in the cells treated with tunicamycin (Fig. 1B). Wnt-3a N298Q exhibited a single slowly migrated band as compared to Wnt-3a NQ, while Wnt-3a N87Q showed two bands (Fig. 1B). These results indicate that Wnt-3a is glycosylated at both Asn87 and Asn298 and suggest that Asn87 is more readily glycosylated than Asn298. To know which type of glycans were attached to Wnt-3a, purified Wnt-3a or Wnt-3a in the cell lysates were immunoprecipitated with anti-Wnt-3a antibody, and then the immunoprecipitates were treated with Endoglycosidase Hf (Endo Hf) (Fig. 1C). While IgG heavy chains (having N-glycans) were resistant to Endo Hf treatment, the sugar chains attached to Wnt-3a were sensitive to Endo Hf both in cells and after secretion (Fig. 1C), indicating that mature Wnt-3a protein has high mannose or hybrid type N-glycans.


Figure 1
View larger version (20K):
[in this window]
[in a new window]

  		Figure 1  Glycosylation of Wnt-3a. (A) Purified Wnt-3a was incubated with the indicated amounts of Peptide: N-glycosidase F (PNGase F) for 16 h at 37 °C, and analyzed by SDS-PAGE and silver staining. (B) HEK-293T cells were transiently transfected with plasmids expressing wild-type Wnt-3a (Wnt-3a WT) or its asparagine mutants. Cell lysates were probed with anti-Wnt-3a antibody, and the mobility of the stained bands on SDS-PAGE was compared with that in lysates from Wnt-3a WT-expressing cells treated with 10 µg/mL tunicamycin for 24 h (left lane). (C) Lysates of HEK-293T cells expressing Wnt-3a WT (left panel) or 50 ng of purified Wnt-3a diluted in PBS containing 0.5% CHAPS (right panel) were precipitated with 0.5 µg of anti-Wnt-3a antibody and protein G Sepharose. One-third of the precipitates were left untreated, one-third were treated with 1000 U of Endo Hf, and the remaining one-third were treated with 1 mU of PNGase F under denaturing conditions at 37 °C for 12 h. Then each sample was probed with anti-Wnt-3a antibody. An arrow indicates the IgG heavy chains of anti-Wnt-3a antibody. IB, immunoblotting; Ab, antibody.

 
Requirement of glycosylation for secretion and action of Wnt-3a

We next analyzed the effects of glycosylation on the secretion of Wnt-3a. Consistent with the previous observations (Burrus & McMahon 1995), some of Wnt-3a WT bound to components to the extracellular matrix (ECM) and the remainder was recovered in the conditioned medium (CM) (Fig. 2A). It is notable that only the highest molecular weight band (a mature form) was secreted. However, Wnt-3a NQ was hardly detected in the ECM or CM even though it was present at a high level in the cell lysates (Fig. 2A). Furthermore, at 12 h after treatment of the cells with cycloheximide (CHX), which prevents protein synthesis, most of Wnt-3a WT disappeared from the cell lysates and it was instead observed in the CM, while Wnt-3a NQ was still retained in the cell lysates (Fig. 2B). Therefore, it is conceivable that glycosylation plays an important role in the secretion of Wnt-3a. It was recently reported that Wntless binds to Wnt proteins and stimulates their secretion (Bänziger et al. 2006). Wnt-3a NQ formed a complex with Wntless (hWls–MycHis) with similar efficiency to Wnt-3a WT (Fig. 2C), indicating that glycosylation of Wnt-3a is not required for the binding to Wntless.


Figure 2
View larger version (30K):
[in this window]
[in a new window]

  		Figure 2  Impairment of secretion of unglycosylated form of Wnt-3a. (A) Lysates, ECM and CM from L cells stably expressing Wnt-3a WT or Wnt-3a NQ were probed with anti-Wnt-3a antibody. (B) L cells stably expressing Wnt-3a WT or Wnt-3a NQ were treated with 50 µg/mL cycloheximide (CHX) for the indicated periods of time, and then CM and cell lysates were probed with anti-Wnt-3a antibody. (C) Wnt-3a WT or its mutants were expressed with or without human Wntless (hWls)–MycHis in HEK-293T cells, and the cell lysates were immunoprecipitated with anti-Myc-tag antibody. The lysates and precipitates were probed with anti-Myc-tag or anti-Wnt-3a antibody. (D) 1 µg of vector expressing Wnt-3a WT or mutants were transfected into HEK-293T cells (35-mm diameter dishes) with 0.5 µg of TOP-fos-Luc or FOP-fos-Luc, 0.5 µg of pME18S/lacZ and 0.1 µg of pEF-BOS/hTcf-4E. At 36 h after transfection, cells were lyzed and luciferase activity (normalized to ß-galactosidase activity) was measured. The results shown are means ± SD of three independent experiments. (E) Purified Wnt-3a was treated with or without PNGase F, then the former was incubated with Con A Sepharose to recover only the deglycosylated form of Wnt-3a. Then 5 nM deglycosylated or control Wnt-3a proteins were incubated with 0.1 µM IgG or mFz8 CRD-IgG. The precipitates and input samples (left lanes) were probed with anti-Wnt-3a antibody. (F) Upper panel, purified Wnt-3a was deglycosylated as described in Fig. 2E. One-tenth amount of each sample was probed with anti-Wnt-3a antibody. Lower panels, L cells were incubated with deglycosylated (100 ng/mL, right lane) or control (50 or 100 ng/mL, middle lanes) Wnt-3a proteins for 1 h. Lysates from L cells not treated with Wnt-3a were used as a negative control (left lane). The cell lysates were probed with anti-ß-catenin or anti-GSK-3ß (loading control) antibody. The results shown are representative of three independent experiments. IP, immunoprecipitation.

 
The Wnt-3a mutants were expressed and their ability to activate Tcf transcriptional activity was assayed. The ability of Wnt-3a NQ and Wnt-3a N87Q was severely suppressed as compared with that of Wnt-3a WT, while that of Wnt-3a N298Q was slightly reduced (Fig. 2D). To test whether glycosylation is necessary for the action of Wnt-3a after secretion, we treated purified Wnt-3a with PNGase F and examined the ability of Wnt-3a to bind to cysteine-rich domain of Frizzled8 tagged with immunoglobulin-{gamma} Fc epitope (Fz8 CRD-IgG) or to induce the accumulation of ß-catenin. Although the deglycosylated form of Wnt-3a showed a weak interaction with Fz8-CRD (Fig. 2E), it hardly maintained the ability to induce the accumulation of ß-catenin (Fig. 2F). Therefore, glycosylation would be partly involved in maintaining the activity of Wnt-3a after secretion, if at all.

Palmitoylation of Wnt-3a

Wnt-3a has been shown to be modified with palmitate at Cys77 (Willert et al. 2003). Wnt-3a C77A, in which Cys77 is mutated to Ala, also formed a complex with Wntless (see Fig. 2C). Wnt-3a C77A was recovered in both the ECM and CM, but a relatively greater proportion of it was present in the ECM as compared to Wnt-3a WT (Fig. 3A), indicating that palmitoylation of Wnt-3a at Cys77 is not essential for Wnt-3a secretion.


