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

J. Biol. Chem., Vol. 277, Issue 44, 41762-41769, November 1, 2002

This Article Abstract Full Text (PDF) An addition or correction has been published All Versions of this Article: 277/44/41762    most recent M207850200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wu, C.-h. Articles by Nusse, R. Search for Related Content PubMed PubMed Citation Articles by Wu, C.-h. Articles by Nusse, R. Social Bookmarking                      What's this?

Ligand Receptor Interactions in the Wnt Signaling Pathway in Drosophila^*

Chi-hwa Wu and Roel Nusse

From the Howard Hughes Medical Institute, Department of Developmental Biology, Stanford University Medical School, Stanford, California 94305-5323

Received for publication, August 2, 2002, and in revised form, August 29, 2002

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Secreted Wnt proteins have numerous signaling functions during development, mediated by Frizzled molecules that act as Wnt receptors on the cell surface. In the genome of Drosophila, seven Wnt genes (including wingless; wg), and five frizzled genes have been identified. Relatively little is known about signaling and binding specificities of different Wnt and Frizzled proteins. We have developed an assay to determine the strength of binding between membrane-tethered Wnts and ligand binding domains of the Frizzled receptors. We found a wide spectrum of binding affinities, reflecting known genetic interactions. Most Wnt proteins can bind to multiple Frizzleds and vice versa, suggesting redundancy in vivo. In an extension of these experiments, we tested whether two different subdomains of the Wg protein would by themselves bind to Frizzled and generate a biological response. Whereas these two separate domains are secreted from cells, suggesting that they form independently folded parts of the protein, they were only able to evoke a response when co-transfected, indicating that both are required for function. In addition to the Frizzleds, members of the LRP family (represented by the arrow gene in Drosophila) are also necessary for Wnt signal transduction and have been postulated to act as co-receptors. We have therefore examined whether a soluble form of the Arrow molecule can bind to Wingless and Frizzled, but no interactions were detected.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Among the many signaling events occurring during animal development, the interactions between members of the secreted Wnt protein family and their receptors, the Frizzleds, are prominent (1-5). In Drosophila, there are seven Wnt genes (including wingless; wg)¹ and five members of the frizzled (fz) family (6, 7). Given that there are multiple genes for ligands and receptors, the phenotypic consequences of Wnt signaling in vivo are probably determined not only by where the genes are expressed but also by the strength of binding between these molecules. However, very few studies have examined to what extent different Wnt proteins bind to their receptors (1, 8-10).

Structurally, each Frizzled protein is composed of an extracellular cysteine-rich domain (CRD), a seven-transmembrane region, and a cytoplasmic tail (11). The CRD domain, when overexpressed with a cell surface anchoring sequence, is necessary and sufficient for Wnt binding (1, 12). CRDs are also found as the Wnt binding modules in secreted Frizzled-related proteins (sFRP or FRP), a group of secreted Wnt inhibitors (8, 13-15). The crystal structure of the CRD has been obtained recently (16). The structures are predominately -helical with all cysteines forming disulfide bonds. The ligand binding surface of the CRD has been defined by combining the information from the crystal structure with mutagenesis studies (9, 16).

With respect to the genetics of frizzled genes in Drosophila, there are mutants in four of the five genes, only Dfz4 remains to be mutated. Special attention has been given to fz, the first frizzled gene found. This gene is essential in tissue polarity formation (11). fz does not have phenotypes similar to wg mutants (17, 18), but when combined with mutants in Dfz2, the resulting embryos are identical to wg mutants (19-21). Tissue culture experiments demonstrated that overexpressing fz or Dfz2 in S2 cells allows the cells to respond to Wg protein (1). DFz3 mutants do not have obvious developmental defects (22, 23) but absence of the Dfz3 gene can modify some wg phenotypes (22). Finally, smoothened (smo), a distantly related fz gene is required for transducing Hedgehog (Hh) signals (24, 25).

Additional complexity in Wnt-receptor interactions arose when Arrow (Arr), a low density lipoprotein-receptor-related protein (LRP), was found to be required for Wnt signaling (26, 27). The suggestion has been made that LRP can bind to Wnt directly (28), but this important aspect of Wnt-receptor interactions has not always been confirmed (29). The role of LRP in Wnt signaling became more intriguing after several groups showed that LRP is a receptor for Dickkopf (Dkk), a secreted Wnt inhibitor (29-31).

In this work, we analyze the binding between most of the Drosophila Wnts and all of the Frizzled proteins in a quantitative manner. We find significant differences, which reflect genetic interactions between Wnt and Frizzled genes in the fly. We also report on attempts to detect direct binding between Arrow and Wg, using various assays.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Generation of CRDs-AP Fusion Constructs, Cell Culture, Transfection, and CRDs-AP Fusion Protein Production-- We initially found that chimeric proteins made by fusing various Frizzled CRDs (from the first methionine until the 10th conserved cysteine) to alkaline phosphatase were not secreted into the medium after overexpression in 293T cells. However, the alkaline phosphatase fusion for the sFRP-3, which contains the conserved CRD followed by a tail of hydrophilic amino acids (8), is secreted well in 293T cells. Therefore, all CRD-AP fusion constructs were made by replacing the CRD region of the psFRP3-AP construct (8) with CRDs of fz, DFz2, DFz3, DFz4, or Smo.

