• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. Jun 2012; 32(11): 2054–2064.
PMCID: PMC3372227

LGR5 Interacts and Cointernalizes with Wnt Receptors To Modulate Wnt/β-Catenin Signaling

Abstract

LGR5, a seven-transmembrane domain receptor of the rhodopsin family, is a Wnt target gene and a bona fide marker of adult stem cells in the gastrointestinal tract and hair follicle bulge. Recently, we and others demonstrated that LGR5 and its homologues function as receptors of the R-spondin family of stem cell factors to potentiate Wnt/β-catenin signaling. However, the mechanism of how LGR5 enhances the signaling output remains unclear. Here we report that following costimulation with the ligands R-spondin1 and Wnt3a, LGR5 interacts and forms a supercomplex with the Wnt coreceptors LRP6 and Fzd5 which is rapidly internalized and then degraded. Internalization of LGR5 is mediated through a dynamin- and clathrin-dependent pathway. Inhibition of this endocytic process has no effect on LGR5 signaling. Deletion of the C-terminal tail of LGR5 maintains its ability to interact with LRP6, yet this LGR5 mutant exhibits increased signaling activity and a decreased rate of endocytosis in response to R-spondin1 compared to the wild-type receptor. This study provides direct evidence that LGR5 becomes part of the Wnt signaling complex at the membrane level to enhance Wnt/β-catenin signaling. However, internalization of LGR5 does not appear to be essential for potentiating the canonical Wnt signaling pathway.

INTRODUCTION

Leucine-rich repeat-containing, G protein-coupled receptor 5 (LGR5) is a target of Wnt signaling and a bona fide marker of adult stem cells of the intestinal crypts, stomach, and hair follicle bulge (2). Single LGR5-positive crypt stem cells can form organoid structures in vitro, and mouse models have demonstrated that LGR5-positive stem cells are the cells of origin of intestinal tumors (3, 36). Increased expression of LGR5 has been documented for several types of human cancer (3, 24, 38, 40), particularly in colon cancer stem cells (26). LGR5, also referred to as HG38 and GPR49, was initially cloned as an orphan receptor of the glycoprotein hormone receptor subfamily (15, 25). It is most homologous to LGR4 and LGR6, with ~50% amino acid identity between each other (14, 25). The three receptors (LGR4 to -6) form a structurally distinct group with a considerably large extracellular domain (ECD) composed of 17 leucine-rich repeats and a seven-transmembrane (7TM) domain typical of the rhodopsin family of G protein-coupled receptors (GPCRs) (14, 25). Homozygous knockout of either LGR4 or LGR5 is lethal, and LGR4 is essential for the survival of intestinal stem cells (11, 23, 28, 29). Recently, we and others demonstrated that LGR4 to -6 function as receptors of stem cell factor R-spondins to potentiate Wnt/β-catenin signaling (8, 11, 13). R-spondins are a group of four secreted proteins (RSPO1 to -4) that were known to be potent agonists of Wnt signaling with pleiotropic functions in development, stem cell survival and differentiation, and oncogenesis (18, 36, 39, 46). The agonistic effect of RSPOs on β-catenin signaling is mediated by LGR4 to -6 and requires costimulation with a Wnt ligand (8, 11, 13, 17). RSPO1 to -4 can bind to and activate LGR4 and LGR5 with high affinity, and subsequent potentiation of the Wnt3a/β-catenin pathway does not involve activation of heterotrimeric G proteins (8, 11, 13). However, the precise mechanism of how activation of LGR4/5 by RSPOs leads to increased β-catenin signaling remains unknown.

The Wnt/β-catenin signaling pathway plays essential roles in normal development and the maintenance of adult stem cells as well as in tumor initiation and growth (20, 32, 34). In the absence of Wnt ligand stimulation, cytoplasmic levels of β-catenin are kept low due to degradation mediated by a destruction complex consisting of CK1α, glycogen synthase kinase 3 (GSK3), axin, APC, and other factors (7, 20). Binding of Wnt ligands to Frizzled receptors (Fzd) initiates recruitment of the coreceptors LRP5/6 (7, 20). The Wnt-Fzd-LRP5/6 ternary complex interacts with and activates the intracellular scaffold protein Dishevelled (Dvl), forming multiprotein aggregates or endocytic signalosomes that promote LRP6 phosphorylation by CK1γ and GSK3β (7, 20, 27). Phosphorylated LRP6 binds to and inhibits the activity of the destruction complex, leading to β-catenin accumulation, nuclear translocation, and eventually transcriptional activation (20, 27). One of the key sites of LRP6 phosphorylation is at Ser1490 following Wnt stimulation (31, 47). Notably, RSPOs clearly potentiate Wnt3a-induced phosphorylation of LRP6 at this position in an LGR4/5-dependent manner through a yet unknown mechanism (8).

Ligand-induced receptor activation is generally followed by endocytic internalization, which can lead to either enhancement or termination of signaling (37). Typically, receptor endocytosis occurs through a clathrin-dependent or a caveolin-mediated process (37). For most 7TM GPCRs, endocytosis is initiated through phosphorylation of the C-terminal tail by GPCR kinases (GRKs) subsequent to agonist-induced activation (43). Phosphorylated receptors then recruit β-arrestin, which binds AP2 to form clathrin-coated pits and triggers dynamin-dependent endocytosis (43). Internalized receptors are delivered to early endosomes, where the components are either sorted for recycling back to the plasma membrane or targeted to late endosomes for degradation (37, 43). In the case of Wnt/β-catenin signaling in mammalian cells, the role of endocytosis remains unclear (12). Some studies indicated that clathrin-mediated internalization is essential for signaling, while others suggested that the caveolin pathway is more important (6, 12, 45). We have previously shown that LGR4 and LGR5 overexpressed in HEK293 cells are rapidly internalized in the presence and absence of exogenous RSPOs (8). Here we show that LGR5 forms a complex with Wnt coreceptors following ligand stimulation and is internalized through a clathrin- and dynamin-dependent mechanism. However, inhibition of internalization does not affect LGR5-mediated potentiation of Wnt/β-catenin signaling. We also provide evidence that the C-terminal tail of LGR5 serves an important role in driving receptor internalization and degradation. Interestingly, deletion of the tail leads to increased LGR5 activity in potentiating Wnt/β-catenin signaling.

MATERIALS AND METHODS

Plasmids and recombinant proteins.