Figure 3
View larger version (23K):
[in this window]
[in a new window]

  		Figure 3  Palmitoylation of Wnt-3a. (A) Lysates, ECM and CM from L cells stably expressing Wnt-3a WT and Wnt-3a C77A were probed with anti-Wnt-3a antibody. (B) Upper panel, CM from L cells stably expressing Wnt-3a WT or Wnt-3a C77A was subjected to the Triton X-114 phase separation assay. Lower panels, HEK-293T cells transiently expressing Wnt-3a WT or Wnt-3a C77A were treated with 200 µM 2-Bromopalmitate or 0.08% DMSO (control) for 12 h. The cell lysates were subjected to the Triton X-114 phase separation assay. T, total turbid solution before centrifugation; Aq, aqueous phase; De, detergent-enriched phase. (C) Wnt-3a WT and Wnt-3a C77A secreted into CM from L cells were labeled with [3H]palmitate. AP, affinity precipitation. (D) Lysates of HEK-293T cells expressing Wnt-3a WT or C77A were subjected to acyl–biotinyl exchange (ABE) analysis. HEK-293T cells transfected with empty vector (pPGK-neo) were used as control. Biotinylated proteins were precipitated with Neutravidin–agarose and the precipitates were probed with anti-Wnt-3a antibody or anti-caveolin antibody. Endogenous caveolin was used as a positive control of a palmitoylated protein.

 
Although secreted Wnt-3a C77A in the CM was recovered in the aqueous phase by the phase separation assay (Fig. 3B, upper panel), a part of Wnt-3a C77A in the cell lysates was partitioned into the detergent phase to a similar extent to Wnt-3a WT (Fig. 3B, lower panel). These results suggest that cysteine residues other than Cys77 are also modified with lipids. To test this possibility, we used 2-Bromopalmitate, which blocks the transfer of palmitate from palmitoyl-CoA to cystenyl thiols and is a general inhibitor of O-acyltransferase activity (Webb et al. 2000). By the treatment of HEK-293T cells with 2-Bromopalmitate, both Wnt-3a WT and Wnt-3a C77A of the cell extracts disappeared in the detergent phase and all of them were observed in the aqueous phase (Fig. 3B, lower panel).

Furthermore, when Wnt-3a WT and Wnt-3a C77A were prepared from the CM of the cells cultured in the presence of [3H]palmitate, both of these Wnt-3a proteins were labeled with palmitate (Fig. 3C). It was difficult to find Wnt-3a labeled with [3H]palmitate in cell lysates, probably because fully matured Wnt-3a was easily secreted. Then we analyzed the extent of palmitoylation of Wnt-3a in cell lysates with the acyl–biotinyl exchange (ABE) method, in which palmitate attached to cysteine residues through thioester linkage was replaced by thiol-specific biotinylation reagent by concomitant hydroxylamine treatment. Both Wnt-3a WT and Wnt-3a C77A were biotinylated only when treated with hydroxylamine, but the proportion of biotinylation was small when compared to that of endogenous caveolin protein, which is known to be palmitoylated (Fig. 3D). Therefore, Wnt-3a recovered in the detergent phase of the cell extracts is indeed modified with palmitate, strongly suggesting that Wnt-3a can be modified with palmitate at cysteine residues other than Cys77. Since Wnt-3a C77A lost its hydrophobic property after secretion, as far as examined by the Triton X-114 phase separation assay (see Fig. 3B), it is likely that Cys77 is a specific site of palmitoylation for maintaining the hydrophobicity of Wnt-3a after secretion.

Requirement of palmitoylation for action of Wnt-3a

It has been shown that palmitoylation at Cys77 of Wnt-3a is important for its binding to receptors in a binding assay in which Wnt-3a CM is mixed with CM containing the extracellular domain of LRP6 tagged with immunoglobulin-{gamma} Fc epitope (LRP6N-IgG) or Fz8 CRD-IgG (Cong et al. 2004). However, since CM was used in the experiment, the possibility that factors other than Wnt-3a and its receptors affect their interaction cannot be ruled out. To clarify the roles of palmitoylation definitively, we used purified Wnt-3a, LRP6N-IgG and Fz8 CRD-IgG proteins. By the treatment of purified Wnt-3a with GST-acyl-protein thioesterase 1 (GST-APT1), which removes palmitate from thioacyl substrates, Wnt-3a was indeed detected in the aqueous phase in the Triton X-114 phase separation assay (data not shown). Wnt-3a WT directly bound to Fz8 CRD-IgG and LRP6N-IgG, whereas GST-APT1-treated Wnt-3a did not (Fig. 4A,B). sFRP2 is known to bind to Wnt and to inhibit the Wnt signaling (Kawano & Kypta 2003). Palmitoylation was also required for the binding of Wnt-3a to sFRP2 (Fig. 4C).


Figure 4
View larger version (30K):
[in this window]
[in a new window]

  		Figure 4  Requirement of palmitoylation for the binding of Wnt-3a to receptors. (A–C) Purified Wnt-3a protein treated with GST or GST-rAPT1 was incubated with 300 nM mFz8 CRD-IgG (A), 10 nM hLRP6N-IgG (B) or 70 nM sFRP2-FLAG (C), and the Wnt-3a binding proteins were precipitated by centrifugation. The precipitates were probed with anti-Wnt-3a antibody. (D) HEK-293 cells expressing LRP6-GFP and FLAG-Mesd were subjected to the receptor internalization assay using the Wnt-3a WT or Wnt-3a C77A CM, and the fixed cells were directly processed for the microscopy. Upper panels, confocal images; lower panel, quantification of internalized LRP6-GFP. Scale bar, 10 µm. (E) L or NIH3T3 cells were incubated with depalmitoylated or control Wnt-3a (100 ng/mL) for 1 h. An equal amount of Wnt-3a without incubation with GST-rAPT1 was used as a positive control (right lane). Whole cell lysates of L cells and cytosol fractions of NIH3T3 cells were probed with anti-ß-catenin or anti-Dvl antibody, respectively. An arrow and arrowhead indicate non-phoshorylated Dvl and phosphorylated Dvl, respectively.

 
We have recently found that Wnt-3a induces the internalization of LRP6 (Yamamoto et al. 2006). When HEK-293 cells expressing green fluorescent protein (GFP)-tagged LRP6 (LRP6-GFP) were stimulated with Wnt-3a WT, the receptor was internalized and observed in the intracellular vesicles (Fig. 4D). However, Wnt-3a C77A lost this activity (Fig. 4D). Consistent with these findings, the GST-APT1-treated Wnt-3a completely lost the ability to induce the accumulation of ß-catenin or to stimulate the phosphorylation of Dvl (Fig. 4E). Taken together with the previously reported observations (Cong et al. 2004), palmitoylation at Cys77 is essential for the ability of Wnt-3a to bind to its receptors and to activate the subsequent signaling.

Coupling of glycosylation and palmitoylation

We next investigated the relationship between glycosylation and palmitoylation of Wnt-3a. Wnt-3a WT and Wnt-3a C77A were expressed in HEK-293T cells, and the cells were treated with tunicamycin. Both proteins exhibited a time-dependent increase in electrophoretic mobility in a similar manner (Fig. 5A). At 12 h after treatment, Wnt-3a C77A exhibited the same mobility as Wnt-3a NQ (Fig. 5A). Furthermore, as shown in Fig. 3B, Wnt-3a WT and Wnt-3a C77A from the cells treated with 2-Bromopalmitate showed an electrophoretic mobility similar to that of Wnt-3a from 2-Bromopalmitate-untreated cells. These results indicate that palmitoylation of Wnt-3a is not essential for glycosylation of Wnt-3a.