To easily swap CRDs for these frizzled constructs, MluI sites were generated after the 10th conserved cysteines using the QuikChange site-directed mutagenesis kit (Stratagene). After MluI sites were created in the cDNAs of sFRP3-AP, fz, DFz2, DFz3, Dfz4, and Smo constructs, the CRD of sFRP3-AP was replaced by the CRDs of Drosophila frizzleds. Changes of DNA and amino acid sequences are as follows (changed sequences are bold): sFRP3-AP, TGCATCTACGCGTTGGCC, CIYALA; fzCRD-AP, TGCGTGGACGCGTTGGCC, CVDALA; Dfz2CRD-AP, TGCATGGACGCGTTGGCC, CMDALA; Dfz3CRD-AP, TGCATGCACGCGTTGGCC, CMHALA; Dfz4CRD-AP, TTCACAAACGCGTTGGCC, FTNALA; SmoCRD-AP, TGTTTAAACGCGTTGGCC, CLNALA.

All CRD-AP fusion constructs were cloned into the pRK5 vector (1) for expression in 293T cells. 293T cells were cultured with Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum (HyClone), penicillin, and streptomycin. Transfection was done using calcium phosphate precipitation. Four days after transfection, AP fusion proteins were collected from the conditioned medium and concentrated using Centriprep 50 columns (Amicon). To confirm the expression and concentration of the CRD-AP fusion proteins, concentrated conditioned medium was immunoprecipitated with anti-alkaline phosphatase antibody (Genzyme) and the immunocomplexes were resolved on an SDS-PAGE. The gel was then stained with Coomassie Blue.

Generation of the Arr-Ig Fusion Construct and Arr-Ig Fusion Protein-- The Arr-Ig fusion construct was made by replacing the Lrp6 portion in a Lrp6-Ig fusion construct for immunoprecipitation-Western experiments (28). The construct contains the extracellular domain of Arr, which starts from the first amino acid to the 1447th amino acid. A HindIII site at the 5' and a XbaI site at the junction between Lrp6 and Ig were used for swapping. After replacing the Lrp6 with the extracellular domain of Arr, the sequences of the new junction site between Arr and Xba site become RMAPATSLG (Arr sequences are bold).

Arr-Ig and Ig were produced in 293T cells that were transiently transfected using LipofectAMINE (Invitrogen). One day after transfection, cells were transferred to serum-free Dulbecco's modified Eagle's medium, and the secreted proteins were harvested after an additional 24 h. Control conditioned medium was obtained from untransfected 293T cells. These conditioned medium were then concentrated through Centriprep (Amicon) columns. Concentrated proteins were immunoprecipitated with protein G-Sepharose (Amersham Biosciences) and resolved on SDS-PAGE gels. Proteins were transferred to a nitrocellulose filter and the filter was probed with the horseradish peroxidase-conjugated horse anti-human antibody (Bio-Rad) overnight. The blot was detected using ECL Western blot detecting reagent (Amersham Biosciences).

Generation of Neurotactin (Nrt)-Wnts Fusion Constructs, Insect Cell Culture, and Stable Line Selection-- Nrt-Wnt fusion constructs were made by swapping the Wg portion in the Nrt-Wg construct (32) with other Wnts. The Nrt-Wg construct has HA sequences between the Nrt and Wg cDNA. The wnt cDNAs excluding the signal sequences were cloned by PCR. For DWnt3, only the Wnt homology region was cloned into the fusion construct (starting from amino acid 559 to 1010). The sequences around the regions linking HA and Wnts are (Wnt sequences are bold): Nrt-DWnt2, WEDEEASMEIRLVS; Nrt-DWnt3, WEDEEASMLHLTAR; Nrt-DWnt4, WEDEEASAGGQGLP; and Nrt-DWnt8, WEDEEASVLEPMSY.

All Nrt-wnt fusions were cloned into the pMK33HS vector (33) for expression in Schneider 2 (S2) insect cells. Transfections were done by calcium phosphate precipitation. Transfected S2 cells were cultured in Schneider 2 medium (Invitrogen) supplemented with 15% fetal bovine serum (Sigma), penicillin, streptomycin, and 125 µg/ml hygromycin (Sigma) for selection.

Antibody Staining-- S2 cells were heat shocked at 37 °C for 40 min and then cultured at 25 °C for 2 h. Cells were fixed in 2% methanol-free formaldehyde in phosphate-buffered saline for 20 min and then probed with a mouse monoclonal antibody against HA (Roche Molecular Biochemicals) overnight as a 1:25 dilution supplemented with normal donkey serum as the blocking reagent. Cells were washed 3 times in phosphate-buffered saline before being hybridized with donkey anti-mouse antibodies labeled with Cy3. Cells were again washed with phosphate-buffered saline and then mounted with Vectashield (Vector). A confocal microscope (Bio-Rad) was used for observing and taking images of the stained cells.

