Wnt/β-catenin signaling plays an essential role in colon carcinogenesis. Galectin-3, a β-galactoside–binding protein, has been implicated in Wnt signaling, but the precise mechanisms by which galectin-3 modulates the Wnt pathway are unknown. In the present study, we determined the effects of galectin-3 on the Wnt/β-catenin pathway in colon cancer cells, as well as the mechanisms involved. Galectin-3 levels were manipulated in human colon cancer cells by stable transfection of galectin-3 antisense, short hairpin RNA, or full-length galectin-3 cDNA, and effects on β-catenin levels, subcellular distribution, and Wnt signaling were determined. Galectin-3 levels correlated with β-catenin levels in a variety of colon cancer cell lines. Down-regulation of galectin-3 resulted in decreased β-catenin protein levels but no change in β-catenin mRNA levels, suggesting that galectin-3 modulates β-catenin by another mechanism. Reduction of galectin-3 led to reduced nuclear β-catenin with a concomitant decrease in TCF4 transcriptional activity and expression of its target genes. Conversely, transfection of galectin-3 cDNA into colon cancer cells increased β-catenin expression and TCF4 transcriptional activity. Down-regulation of galectin-3 resulted in AKT and glycogen synthase kinase-3β (GSK-3β) dephosphorylation and increased GSK activity, increasing β-catenin phosphorylation and degradation. Ly294002, an inhibitor of phosphatidylinositol 3-kinase, and dominant-negative AKT, suppressed TCF4 transcriptional activity induced by galectin-3 whereas LiCl, a GSK-3β inhibitor, increased TCF4 activity, mimicking the effects of galectin-3. These results suggest that galectin-3 mediates Wnt signaling, at least in part, by regulating GSK-3β phosphorylation and activity via the phosphatidylinositol 3-kinase/AKT pathway, and, thus, the degradation of β-catenin in colon cancer cells. [Cancer Res 2009;69(4):1343–9]

Wnt/β-catenin signaling plays an essential role in both gastrointestinal development and carcinogenesis. More than 90% of colorectal cancers have an activating mutation in the Wnt signaling pathway [80% in the adenomatous polyposis coli (APC) gene and 10% to 15% in the β-catenin gene; ref. 1]. One of the hallmarks of activated Wnt signaling is the accumulation of nuclear β-catenin (2), a downstream component of the Wnt signaling pathway. β-Catenin is phosphorylated by CK1α and glycogen synthase kinase-3β (GSK-3β) in a complex that contains APC and axin, targeting β-catenin for ubiquitination and degradation (3). Mutations in genes encoding APC, axin, or β-catenin prevent β-catenin phosphorylation by GSK-3β, which in turn leads to β-catenin accumulation in the nucleus resulting in activation of the transcription of Wnt-target genes including cyclin D1 and c-Myc (4). Among the components of the β-catenin destruction complex, GSK-3β is one of the key molecules that mediates either β-catenin degradation or accumulation in the nucleus. It is known that when a Wnt signal is present, GSK-3β activity is inhibited (5). However, the mechanisms underlying the regulation of GSK-3β are poorly understood. It has been reported that inactivation of GSK-3β by oncogenic processes dependent on the phosphatidylinositol 3-kinase (PI3K) pathway leads to β-catenin dephosphorylation and its accumulation and translocation into the nucleus (6). Some reports suggest that inhibition of GSK-3β activity is mediated through phosphorylation at Ser9, indicating a link between GSK-3β phosphorylation and the downstream activation of Wnt target genes (6, 7). Nevertheless, modulation of GSK-3β activity may occur by other mechanisms for cancer cell survival and proliferation. Although inactivation of APC by mutation or deletion is thought to be the “gatekeeper” event in colon carcinogenesis, additional genetic or epigenetic events cooperate to lead to tumor formation and progression in the colon and other organs. Galectin-3 is one of the molecules which may contribute to this process.