Figure 5
View larger version (35K):
[in this window]
[in a new window]

  		Figure 5  Prerequisite of glycosylation for palmitoylation of Wnt-3a. (A) HEK-293T cells expressing Wnt-3a WT or Wnt-3a C77A were treated with 10 µg/mL tunicamycin for the indicated periods of time. Then the cell lysates were probed with anti-Wnt-3a antibody, and the mobility of Wnt-3a in the lysates was compared with that of Wnt-3a NQ (right lane). (B) Lysates of HEK-293T cells expressing the indicated proteins were subjected to the Triton X-114 phase separation assay. (C) Wnt-3a WT or Wnt-3a NQ secreted from L cells into DMEM/Ham's F12 supplemented with 2% FBS was concentrated about eightfold by centrifugation at 3000 g at 4 °C using a Macrosep 30K centrifugal device (Pall Corporation). CM containing Wnt-3a WT was diluted with concentrated CM from L cells so that Wnt-3a and other proteins such as bovine serum albumin (BSA) were equally present in the CM of Wnt-3a WT and Wnt-3a NQ. Concentrated CM was subjected to the Triton X-114 phase separation assay. BSA was cross-reacted with anti-Wnt-3a antibody. (D) L cells were incubated with concentrated control CM, Wnt-3a WT CM or Wnt-3a NQ CM prepared in Fig. 5C for 1 h. The cell lysates were probed with anti-ß-catenin or anti-GSK-3ß antibody. (E) Purified Wnt-3a protein (80 ng of protein) with or without PNGase F treatment was subjected to the Triton X-114 phase separation assay. (F) The lysates of HEK-293T cells expressing Wnt-3a WT or its mutants with or without HA-Mporc C were immunoprecipitated with anti-HA antibody. The precipitates were probed with anti-HA or anti-Wnt-3a antibody. (G) Upper panel, lysates of HEK-293T cells expressing Wnt-3a with or without Mporc C were probed with anti-HA or anti-Wnt-3a antibody. Lower panel, the same lysates were subjected to the Triton X-114 phase separation assay, and probed with the indicated antibodies.

 
To examine whether glycosylation affects lipidation, we analyzed the hydrophobicity of the Wnt-3a mutants, which lack the glycosylation sites, in the cell extracts. The ratio of Wnt-3a N87Q or Wnt-3a N298Q distributed in the detergent phase was reduced as compared with that of Wnt-3a WT, and none of Wnt-3a NQ was partitioned into the detergent phase (Fig. 5B). Since a small amount of Wnt-3a NQ was secreted in the CM, we concentrated CM containing Wnt-3a NQ and subjected it to the phase separation assay. All of the secreted Wnt-3a NQ was recovered in the aqueous phase (Fig. 5C), and secreted Wnt-3a NQ failed to induce the accumulation of ß-catenin like Wnt-3a C77A or GST-APT1-treated Wnt-3a (Fig. 5D). Considering that hydrophobicity of Wnt-3a in CM was lost by lack of palmitoylation at Cys77 (see Fig. 3B), this result indicates that Wnt-3a NQ is not palmitoylated, at least at Cys77. On the other hand, when purified Wnt-3a was treated with PNGase F, all of the deglycosylated form of Wnt-3a was recovered in the detergent phase as well as the glycosylated form of Wnt-3a (Fig. 5E), indicating that glycosylation itself is not responsible for the hydrophobicity of Wnt-3a. Taken together, these results suggest that glycosylation is required for lipidation during the processing of Wnt-3a in the cells.

The protein (Porc) encoded by the porcupine gene in Drosophila has been suggested to be a membrane-bound acyltransferase (Hofmann 2000). It has been found that four different types of mouse Porc (Mporc A, B, C and D) are generated from a single gene by alternative splicing (Tanaka et al. 2000). NIH3T3 cells express multiple types of Mporc mRNA (Tanaka et al. 2000), suggesting that alternative splicing can occur in a single cell. When Mporc C was co-expressed with Wnt-3a, it formed a complex with Wnt-3a WT, Wnt-3a C77A and Wnt-3a NQ with similar efficiency (Fig. 5F), suggesting that neither glycosylation nor palmitoylation at Cys77 is necessary for the association with Mporc C. These results are consistent with the previous observations that fly Porc interacts with the N-terminal region of wingless (Tanaka et al. 2002). By over-expressing Mporc C, the fully processed forms of Wnt-3a WT and Wnt-3a C77A were increased (Fig. 5G, upper panel). Furthermore, expression of Mporc C increased the amounts of Wnt-3a WT and Wnt-3a C77A recovered in the detergent phase (Fig. 5G, lower panel). However, Mporc C slightly induced the distribution of Wnt-3a NQ to the detergent phase (Fig. 5G, lower panel). These results indicate that glycosylated Wnt-3a is palmitoylated more efficiently than the unglycosylated form.

Glycosylation precedes palmitoylation

To examine the order of glycosylation and palmitoylation of Wnt-3a, we fractionated post-nuclear supernatant (PNS) of HEK-293T cells on a discontinuous sucrose gradient. Major peaks of GSK-3ß (a cytosol marker), adaptin-{gamma} (a Golgi marker) and calnexin (an endoplasmic reticulum (ER) marker) were distributed in Fractions 3–5, 6–8 and 11–12, respectively (Fig. 6A). Wnt-3a WT and Wnt-3a C77A were observed mainly in the ER fraction and were also present in the Golgi fraction (Fig. 6A). However, almost all of Wnt-3a NQ was detected in the ER fraction, and it was not observed in the other fractions (Fig. 6A). These results suggest that glycosylation of Wnt-3a is necessary for its transport from the ER to the Golgi and that palmitoylation at Cys77 is not essential for this process.


Figure 6
View larger version (34K):
[in this window]
[in a new window]

  		Figure 6  Glycosylation precedes palmitoylation of Wnt-3a. (A) PNS expressing Wnt-3a WT, Wnt-3a C77A or Wnt-3a NQ was fractionated using a discontinuous sucrose density gradient. An aliquot of each fraction was probed with the indicated antibodies. (B) COS7 cells expressing GFP-KDEL (retained in the ER) and Wnt-3a WT were treated with 10 µg/mL Brefeldin A or 0.05% DMSO (control) for 1 h, and then the cells were stained with anti-Wnt-3a and anti-adaptin-{gamma} antibodies. Scale bar, 10 µm. (C and D) Brefeldin A treatment alters the post-translational status of Wnt-3a. (C) HEK-293T cells expressing Wnt-3a WT or its mutants were treated with 10 µg/mL Brefeldin A or 0.05% DMSO (control) for 12 h, and then the cell lysates were probed with anti-Wnt-3a antibody. (D) The same lysates used in Fig. 6C were subjected to the Triton X-114 phase separation assay, followed by Western blotting with anti-Wnt-3a antibody.

 
We treated the cells with Brefeldin A, which induces the rapid disassembly of the Golgi complex and the redistribution of resident Golgi markers into the ER, thereby blocking steps of many transport, including between the Golgi complex and the ER (Klausner et al. 1992). Immunocytochemical analyses showed that Wnt-3a was distributed in a broad pattern, reflecting the entire secretory pathway, and a large part was co-localized with the ER, which was indicated by expression of GFP-KDEL (Fig. 6B). The Golgi apparatus, which was indicated by the presence of adaptin-{gamma}, showed a different distribution from the ER (Fig. 6B). Brefeldin A induced the breakdown of the Golgi apparatus but did not influence the morphology of the ER, and a similar pattern of distribution of Wnt-3a between Brefeldin A-treated and untreated cells was observed (Fig. 6B). Biochemical analyses revealed that only a small fraction of Wnt-3a WT and Wnt-3a C77A showed an increase in electrophoretic mobility in Brefeldin A-treated cells (Fig. 6C,D). On the other hand, the partition of Wnt-3a WT and Wnt-3a C77A in the cell lysates into the detergent phase was completely lost upon the treatment with Brefeldin A (Fig. 6D). Thus, these results suggest that glycosylation and palmitoylation occur in different compartments.