Binding Assay for AP Fusion Proteins and S2 Cells Expressing Nrt-Wnts-- The protocol for the binding assay was performed as previously published (34, 35). CRD-AP fusion proteins produced from conditioned medium were first checked for their specific activities. All CRD-AP fusions have similar specific activities compared with alkaline phosphatase at the same assay condition (34).

S2 cells overexpressing Nrt-Wnts were heat shocked at 37 °C for 45 min and then cultured at 25 °C for 2 h. Different concentrations of CRD-AP fusions were incubated with S2 cells for 90 min at room temperature. Cells were washed with Hanks' balanced salt solution 3 times before being lysed in 1% Triton with vortexing. The cell lysate was centrifuged at 1000 × g for 15 min to remove nuclei. The supernatant of the lysate was heated at 85 °C for 15 min to inactivate background phosphatases activities. The assay was then performed by measuring the A₄₀₅ after incubating the lysate with 1 M diethanolamine (pH 9.8), 0.5 mM MgCl₂, 10 mM L-homoarginine, 0.5 mg/ml bovine serum albumin (Sigma), and 12 mM p-nitrophenyl phosphate (Sigma).

Affinity Measurements-- The total AP activities added to each experiment and the bound AP activities on the S2 cells can be measured as A₄₀₅ changes over an hour. Saturable binding curves were plotted by fitting the bound AP activities and free AP activities in each experiment with the saturation binding curve, which is Y = A·X/(B + X). Scatchard analysis was done by plotting the ratios of the bound AP activity to the free AP activity against the concentrations of the bound AP in each experiment. All data were analyzed in Excel and KaleidaGraph.

Cell Surface Binding between Arr-Ig and Nrt-Wg/S2 Cells-- Arr-Ig-conditioned medium from 293T cells were concentrated with Centriprep (Amicon). Concentrated Arr-Ig was then added to Nrt-Wg/S2 cells after the cells were heat-shocked at 37 °C for 1 h and then cultured at 18 °C for 2 h. The Ig fusion protein was incubated with S2 cells for 2 h at room temperature. S2 cells were then washed with Hanks' balanced salt solution, lysed, and assayed for the AP activity according to the protocol described in the previous section.

Generation of WgA and WgB Constructs-- The WgA construct was generated by inserting two continuous stop codons at the AatII site of the Wg cDNA. DNA oligomers 5'-CGTTGATAAGCTTACGT-3' and 5'-AAGCTTATCAAGCACGT-3' were annealed and then ligated with the Wg cDNA construct cut with AatII. This construct generates the amino-terminal part of Wg, which contains the first amino acid to the 359th amino acid in the coding region.

To generate the WgB construct, which makes a secreted version of the carboxyl-terminal part of the Wg protein, the Wg cDNA construct was cut with NarI and NdeI enzymes to remove the amino-terminal part of Wg. DNA oligomers containing HA tag sequences flanking with NarI and NdeI sites (sense, 5'-CGCCATGCATTACCCATATGATGTTCCAGATTACGCTTCCGC-3'; antisense, 5'-TAGCGGAAGCGTAATCTGGAACATCATATGGGTAATGCATGG-3') were used in ligating the Wg cDNA cut with NarI and NdeI. Therefore, the amino acid sequence at the junction site becomes (Wg sequences are bold) VKGAMHYPYDVPDYASAMPD. To express WgA and WgB in Schneider 2 (S2) cells, WgA and WgB cDNAs were cloned into the pMK33HS plasmid that contains the hygromycin marker for selecting cells expressing the constructs.

Tissue Culture, Transfection, Conditioned Medium Collection-- S2 cells were cultured in Schineider's Drosophila medium (Invitrogen) supplemented with 12.5% fetal bovine serum (Sigma), penicillin, and streptomycin. pMK33HS-WgA, pMK33HS-WgB, pMK33-Dfz2, and pMKHS-Wg were introduced into S2 cells by calcium phosphate-mediated transfection followed by selection with 125 µg/ml hygromycin B (Sigma) until hygromycin-resistant cell lines were established.

Transgenes in the stable S2 cells were induced by a 45-min heat shock at 37 °C and then cultured for 2 h before lysate preparation. For WgA- or WgB-conditioned medium collection, after heat shock, S2 cells were rested at 25 °C for 30 min before being transferred into serum-free Schineider's Drosophila medium. S2 cells were then cultured in serum-free medium for 4 h. This serum-free conditioned medium containing WgA or WgB was collected by first spinning down S2 cells and then centrifuging the soluble medium at 10,000 × g for 20 min.

Cell Lysate Preparation and Immunoblotting-- Cells were first washed with phosphate-buffered saline and then lysed on ice with lysis buffer (50 mM Tris-HCl, 150 mM NaCl, 1% Nonidet P-40, pH 8) supplemented with protease inhibitors (0.5 µg/ml leupeptin, 1 µg/ml pepstatin A, and 100 µg/ml phenylmethylsulfonyl fluoride). Protein concentration of the lysates was determined using the Bio-Rad Protein assay dye reagent.