Plasmids encoding human Myc-LGR5 (amino acids [aa] 21 to 907, GenBank accession no. NP_003658), HA-LRP6 (aa 19 to 1613, GenBank accession no. NP_002327), and HA-LRP6ECD (aa 19 to 1408) were generated as previously described (8). Myc-LGR5ΔC was constructed by using primers designed to PCR amplify the portion of the LGR5 nucleotide sequence encoding amino acids 21 to 836 from the Myc-LGR5 plasmid. This amplicon was then ligated into the pIRESpuro3 vector (Clontech) modified to incorporate an N-terminal Myc-tag and CD8 signal peptide sequence (pIP3SP). LGR5-ECDTM was constructed in the same vector by fusing the transmembrane domain of CD4 (aa 216 to 246) to the ECD of LGR5 (aa 21 to 541). The rat Fzd5-eGFP plasmid was generated by subcloning Fzd5 from the rat Fzd5 V5/6×His pcDNA3.1 construct (10) into peGFP-N1 (Clontech). Super 8× TopFlash was purchased from Addgene. pRL-SV40 was purchased from Promega. DN-DynK44A was a gift from the laboratory of John Hancock (UT Houston Medical School, TX). Recombinant human RSPO1 and RSPO2 were purchased from R&D Systems.

Cell culture, stable cell line generation, and production of Wnt3aCM.

HEK293 and HEK293T cells were grown in Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal calf serum (FCS) and penicillin and streptomycin in a 37°C incubator with 95% humidity and 5% CO2. For stable cell line generation, the Myc-LGR5, Myc-LGR5ΔC, and vector control plasmids were transfected into HEK293 cells using Fugene 6 (Roche), and bulk stable cells were selected and maintained with puromycin at 1 μg/ml. Wnt3a conditioned medium (Wnt3aCM) was produced from L cells as previously described (42). The concentration of Wnt3a in Wnt3aCM is estimated to be approximately 25 to 50 ng/ml based on previous data (8, 42).

Immunofluorescence and confocal microscopy.

Transient transfection of stable cell lines or HEK293T cells with vector (pIP3SP), HA-LRP6, or Fzd5-eGFP was performed in six-well plates using Fugene HD (Roche). For internalization assays, cells were reseeded into poly-d-lysine-coated eight-well chamber slides (BD Biosciences) and allowed to adhere overnight. The next day, cells were preincubated with mouse anti-Myc-Cy3 antibody (Sigma) at 4°C for 2 h, washed, and then treated with either prewarmed RSPO1, Wnt3aCM, or RSPO1 plus Wnt3aCM for the indicated length of time. Cells were then washed, fixed, and when necessary, permeabilized. Cells cotransfected with HA-LRP6 were incubated with rabbit antihemagglutinin (anti-HA) (Novus) for 45 min followed by goat anti-rabbit antibody–Alexa 488 (Invitrogen) for 1 h at room temperature, respectively. Nuclei were counterstained with TO-PRO-3. To detect the effects of endocytic inhibitors on LGR5 internalization, cells were treated with dimethyl sulfoxide (DMSO) vehicle, 10 μM filipin III, 40 μM dynasore, or 100 μM monodansylcadaverine (MDC) (Sigma) for 30 min prior to 4°C incubation with mouse anti-Myc-Cy3 antibody for 2 h and ligand stimulation at 37°C for 30 min. Images were recorded and analyzed using confocal laser scanning microscopy (Leica TCS SP5 microscope) with the LAS AF Lite software. Quantification of LGR5 or LGR5ΔC internalization was performed using an algorithm in the software BlobFinder (1). The translocation of internalized puncta of LRP6, Fzd5, and LGR5 was assessed manually with the constraints of >15 puncta per cell representative of cytoplasmic localization. Approximately 100 cells were quantified per treatment and/or time point from three individual experiments. Data were analyzed in GraphPad Prism 5 using analysis of variance (ANOVA) and Bonferroni post hoc statistical analysis.

FRET.

HEK293T cells transiently transfected with 1 μg HA-LRP6 (donor) and 3 μg Myc-LGR5 (acceptor) or vector were reseeded into a 96-well black poly-d-lysine-coated flat-bottom plate. The next day, cells were treated with 100 ng/ml RSPO1, Wnt3aCM (1:5), or 100 ng/ml RSPO1 plus Wnt3aCM (1:5) in DMEM or with DMEM alone for 30 min. Cells were then washed, fixed for 10 min in 4% paraformaldehyde, and permeabilized with 0.1% saponin for 10 min. To label the donor and acceptor pair, cells were incubated with rabbit anti-HA for 1 h, followed by mouse anti-Myc-Cy3 and goat anti-rabbit antibody–Alexa 488 for an additional hour. Cells were washed in phosphate-buffered saline (PBS), and fluorescence was measured on the Tecan Infinite M1000 plate reader for donor (470 nm excitation/520 nm emission) and fluorescence resonance energy transfer (FRET) (470 nm excitation/570 nm emission). This experiment was performed three times with at least nine replicate wells per sample. For data analysis, the average FRET was obtained by normalizing the individual FRET values to the corresponding donor values for each well and then calculating the mean for the entire sample set. The percent increase in FRET was calculated from the equation (FRETtreated−FRETcontrol/FRETcontrol) × 100, where FRETcontrol and FRETtreated represent the average FRET ratios. Data were analyzed in GraphPad Prism 5 using ANOVA and Bonferroni post hoc statistical analysis.

SuperTopFlash β-catenin reporter assays.

SuperTopFlash assays were performed as before with slight modifications (8). Briefly, HEK293T cells were transiently transfected in six-well culture plates with 1 μg vector plasmid, Myc-LGR5, or Myc-LGR5ΔC, 1 μg Super 8× TopFlash and 100 ng pRL-SV40-renilla luciferase reporter using FuGene HD. HA-LRP6 and HA-LRP6ECD were transfected at 0.5 μg each. DN-DynK44A was transfected at 1 μg per well. Cells were reseeded the next day into 384-well plates, treated with various concentrations of RSPO1 in Wnt3aCM (1:5 endpoint dilution), and incubated overnight. Luciferase assay measurements were carried out using the Dual-Glo luciferase assay kit (Promega) according to the manufacturer's protocol. All experiments were performed at least three times with quadruplicates in each experiment. Data were analyzed in GraphPad Prism 5 using ANOVA and Bonferroni post hoc statistical analysis.

Coimmunoprecipitation.

The Myc-LGR5 HEK293T stable cell line was transiently transfected with HA-LRP6 in a 10-cm culture dish using Fugene HD. The next day, cells were reseeded into three separate 10-cm dishes and allowed to adhere overnight. Cells were then incubated at 4°C for 45 min and then treated with warm DMEM containing vehicle or 100 ng/ml RSPO1 plus Wnt3aCM (1:5) for 30 min at 37°C. The cells were lysed with RIPA buffer (50 mM Tris-Cl, pH 7.4, 150 mM NaCl, 1 mM dithiothreitol [DTT], 1% Triton X-100, 0.5% sodium deoxycholate, 0.1% SDS) supplemented with protease and phosphatase inhibitors. Rabbit anti-Myc antibody or control rabbit IgG (Santa Cruz) was added to the cleared lysate, and the samples were mixed for 1 h at 4°C. Protein A/G agarose beads were then added, and the samples were mixed overnight at 4°C and then washed twice with ice-cold lysis buffer and once with PBS. Proteins were eluted from the beads with 2× SDS sample buffer, and Western analysis was carried out.