   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 References
 
Roles of glycosylation of Wnt-3a

Asparagine-linked glycosylation plays pivotal roles in protein folding, oligomerization, quality control and transport (Helenius & Aebi 2001). The unglycosylated Wnt-3a remained in the ER and consequently its secretion was impaired, although the mechanism by which glycosylated Wnt-3a is transported to the cell surface from the ER is not known. It was recently reported that Wntless/Evi is required for the secretion of Wnt proteins in Drosophila and mammalian cells (Bänziger et al. 2006). Wntless is localized in the Golgi apparatus and physically associates with Wnt-3a (Bänziger et al. 2006). Wnt-3a NQ formed a complex with Wntless with similar efficiency to Wnt-3a WT, suggesting that the unglycosylated form of Wnt-3a does not lose the ability to bind to Wntless but cannot be transported to the region where Wntless is present. Although Wnt-3a NQ was hardly detected in the ECM or CM, a small amount of Wnt-3a may be secreted to act in an autocrine manner. Wnt-3a NQ was indeed able to activate Tcf-4 slightly when expressed at a high level in an autocrine manner. These results are consistent with the report that the glycosylation sites of Wnt-1 are dispensable for its ability to transform C57MG cells in an autocrine assay (Mason et al. 1992).

It is noteworthy that glycosylation is not usually essential for maintaining the overall folded structure once a glycoprotein has folded (Imperiali & O’Connor 1999). However, this is not the case for Wnt-3a. When purified Wnt-3a was deglycosylated by PNGase F, the deglycosylated form of Wnt-3a showed reduced ability to bind to Fz8 and to induce the accumulation of ß-catenin. Since the presence of glycans influences the properties of the polypeptide moiety, for example, by increasing stability, removal of glycans may affect the biological activity of a folded glycoprotein. Thus, glycosylation of Wnt-3a is important for secretion and action of Wnt-3a.

Roles of palmitoylation of Wnt-3a

Lipid attachment is a common modification of cytoplasmic proteins and is important for the membrane targeting of signaling molecules and protein-protein interaction in the cells (Dunphy & Linder 1998). However, it is rare in proteins that operate outside cells. Wnt proteins, at least Drosophila Wnt8 and mouse Wnt-3a, are such exceptional cases (Willert et al. 2003).

We also found that Wnt-3a C77A shows hydrophobicity in the cell lysates, like Wnt-3a WT, and that Mporc C increases the proportion of Wnt-3a C77A partitioned into the detergent phase. Furthermore, the hydrophobicity of Wnt-3a C77A in the lysates disappeared as a result of the treatment of the cells with 2-Bromopalmitate, and Wnt-3a C77A was labeled with [3H]palmitate in CM and thiol-specific Biotin in cell lysates. These results suggest that amino acid residues of Wnt-3a other than Cys77 are also modified with palmitate. Alternatively, failure to be palmitoylated at Cys77 may cause Wnt-3a to be palmitoylated at different amino acids. In addition, our result suggest that palmitoylation of Wnt-3a at Cys77 but not at other amino acids is involved in the hydrophobicity of Wnt-3a after secretion. Since it has been reported that fly porc is required for wingless-producing cells for generating the fully functional protein signal (van den Heuvel et al. 1993) and for secreting wingless protein (Tanaka et al. 2000), palmitoylation at other amino acids may instead have other distinct roles. During submitting this paper, Wnt-3a has been shown to be modified with palmitoleic acid at Ser209 (Takada et al. 2006). This modification occurs through oxyester linkage, while palmitoylation at Cys77 does via thioester linkage. Since acylation at Ser209 is necessary for the secretion of Wnt-3a (Takada et al. 2006), it is likely that multiple lipidation at different amino acids of Wnt-3a has distinct roles in exerting the activity of Wnt-3a.

Palmitoylation was shown to be definitively required for the abilities of Wnt-3a to induce the phosphorylation of Dvl and accumulation of ß-catenin by experiments in which purified Wnt-3a was treated with rAPT, consistent with the previous observations (Willert et al. 2003). In addition, depalmitoylation of purified Wnt-3a caused loss of the ability to bind to Fz8, LRP6 and sFRP2. Considering the previous report showing that Wnt-3a C77A in CM failed to bind to Fz8 and LRP6 (Cong et al. 2004), palmitoylation at Cys77 of Wnt-3a is essential for the binding to its receptors. Palmitate attached to Cys77 of Wnt-3a may maintain the tertiary structure of Wnt-3a needed to bind to the receptors, or this lipid may directly bind to the receptors. However, Wnt-3a C77A retained some activity to activate Tcf-4 when expressed at high levels. Therefore, it is also possible that palmitoylation at Cys77 enhances the affinity between Wnt-3a and the receptors. Alternatively, palmitoylation at Cys77 may be important for the anchoring of Wnt-3a to the cell surface membrane, because it has been reported that lipidation of wingless is required for its targeting to the lipid raft microdomains in Drosophila S2 cells (Zhai et al. 2004).

It has also been reported that wingless and Wnt-3a induce the internalization of Frizzled and LRP6/Arrow (Seto & Bellen 2006; Yamamoto et al. 2006). Considering the roles of palmitoylation at Cys77 in the binding of Wnt-3a and receptors, it is reasonable that Wnt-3a C77A lost the ability to induce the internalization of LRP6. Although whether the internalization of Wnt receptors is necessary for the activation of the ß-catenin pathway is not clear, the role of palmitoylation in the interaction of Wnt proteins with their receptors on the cell surface is important for transducing an extracellular signal into intracellular responses.

Coupling of glycosylation and palmitoylation of Wnt-3a

It is well known that newly synthesized polypeptides are first glycosylated in the ER. It has also been reported that Porc was located to the ER and that over-expression of Porc enhances the extent of glycosylation of some Wnt proteins, including Wnt-3a (Tanaka et al. 2000). Thus, which modification, namely glycosylation or palmitoylation, occurs first has not been elucidated. The following results indicate that glycosylation precedes palmitoylation. First, in both cell extracts and CM, Wnt-3a NQ was not partitioned into the detergent phase, but rather all of it was present in the aqueous phase, while Wnt-3a C77A or Wnt-3a from the cells treated with 2-Bromopalmitate was glycosylated to an extent similar to Wnt-3a WT. Furthermore, Wnt-3a N87Q and Wnt-3a N298Q were distributed into the detergent phase to a lesser extent than Wnt-3a WT, indicating that the hydrophobicity of Wnt-3a is dependent on the extent of glycosylation. On the other hand, when purified Wnt-3a was deglycosylated by PNGase F in vitro, the unglycosylated form of Wnt-3a retained hydrophobicity. These results suggest that preceding glycosylation is important for palmitoylation of Wnt-3a in the maturation process in the cells but that palmitoylation is not essential for glycosylation of Wnt-3a. Second, expression of Mporc C enhanced the hydrophobicity of Wnt-3a WT more efficiently than that of Wnt-3a NQ. Therefore, it is likely that the increase of the hydrophobicity of Wnt-3a by Porc is dependent on glycosylation in mammalian cells. Lastly, subcellular fractionation analyses revealed that Wnt-3a WT is distributed in both the ER and Golgi fractions but Wnt-3a NQ is located almost exclusively to the ER. Brefeldin A treatment suppressed the hydrophobicity of Wnt-3a, but its effect on the glycosylation of Wnt-3a was slight. Brefeldin A specifically blocks protein transport from the ER to the Golgi apparatus by interfering with the action of a guanine–nucleotide exchange factor for ADP ribosylation factor (Klausner et al. 1992). Thus, it is intriguing to speculate that Wnt-3a is modified with glycans in the ER, followed by palmitoylation in the Golgi apparatus.