Sample buffer for SDS-PAGE was added to the cell lysate or the conditioned medium before samples were resolved by SDS-PAGE. Proteins were then transferred to a nitrocellulose membrane, and the blots were blocked in blocking buffer (3% nonfat dry milk, 1% bovine serum albumin in TBST (20 mM Tris-HCl, 150 mM NaCl, 0.2% Tween 20, pH 8)), and then incubated overnight at 4 °C in blocking buffer containing antibodies. The rabbit anti-Wg antibody (1:1000 dilution) and the mouse monoclonal antibody against HA (Roche Molecular Biochemicals) (1:1000 dilution) were used for immunoblotting. Proteins were detected using horseradish peroxidase-conjugated secondary antibodies (Bio-Rad) with the ECL Western blot detection reagents (Amersham Biosciences).

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Whereas some progress has been made to produce soluble Wnt molecules (9, 36), in particular Wg (37), it is problematic to purify significant amounts of Wnt proteins. This complication limits the possibility of doing conventional binding experiments in which saturating quantities of ligands are necessary. Our approach to measure binding between Drosophila Wnts and Frizzleds is therefore based on the "reverse binding" assay we developed, in which Wnts are presented on the surface of the cell in the form of type II transmembrane proteins, i.e. with the COOH terminus outside the cells (32). The cells are then incubated with the ligand-binding domain of Frizzled (the CRD) tagged with alkaline phosphatase (10).

Generation of S2 Cells Expressing Neurotactin-Wnt Fusion Proteins-- Nrt is a single transmembrane type II cell surface molecule (38, 39). As we described previously (10), we utilized a clone in which Wg (without its own signal sequence) was fused to the carboxyl-terminal end of the Nrt cDNA to make a membrane-tethered form of Wg (32). Expression from this construct is capable of inducing wg target genes and rescuing wg mutant phenotypes in vivo (32). Other Nrt-Wnt fusions were generated by exchanging the coding regions of the Wnts with the Wg region in the Nrt-Wg construct. The DWnt3 protein, however, has an extension at the amino terminus that is cleaved to generate the mature and secreted protein (40). To eliminate possible cleavage of the fusion protein, we used only the region where DWnt3 is homologous to other Wnts.

The Nrt-Wnt fusion constructs were tagged with HA sequences to confirm the correct expression of the fusion proteins in S2 cells. All Nrt-Wnt fusion proteins were detected at the predicted sizes (data not shown). We also stained the S2 cells with the anti-HA antibody, detecting cell surface staining for each of the fusion proteins indicating that they are correctly presented on the cell surface (Fig. 1). We successfully generated these fusions for five of the seven Drosophila Wnts, but we were unable to synthesize complete cDNAs for DWnt6 and DWnt10 (41), which therefore remain to be examined. Because these genes as presented in the Drosophila genome (7) lack good signal sequences, we suspect that the amino-terminal sequences are incorrect.

View larger version (23K): [in this window] [in a new window]   Fig. 1.   Nrt-Wnt fusion constructs are localized on the cell surface. S2 cells stably expressing different Nrt-Wnt constructs (as labeled on each panel) tagged with the HA sequences were stained with an anti-HA antibody. No detergent was applied during the antibody staining to keep the plasma membrane intact. Compared with untransfected S2 cells, HA antibody generates cell surface staining in all Nrt-Wnt-transfected cells.

Generation of Frizzled CRD-AP Fusion Proteins-- To measure binding between the cell surface-bound Wnt molecules and the Frizzled CRDs, we had to tag the CRD in such a way that its concentration could be established. We used the AP protein to do so (34, 35). All of the CRD-AP fusion proteins were secreted into the medium as shown by immunoprecipitation experiments (Fig. 2).

View larger version (26K): [in this window] [in a new window]   Fig. 2.   CRD-AP fusion proteins containing AP activities are secreted into the conditioned medium of 293T Cells. Conditioned medium from 293T cells transiently transfected with the CRD-AP fusion constructs was collected and concentrated for immunoprecipitation with the anti-AP antibody. Immunocomplexes from mock transfected (lane 1), AP (lane 2), FzCRD-AP (lane 3), Dfz2CRD-AP (lane 4), Dfz3CRD-AP (lane 5), SmoCRD-AP (lane 6), and Dfz4CRD-AP (lane 7) were resolved by SDS-PAGE and the gel was stained with Coomassie Blue.

By incubating the CRD-AP fusion proteins with a substrate for alkaline phosphatase, specific activities of these fusion proteins were obtained (Table I). Considering the variations in sizes of different CRDs, the specific activities of the various CRD-AP fusion proteins are not significantly different from nonfused AP on a molar basis.

Binding Affinities between Wnt and Frizzled Molecules in Drosophila-- We previously presented the binding affinities of FzCRD-AP and Dfz2CRD-AP to Nrt-Wg (10). By using the same technique, we measured binding affinities of other Wnts and Frizzleds in Drosophila (Fig. 3 and Table II). Wg binds to Fz, DFz2, and DFz3 with the highest affinity for Dfz2 (10). DWnt2 binds to the same set of three Frizzleds with approximately the same affinity, whereas DWnt3 does not bind any of the Frizzleds. DWnt8 binds to DFz4 only. Fig. 4 summarizes the affinities (1/K_d) between different Wnts and Frizzleds.