Galectin-3 is a member of a family of β-galactoside–specific lectins found in many species and cell types. Galectin-3 exhibits pleiotropic biological functions extracellularly, by interacting with cell surface and extracellular matrix glycoproteins and glycolipids, and intracellularly, by interacting with cytoplasmic and nuclear proteins to modulate signaling pathways (8). Substantial evidence indicates that galectin-3 contributes to neoplastic transformation, tumor cell survival, angiogenesis, and tumor metastasis (911). Inhibition of galectin-3 in breast carcinoma and thyroid papillary carcinoma cells leads to loss of their characteristic transformed phenotypes in cell culture (12). Conversely, the introduction of galectin-3 cDNA induces a transformed phenotype in normal thyroid follicular cell lines and causes human lymphoma Jurkat T cells to grow faster (12). However, the mechanisms by which galectin-3 are involved in cell transformation and proliferation are not yet fully understood. Kloog and colleagues found that galectin-3 promotes the activation of RAF1 and PI3K and contributes to the selective activation of signaling cascades and regulation of gene expression at the transcriptional level (13). Recent studies have shown that galectin-3 binds to β-catenin and TCF4 and activates Wnt signaling target gene cyclin D1 and c-Myc in breast cancer cells (14, 15). More recently, Shi and colleagues (16) showed that inhibition of both Wnt-2 and galectin-3 had synergistic effects on suppressing Wnt signaling and inducing apoptosis. However, whether galectin-3 modulates β-catenin expression and the mechanisms by which galectin-3 might activate Wnt signaling in colon cancer cells remains unclear.

The aim of the present study was to determine whether galectin-3 modulates Wnt signaling in colon cancer cells, and if so, by what mechanisms. We found that galectin-3 regulates β-catenin protein levels and its nuclear accumulation via a posttranslational mechanism. This results in an increase in TCF4 transcriptional activity and up-regulation of Wnt target gene expression. We have shown, for the first time, that galectin-3 modulates β-catenin levels and Wnt signaling by regulating GSK-3β phosphorylation/activity through the PI3K/AKT pathway.

Cells and reagents. LS174T and LiM6 human colon cancer cell lines have been characterized and described previously (17). Other human colorectal adenocarcinoma cell lines (HT29, SW480, Caco2, HCT116, and RKO) were purchased from American Type Culture Collection. Cell lines were maintained at 37°C in a 5% CO2 atmosphere in DMEM containing 10% heat-inactivated fetal bovine serum with penicillin and streptomycin. Anti–galectin-3 antibody was obtained as described previously (17). Anti–β-catenin and phosphorylated β-catenin (Ser33, Ser37, and Thr41) antibodies, anti-AKT and phosphorylated AKT antibodies, and anti–GSK-3β and phosphorylated GSK-3β at Ser9 antibodies were from Cell Signaling Technology, Inc. Anti–cyclin D1 antibody was from Santa Cruz Biotechnology, Inc.

Production of derivative cell lines. MC1 and M22 were generated by stable transfection of LiM6 colon cancer cells with a control vector and a galectin-3 antisense plasmid, respectively, as previously described (17).

GV3, GB2, and E2 are stable derivatives of the LiM6 cell line and were infected with a retrovirus containing galectin-3 short hairpin RNA (shRNA) in pSilencer 5.1-U6 Retro Vector (Ambion). GB2 and E2 cells contain galectin-3 shRNA in pSilencer 5.1-U6 Retro Vector. GV3 cells were used as control cells and contained a scrambled shRNA in pSilencer 5.1-U6 Retro Vector. These cells were selected and maintained in puromycin (10 μg/mL).

AG1, a derivative of colon cancer cell line HM7 containing the antisense galectin-3 sequence under control of an inducible promoter was generated using a tetracycline-inducible system (pTet-On; Clontech), as previously described (17). Cells were incubated with doxycycline (1 μg/mL) for 48 h to induce galectin-3 antisense.

RKO-v and RKO-gl3 are stable derivatives of the RKO cell line and were generated by stable transfection of galectin-3 cDNA in pIRES2-AcGFP1 (Clontech). These cells were selected and maintained in G418 (300 μg/mL).

Protein isolation and immunoblot analysis. Total cell lysates were prepared in 2% SDS lysis buffer containing 330 mmol/L Tris-HCl (pH 8.8), 2% SDS, 10% glycerol, and one mini tablet protease inhibitor cocktail (Roche Diagnostics, Corp.). Cytoplasmic and nuclear extractions were prepared using NE-PER Nuclear and Cytoplasmic Extraction Reagents (Pierce Biotechnology) according to the manufacturer's instructions. Equal amounts of protein were subjected to electrophoresis on 10% Tris-glycine gels. Western blot analyses were performed as previously described (17), and immunoreactive bands were visualized by chemiluminescence detection.

Immunocytochemistry. Cells were harvested for immunocytochemical staining; cytospin slides were prepared using a Cytospin cytological centrifuge. Galectin-3 and β-catenin were detected as described previously (17).