Although over-expressed Porc is reported to be located to the ER (Tanaka et al. 2000), other acyltransferases, such as palmitoyl tranferase(s) that acylate endothelial nitric oxide synthase, have been shown to be present in the Golgi apparatus (Fernandez-Hernando et al. 2006). However, it is known that Brefeldin A also causes protein(s) normally present in the Golgi apparatus to redistribute into the ER (Lippincott-Schwartz et al. 1989). Therefore, we cannot exclude the possibility that palmitoylation of Wnt-3a, which normally occurs in the ER, is impaired by redistribution of the Golgi proteins. To identify acyltransferase(s) of Wnt-3a other than Porc would make it possible to understand the whole picture of the physiological significance of palmitoylation of Wnt proteins.

Wnts in human, mouse, Drosophila and Hydra contain an average of 23–25 cysteines, some of which are highly conserved among different species. Cys77 in mouse Wnt-3a and Cys51 in Drosophila Wnt-8 are modified with palmitate (Willert et al. 2003). In addition, we have recently found that mouse Wnt-5a is palmitoylated at Cys104 (Kurayoshi et al. 2006). These imply that palmitoylated cysteines are the most amino-terminally conserved cysteines of the Wnt proteins (Fig. 7). Other Wnt proteins also contain this conserved cysteine, suggesting that Wnt family proteins may be modified with palmitate at the same site and that the palmitoylation at this cysteine residue is important for exerting the actions of Wnts. Each Wnt protein has a number of potential N-glycosylation sites. We also identified four N-glycosylation sites (Asn114, Asn120, Asn311 and Asn325) in mouse Wnt-5a (Kurayoshi et al. 2007) and two of them corresponded to Asn87 and Asn298 in Wnt-3a (Fig. 7). Furthermore, these two asparagine residues are conserved in Drosophila Wg and Hydra Wnt. Therefore, glycosylation at these asparagine residues may have similar functions. Thus, the sequential glycosylation and palmitoylation of Wnt protein could be important for the secretion of not only Wnt-3a but also other Wnt proteins in an active form.


Figure 7
View larger version (20K):
[in this window]
[in a new window]

  		Figure 7  Palmitoylation and N-glycosylation sites of Wnt-3a are conserved among Wnt proteins of several species. The amino acid sequence around Cys77 and Asn87 (upper panel) or Asn298 (lower panel) in Wnt-3a is aligned. Conserved cysteine or asparagine residues are indicated by an arrowhead or asterisks, respectively.

 

   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
References
 
Materials and chemicals

pPGK-neo and pPGK-neo/mouse Wnt-3a were provided by Dr S. Takada (Okazaki Institute for Integrative Biosciences, Okazaki, Japan). TOP-fos-Luc and FOP-fos-Luc and pCMV/sFRP2-FLAG were provided by Dr H. Clevers (Hubrecht Institute, Utrecht, the Netherlands) and Dr T. Akiyama (University of Tokyo, Tokyo, Japan), respectively. pMKIT/HA-Mporc C was supplied by Dr T. Kadowaki (Nagoya University, Nagoya, Japan). pRK5/IgG and pRK5/mFz8 CRD (1-173 amino acids)-IgG, and pcDNA3.1/hLRP6N-IgG were provided by Drs J. C. Hsieh (State University of New York, Stony Brook, NY) and M. Semënov (Harvard Medical School, Boston, MA), respectively. pQE-60/His6-rAPT1 was from Dr A. G. Gilman (University of Texas South-Western Medical Center, Dallas, TX). pcDNA3.1/GFP-KDEL was provided by Dr T. Inoue (University of Tokyo, Tokyo, Japan). Recombinant GST-tagged rat acyl-protein thioesterase 1 (GST-rAPT1) protein was purified from Escherichia coli transformed with pGEX-KG/rAPT1 according to the suppliers’ instructions (Duncan & Gilman 1998). Anti-Wnt-3a polyclonal antibody (Kishida et al. 2004) and anti-Dvl polyclonal antibody (Kishida et al. 2001) were prepared as described, and anti-LRP6 polyclonal antibody was generated in rabbits by immunization with a synthetic peptide corresponding to residues 1360–1376 of human LRP6. Anti-ß-catenin monoclonal antibody, anti-GSK-3ß monoclonal antibody, anti-caveolin polyclonal antibody, anti-adaptin-{gamma} monoclonal antibody and anti-calnexin monoclonal antibody were purchased from BD Biosciences (San Jose, CA), and anti-FLAG M2 monoclonal antibody and anti-FLAG M2 affinity gel were purchased from Sigma-Aldrich (St. Louis, MO). Anti-HA monoclonal antibody was prepared from 12CA5 cells and anti-HA monoclonal antibody (16B12) was purchased from Covance (Princeton, NJ). Anti-Myc-tag monoclonal antibody was generated from 9E10 cells. Peptide: N-glycosidase F (PNGase F), Endoglycosidase Hf (Endo Hf), 2-Bromopalmitate and Brefeldin A were purchased from Takara Bio (Otsu, Japan), New England Biolabs (Beverly, MA), Wako (Osaka, Japan) and Nacalai Tesque (Kyoto, Japan), respectively. Biotin–HPDP and Neutravidin–agarose were purchased from Pierce Biotechnology (Rockford, IL). Other materials were obtained from commercial sources.

Plasmid construction

Standard recombinant DNA techniques were used to construct pPGK-neo/Wnt3a C77A, pPGK-neo/Wnt3a N87Q, pPGK-neo/Wnt3a N298Q and pPGK-neo/Wnt3a N87QN298Q (NQ). A cDNA fragment encoding rAPT1 was inserted into pGEX-KG to generate pGEX-KG/rAPT1. Human Wntless (hWls) cDNA was cloned from a human cDNA library, and subcloned into pcDNA3.1-MycHis vector.