Receptor degradation and Western blot analysis.

HEK293T cells were transfected with 1 μg Myc-LGR5 and/or HA-LRP6 and the appropriate amount of control vector to equalize the total DNA transfected. Cells were reseeded into six-well dishes and treated overnight with vehicle, 100 ng/ml RSPO2, Wnt3aCM, or RSPO2 plus Wnt3aCM. For the LGR5 turnover experiment, LGR5 and LGR5ΔC stable cells were seeded into six-well plates. The next day, cells were pretreated with DMSO vehicle or 100 μM cycloheximide for 30 min at 37°C. Cells were then kept at 4°C for 45 min to recruit receptors to the plasma membrane. For internalization to ensue, cells were treated with either PBS vehicle in control CM or RSPO1 (100 ng/ml) in Wnt3aCM for the time points indicated. To detect levels of active β-catenin, HEK293T cells transiently transfected with vector, LGR5, or LGR5ΔC were serum starved for 3 h and then treated with RSPO1 (100 ng/ml) in Wnt3aCM (1:5 dilution) for 0, 2, and 6 h. For Western blot analysis, cells were lysed with RIPA buffer (50 mM Tris-Cl, pH 7.4, 150 mM NaCl, 1 mM DTT, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS) supplemented with protease and phosphatase inhibitors. Immunoblotting was performed using mouse anti-active β-catenin (Millipore), mouse anti-Myc (Cell Signaling) and rabbit anti-HA (Novus) to probe levels of recombinant receptors, and mouse anti-β-actin (Cell Signaling) antibody for normalization. Horseradish peroxidase (HRP)-labeled goat anti-mouse and goat anti-rabbit secondary antibodies were utilized for detection along with the standard ECL protocol. Quantification of Western analysis was performed using the software ImageJ.

RESULTS

Internalization of LGR5 is enhanced by costimulation with RSPO1 and Wnt3a.

We have previously shown that LGR5 is highly internalized in the absence of exogenous RSPOs (8). As receptor internalization is often induced by ligand stimulation, we examined whether the addition of RSPOs would affect this process using HEK293 cells stably expressing LGR5 with an N-terminal Myc tag. Cells were preincubated with Cy3-labeled anti-Myc antibody for 2 h at 4°C to label the receptors on the cell surface. Free antibody was washed away, and the cells were either fixed immediately or incubated at 37°C for 30 min with medium alone, Wnt3a, RSPO1 (100 ng/ml), or RSPO1 (100 ng/ml) plus Wnt3a in prewarmed DMEM. The cells were washed, fixed, and then imaged by confocal fluorescence microscopy. In order to quantify the extent of internalization, a series of images were acquired and an algorithm was implemented to assess the number of internalized puncta per cell (1). High levels of LGR5 were detected on the cell surface following antibody labeling at 4°C (Fig. 1A, panel a). Incubation at 37°C for 30 min with medium alone resulted in significant internalization, as evident from the appearance of labeled intracellular vesicles (Fig. 1A, panel b). The addition of either Wnt3a or RSPO1 alone had no significant effect on internalization (Fig. 1A, panels c and d versus panel b; Fig. 1B). Interestingly, simultaneous treatment with RSPO1 and Wnt3a significantly increased the extent of internalization (Fig. 1A, panel e versus panels c and d). Quantification of the images indicated that the cotreatment increased the number of cytoplasmic puncta by ~2.5-fold versus vehicle (Fig. 1B). The data indicate that at the specified ligand concentration and time point, LGR5 internalization is considerably enhanced when cells are simultaneously treated with RSPO1 and Wnt3a.

Fig 1
Internalization of LGR5 is increased by cotreatment with RSPO1 and Wnt3a. (A) Confocal images of HEK293 cells stably expressing Myc-LGR5 that were prelabeled with Cy3-anti-Myc antibody at 4°C (baseline) and incubated at 37°C for 30 min ...

LGR5 interacts and cointernalizes with LRP6 in response to ligand stimulation.

Complex formation and endocytosis of Wnt coreceptors and other signaling molecules are integral processes of the Wnt/β-catenin signaling pathway (4, 20, 27). Since RSPO1 increases internalization of LGR5 in the presence of Wnt3a, we set out to test if LGR5 interacts and cointernalizes with the Wnt3a coreceptor LRP6 as part of the mechanism responsible for the accelerated internalization and LGR5 potentiation of β-catenin signaling. LGR5 was prelabeled with Cy3-anti-Myc antibody at 4°C as before, while N-terminal HA-tagged LRP6 was detected using anti-HA antibody after the cells were fixed and permeabilized. At baseline, both receptors are found on the cell surface (Fig. 2A, panels a to c). In the absence of exogenous ligands, LGR5 was rapidly internalized whereas LRP6 was still mostly on the cell surface after the cells were incubated at 37°C for 30 min (Fig. 2A, panels d to f). When cells were cotreated with RSPO1 and Wnt3a, both LGR5 and LRP6 were found in intracellular vesicle with near-complete colocalization between the two receptors (Fig. 2A, panels g to i). Treatment with RSPO1 alone had little effect on the distribution of LRP6 (see Fig. S1a to c in the supplemental material), consistent with the finding that RSPO1 is not a ligand of LRP6. In contrast, addition of Wnt3a induced internalization of LRP6 which are colocalized with LGR5 in cells expressing both receptors (see Fig. S1d to f in the supplemental material). These data imply that stimulation of LGR5 and LRP6 with their cognate ligands leads to cointernalization of the two receptors, suggesting that LGR5 becomes part of the Wnt-LRP-Fzd signaling complex.

Fig 2
LGR5 forms a complex with LRP6. (A) Confocal imaging of Myc-LGR5 (red) colocalization with HA-LRP6 (green) at 4°C (a to c) or postincubation at 37°C with vehicle (d to f) or RSPO1 plus Wnt3a (g to i) for 30 min. (B) Measurement of FRET ...