Indirect immunofluorescence staining and confocal laser microscopy. Cells were seeded onto two-chamber slides, fixed, permeabilized, and blocked in flow cytometry buffers (Santa Cruz Biotechnology) following the manufacturer's instructions. Cells were incubated with anti–galectin-3 rat antibody and anti–β-catenin rabbit antibody at 1:100 dilutions for 1 h at room temperature. Secondary antibodies, Cy3-conjugated goat anti-rat and FITC-conjugated goat anti-rabbit, were added at 1:100 dilutions, and the cells were then incubated for another 30 min. Nuclei were stained with ProLong Gold antifade reagent with 4′,6-diamidino-2-phenylindole (DAPI; Invitrogen). Expression and localization of the proteins were observed under a confocal microscope system (FluoView FV500; Olympus) and analyzed by CellQuest PRO software (BD Biosciences) at the Flow Cytometry and Image Analysis Core Laboratory at The University of Texas M. D. Anderson Cancer Center.

Plasmids and transient transfection and luciferase reporter assays. SuperTop TCF4 luciferase reporter plasmid under the control of eight TCF4 consensus and pFOPFLASH plasmids were provided by Dr. C. Liu (The University of Texas Medical Branch, Galveston, TX). The cDNA plasmid for dominant-negative AKT mutant (AKT AAA) was provided by Dr. D. Koul (M. D. Anderson Cancer Center, Houston, TX; ref. 18). Galectin-3 full-length cDNA expression vector pCNC10gal3 and control vector pCNC10 were provided by Dr. A. Raz (Wayne State University Medical School, Detroit, MI). β-Galactoside expression vector pCH110 (Amersham-Pharmacia) was used to normalize for transfection efficiency.

To examine the effects of galectin-3 on TCF4 transcriptional activity, cells with different levels of galectin-3 in six-well plates were transiently transfected with 1 μg of SuperTop TCF4 reporter and pFOPFLASH plasmids and 0.2 μg of pCH110 mixed with 3 μL of FuGENE 6 (Roche Diagnostics) according to the manufacturer's protocol. After 48 h, cells were harvested for determination of luciferase activity, which was measured in relative light units (normalized to β-galactosidase) using a TD-20/20 luminometer (Turner Designs) and the Promega luciferase assay system. The values reported here represent the mean and SD of at least three independent experiments.

Real-time PCR. To quantify the change in the β-catenin mRNA level, real-time reverse transcription-PCR was performed on the ABI Prism 7900 (Applied Biosystems) using the commercially available gene expression assay for β-catenin CTNNB1 (Hs00170025_m1), and the cyclophilin A Vic-labeled predeveloped assay reagent (4326316E; Applied Biosystems). A 25 μL final reaction volume containing 1× TaqMan Universal PCR master mix (Applied Biosystems), 1× Multiscribe with RNase inhibitors, and 1× gene expression assay to amplify ∼50 ng of total RNA with the following cycling conditions: 30 min at 48°C, 10 min at 95°C, and 50 cycles at 95°C for 15 s and 60°C for 1 min. The 7900 Sequence Detection System 2.2 software (Applied Biosystems) automatically determined the fold change for CTNNB1 in each sample by using the δδCt method with 95% confidence.

In vitro kinase assay. Endogenous GSK-3β was immunoprecipitated by 2 μg of rabbit anti–GSK-3β antibody from 1,000 μg of galectin-3 varied cell lysates, then specific GSK-3β substrate peptide (Calbiochem) was used to perform in vitro kinase assay, the kinase activity was read in scintillation counter (cpm). Purified GSK-3β (0.2 μg; Upstate Biotechnology) was used as a positive kinase control, and negative GSK-3β substrate peptide (Calbiochem) was also used to detect background phosphorylation (GSK-3β autophosphorylation). The data indicated in the figure was presented after the background phosphorylation was subtracted.

Statistical analysis. Promoter assays are presented in figures as mean ± SD and represent the results of at least three experiments. The significance of differences between the groups was judged using a two-tailed Student's t test. Results were considered statistically significant at P < 0.05.