Purification of Wnt-3a protein

Wnt-3a was purified as described (Kishida et al. 2004) with modifications. One litre of Wnt-3a-conditioned medium (CM) (3770 mg of protein) was adjusted to 1% Triton X-100 and applied to Blue Sepharose HP in an XK 26/20 column (column volume (CV) 50 (mL) (GE Healthcare Bio-Sciences, Piscataway, NJ) equilibrated with binding buffer (20 mM Tris–HCl, pH 7.5 and 1% 3-[(3-cholamidopropyl) dimethylammonio] propanesulfonic acid (CHAPS)) containing 150 mM KCl. After the column was washed with 250 mL of binding buffer, elution was performed in a step-wise manner with 250 mL of elution buffer containing 1.5 M KCl at a flow rate of 5 mL/min. Fractions of 25 mL were collected. An aliquot of each fraction was probed with anti-Wnt-3a antibody. Wnt-3a appeared in Fractions 2–5, and these fractions (100 mL, 11.7 mg of protein) were pooled and concentrated to 6 mL using an OMEGA ultrafiltration membrane (Pall Corporation, East Hills, NY). The concentrate (6 mL, 10.1 mg of protein) was applied to a HiLoad 26/60 Superdex 200 column (CV 320 mL) (GE Healthcare Bio-Sciences) equilibrated with phosphate buffered saline (PBS) and 1% CHAPS. Elution was performed with the same buffer at a flow rate of 2.5 mL/min, and fractions of 2.5 mL were collected after the void volume (30% of CV) had been eluted. Wnt-3a appeared in Fractions 31–43, and these fractions (32.5 mL, 0.78 mg of protein) were applied to a HiTrap Chelating HP (CV 1 mL) column (GE Healthcare Bio-Sciences) loaded with 1 mL of 0.1 M CuSO4 and then equilibrated with binding buffer (20 mM sodium phosphate, pH 7.4, 0.5 M NaCl, 10 mM imidazole, and 1% CHAPS). After the column was washed with 10 mL of binding buffer, elution was performed with a 10 mL linear gradient of imidazole (10–100 mM) elution buffer (20 mM sodium phosphate, pH 7.4, 0.5 M NaCl, and 1% CHAPS) at a flow rate of 1 mL/min. Fractions of 1 mL were collected, and Wnt-3a appeared mainly in Fractions 3–6. The purity of Wnt-3a was almost 100% and its maximum concentration was about 30 µg/mL in Fractions 4 and 5 as assessed by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and Coomassie staining. All these chromatography procedures were performed using ÄKTA explorer 10S (GE Healthcare Bio-Sciences). In assays using purified Wnt-3a, 20 mM sodium phosphate, pH 7.4 containing 0.5 M NaCl, 30 mM imidazole and 1% CHAPS was used as a control.

Deglycosylation and depalmitoylation of purified Wnt-3a in vitro

To prepare deglycosylated Wnt-3a under native conditions, 600 ng of purified Wnt-3a protein was incubated for 30 min at 37 °C with or without 5 mU of PNGase F (reconstituted with 20 mM sodium phosphate, pH 7.2 and 50 mM EDTA), in the presence of 0.1 M Tris–HCl, pH 8.6 and 1% CHAPS, in a total volume of 35 µL. To remove the free sugar chains and Wnt-3a which failed to be deglycosylated and retained glycans, Wnt-3a samples treated with PNGase F were diluted with 20 mM Tris–HCl, pH 7.5 containing 0.5 M NaCl and 1% CHAPS and incubated with Con A Sepharose (GE Healthcare Bio-Sciences) for 30 min on ice, then the supernatant after centrifugation was recovered. When indicated, deglycosylation was performed under denaturing conditions according to the manufacturer's instructions.

To prepare depalmitoylated Wnt-3a, 200 ng of purified Wnt-3a was incubated for 8–12 h at 30 °C with 1 µg of GST-rAPT1 or 0.5 µg of GST (as a control) and 2 µg of bovine serum albumin (BSA) in the presence of 1% CHAPS, in a total volume of 12 µL. Triton X-114 phase separation assay was performed to confirm depalmitoylation.

Receptor binding assay

CM from HEK-293T cells expressing mFz8 CRD-IgG, hLRP6N-IgG or sFRP2-FLAG was incubated with protein A Sepharose or anti-FLAG M2 affinity gel, and each recombinant protein having the Wnt-3a-binding domain was prepared. Purified Wnt-3a protein was treated with PNGase F or GST-rAPT1 for deglycosylation or depalmitoylation, respectively, and diluted to the indicated concentration with PBS containing 0.5% CHAPS. Wnt-3a samples (200 µL) were incubated with indicated concentration of mFz8 CRD-IgG, hLRP6N-IgG or sFRP2-FLAG for 1 h on ice, and then the Wnt-3a-binding domain was precipitated by centrifugation. The precipitates were probed with anti-Wnt-3a antibody.

Wnt-3a secretion assay

L cells (1 x 106 cells) stably expressing Wnt-3a WT or mutants were seeded in culture dishes (60-mm diameter) and grown in 3.5 mL of culture medium for 72 h. After CM was harvested, cells were washed with cold PBS, collected using a scraper and suspended in cold PBS. After the suspension was centrifuged at 800 g for 10 min and the supernatant was discarded, cells were lyzed with 250 µL of NP40 buffer (20 mM Tris–HCl, pH 7.5, 137 mM NaCl, 10% glycerol, and 1% NP40) supplemented with protease inhibitors and the samples were used as a lysate faction. The culture dishes from which the cells had been scraped were washed with cold PBS three times, then washed with NP40 buffer three times. After 375 µL of Laemmli buffer was added on to the dishes, the dishes were heated for 5 min at 100 °C using a heat block, and the liquid on the dish was collected and used as an extracellular matrix (ECM) fraction. Then each fraction was probed with anti-Wnt-3a antibody.

Triton X-114 phase separation assay

The Triton X-114 phase separation assay was performed as described (Willert et al. 2003) with modifications. 100 µL of Wnt-3a CM was mixed with the same volume of 10 mM Tris–HCl, pH 7.5, 150 mM NaCl and 4.5% Triton X-114. When cell lysates were used for this assay, cells were lyzed with 10 mM Tris–HCl, pH 7.5, 150 mM NaCl and 1% Triton X-114 with protease inhibitors, and the supernatant was mixed with an equal amount of 10 mM Tris–HCl, pH 7.5, 150 mM NaCl and 3.5% Triton X-114. Purified Wnt-3a with or without various treatments was diluted with 10 mM Tris–HCl, pH 7.5, and 150 mM NaCl (final CHAPS concentration was less than 0.05%), and then mixed with an equal volume of 10 mM Tris–HCl, pH 7.5, 150 mM NaCl and 4.5% Triton X-114. The mixtures were incubated on ice for 5 min and then at 37 °C for 5 min. The turbid solution was centrifuged at 2000 g at room temperature for 5 min. After centrifugation, the sample was separated into the aqueous (top) and detergent (bottom) phases, and an aliquot of each phase was analyzed by Western blotting.

[3H]palmitate labeling and autoradiography

L cells (1 x 106 cells) stably expressing Wnt-3a WT or C77A (L cells stably expressing pPGK-neo were used as a control) were seeded into 60-mm diameter dishes and grown for 36 h. Then the cells were washed with PBS and labeled with 0.2 mCi/mL [3H]palmitate (GE Healthcare Bio-Sciences) in 2 mL of DMEM/Ham's F12 supplemented with 0.1% fatty acid-free BSA (Sigma-Aldrich) for 12 h. CM was adjusted to 1% Triton X-100 and incubated for 3 h at 4 °C with 30 µL of slurry of Blue Sepharose FF beads (GE Healthcare Bio-Sciences) equilibrated with 20 mM Tris–HCl, pH 7.5 containing 150 mM KCl and 1% CHAPS to concentrate Wnt-3a. A quarter of each precipitate was probed with anti-Wnt-3a antibody. The remainder was subjected to SDS-PAGE, and the gels were incubated for 30 min in an autoradiography enhancer (Amplify; GE Healthcare Bio-Sciences) after fixation and then exposed to Biomax MS film (Kodak, Rochester, NY) for 1 month at –80 °C.