View larger version (12K): [in this window] [in a new window]   Fig. 3.   FzCRD-AP binds to Nrt-DWnt2/S2 cells. Nrt-DWnt2/S2 cells were incubated with different concentrations of FzCRD-AP-conditioned medium generated from 293T cells. The cells were lysed and the bound AP activities were measured colorimetrically at A405 after incubating with the AP substrate. A saturable binding curve is observed when bound AP activities were plotted against the total AP activities added into each experiment (panel A). The data were transformed into the Scatchard plot as shown in panel B. The Kd between FzCRD and DWnt2 is 6.918 × 108 M.

View larger version (28K): [in this window] [in a new window]   Fig. 4.   Diagram displaying the relative strength of the binding affinities (1/Kd) between Wnts/Frizzleds tested in the assay. The scale is arbitrary.

Separate Domains of Wg Can Be Secreted But Do Not Bind to Frizzled or Evoke a Response-- There are several natural wg alleles in which the Wg protein is truncated. Most of these mutant proteins are misfolded and retained in the endoplasmic reticulum (42-45). However, some of these mutations lead to a shortened protein that is secreted. Interestingly, those mutations truncate Wg close to a region that is uniquely present in Wg, an 85-amino acid insert that is dispensable for Wg function and evolutionary not conserved (44). This region also contains strong antigenic determinants and has served as an epitope for several anti-Wg antibodies (43, 46). The fact that the Wg protein contains a large insert and can be truncated at the site of the insert to generate a folded and secreted variant may be taken as evidence that there are separately folding domains on Wg, perhaps even functionally different. Hays and Bejsovec (44) have reported that such a truncated version of Wg, when overexpressed, can partially rescue wg mutant embryos, indicating that this domain has a function by itself (43).

We were therefore interested in generating different domains of the Wg protein in vitro, to test whether any of them would be able to bind to the Frizzled CRD and would be functional. Thus, the insert region was used as the dividing point for making WgA (the amino-terminal part of Wg) and WgB (carboxyl-terminal part of Wg) constructs (Fig. 5A). The WgB construct included the signal sequence of Wg to allow secretion. An HA sequence was added between the signal sequence and the first cysteine for detection of expression.

View larger version (29K): [in this window] [in a new window]   Fig. 5.   WgA and WgB domains complement each others activities. A, diagrams of WgA (containing the amino-terminal region of Wg) and WgB (containing the carboxyl-terminal region of Wg) and the insert region of Wg. B, combinations of DFz2, WgA, WgB, or Wg were transfected into S2 cells. S2 cells stably expressed DFz2 (lane 1), WgA (lane 2), WgB (lane 3), WgA and DFz2 (lane 4), WgB and DFz2 (lane 5), WgA, WgB, and DFz2 (lane 6), and Wg and DFz2 (lane 7), and were lysed for a Western blot probed with an anti-Armadillo antibody. As shown in lanes 6 and 7, Armadillo proteins were stabilized in WgA + WgB + DFz2 triple transfected cells and Wg + DFz2 double transfected S2 cells.

When we co-transfected plasmids independently generating the WgA and WgB domains, together with the Dfz2 protein in S2 cells, the cells accumulated the Armadillo protein, just like when the full-length Wg was co-transfected with the Dfz2 receptor (Fig. 5B). This indicated that the WgA and WgB expression plasmids were functional. When those same plasmids were transfected separately (WgA or WgB), with Dfz2, no response was seen. We did observe that both the WgA and WgB domains were secreted into the medium (data not shown). However, the secreted forms were not active in eliciting a response in Dfz2-expressing S2 cells or in clone-8 cells, either by themselves or when mixed together (data not shown). Similarly, we were unable to detect binding of either Wg domain to Dfz2 in assays in which full-length Wg did bind (data not shown).

Thus, our data support the view that these two portions of the Wg protein can fold separately, but we did not obtain evidence for separate functions, because both the WgA and WgB domains are required to generate a response when transfected. The lack of a response when these two domains were jointly added to cells in a soluble form might be due to concentration effects, the local concentrations of WgA and WgB in the endoplasmic reticulum of transfected cells is probably higher than in the medium.

No Detectable Binding between Wg and Arr-- Genetic evidence has implicated arrow (LRP5 and LRP6 in vertebrate species) as being required for Wnt signaling in Drosophila and mice (26, 27). Moreover, it has been reported that the LRP6 extracellular domain, as a fusion with the constant region of immunoglobulin, is able to bind to mammalian secreted Wnt1 and form a ternary complex with a Frizzled CRD (28). This has led to a model in which Arrow/LRP acts as a co-receptor for Wnt (28, 31). However, other workers failed to see direct interactions between Arrow/LRP and secreted vertebrate Wnts (29). In vertebrates, such differences between experimental results might be explained by the use of different family members and not having the right cognate molecules. In Drosophila, however, genetic evidence has strongly implicated wg and arrow to be a matching pair and we therefore set out to detect binding between those molecules.

We generated an Arr-Ig construct similar to the LRP6-Ig fusion used by Tamai et al. (28) to obtain the fusion protein from the medium of transiently transfected 293T cells (Fig. 6C). To determine whether Arr-Ig can bind to cell surface-bound Wg, the fusion protein was added to Nrt-Wg/S2 cells, followed by a conjugated mouse anti-human Ig antibody. In comparison to the binding of Dfz2CRD-AP to Nrt-Wg/S2 cells, we did not detect any binding of Arr-Ig (Fig. 6A).