Down-regulation of galectin-3 reduces levels of β-catenin and cyclin D1. It was recently reported that galectin-3 plays a role in Wnt signaling in breast cancer cells by interacting with β-catenin (14, 15). We therefore examined whether galectin-3 modulates β-catenin levels and Wnt signaling in human colon cancer cells. We first examined the levels of galectin-3 and β-catenin in a variety of colon cancer cells by immunoblotting and determined that β-catenin levels correlated with levels of galectin-3 (Fig. 1A). To determine whether galectin-3 mediates β-catenin expression in colon cancer cells, we manipulated galectin-3 levels by stable transfection of galectin-3 antisense or retrovirus galectin-3 shRNA, and examined the expression of β-catenin and its target gene, cyclin D1. Immunoblot analysis of total cell lysates showed that β-catenin levels were dramatically lower in M22 galectin-3 antisense cells than that in MC1 control cells containing vector only, and that this correlated with galectin-3 levels in the cells. Similar findings were observed in E2 galectin-3 shRNA knockdown cells compared with GV3 control cells containing a scrambled shRNA sequence. In line with these findings, the reduction of galectin-3 was also associated with a decrease in the expression of cyclin D1 in both M22 and E2 cells relative to the control MC1 and GV3 cells, respectively (Fig. 1B). Immunocytochemical staining of MC1 and M22 cells confirmed that a decrease in galectin-3 expression in M22 antisense cells was accompanied by an overall reduction of β-catenin expression (Fig. 1C). Similar patterns of β-catenin expression were found in E2 galectin-3 shRNA knockdown cells compared with GV3 scrambled control cells (Fig. 1D). To further confirm regulation of β-catenin levels by galectin-3, we examined β-catenin and cyclin D1 levels in newly established stably transfected galectin-3 cDNA sense cells (RKO-gl3) and vector-transfected control (RKO-v) cells. Immunoblot analysis of total cell lysates showed that β-catenin and cyclin D1 levels were higher in RKO-gl3 galectin-3 sense cells than that in RKO-v control cells and correlated with galectin-3 levels in the cells (Fig. 2A). These findings suggest that stable alterations of galectin-3 are associated with corresponding changes in β-catenin protein levels.

Figure 1.

Down-regulation of galectin-3 reduces β-catenin levels in human colon cancer cells. A, the expression of galectin-3 and β-catenin in a variety of colon cancer cell lines was determined by immunoblotting. B, protein levels of galectin-3, β-catenin, and cyclin D1 were assessed by Western blotting in paired MC1 (vector control) and M22 (galectin-3 antisense) cells and paired GV3 (control) and E2 (galectin-3 shRNA) cells. C and D, MC1/M22 and GV3/E2 cells were prepared in cytospin slides and stained for galectin-3 and β-catenin as described in Materials and Methods.

Figure 1.

Down-regulation of galectin-3 reduces β-catenin levels in human colon cancer cells. A, the expression of galectin-3 and β-catenin in a variety of colon cancer cell lines was determined by immunoblotting. B, protein levels of galectin-3, β-catenin, and cyclin D1 were assessed by Western blotting in paired MC1 (vector control) and M22 (galectin-3 antisense) cells and paired GV3 (control) and E2 (galectin-3 shRNA) cells. C and D, MC1/M22 and GV3/E2 cells were prepared in cytospin slides and stained for galectin-3 and β-catenin as described in Materials and Methods.

Close modal
Figure 2.

Up-regulation of galectin-3 increases β-catenin level and TCF4 activity in human colon cancer cells. A, protein levels of galectin-3, β-catenin, and cyclin D1 were assessed by Western blotting in RKO cells stably transfected with galectin-3 sense (RKO-gl3) or a vector control plasmid (RKO-v). B, RKO-gl3 and RKO-v cells were transfected with 1 μg of SuperTop TCF4 reporter and 0.2 μg of pCH110. Luciferase reporter activity was measured after 48 h. Columns, mean of at least triplicate assays for all experiments; bars, SD (*, P < 0.005).

Figure 2.

Up-regulation of galectin-3 increases β-catenin level and TCF4 activity in human colon cancer cells. A, protein levels of galectin-3, β-catenin, and cyclin D1 were assessed by Western blotting in RKO cells stably transfected with galectin-3 sense (RKO-gl3) or a vector control plasmid (RKO-v). B, RKO-gl3 and RKO-v cells were transfected with 1 μg of SuperTop TCF4 reporter and 0.2 μg of pCH110. Luciferase reporter activity was measured after 48 h. Columns, mean of at least triplicate assays for all experiments; bars, SD (*, P < 0.005).