Acyl-biotinyl exchange (ABE) analysis

ABE analysis was performed as previously described (Roth et al. 2006). Briefly, HEK-293T cells expressing Wnt-3a (two 60-mm diameter dishes) were lyzed with 100 µL of lysis buffer (LB: 50 mM Tris–HCl, pH 7.5, 150 mM NaCl, 5 mM EDTA) containing 1% SDS and 10 mM N-ethylmaleimide (NEM), then diluted with 900 µL of LB containing 2% Triton X-100 and 10 mM NEM. The lysates were incubated for 1 h at 4 °C and precipitated by the chloroform–methanol method. The precipitates were solubilized with 200 µL of SB buffer (50 mM Tris–HCl, pH 7.5, 4% SDS, 5 mM EDTA) containing 10 mM NEM for 10 min at 37 °C, then diluted with 800 µL of LB containing 0.2% Triton X-100 and 1 mM NEM. The mixtures were incubated overnight at 4 °C, then precipitated with chloroform–methanol method. The pellets were solubilized with 200 µL of SB at 37 °C and diluted with 800 µL of LB containing 0.2% Triton X-100. Then additional two chloroform–methanol precipitations were performed to remove NEM from the samples completely. After the third precipitation, the pelleted proteins were dissolved with 200 µL of SB. One of the duplicated samples was mixed with 800 µL of 1 M Hydroxylamine, pH 7.4, 150 mM NaCl, 0.2% Triton X-100 and 1 mM Biotin–HPDP. The other was mixed with 1 M Tris–HCl, pH 7.4, 150 mM NaCl, 0.2% Triton X-100 and 1 mM Biotin–HPDP (used for a negative control). The mixtures were incubated for 1 h at room temperature with gentle shaking, then precipitated with chloroform–methanol method. The precipitated proteins were solubilized with 50 µL of TB buffer (50 mM Tris–HCl, pH 7.5, 2% SDS, 5 mM EDTA) and diluted with 450 µL of LB containing 0.2% Triton X-100. The samples were incubated at 4 °C for 30 min and centrifugated at 20 000 g for 5 min. 50 µL of supernatant was mixed with 25 µL of 3x Laemmli buffer supplemented with 6% 2-mercaptoethanol and incubated at 95 °C for 5 min (input samples). The remnants were mixed with 30 µL slurry of Neutravidin–agarose equilibrated with LB containing 0.1% SDS and 0.2% Triton X-100. The precipitates were analyzed by Western blotting.

Sucrose density gradient fractionation

Subcellular fractionation by sucrose density gradient centrifugation was performed as previously described (Waugh et al. 2003). Briefly, HEK-293T cells transiently expressing Wnt-3a WT, Wnt-3a C77A or Wnt-3a NQ were washed with cold PBS and suspended in 10 mM Tris–HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA and 0.25 M (8.6% (w/v)) sucrose containing protease inhibitors. Post-nuclear supernatant (PNS) (0.64 mL) was layered onto a discontinuous sucrose gradient (successive layers of 0.32 mL of 40% (w/v), 0.32 mL of 30%, 0.64 mL of 25%, 0.64 mL of 20%, 0.64 mL of 15% and 0.64 mL of 10% sucrose). After centrifugation at 175 000 g at 4 °C for 16 h in an RPS56T rotor (Hitachi Koki, Tokyo, Japan), fractions of 0.32 mL each were collected from the top of the gradient.

Others

The ability of Wnt-3a to induce the phospshorylation of Dvl in NIH3T3 cells and Tcf-4 transcriptional activity were measured as described (Kishida et al. 2001; Yamamoto et al. 2003). Immunocytochemical analyses of the cultured cells and internalization assay of LRP6 were performed as described (Yamamoto et al. 2003, 2006).


   Acknowledgements
 
We thank Drs S. Takada, H. Clevers, T. Akiyama, T. Kadowaki, J. C. Hsieh, M. Semënov, A. G. Gilman and T. Inoue for donating plasmids. We are also grateful to Dr M. Fukata for helpful discussion and advice about protein palmitoylation. This work was supported by Grants-in-Aid for Scientific Research and for Scientific Research on Priority Areas from the Ministry of Education, Culture, Sports, Science and Technology of Japan (2004, 2005, 2006).


   Footnotes
 
Communicated by: Kozo Kaibuchi

* Correspondence: E-mail: akikuchi{at}hiroshima-u.ac.jp


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Bänziger, C., Soldini, D., Schütt, C., Zipperlen, P., Hausmann, G. & Basler, K. (2006) Wntless, a conserved membrane protein dedicated to the secretion of Wnt proteins from signaling cells. Cell 125, 509–522.[CrossRef][Medline]

Burrus, L.W. & McMahon, A.P. (1995) Biochemical analysis of murine Wnt proteins reveals both shared and distinct properties. Exp. Cell Res. 220, 363–373.[CrossRef][Medline]

Cong, F., Schweizer, L. & Varmus, H. (2004) Wnt signals across the plasma membrane to activate the ß-catenin pathway by forming oligomers containing its receptors, Frizzled and LRP. Development 131, 5103–5115.[Abstract/Free Full Text]

Duncan, J.A. & Gilman, A.G. (1998) A cytoplasmic acyl-protein thioesterase that removes palmitate from G protein alpha subunits and p21RAS. J. Biol. Chem. 273, 15830–15837.[Abstract/Free Full Text]

Dunphy, J.T. & Linder, M.E. (1998) Signalling functions of protein palmitoylation. Biochim. Biophys. Acta 1436, 245–261.[Medline]

Fernandez-Hernando, C., Fukata, M., Bernatchez, P.N., Fukata, Y., Lin, M.I., Bredt, D.S. & Sessa, W.C. (2006) Identification of Golgi-localized acyl transferases that palmitoylate and regulate endothelial nitric oxide synthase. J. Cell Biol. 174, 369–377.[Abstract/Free Full Text]

Galceran, J., Farinas, I., Depew, M.J., Clevers, H. & Grosschedl, R. (1999) Wnt3a–/–-like phenotype and limb deficiency in Lef1–/–Tcf1–/– mice. Genes Dev. 13, 709–717.[Abstract/Free Full Text]

Habas, R., Dawid, I.B. & He, X. (2003) Coactivation of Rac and Rho by Wnt/Frizzled signaling is required for vertebrate gastrulation. Genes Dev. 17, 295–309.[Abstract/Free Full Text]

He, X., Semënov, M., Tamai, K. & Zeng, X. (2004) LDL receptor-related proteins 5 and 6 in Wnt/ß-catenin signaling: arrows point the way. Development 131, 1663–1677.[Abstract/Free Full Text]

Helenius, A. & Aebi, M. (2001) Intracellular functions of N-linked glycans. Science 291, 2364–2369.[Abstract/Free Full Text]

van den Heuvel, M., Harryman-Samos, C., Klingensmith, J., Perrimon, N. & Nusse, R. (1993) Mutations in the segment polarity genes wingless and porcupine impair secretion of the wingless protein. EMBO J. 12, 5293–5302.[Medline]

Hofmann, K. (2000) A superfamily of membrane-bound O-acyltransferases with implications for Wnt signaling. Trends Biochem. Sci. 25, 111–112.[CrossRef][Medline]

Hurlstone, A. & Clevers, H. (2002) T-cell factors: turn-ons and turn-offs. EMBO J. 21, 2303–2311.[CrossRef][Medline]