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Dfz2CRD but not the extracellular domain of Arr binds Wg. A, S2 cells and Nrt-Wg/S2 cells were heat-shocked and then incubated with concentrated mediums containing alkaline phosphatase (negative control), Ig, Arr-Ig, or Dfz2CRD-AP proteins. Cells incubated with Ig or Arr-Ig were then washed, and then incubated with the alkaline phosphatase-conjugated mouse anti-human Ig antibody. All cells were then lysed and the bound alkaline phosphatase activity was assayed after incubation with the AP substrate. Dfz2CRD-AP is shown to bind to S2 cells expressing Nrt-Wg. However, Arr-Ig does not bind detectable to Nrt-Wg. B, conditioned medium containing Ig, Arr-Ig, Wg, alkaline phosphatase, or Dfz2CRD-AP were used in this immunoprecipitation-Western blot experiment. Protein G-Sepharose was incubated with Ig and S2-conditioned medium (lane 1), Ig and Wg-conditioned medium (lane 2), Arr-Ig and S2-conditioned medium (lane 3), and Arr-Ig and Wg-conditioned medium (lane 4). Protein A-Sepharose conjugated with the anti-AP antibody was incubated with conditioned medium containing alkaline phosphatase (lane 5) or Dfz2CRD-AP (lane 6). The immunocomplexes were then mixed with Wg-conditioned medium (lanes 5 and 6). Protein G-Sepharose was incubated with Ig, Dfz2CRD-AP, and Wg (lane 7) or Arr-Ig, Dfz2CRD-AP, and Wg (lane 8). All the immunocomplexes were separated by SDS-PAGE before being transferred to a Western blot probed with an anti-Wingless antibody. Wg is shown to bind to Dfz2CRD-AP (lane 6) but not to Arr (lane 4). After mixing Dfz2CRD-AP, Arr-Ig, and Wg, there are no detectable trimeric complexes (lane 8). The band co-migrating with Wg in lane 5 is a background band caused by reaction with the Ig heavy chain from the anti-AP antibody (the Wg protein and the Ig heavy chain are both around 50 kDa). C, the blot was probed with the anti-human Ig antibody. Ig and Arr-Ig are seen at the predicted size.

To test whether soluble Wg could bind to soluble Arrow, we mixed S2-conditioned medium containing Wg with medium containing the secreted Arr-Ig protein. Protein G-Sepharose was added to immunoprecipitate the Arr-Ig fusion. As shown in Fig. 6B, Wg was not detected in this immunocomplex, whereas Wg could be pulled down by the Dfz2CRD-AP, as expected. Finally, to test whether a ternary complex could be formed, we mixed Arrow-Ig, Dfz2CRD-AP, and Wg. However, a ternary complex could not be detected (Fig. 6B).

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Implications of the Interactions between Different Wnts and Frizzleds-- The quantitative measurements of the binding between Wnts and Frizzleds are generally in good agreement with available genetic data in the fly. For example, we have not found any of the five Drosophila Wnt molecules tested here to bind to Smo. Smo is a distant member of the Frizzled family and is implicated in Hedgehog rather than in Wnt signaling. In the Hedgehog pathway, Smo is regulated not by an extracellular ligand, but by an inhibitory interaction with the multipass membrane protein Patched, which serves as the Hh binding receptor (47, 48). When Patched is inactive, Smo generates a signal constitutively (49-52).

With respect to Wg, the best understood member of the Wnt gene family, we found that it binds to three of the five Frizzleds: Fz, Dfz2, and Dfz3. This suggests that these receptors may act redundantly and indeed, wg-like phenotypes in Drosophila embryos and other tissues are only seen when fz and Dfz2 are removed (19-21). DFz3 mutants are viable with no obvious phenotypes (22). Whereas this suggests that Dfz3 is not genetically required for Wg signaling, it should be noted that Dfz3 expression is dependent on active Wg signaling (23). Hence, in flies mutant for fz and Dfz2, where Wg signaling is blocked, Dfz3 is not expressed and the phenotype of the first two receptors may be caused by functional absence of all three.

In addition to wg, there are mutants in two other Drosophila Wnt genes: DWnt2 and DWnt4. DWnt2 is required for pigment cell and direct flight muscle formation in the Drosophila testes and wings, respectively (53, 54). According to our data, DWnt2 binds to Fz, DFz2, and DFz3. The double mutant fz,Dfz2 dies in embryogenesis, making it impossible to examine a possible later phenotype in the testis, but it is likely that the multiple receptors are redundant in DWnt2 signaling, just as they are for Wg. With respect to DWnt4, we find that it binds to Fz, Dfz2, and Dfz4. This agrees with the phenotypes of DWnt4 and Dfz2 in the ovary: a cell migration defect seen for both the ligand and the receptor (55).