Close modal

Galectin-3 mediates nuclear accumulation of β-catenin. One of the hallmarks of activated Wnt signaling is the accumulation of nuclear β-catenin (2). The subcellular distribution of β-catenin in MC1 and M22 cells was examined by Western blot analysis of cytoplasmic and nuclear subcellular fractions and confocal immunofluorescence staining (Fig. 3). β-Catenin was present in the nucleus of MC1 cells containing high galectin-3 levels but was undetectable in the nuclear compartment of M22 galectin-3 antisense cells (Fig. 3A). This result was confirmed by immunofluorescence analysis (Fig. 3B). Confocal microscopy showed that in MC1 cells with high galectin-3 levels, both galectin-3 and β-catenin proteins were distributed in the cytoplasm and the nucleus, whereas in M22 galectin-3 antisense cells, β-catenin was localized to the cytoplasm and membrane (Fig. 3B). This finding is consistent with the results of Shimura and colleagues (12) in human breast cancer cells. Similar β-catenin subcellular distribution patterns were also found in E2 galectin-3 shRNA knockdown cells compared with GV3 control cells (data not shown).

Figure 3.

Subcellular localization of β-catenin in galectin-3 antisense M22 cells and vector control MC1 cells. A, cytoplasmic and nuclear fractions of colon cancer cells were subjected to SDS-PAGE and then immunoblotted with anti–β-catenin, β-actin, and histone-1. β-Actin was used as a cytoplasmic protein loading control, and histone-1 was used for nuclear protein loading control. B, indirect immunofluorescence was performed on MC1 and M22 cells using anti–galectin-3 antibody (TIB166 1:100; red) and anti–β-catenin antibody (1:100, green), followed by DAPI nuclear counterstaining (blue). The merge of galectin-3 (red) and β-catenin (green) with DAPI (blue) is also shown.

Figure 3.

Subcellular localization of β-catenin in galectin-3 antisense M22 cells and vector control MC1 cells. A, cytoplasmic and nuclear fractions of colon cancer cells were subjected to SDS-PAGE and then immunoblotted with anti–β-catenin, β-actin, and histone-1. β-Actin was used as a cytoplasmic protein loading control, and histone-1 was used for nuclear protein loading control. B, indirect immunofluorescence was performed on MC1 and M22 cells using anti–galectin-3 antibody (TIB166 1:100; red) and anti–β-catenin antibody (1:100, green), followed by DAPI nuclear counterstaining (blue). The merge of galectin-3 (red) and β-catenin (green) with DAPI (blue) is also shown.

Close modal

Galectin-3 increases TCF4 transcriptional activity. To determine whether galectin-3 affects β-catenin and TCF signaling, the SuperTop TCF4 luciferase reporter and pFOPFLASH plasmid were transfected into cell lines in which the levels of galectin-3 were manipulated. TCF4 activity in M22 galectin-3 antisense cells was reduced more than 10-fold compared with that in MC1 control cells (Fig. 4A,, left). A similar reduction in TCF4 activity occurred in AG1 doxycycline-inducible galectin-3 antisense cells treated with doxycycline (1 μg/mL) for 48 hours (Fig. 4A,, right). To further confirm that down-regulation of galectin-3 leads to reduced TCF4 activity, we transfected TCF4 luciferase reporter into the newly established retrovirus galectin-3 shRNA GB2 and E2 cells and their control GV3 cells. Knockdown of galectin-3 resulted in reduced TCF4 activity in GB2 and E2 cells relative to that in GV3 cells (Fig. 4B). In contrast, up-regulation of galectin-3 by cotransfection of a galectin-3 full-length cDNA expression vector (pCNC10gal-3) with the TCF4 reporter plasmid led to greatly increased TCF4 activity in LS174T and LiM6 colon cancer cells (Fig. 4C). This is further confirmed in RKO-v and RKO-gl3 cells (Fig. 2B). Our results indicate that manipulation of galectin-3 expression in colon cancer cells leads to a parallel change in TCF4 activity, further suggesting that galectin-3 is a positive regulator of Wnt signaling.

Figure 4.