We then tested if LGR5 interacts with LRP6 using the fluorescence resonance energy transfer (FRET) method. FRET occurs when the fluorescent donor and acceptor molecules are in close proximity to one another (typically 10 to 100 Å) and is commonly used to detect direct interactions between two molecules (44). HEK293T cells were transfected to achieve the most favorable transfected levels of LRP6 (donor) and LGR5 (acceptor) and then treated with vehicle, RSPO1, Wnt3a, or RSPO1 plus Wnt3a for 30 min. The cells were incubated with rabbit anti-HA followed by incubation with Alexa 488-labeled goat anti-rabbit IgG and Cy3-labeled mouse anti-Myc antibodies. This pair of dyes (Alexa 488 and Cy3) was shown to be an excellent donor-acceptor pair for FRET analysis (33). Cells transfected with LRP6 (donor) alone exhibited a baseline FRET signal which was unaffected by the addition of exogenous ligands (Fig. 2B). Cotransfection of LGR5 and LRP6 led to a significant increase in the amount of FRET signal, which was further enhanced when both Wnt3a and RSPO1 were added. The increase of the FRET signal in vehicle- or single ligand-treated cells is most likely due to the presence of endogenous RSPO and Wnt ligands, since HEK293T cells are known to express both Wnts and RSPOs (5, 8, 16). Only the cotreatment with RSPO1 and Wnt3a showed a statistically significant increase in FRET between LRP6 and LGR5 compared to vehicle-treated cells (Fig. 2B, 64% for RSPO1 plus Wnt3a versus 21% and 23% for RSPO1 and Wnt3a, respectively). These results indicate that dual stimulation with Wnt3a and RSPO1 brings LRP6 and LGR5 into close proximity to each other.

The ligand-induced interaction between LGR5 and LRP6 was further validated using coimmunoprecipitation (co-IP) of LGR5 with LRP6. LGR5 HEK293 cells transfected with HA-LRP6 were incubated at 4°C for 45 min and then cotreated with RSPO1 plus Wnt3a or vehicle and placed in a 37°C incubator for 30 min. Cells were harvested and LGR5 was immunoprecipitated with rabbit anti-Myc antibody and probed with anti-HA for detection of LRP6. As shown in Fig. 2C, a strong LRP6 signal was detected when both RSPO1 and Wnt3a were present, suggesting that LRP6 directly interacts with LGR5 or resides in the same signaling complex following ligand stimulation. The fact that a lesser amount of LRP6 was detected in the absence of ligands implies that the coprecipitation is highly specific. Taken together with the results from the cointernalization and FRET experiments, these data indicate that LGR5 forms a complex with LRP6 through either direct or indirect interaction, and the two receptors cointernalize following stimulation with their cognate ligands.

LGR5 accelerates the rate of internalization and degradation of LRP6.

Given that LGR5 is rapidly internalized in the absence of exogenous ligands, whereas LRP6 is not, we investigated if LGR5 affects the rate of LRP6 internalization and degradation as a result of complex formation between the two receptors. LGR5 and vector HEK293 stable cell lines were transfected with LRP6 and treated with RSPO1 and Wnt3a for 0, 10, 20, 30, 60, and 120 min and then fixed and stained for confocal imaging. A series of images were taken at each time point, and the number of puncta or vesicles per cell was determined by automated image analysis (1). To better define the process of internalization, the cells were classified based on the localization of LRP6 into three subtypes: membrane only, membrane/cytoplasmic with <15 puncta per cell, and cytoplasmic with >15 puncta per cell. In vector cells treated with both RSPO1 and Wnt3a, significant internalization of LRP6 occurred at 30 min postincubation (Fig. 3A). In cells coexpressing LGR5, initiation of LRP6 internalization in response to RSPO1 and Wnt3a treatment was apparent at a much earlier time point (10 min versus 30 min) (Fig. 3B versus A). Virtually all of the recombinant LRP6 receptor was completely internalized at 30 min in LGR5 cells (Fig. 3B). In contrast, LRP6 did not reach peak internalization in vector control cells until approximately 1 h and still failed to completely internalize (Fig. 3A). Furthermore, LRP6 started to reappear at the membrane after 1 h of incubation, which is most likely a consequence of either recycled or newly synthesized protein (Fig. 3A and B). LRP6 internalization induced by RSPO1 or Wnt3aCM alone was also slightly accelerated in LGR5 cells (see Fig. S1B to E in the supplemental material). These data quantitatively confirm that internalization of LRP6 in cells coexpressing LGR5 is significantly increased when both receptors are stimulated by their cognate ligands.

Fig 3
LGR5 increases the internalization and degradation of LRP6. Quantification of LRP6 internalization in vector (A) and LGR5 stable HEK293 (B) cells treated with RSPO1 and Wnt3a. Cytoplasmic localization was gated at a translocation of >15 receptor ...

Interestingly, through the course of LRP6-LGR5 cointernalization experiments, we noticed that when LRP6 was overexpressed in the LGR5 stable cell line, LGR5 levels were significantly reduced or completely excluded in cells with high levels of LRP6 and vice versa. This phenomenon was significantly enhanced over time following transfection. Therefore, we tested if these two receptors are codegraded as a potential mechanism of negative feedback due to overactivation of Wnt signaling. HEK293T cells were transiently transfected to overexpress LGR5, LRP6, or both and then treated overnight with RSPO2, Wnt3a, or RSPO2 plus Wnt3a. RSPO2 was used here since it is generally interchangeable with RSPO1 in potentiating Wnt/β-catenin signaling (8, 11, 17). LGR5 levels were significantly lower when coexpressed with LRP6, and both receptors were codegraded in response to ligand treatment, especially with RSPO2 plus Wnt3a (Fig. 3C). In contrast, receptor levels were not affected by ligand treatment in cells expressing LGR5 or LRP6 alone (Fig. 3C). To exclude the possibility that this observed codegradation could potentially be an artifact of receptor overexpression, we performed a similar experiment replacing wild-type LGR5 with a truncated form of LGR5 (LGR5-ECDTM) that expresses the ECD of LGR5 fused to the single transmembrane domain of CD4. This membrane-anchored ECD was well expressed on the cell surface and did not undergo internalization (see Fig. S2A, panels a, d, and g, in the supplemental material). When LRP6 was cotransfected with LGR5-ECDTM, no internalization was induced following stimulation with ligands (see Fig. S2A, panels b, e, and h). Importantly, codegradation of LRP6 and LGR5-ECDTM was not observed in either the presence or absence of exogenous ligand (see Fig. S2b). These data indicate that coexpression of LGR5 with LRP6 increases the rate of LRP6 internalization and enhances receptor codegradation.

LGR5 cointernalizes with Fzd5.