There are two Wnts tested here for which there are no mutants available, DWnt3/5 (40, 56) and DWnt8. Surprisingly, DWnt3 did not bind to any of the Frizzleds. It is possible that the fusion protein we used, although expressed on the cell surface, was not functional. Because of the proteolytic processing of the amino-terminal part of the DWnt3 protein, we fused the "Wnt homology region" of DWnt3 to Nrt, but there are no assays to demonstrate that Nrt-DWnt3 encodes a functional protein.

DWnt8 is unique in the sense that it binds to only one Frizzled, DFz4. Both genes are expressed in the embryonic central nervous system (41)² and the fact that they interact biochemically suggests that they are likely to share a function, to be explored when mutants are available.

Is There a Wnt Ligand for Frizzled in the Planar Polarity Pathway?-- frizzled was originally found as a gene involved in the planar polarity pathway, but it has yet to be shown that it interacts with a Wnt during that process. Our binding data indicate that a number of different Wnts, including Wg, DWnt2, and DWnt4, can bind to Frizzled, with approximately equal affinity. All of these Wnt genes are expressed in tissues during the time when planar polarity is established (57).³ Whether they are functional as ligands for Fz remains to be seen, but as with other Wnt-Frizzled interactions, it seems likely that genetic redundancy will be an important consideration.

Do Wnts Interact with Arrow/LRP?-- In contrast to the claim that soluble forms of LRP can bind to Wnt1 and form a ternary complex with a Frizzled CRD, we were unable to detect any direct interactions between the Drosophila LRP homologs Arr and Wg. This discrepancy might be explained by technical differences between our experiments and the ones reported by Tamai et al. (28). However, we note that Mao et al. (29) also failed to see any direct binding between LRP6 and Wnt, in experiments in which robust direct binding between Dickkopf and LRP was detected. At present, we cannot rule out that LRP does act as a co-receptor for Wnt as a direct binding partner, but our experiments do suggest that other mechanisms should also be considered.

	ACKNOWLEDGEMENTS

We thank various colleagues in the lab for comments on the manuscript.

	FOOTNOTES

^* This work was supported by the Howard Hughes Medical Institute and National Institutes of Health Grant 1R01GM/CA60388-01.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed. Tel.: 650-723-7769; Fax: 650-723-1399; E-mail: rnusse@cmgm.stanford.edu.

Published, JBC Papers in Press, August 29, 2002, DOI 10.1074/jbc.M207850200

² C.-H. Wu and Y. K. Xu, unpublished results.

³ C. Logan, unpublished results.

	ABBREVIATIONS

The abbreviations used are: wg, wingless; fz, frizzled; CRD, cysteine-rich domain; sFRP, secreted Frizzled-related protein; FRP, Frizzled-related protein; LRP, low density lipoprotein-related protein; AP, alkaline phosphatase; HA, hemagglutinin; Nrt, neurotactin.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This article has been cited by other articles:

				J.-A Kim, Y.-J. Kang, G. Park, M. Kim, Y.-O. Park, H. Kim, S.-H. Leem, I.-S. Chu, J.-S. Lee, E.-H. Jho, et al. Identification of a Stroma-Mediated Wnt/{beta}-Catenin Signal Promoting Self-Renewal of Hematopoietic Stem Cells in the Stem Cell Niche Stem Cells, June 1, 2009; 27(6): 1318 - 1329. [Abstract] [Full Text] [PDF]

				H. Kim, S.-M. Cheong, J. Ryu, H.-J. Jung, E.-h. Jho, and J.-K. Han Xenopus Wntless and the Retromer Complex Cooperate To Regulate XWnt4 Secretion Mol. Cell. Biol., April 15, 2009; 29(8): 2118 - 2128. [Abstract] [Full Text] [PDF]

				H.-X. Wang, F. R. Tekpetey, and G. M. Kidder Identification of WNT/{beta}-CATENIN signaling pathway components in human cumulus cells Mol. Hum. Reprod., January 1, 2009; 15(1): 11 - 17. [Abstract] [Full Text] [PDF]

				C.-H. Xia, H. Liu, D. Cheung, M. Wang, C. Cheng, X. Du, B. Chang, B. Beutler, and X. Gong A model for familial exudative vitreoretinopathy caused by LPR5 mutations Hum. Mol. Genet., June 1, 2008; 17(11): 1605 - 1612. [Abstract] [Full Text] [PDF]

				P. Bovolenta, P. Esteve, J. M. Ruiz, E. Cisneros, and J. Lopez-Rios Beyond Wnt inhibition: new functions of secreted Frizzled-related proteins in development and disease J. Cell Sci., March 15, 2008; 121(6): 737 - 746. [Abstract] [Full Text] [PDF]

				J. L. Green, T. Inoue, and P. W. Sternberg The C. elegans ROR receptor tyrosine kinase, CAM-1, non-autonomously inhibits the Wnt pathway Development, November 15, 2007; 134(22): 4053 - 4062. [Abstract] [Full Text] [PDF]

				K. E. Harris and S. K. Beckendorf Different Wnt signals act through the Frizzled and RYK receptors during Drosophila salivary gland migration Development, June 1, 2007; 134(11): 2017 - 2025. [Abstract] [Full Text] [PDF]