Ikeda, S., Kishida, S., Yamamoto, H., Murai, H., Koyama, S. & Kikuchi, A. (1998) Axin, a negative regulator of the Wnt signaling pathway, forms a complex with GSK-3ß and ß-catenin and promotes GSK-3ß-dependent phosphorylation of ß-catenin. EMBO J. 17, 1371–1384.[CrossRef][Medline]

Imperiali, B. & O’Connor, S.E. (1999) Effect of N-linked glycosylation on glycopeptide and glycoprotein structure. Curr. Opin. Chem. Biol. 3, 643–649.[CrossRef][Medline]

Kawano, Y. & Kypta, R. (2003) Secreted antagonists of the Wnt signalling pathway. J. Cell Sci. 116, 2627–2634.[Abstract/Free Full Text]

Kikuchi, A. (1999) Roles of Axin in the Wnt signalling pathway. Cell. Signal. 11, 777–788.[CrossRef][Medline]

Kishida, M., Hino, S.-I., Michiue, T., Yamamoto, H., Kishida, S., Fukui, A., Asashima, M. & Kikuchi, A. (2001) Synergistic activation of the Wnt signaling pathway by Dvl and casein kinase I{varepsilon}. J. Biol. Chem. 276, 33147–33155.[Abstract/Free Full Text]

Kishida, S., Yamamoto, H. & Kikuchi, A. (2004) Wnt-3a and Dvl induce neurite retraction by activating Rho-associated kinase. Mol. Cell. Biol. 24, 4487–4501.[Abstract/Free Full Text]

Klausner, R.D., Donaldson, J.G. & Lippincott-Schwartz, J. (1992) Brefeldin A: insights into the control of membrane traffic and organelle structure. J. Cell Biol. 116, 1071–1080.[Free Full Text]

Kurayoshi, M., Yamamoto, H., Izumi, S. & Kikuchi, A. (2007) Post-translational palmitoylation and glycosylation of Wnt-5a are necessary for its signaling. Biochem. J. in press.

Lippincott-Schwartz, J., Yuan, L.C., Bonifacino, J.S. & Klausner, R.D. (1989) Rapid redistribution of Golgi proteins into the ER in cells treated with brefeldin A: evidence for membrane cycling from Golgi to ER. Cell 56, 801–813.[CrossRef][Medline]

Mason, J.O., Kitajewski, J. & Varmus, H.E. (1992) Mutational analysis of mouse Wnt-1 identifies two temperature-sensitive alleles and attributes of Wnt-1 protein essential for transformation of a mammary cell line. Mol. Biol. Cell 3, 521–533.[Abstract]

Patel, M.S. & Karsenty, G. (2002) Regulation of bone formation and vision by LRP5. N. Engl. J. Med. 346, 1572–1574.[Free Full Text]

Polakis, P. (2000) Wnt signaling and cancer. Genes Dev. 14, 1837–1851.[Free Full Text]

Roth, A.F., Wan, J., Bailey, A.O., Sun, B., Kuchar, J.A., Green, W.N., Phinney, B.S., Yates, R. III & Davis, N.G. (2006) Global analysis of protein palmitoylation in yeast. Cell 125, 1003–1013.[CrossRef][Medline]

Seto, E.S. & Bellen, H.J. (2006) Internalization is required for proper Wingless signaling in Drosophila melanogaster. J. Cell Biol. 173, 95–106.[Abstract/Free Full Text]

Smolich, B.D., McMahon, J.A., McMahon, A.P. & Papkoff, J. (1993) Wnt family proteins are secreted and associated with the cell surface. Mol. Biol. Cell 4, 1267–1275.[Abstract]

Takada, R., Satomi, Y., Kurata, T., Ueno, N., Norioka, S., Kondoh, H., Takao, T. & Takada, S. (2006) Monounsaturated fatty acid modification of Wnt protein: its role in Wnt secretion. Dev. Cell 11, 791–801.[CrossRef][Medline]

Takada, S., Stark, K.L., Shea, M.J., Vassileva, G., McMahon, J.A. & McMahon, A.P. (1994) Wnt-3a regulates somite and tailbud formation in the mouse embryo. Genes Dev. 8, 174–189.[Abstract/Free Full Text]

Tanaka, K., Kitagawa, Y. & Kadowaki, T. (2002) Drosophila segment polarity gene product porcupine stimulates the posttranslational N-glycosylation of wingless in the endoplasmic reticulum. J. Biol. Chem. 277, 12816–12823.[Abstract/Free Full Text]

Tanaka, K., Okabayashi, K., Asashima, M., Perrimon, N. & Kadowaki, T. (2000) The evolutionarily conserved porcupine gene family is involved in the processing of the Wnt family. Eur. J. Biochem. 267, 4300–4311.[Medline]

Veeman, M.T., Axelrod, J.D. & Moon, R.T. (2003) A second canon. Functions and mechanisms of ß-catenin-independent Wnt signaling. Dev. Cell 5, 367–377.[CrossRef][Medline]

Waugh, M.G., Minogue, S., Anderson, J.S., Balinger, A., Blumenkrantz, D., Calnan, D.P., Cramer, R. & Hsuan, J.J. (2003) Localization of a highly active pool of type II phosphatidylinositol 4-kinase in a p97/valosin-containing-protein-rich fraction of the endoplasmic reticulum. Biochem. J. 373, 57–63.[CrossRef][Medline]

Webb, Y., Hermida-Matsumoto, L. & Resh, M.D. (2000) Inhibition of protein palmitoylation, raft localization, and T cell signaling by 2-bromopalmitate and polyunsaturated fatty acids. J. Biol. Chem. 275, 261–270.[Abstract/Free Full Text]

Willert, K., Brown, J.D., Danenberg, E., Duncan, A.W., Weissman, I.L., Reya, T., Yates, J.R.I. & Nusse, R. (2003) Wnt proteins are lipid-modified and can act as stem cell growth factors. Nature 423, 448–452.[CrossRef][Medline]

Wodarz, A. & Nusse, R. (1998) Mechanisms of Wnt signaling in development. Annu. Rev. Cell Dev. Biol. 14, 59–88.[CrossRef][Medline]

Yamamoto, H., Ihara, M., Matsuura, Y. & Kikuchi, A. (2003) Sumoylation is involved in ß-catenin-dependent activation of Tcf-4. EMBO J. 22, 2047–2059.[CrossRef][Medline]

Yamamoto, H., Komekado, H. & Kikuchi, A. (2006) Caveolin is necessary for Wnt-3a-dependent internalization of LRP6 and accumulation of ß-catenin. Dev. Cell 11, 213–223.[CrossRef][Medline]

Zhai, L., Chaturvedi, D. & Cumberledge, S. (2004) Drosophila wnt-1 undergoes a hydrophobic modification and is targeted to lipid rafts, a process that requires porcupine. J. Biol. Chem. 279, 33220–33227.[Abstract/Free Full Text]

Received: 14 December 2006
Accepted: 17 January 2007




This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
W. Ching, H. C. Hang, and R. Nusse
Lipid-independent Secretion of a Drosophila Wnt Protein
J. Biol. Chem., June 20, 2008; 283(25): 17092 - 17098.
[Abstract] [Full Text] [PDF]

Home page
 DevelopmentHome page
L. M. Galli, T. L. Barnes, S. S. Secrest, T. Kadowaki, and L. W. Burrus
Porcupine-mediated lipid-modification regulates the activity and distribution of Wnt proteins in the chick neural tube
Development, September 15, 2007; 134(18): 3339 - 3348.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Komekado, H.
Right arrow Articles by Kikuchi, A.
Right arrow Search for Related Content