LRP5/6 interacts with frizzled receptors (Fzds) in the presence of Wnt ligands to form ternary complexes (LPR5/6-Wnt-Fzd) in order to activate and transmit canonical Wnt/β-catenin signaling (7, 20). Since LGR5 forms a complex with LRP6 and requires the presence of Wnt3a for efficient signaling, we examined if LGR5 also colocalized with Fzd5, a receptor for Wnt3a (9). LGR5 HEK293 cells transiently transfected with Fzd5-eGFP were preincubated with Cy3-anti-Myc at 4°C as before and were then treated with RSPO1 and Wnt3a at 37°C for 30 min. Similar to our findings with LRP6, Fzd5-eGFP was colocalized with LGR5 at the cell membrane at the 4°C baseline and cointernalized with LGR5 with the addition of both ligands (Fig. 4A, panels a to c and g to h). In untreated cells, cointernalization was much less apparent since Fzd5 internalization is mainly ligand dependent and LGR5 is internalized in the absence of exogenous ligand (Fig. 4A, panels d to f). We did observe LGR5 cointernalization with Fzd5 when treated with Wnt3a alone, though the extent of Fzd5 internalization was much less (see Fig. S3 in the supplemental material). To assess whether LGR5 overexpression also alters the rate of Fzd5 internalization, vector and LGR5 HEK293 stable cell lines were transfected with Fzd5-eGFP and treated with RSPO1 plus Wnt3aCM for 0, 15, 30, and 60 min and then were fixed and stained for confocal imaging. Approximately 50% of the LGR5 cells coexpressing Fzd5-eGFP exhibited some degree of Fzd5 internalization after 15 min versus 20% of vector cells (Fig. 4B and C). Peak internalization of Fzd5-eGFP occurred around 30 min and 45 min-1 h for LGR5 and vector cells, respectively. Treatment with either RSPO1 or Wnt3a alone also triggered an increase in the rate of Fzd5 internalization in LGR5 cells (see Fig. S3B to E in the supplemental material). These results demonstrated that overexpression of Fzd5 with LGR5 leads to accelerated internalization of Fzd5, which is further increased upon treatment with ligands. These results, taken together with the data demonstrating the LGR5-LRP6 interaction, indicate that subsequent to RSPO1 and Wnt3a stimulation, LGR5 becomes part of the signaling complex comprising Wnt coreceptors LRP6 and Fzd5. This complex formation may provide the basis for how LGR5 enhances Wnt/β-catenin signaling. Furthermore, these data suggest that the three receptors are ultimately internalized and degraded together.

Fig 4
LGR5 colocalizes and internalizes with Fzd5. (A) Colocalization of Cy3-labeled Myc-LGR5 (red) with Fzd5-eGFP (green) at 4°C (a to c) or postincubation at 37°C for 30 min with vehicle (d to f) or RSPO1 plus Wnt3a (g to i). (B and C) Quantification ...

Internalization of LGR5 is mediated by a clathrin-dependent pathway.

Receptor internalization is generally mediated through either a clathrin-dependent or -independent pathway (37). To uncover the route of the LGR5 internalization, HEK293T cells overexpressing LGR5 were preincubated with endocytic inhibitors that block distinct pathways of endocytosis and then subsequently treated with RSPO1 and Wnt3a. LGR5 internalization/endocytosis was unaffected by the presence of 10 μM filipin III (Fig. 5A and B), a standard agent used to suppress caveolin-mediated, clathrin-independent endocytosis (35). On the other hand, receptor endocytosis was completely inhibited in cells treated with 100 μM monodansylcadaverine (MDC) and ~70% inhibited in cells with 40 μM dynasore (Fig. 5A and B). MDC inhibits the clustering and internalization of ligand-receptor complexes into clathrin-coated vesicles and thus abolishes only clathrin-dependent endocytosis. Dynasore is a potent inhibitor of dynamin GTPase activity and thus blocks both clathrin-dependent and caveolin-dependent pathways (21). Endocytosis was also blocked by treatment with 4 M hypertonic sucrose (see Fig. S4 in the supplemental material), another commonly used method to inhibit the clathrin-mediated pathway. As receptor internalization often affects signaling (12, 37), it is important to determine if inhibition of LGR5 internalization leads to changes in its functional activity in potentiating Wnt/β-catenin signaling. Unfortunately, treatment with MDC, dynasore, or sucrose led to severe cytotoxicity under these conditions (data not shown), rendering them unsuitable for functional evaluation. As an alternative approach, we utilized a construct to express a dominant negative form of dynamin (DN-DynK44A) in order to test the effect of blocked internalization on signaling. HEK293T cells transfected with vector, DN-DynK44A, LGR5, or LGR5 plus DN-DynK44A were subjected to an RSPO1 dose-response assessment in the presence of Wnt3a. Surprisingly, the dominant negative mutant of dynamin had no effect on RSPO1-LGR5-induced potentiation of Wnt/β-catenin signaling (Fig. 5C). On the other hand, the dynamin mutant clearly blocked receptor internalization, as shown by the confocal imaging analysis data (Fig. 5D and E). The small number of puncta located close to the membrane in dynamin mutant cells may represent “stalled” endocytosis, since dynamin is required for the pinching off of clathrin-coated pits, or incomplete suppression of internalization. Similarly, cotransfection of DN-DynK44A and LGR5 in HEK293 cells stably expressing HA-LRP6 blocked ligand-induced cointernalization of LGR5 and LRP6 (see Fig. S5 in the supplemental material). The latter finding suggests that under these specified conditions, LRP6 is internalized via the same endocytic route as LGR5. These results demonstrate that LGR5 internalization occurs through a clathrin-dependent pathway, but the process is not essential for LGR5's function in potentiating Wnt/β-catenin signaling.

Fig 5
LGR5 is internalized through a clathrin-dependent pathway. (A) HEK293 cells stably expressing Myc-LGR5 were labeled with Cy3-anti-Myc, pretreated with the indicated endocytic inhibitors (10 μM filipin, 40 μM dynasore, 100 μM MDC) ...

The C-terminal tail of LGR5 is important for receptor internalization and degradation.

For classical G protein-coupled 7TM receptors, clathrin-dependent endocytosis is mediated through β-arrestin binding to the receptor following phosphorylation that generally occurs at the C-terminal tail of the receptor (19). LGR5 contains a 7TM domain that is typical of classical GPCRs followed by a short C-terminal tail. However, LGR5 does not induce β-arrestin translocation after RSPO stimulation (8). To test if the LGR5 C-terminal tail is still critical for receptor internalization, we generated a truncation mutant (LGR5ΔC) that lacks the sequence beyond 7TM. HEK293 cell lines stably overexpressing either LGR5 or LGR5ΔC were labeled with Cy3-anti-Myc antibody as before and then were incubated with RSPO1 and Wnt3a for 0, 15, 30, 60, 120, and 240 min at 37°C. With regard to wild-type LGR5 (LGR5 for short), ~20% of the cells showed complete internalization at 15 min, 100% showed complete internalization at 30 min, and ~90% of the cells displayed receptor degradation at 120 min in the absence of ligand stimulation (Fig. 6A, 1st row; Fig. 6B). Treatment of LGR5 cells with RSPO1 and Wnt3a clearly accelerated the process of internalization and degradation, as evident from the data indicating that ~60% of the cells showed complete internalization at 15 min (3-fold increase versus untreated cells) and 90% of the cells showed degradation at 60 min (Fig. 6A, 3rd row versus 1st row; Fig. 6C versus B). For LGR5ΔC, no significant internalization occurred at 15 min (versus 80% for LGR5) and only ~30% of the cells showed complete internalization at 60 min (versus 60% for LGR5) in the absence of ligand stimulation (Fig. 6A, 2nd row versus 1st row; Fig. 6D versus B). LGR5ΔC treated with RSPO1 plus Wnt3a showed accelerated internalization (Fig. 6A, 4th row versus 2nd row; Fig. 6E versus D) but still did not reach the level of LGR5 (Fig. 6A, 4th row versus 3rd and 1st rows; Fig. 6E versus B and D). These data suggest that the C-terminal tail of LGR5 increases the rate of internalization but is not essential for the process.