				Q. Wei, C. Yokota, M. V. Semenov, B. Doble, J. Woodgett, and X. He R-spondin1 Is a High Affinity Ligand for LRP6 and Induces LRP6 Phosphorylation and beta-Catenin Signaling J. Biol. Chem., May 25, 2007; 282(21): 15903 - 15911. [Abstract] [Full Text] [PDF]

				P. M. Smallwood, J. Williams, Q. Xu, D. J. Leahy, and J. Nathans Mutational Analysis of Norrin-Frizzled4 Recognition J. Biol. Chem., February 9, 2007; 282(6): 4057 - 4068. [Abstract] [Full Text] [PDF]

				K. M. Cadigan and Y. I. Liu Wnt signaling: complexity at the surface J. Cell Sci., February 1, 2006; 119(3): 395 - 402. [Abstract] [Full Text] [PDF]

				E. Piddini, F. Marshall, L. Dubois, E. Hirst, and J.-P. Vincent Arrow (LRP6) and Frizzled2 cooperate to degrade Wingless in Drosophila imaginal discs Development, December 15, 2005; 132(24): 5479 - 5489. [Abstract] [Full Text] [PDF]

				R. Takada, H. Hijikata, H. Kondoh, and S. Takada Analysis of combinatorial effects of Wnts and Frizzleds on {beta}-catenin/armadillo stabilization and Dishevelled phosphorylation Genes Cells, September 1, 2005; 10(9): 919 - 928. [Abstract] [Full Text] [PDF]

				A. Ganguly, J. Jiang, and Y. T. Ip Drosophila WntD is a target and an inhibitor of the Dorsal/Twist/Snail network in the gastrulating embryo Development, August 1, 2005; 132(15): 3419 - 3429. [Abstract] [Full Text] [PDF]

				Z. Wang, W. Shu, M. M. Lu, and E. E. Morrisey Wnt7b Activates Canonical Signaling in Epithelial and Vascular Smooth Muscle Cells through Interactions with Fzd1, Fzd10, and LRP5 Mol. Cell. Biol., June 15, 2005; 25(12): 5022 - 5030. [Abstract] [Full Text] [PDF]

				C. Han, D. Yan, T. Y. Belenkaya, and X. Lin Drosophila glypicans Dally and Dally-like shape the extracellular Wingless morphogen gradient in the wing disc Development, February 15, 2005; 132(4): 667 - 679. [Abstract] [Full Text] [PDF]

				W. C. Forrester, C. Kim, and G. Garriga The Caenorhabditis elegans Ror RTK CAM-1 Inhibits EGL-20/Wnt Signaling in Cell Migration Genetics, December 1, 2004; 168(4): 1951 - 1962. [Abstract] [Full Text] [PDF]

				C.-m. Chen, W. Strapps, A. Tomlinson, and G. Struhl Evidence that the cysteine-rich domain of Drosophila Frizzled family receptors is dispensable for transducing Wingless PNAS, November 9, 2004; 101(45): 15961 - 15966. [Abstract] [Full Text] [PDF]

				F. Cong, L. Schweizer, and H. Varmus Wnt signals across the plasma membrane to activate the {beta}-catenin pathway by forming oligomers containing its receptors, Frizzled and LRP Development, October 15, 2004; 131(20): 5103 - 5115. [Abstract] [Full Text] [PDF]

				O. G. Kelly, K. I. Pinson, and W. C. Skarnes The Wnt co-receptors Lrp5 and Lrp6 are essential for gastrulation in mice Development, June 15, 2004; 131(12): 2803 - 2815. [Abstract] [Full Text] [PDF]

				P. Maye, J. Zheng, L. Li, and D. Wu Multiple Mechanisms for Wnt11-mediated Repression of the Canonical Wnt Signaling Pathway J. Biol. Chem., June 4, 2004; 279(23): 24659 - 24665. [Abstract] [Full Text] [PDF]

				X. He, M. Semenov, K. Tamai, and X. Zeng LDL receptor-related proteins 5 and 6 in Wnt/{beta}-catenin signaling: Arrows point the way Development, April 15, 2004; 131(8): 1663 - 1677. [Abstract] [Full Text] [PDF]

				R. Nusse Wnts and Hedgehogs: lipid-modified proteins and similarities in signaling mechanisms at the cell surface Development, November 15, 2003; 130(22): 5297 - 5305. [Abstract] [Full Text] [PDF]

				G. Liu, A. Bafico, V. K. Harris, and S. A. Aaronson A Novel Mechanism for Wnt Activation of Canonical Signaling through the LRP6 Receptor Mol. Cell. Biol., August 15, 2003; 23(16): 5825 - 5835. [Abstract] [Full Text] [PDF]

				X. Ai, A.-T. Do, O. Lozynska, M. Kusche-Gullberg, U. Lindahl, and C. P. Emerson Jr. QSulf1 remodels the 6-O sulfation states of cell surface heparan sulfate proteoglycans to promote Wnt signaling J. Cell Biol., July 21, 2003; 162(2): 341 - 351. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) An addition or correction has been published All Versions of this Article: 277/44/41762    most recent M207850200v1 Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Wu, C.-h. Articles by Nusse, R. Search for Related Content PubMed PubMed Citation Articles by Wu, C.-h. Articles by Nusse, R. Social Bookmarking                      What's this?