Alterations in galectin-3 levels lead to corresponding changes in TCF4 transcriptional activity. A, MC1 and M22 cells were cotransfected with 1 μg of SuperTop TCF4 reporter and 0.2 μg of pCH110. Luciferase reporter activity was measured after 48 h (left). AG1 cells containing galectin-3 antisense under control of a tetracycline-inducible promoter were cotransfected with 1 μg of SuperTop TCF4 reporter and 0.2 μg of pCH110 and treated with or without doxycycline (1 μg/mL) for 48 h (right). B, cell lines GB2 and E2 (containing galectin-3 shRNA) and vector-transfected GV3 control cells were cotransfected with 1 μg of SuperTop TCF4 reporter and 0.2 μg of pCH110. Luciferase reporter activity was measured after 48 h. C, LS174T and LiM6 cells were cotransfected with 1 μg of pCNC10gal-3 vector or control vector pCNC10 and 1 μg of SuperTop TCF4 reporter and 0.2 μg of pCH110. Luciferase reporter activity was measured after 48 h. Columns, mean of at least triplicate assays for all transfection experiments, experiments were repeated at least twice; bars, SD (*, P < 0.001; **, P < 0.05).

Figure 4.

Alterations in galectin-3 levels lead to corresponding changes in TCF4 transcriptional activity. A, MC1 and M22 cells were cotransfected with 1 μg of SuperTop TCF4 reporter and 0.2 μg of pCH110. Luciferase reporter activity was measured after 48 h (left). AG1 cells containing galectin-3 antisense under control of a tetracycline-inducible promoter were cotransfected with 1 μg of SuperTop TCF4 reporter and 0.2 μg of pCH110 and treated with or without doxycycline (1 μg/mL) for 48 h (right). B, cell lines GB2 and E2 (containing galectin-3 shRNA) and vector-transfected GV3 control cells were cotransfected with 1 μg of SuperTop TCF4 reporter and 0.2 μg of pCH110. Luciferase reporter activity was measured after 48 h. C, LS174T and LiM6 cells were cotransfected with 1 μg of pCNC10gal-3 vector or control vector pCNC10 and 1 μg of SuperTop TCF4 reporter and 0.2 μg of pCH110. Luciferase reporter activity was measured after 48 h. Columns, mean of at least triplicate assays for all transfection experiments, experiments were repeated at least twice; bars, SD (*, P < 0.001; **, P < 0.05).

Close modal

Galectin-3 increases β-catenin expression and activates Wnt signaling by regulating GSK-3β activity via the PI3K/AKT pathway. In order to determine whether galectin-3 modulates β-catenin at the transcription level, quantitative real-time PCR was used to examine β-catenin mRNA levels in paired MC1/M22 cells and GV3/E2 cells. No difference in mRNA levels were detected between these cell lines (data not shown), suggesting that galectin-3 may modulate β-catenin expression at the posttranslational level perhaps by affecting its degradation. We therefore examined whether galectin-3 stabilizes β-catenin and reduces its degradation by affecting GSK-3β activity or phosphorylation. Results in Fig. 5A (left) show that decreased galectin-3 in M22 antisense cells is associated with decreased phosphorylation of AKT and GSK-3β (at Ser9), and a decrease in total β-catenin, whereas total GSK-3β and AKT levels were unchanged. In contrast, increased galectin-3 in RKO-gl3 cells led to increased phosphorylation of AKT and GSK-3β (at Ser9) and an increase in β-catenin levels (Fig. 5A,, right). To confirm that galectin-3 alters GSK-3β activity, we measured GSK-3β activity in MC1 control cells and M22 antisense cells using an in vitro kinase assay (19). GSK-3β activity was 2.0-fold higher in M22 antisense cells compared with MC1 control cells, corresponding to differences in GSK-3β phosphorylation (inactivated phosphorylation; Fig. 5B). Similarly, GSK-3β activity was increased in E2 galectin-3 shRNA cells compared with control GV3 cells (data not shown). Our data suggest that reduction in galectin-3 decreases GSK-3β phosphorylation and increases GSK-3β activity, which may mediate β-catenin accumulation and Wnt signaling in colon cancer cells.

Figure 5.