Fig 6
Truncation of the LGR5 C-terminal tail decreases the rate of receptor internalization. (A) Confocal images of the time course of internalization of Cy3-labeled Myc-LGR5 and Myc-LGR5ΔC without and with RSPO1 plus Wnt3a. (B to E) Quantification ...

It was also notable that cells expressing LGR5ΔC retained many more receptors at later time points following ligand stimulation (Fig. 6A). Therefore, we compared the extent of turnover between the two receptor forms. LGR5 and LGR5ΔC HEK293 cells were pretreated with vehicle or 100 μM cycloheximide (CHX) for 30 min at 37°C to prevent synthesis of new receptors. Cells were then incubated at 4°C for 45 min, to ensure receptor recruitment to the plasma membrane, and then were allowed to internalize at 37°C in the presence of RSPO1 and Wnt3a. In the absence of CHX, no change in receptor degradation was apparent, since new receptors are constantly being synthesized (Fig. 7A). In the presence of CHX, LGR5ΔC exhibited a decreased rate of degradation compared to LGR5 (Fig. 7A). Quantification of the blot revealed that at 1 h posttreatment, total levels of LGR5 decreased by ~40% whereas LGR5ΔC levels remained essentially unchanged over the 1-h time period (Fig. 7B). At 4 h, LGR5 lost ~60% of its receptors while LGR5ΔC lost ~40% (Fig. 7B). In addition, the steady-state level of LGR5ΔC was clearly higher than that of the wild-type receptor in the absence of CHX treatment (Fig. 7A), another indication of the C-terminal tail regulating the stability of the receptor. These findings confirm the results of our image analysis and demonstrate that the C-terminal tail of LGR5 is important but not essential for receptor degradation.

Fig 7
LGR5 without its C-terminal tail is degraded more slowly than the wild type. (A) Immunoblot analysis of RSPO1/Wnt3a-induced receptor turnover of LGR5 versus LGR5ΔC in the presence and absence of 100 μM cycloheximide. (B) Quantification ...

Deletion of the C-terminal tail of LGR5 enhances receptor activity.

Since we have demonstrated that the C-terminal tail of LGR5 plays a positive role in receptor internalization and degradation, we next wanted to address if the tail is required for LGR5 interaction with LRP6. Coimmunoprecipitation was performed as described earlier for LGR5. LGR5ΔC was still able to interact with LRP6 when treated with RSPO1 plus Wnt3aCM for 30 min (Fig. 8A). We then investigated whether LGR5ΔC alters the rate of LRP6 internalization in the presence of exogenous ligands. Quantification of confocal imaging shows that stable overexpression of LGR5ΔC did not accelerate LRP6 internalization as was observed with wild-type LGR5 (Fig. 8B versus Fig. 3B). In fact, LRP6 internalization occurred at a slightly lower rate in LGR5ΔC cells than in vector cells (Fig. 8B versus Fig. 3A). These data indicate that the C-terminal tail of LGR5 is not essential for interaction with LRP6 but plays a significant role in promoting LRP6 internalization.

Fig 8
LGR5 without its C-terminal tail exhibits increased activity in potentiating Wnt signaling. (A) Immunoprecipitation (IP) of Myc-LGR5ΔC with anti-Myc antibody followed by immunoblotting (IB) with anti-HA (LRP6) and anti-Myc (LGR5ΔC) antibodies. ...

To assess if truncating the C-terminal tail of LGR5 has any effect on receptor activity in the potentiation of Wnt/β-catenin signaling, we compared the mutant with wild-type LGR5 in the SuperTopFlash reporter enzyme assay. Interestingly, LGR5ΔC showed a significant increase in both the potency and maximum efficacy of RSPO1 compared to LGR5 (Fig. 8C). The increase was observed consistently in multiple experiments. Furthermore, in the absence of exogenous RSPO1, LGR5ΔC had an 8-fold and 4-fold increase in reporter activity compared to vector and LGR5, respectively (Fig. 8C, leftmost point of dose-response curve). The overall levels of active β-catenin in response to RSPO1 plus Wnt3a stimulation were also compared in HEK293T cells transfected with LGR5, LGR5ΔC, or vector alone. LGR5 or LGR5ΔC overexpressing cells exhibited both higher baseline and at least 2-fold-higher ligand-induced levels of β-catenin compared to vector control cells (Fig. 8D and E). Furthermore, the baseline levels of active β-catenin were augmented in LGR5ΔC compared to LGR5 cells and the maximal level was reached much faster (2 h in LGR5ΔC-transfected cells versus 6 h in LGR5-transfected cells). This additional increase of β-catenin levels in LGR5ΔC-transfected cells is consistent with the enhanced response of LGR5ΔC to RSPO1 in the reporter enzyme assay. Taken together with the data showing that LGR5ΔC has a reduced rate of internalization and that DN-DynK44A does not suppress LGR5 activity, these results indicate that internalization is not essential for RSPO1-induced potentiation of Wnt/β-catenin signaling.

DISCUSSION

Complex formation of LGR5 with Wnt coreceptors.

Recently, we and others reported that LGR4 and LGR5 function as high-affinity receptors of RSPOs to potentiate Wnt/β-catenin signaling (8, 11, 13). However, the precise mechanism of how activation of LGR4/5 leads to increased Wnt/β-catenin signaling remains a mystery. Unlike typical 7TM receptors, heterotrimeric G proteins and β-arrestin do not appear to be involved in LGR4/5 signaling. As part of the initial approach toward delineating this mechanism, our study demonstrates that LGR5 interacts with and promotes internalization of Wnt coreceptors LRP6 and Fzd5 in a ligand-dependent manner. While this work was in progress, de Lau et al. reported that LGR4/5 copurified with LRP5/6, Fzd5, Fzd6, Fzd7, and several other membrane-associated proteins in the absence of ligand stimulation using tandem affinity purification and mass spectrometric analysis (11), consistent with our findings here. Using immunofluorescence, FRET, and co-IP, we further demonstrated that LGR5 forms a complex with the Wnt coreceptors following stimulation with the cognate ligands. Since it is now well established that Wnt ligands bridge Fzd and LRP5/6 together to create a trimeric complex in order to trigger intracellular signaling (20), the question becomes whether RSPOs play a role similar to that of Wnt (i.e., bridging LGRs and the Wnt/LRP/Fzd complex). Interestingly, RSPOs were initially suggested to bind to LRP6 and/or Fzd8 to enhance Wnt/β-catenin signaling (30, 41), yet our group and others have failed to confirm a direct interaction between RSPO1 and LRP6 (5, 8, 11, 16). However, it remains possible that RSPOs directly bind to LGR4/5 and the Wnt-receptor complex to induce a supercomplex only in the presence of other extracellular factors that have not yet been identified. Alternatively, activation of LGR4/5 by RSPOs could potentially lead to the recruitment of distinct cytoplasmic signaling molecules to the receptor in a ligand-dependent manner. In turn, these cytoplasmic mediators could interact with the Wnt coreceptor complex either directly or indirectly through other scaffold proteins, such as DVL, and therefore create a supercomplex. We are actively investigating these and other potential mechanisms to understand the nature of the interactions that mediate the formation of complexes between RSPO/LGR and the Wnt coreceptors.

Endocytosis and degradation of LGR5 and the role of these processes in Wnt/β-catenin signaling.

We consistently found that LGR4 and LGR5 are rapidly internalized in the absence of exogenous RSPOs. This internalization is not due to binding of anti-tag antibodies, since similar intracellular receptor-containing vesicles were observed when the cells were permeabilized first and stained with anti-Myc tag antibodies directly (8). Using inhibitors of endocytosis, we show that LGR5 internalization/endocytosis requires dynamin and is clathrin dependent in HEK293 cells. Glinka et al. also found that clathrin is required for cointernalization of RSPO3 and LGR4 in NTERA2 cells (13). It has been reported that Wnt3a-dependent Fzd5 endocytosis is also clathrin mediated, whereas somewhat conflicting evidence has demonstrated both clathrin- and caveolin-associated endocytic mechanisms for LRP6 (22, 45). Internalization of Wnt ligands in murine L cells was also shown to be mediated by a clathrin- and dynamin-dependent pathway (6). Our results, taken together with these previous findings, suggest that association of LGR5 with the Wnt coreceptors forms a supercomplex which does not alter clathrin-mediated endocytosis of the Wnt signaling complex.

While there is consensus that Wnt coreceptors are internalized through a clathrin-dependent pathway following Wnt ligand-induced activation, the importance and overall mechanism of this endocytic process in Wnt/β-catenin signaling remain far from being elucidated (12). One of the major challenges is the lack of inhibitors that can specifically block endocytosis without general cytotoxicity or broadly affecting intracellular trafficking (12). Another is the lack of receptor mutants that are deficient in internalization but not in ligand binding (12). Here we found that deletion of the C-terminal tail of LGR5 resulted in a reduced rate of internalization, increased protein stability, and importantly, enhanced receptor activity in the potentiation of Wnt/β-catenin signaling. We demonstrated that expression of a dominant negative mutant of dynamin did not affect LGR5 activity but clearly blocked its internalization. Overall, these results indicate that although LGR5 forms a supercomplex with Wnt coreceptors which are rapidly internalized through a clathrin-dependent mechanism, the internalization process is not essential for Wnt/β-catenin signaling, at least in HEK293T cells. If anything, internalization accelerates receptor degradation and thus potentially serves as a negative feedback mechanism. Interestingly, Glinka et al. reported that in NTERA2 cells, clathrin is required for RSPO3-induced nuclear β-catenin translocation (13). This suggests that the role of endocytosis in Wnt/β-catenin signaling depends on the cellular context.

In conclusion, our findings suggest a mechanism by which RSPO1 activates LGR5 to potentiate β-catenin/Wnt signaling through enhancing interaction with members of the Wnt pathway, including LRP6 and Fzd5. LGR5 association accelerates the internalization, endocytosis, and degradation of these components. Removal of the C-terminal tail yields a receptor with an impaired internalization/degradation rate and increased Wnt signaling activity compared to the wild type. Overall, this study provides evidence that RSPO amplification of Wnt signaling is a result of the extent of LGR5-mediated association of Wnt components at the cell surface and is not completely dependent on internalization.

Supplementary Material

Supplemental material:

ACKNOWLEDGMENTS

We thank the laboratory of John Hancock for providing the DN-DynK44A construct.

This work was supported in part by grant RP100678 from the Cancer Prevention and Research Institute of Texas (CPRIT).

Footnotes

Published ahead of print 2 April 2012

Supplemental material for this article may be found at http://mcb.asm.org/.