Galectin-3 regulates GSK-3β activity/phosphorylation via the PI3K/AKT pathway. A, total cell lysates of paired MC1/M22 cells and RKO-v/RKO-gl3 cells were prepared and immunoblots were performed for galectin-3, phosphorylated AKT and total AKT, phosphorylated GSK-3β and total GSK-3β, and total β-catenin and β-actin expression levels. B, in vitro kinase assay. Endogenous GSK-3β was immunoprecipitated by 2 μg of rabbit anti–GSK-3β antibody from 1,000 μg MC1 and M22 cell lysates, then specific GSK-3β substrate peptide (Calbiochem) was used to perform an in vitro kinase assay. Kinase activity was determined by scintillation counting. Purified GSK-3β (0.2 μg; Upstate Biotechnology) was used as a positive kinase control, and negative GSK-3β substrate peptide (Calbiochem) was used to detect background phosphorylation (GSK-3β autophosphorylation). Columns, mean activity after subtraction of background phosphorylation (*, P < 0.005). C, human colon cancer cells (LiM6 or LS174T) were transfected with 1 μg of SuperTop TCF4 reporter alone (columns 1 and 3), or cotransfected with 1 μg of pCNC10 gal-3 vector (columns 2, 4, and 6), cotransfected with 1 μg of dominant-negative AKT construct (AKT AAA; columns 5 and 6) for 48 h. Cells were then treated with Ly294002 5 μmol/L for an additional 24 h (columns 3 and 4) and luciferase reporter activity was measured (*, P < 0.05, **, P < 0.01). D, cells were cotransfected with 1 μg of SuperTop TCF4 reporter and 1 μg of pCNC10gal-3 vector and then treated with 20 mmol/L of LiCl for 24 h. Luciferase reporter activity was measured after 48 h. Columns, mean of at least triplicate assays for all experiments; bars, SD (*, P < 0.005).

Figure 5.

Galectin-3 regulates GSK-3β activity/phosphorylation via the PI3K/AKT pathway. A, total cell lysates of paired MC1/M22 cells and RKO-v/RKO-gl3 cells were prepared and immunoblots were performed for galectin-3, phosphorylated AKT and total AKT, phosphorylated GSK-3β and total GSK-3β, and total β-catenin and β-actin expression levels. B, in vitro kinase assay. Endogenous GSK-3β was immunoprecipitated by 2 μg of rabbit anti–GSK-3β antibody from 1,000 μg MC1 and M22 cell lysates, then specific GSK-3β substrate peptide (Calbiochem) was used to perform an in vitro kinase assay. Kinase activity was determined by scintillation counting. Purified GSK-3β (0.2 μg; Upstate Biotechnology) was used as a positive kinase control, and negative GSK-3β substrate peptide (Calbiochem) was used to detect background phosphorylation (GSK-3β autophosphorylation). Columns, mean activity after subtraction of background phosphorylation (*, P < 0.005). C, human colon cancer cells (LiM6 or LS174T) were transfected with 1 μg of SuperTop TCF4 reporter alone (columns 1 and 3), or cotransfected with 1 μg of pCNC10 gal-3 vector (columns 2, 4, and 6), cotransfected with 1 μg of dominant-negative AKT construct (AKT AAA; columns 5 and 6) for 48 h. Cells were then treated with Ly294002 5 μmol/L for an additional 24 h (columns 3 and 4) and luciferase reporter activity was measured (*, P < 0.05, **, P < 0.01). D, cells were cotransfected with 1 μg of SuperTop TCF4 reporter and 1 μg of pCNC10gal-3 vector and then treated with 20 mmol/L of LiCl for 24 h. Luciferase reporter activity was measured after 48 h. Columns, mean of at least triplicate assays for all experiments; bars, SD (*, P < 0.005).

Close modal

To confirm that the PI3K/AKT pathway is necessary for galectin-3–mediated Wnt signaling, both a pharmacologic inhibitor (Ly294002) and a genetic dominant-negative AKT construct (AKT AAA) were used in the TCF4 transcriptional activity assays. Cotransfection of galectin-3 cDNA and the SuperTop TCF4 reporter increased TCF4 transcriptional activity by approximately 7-fold (P < 0.005) compared with that of the control. Ly294002 (5 μmol/L) blocked galectin-3–induced stimulation of TCF4 activity by 2.6-fold (P < 0.01). Dominant-negative AKT (AKT AAA) blocked galectin-3–mediated TCF4 transcriptional activity by 18-fold (P < 0.005; Fig. 5C). The viability of the cells was not affected by treatment for 24 hours with 5 μmol/L of Ly294002 and cotransfection with AKT AAA for 48 hours. These data suggest that the PI3K/AKT pathway is necessary for galectin-3–mediated Wnt signaling. In order to further confirm that GSK-3β downstream of PI3K/AKT is involved in galectin-3–mediated Wnt signaling, we cotransfected the galectin-3 cDNA and SuperTop TCF4 reporter and treated these with the GSK-3β inhibitor LiCl at 20 mmol/L for 24 hours. Galectin-3 and LiCl alone or in combination increased TCF4 transcriptional activity by 7-fold to 8-fold (Fig. 5D). The viability of the cells was not affected by treatment for 24 hours with LiCl at 20 mmol/L for 24 hours. These data indicate that galectin-3 activates Wnt signaling by inactivating GSK-3β.