REFERENCES

1. Allalou A, Wahlby C. 2009. BlobFinder, a tool for fluorescence microscopy image cytometry. Comput. Methods Programs Biomed. 94:58–65 [PubMed]
2. Barker N, Clevers H. 2010. Leucine-rich repeat-containing G-protein-coupled receptors as markers of adult stem cells. Gastroenterology 138:1681–1696 [PubMed]
3. Barker N, et al. 2009. Crypt stem cells as the cells-of-origin of intestinal cancer. Nature 457:608–611 [PubMed]
4. Bilic J, et al. 2007. Wnt induces LRP6 signalosomes and promotes dishevelled-dependent LRP6 phosphorylation. Science 316:1619–1622 [PubMed]
5. Binnerts ME, et al. 2007. R-Spondin1 regulates Wnt signaling by inhibiting internalization of LRP6. Proc. Natl. Acad. Sci. U. S. A. 104:14700–14705 [PMC free article] [PubMed]
6. Blitzer JT, Nusse R. 2006. A critical role for endocytosis in Wnt signaling. BMC Cell Biol. 7:28. [PMC free article] [PubMed]
7. Cadigan KM, Peifer M. 2009. Wnt signaling from development to disease: insights from model systems. Cold Spring Harb. Perspect. Biol. 1:a002881. [PMC free article] [PubMed]
8. Carmon KS, Gong X, Lin Q, Thomas A, Liu Q. 2011. R-spondins function as ligands of the orphan receptors LGR4 and LGR5 to regulate Wnt/beta-catenin signaling. Proc. Natl. Acad. Sci. U. S. A. 108:11452–11457 [PMC free article] [PubMed]
9. Carmon KS, Loose DS. 2010. Development of a bioassay for detection of Wnt-binding affinities for individual frizzled receptors. Anal. Biochem. 401:288–294 [PMC free article] [PubMed]
10. Carmon KS, Loose DS. 2008. Secreted frizzled-related protein 4 regulates two Wnt7a signaling pathways and inhibits proliferation in endometrial cancer cells. Mol. Cancer Res. 6:1017–1028 [PubMed]
11. de Lau W, et al. 2011. Lgr5 homologues associate with Wnt receptors and mediate R-spondin signalling. Nature 476:293–297 [PubMed]
12. Gagliardi M, Piddini E, Vincent JP. 2008. Endocytosis: a positive or a negative influence on Wnt signalling? Traffic 9:1–9 [PubMed]
13. Glinka A, et al. 2011. LGR4 and LGR5 are R-spondin receptors mediating Wnt/β-catenin and Wnt/PCP signalling. EMBO Rep. 12:1055–1061 [PMC free article] [PubMed]
14. Hsu SY, et al. 2000. The three subfamilies of leucine-rich repeat-containing G protein-coupled receptors (LGR): identification of LGR6 and LGR7 and the signaling mechanism for LGR7. Mol. Endocrinol. 14:1257–1271 [PubMed]
15. Hsu SY, Liang SG, Hsueh AJ. 1998. Characterization of two LGR genes homologous to gonadotropin and thyrotropin receptors with extracellular leucine-rich repeats and a G protein-coupled, seven-transmembrane region. Mol. Endocrinol. 12:1830–1845 [PubMed]
16. Kazanskaya O, et al. 2004. R-Spondin2 is a secreted activator of Wnt/beta-catenin signaling and is required for Xenopus myogenesis. Dev. Cell 7:525–534 [PubMed]
17. Kim KA, et al. 2008. R-Spondin family members regulate the Wnt pathway by a common mechanism. Mol. Biol. Cell 19:2588–2596 [PMC free article] [PubMed]
18. Kim KA, et al. 2006. R-Spondin proteins: a novel link to beta-catenin activation. Cell Cycle 5:23–26 [PubMed]
19. Lefkowitz RJ, Shenoy SK. 2005. Transduction of receptor signals by beta-arrestins. Science 308:512–517 [PubMed]
20. MacDonald BT, Tamai K, He X. 2009. Wnt/beta-catenin signaling: components, mechanisms, and diseases. Dev. Cell 17:9–26 [PMC free article] [PubMed]
21. Macia E, et al. 2006. Dynasore, a cell-permeable inhibitor of dynamin. Dev. Cell 10:839–850 [PubMed]
22. Mao B, et al. 2002. Kremen proteins are Dickkopf receptors that regulate Wnt/beta-catenin signalling. Nature 417:664–667 [PubMed]
23. Mazerbourg S, et al. 2004. Leucine-rich repeat-containing, G protein-coupled receptor 4 null mice exhibit intrauterine growth retardation associated with embryonic and perinatal lethality. Mol. Endocrinol. 18:2241–2254 [PubMed]
24. McClanahan T, et al. 2006. Identification of overexpression of orphan G protein-coupled receptor GPR49 in human colon and ovarian primary tumors. Cancer Biol. Ther. 5:419–426 [PubMed]
25. McDonald T, et al. 1998. Identification and cloning of an orphan G protein-coupled receptor of the glycoprotein hormone receptor subfamily. Biochem. Biophys. Res. Commun. 247:266–270 [PubMed]
26. Merlos-Suarez A, et al. 2011. The intestinal stem cell signature identifies colorectal cancer stem cells and predicts disease relapse. Cell Stem Cell 8:511–524 [PubMed]
27. Metcalfe C, Bienz M. 2011. Inhibition of GSK3 by Wnt signalling—two contrasting models. J. Cell Sci. 124:3537–3544 [PubMed]
28. Morita H, et al. 2004. Neonatal lethality of LGR5 null mice is associated with ankyloglossia and gastrointestinal distension. Mol. Cell. Biol. 24:9736–9743 [PMC free article] [PubMed]
29. Mustata RC, et al. 2011. Lgr4 is required for Paneth cell differentiation and maintenance of intestinal stem cells ex vivo. EMBO Rep. 12:558–564 [PMC free article] [PubMed]
30. Nam JS, Turcotte TJ, Smith PF, Choi S, Yoon JK. 2006. Mouse cristin/R-spondin family proteins are novel ligands for the Frizzled 8 and LRP6 receptors and activate beta-catenin-dependent gene expression. J. Biol. Chem. 281:13247–13257 [PubMed]
31. Niehrs C, Shen J. 2010. Regulation of Lrp6 phosphorylation. Cell. Mol. Life Sci. 67:2551–2562 [PubMed]
32. Nusse R, et al. 2008. Wnt signaling and stem cell control. Cold Spring Harb. Symp. Quant. Biol. 73:59–66 [PubMed]
33. Park JS, et al. 2006. High mobility group box 1 protein interacts with multiple Toll-like receptors. Am. J. Physiol. Cell Physiol. 290:C917–C924 [PubMed]
34. Reya T, Clevers H. 2005. Wnt signalling in stem cells and cancer. Nature 434:843–850 [PubMed]
35. Rothberg KG, et al. 1992. Caveolin, a protein component of caveolae membrane coats. Cell 68:673–682 [PubMed]
36. Sato T, et al. 2009. Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459:262–265 [PubMed]
37. Sorkin A, von Zastrow M. 2009. Endocytosis and signalling: intertwining molecular networks. Nat. Rev. Mol. Cell Biol. 10:609–622 [PMC free article] [PubMed]
38. Tanese K, et al. 2008. G-protein-coupled receptor GPR49 is up-regulated in basal cell carcinoma and promotes cell proliferation and tumor formation. Am. J. Pathol. 173:835–843 [PMC free article] [PubMed]
39. Theodorou V, et al. 2007. MMTV insertional mutagenesis identifies genes, gene families and pathways involved in mammary cancer. Nat. Genet. 39:759–769 [PubMed]
40. Uchida H, et al. 2010. Overexpression of leucine-rich repeat-containing G protein-coupled receptor 5 in colorectal cancer. Cancer Sci. 101:1731–1737 [PubMed]
41. Wei Q, et al. 2007. R-spondin1 is a high affinity ligand for LRP6 and induces LRP6 phosphorylation and beta-catenin signaling. J. Biol. Chem. 282:15903–15911 [PubMed]
42. Willert K, et al. 2003. Wnt proteins are lipid-modified and can act as stem cell growth factors. Nature 423:448–452 [PubMed]
43. Wolfe BL, Trejo J. 2007. Clathrin-dependent mechanisms of G protein-coupled receptor endocytosis. Traffic 8:462–470 [PubMed]
44. Wu P, Brand L. 1994. Resonance energy transfer: methods and applications. Anal. Biochem. 218:1–13 [PubMed]
45. Yamamoto H, Komekado H, Kikuchi A. 2006. Caveolin is necessary for Wnt-3a-dependent internalization of LRP6 and accumulation of beta-catenin. Dev. Cell 11:213–223 [PubMed]
46. Yoon JK, Lee JS. 2012. Cellular signaling and biological functions of R-spondins. Cell. Signal. 24:369–377 [PMC free article] [PubMed]
47. Zeng X, et al. 2005. A dual-kinase mechanism for Wnt co-receptor phosphorylation and activation. Nature 438:873–877 [PMC free article] [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...