Galectin-3, a β-galactoside–binding protein, exhibits multiple biological and pathologic functions including effects on cell growth, apoptosis, immune response, malignant transformation, and metastasis (20). In particular, galectin-3 plays a pivotal role in the regulation of cancer-related gene expression, including cyclin D1, TTF-1, and MUC2 (17, 21, 22). Recent studies have suggested that galectin-3 activates Wnt signaling in human breast cancer cells (14, 15), but the precise mechanisms involved remain unclear. Wnt signaling plays a key role in colon carcinogenesis, but the role of galectin-3 in this pathway in the colon has not been explored. In the current study, we showed that galectin-3 up-regulates β-catenin expression and its nuclear accumulation and augments Wnt signaling in human colon cancer cells by regulating GSK-3β phosphorylation/activity through the PI3K/AKT pathway.

The β-catenin/Wnt pathway plays a key role in development, tissue homeostasis, and cancer susceptibility. Disregulation of β-catenin and other Wnt components leads to nuclear localization of β-catenin, activation of Wnt target genes, and tumor formation (1, 2). In the canonical Wnt pathway, extracellular Wnt proteins bind to and activate Frizzled receptors, which in turn, inhibits phosphorylation of β-catenin by disrupting a complex consisting of the APC, axin, and GSK-3β proteins. Unphosphorylated β-catenin then binds to TCF/Lef family transcription factors and activates their specific target genes including c-Myc, cyclin D1, cyclooxygenase-2, matrix metalloproteinase-7, gastrin, and ITF-2 (2328). Activating mutations in Wnt pathway components, including loss-of-function mutations of APC or less frequently in CTNNB1 (which encodes β-catenin) and AXIN, increase β-catenin protein levels and have been found in numerous human cancers including colorectal, gastric, and ovarian cancer (2). However, mutations of Wnt pathway proteins are not the only factors that contribute to β-catenin activation (29). In the present study, we show that galectin-3, a member of the β-galactoside–binding protein family, up-regulates β-catenin expression, facilitates its nuclear accumulation, and augments Wnt signaling by increasing TCF4 transcriptional activity and its target gene cyclin D1 expression in colon cancer cells.

Although recent data suggests that galectin-3 can mediate Wnt signaling in breast cancer cells, the mechanism by which galectin-3 increases β-catenin expression and activates Wnt signaling is still unclear. Shimura and colleagues (14) found that galectin-3 binds to the β-catenin/TCF complex, colocalizes with β-catenin in the nucleus, and induces transcriptional activity of TCF4 in breast cancer cells. We were unable to show a similar direct interaction between galectin-3 and β-catenin or TCF4 by coimmunoprecipitation in the colon cancer cells we examined (data not shown). We found that in colon cancer cells, galectin-3 mediates β-catenin expression and TCF4 activity by regulation of GSK-3β phosphorylation and activity via the PI3K/AKT pathway. We showed that down-regulation of galectin-3 leads to reduced levels of phosphorylated AKT and phosphorylated GSK-3β at Ser9, and increased GSK-3β activity leading to phosphorylation of β-catenin at sites which are critical for β-catenin recognition by F-box protein β-Trcp, and thus, for its degradation (30).

It has been shown that in response to certain growth stimuli (29, 3133), PI3K-activated AKT can phosphorylate GSK-3β at Ser9, leading to inactivation of GSK-3β and augmentation of β-catenin-TCF4 transcriptional activity (32). Our data suggest that galectin-3 may act in a similar manner. Down-regulation of galectin-3 decreased AKT phosphorylation, and the PI3K inhibitor LY294002 blocked GSK-3β phosphorylation and TCF4 transcriptional activity induced by galectin-3 in colon cancer cells. Furthermore, a dominant-negative AKT construct (AKT AAA) completely suppressed the galectin-3–mediated augmentation in TCF4 activity. Our results suggest that galectin-3 regulates GSK-3β by mediating AKT activation. Our findings are supported by studies from Oka and colleagues (34), who found that cells overexpressing galectin-3 also express a high level of constitutively active Akt. Elad-Sfadia and colleagues found that galectin-3 preferentially binds to K-Ras-GTP and promotes the activation of RAF1 and PI3K (13). Oloumi and colleagues (35) found that integrin-linked kinase modulates prolonged PI3K-dependent effects on Wnt signaling by elevating GSK-3β phosphorylation. Recent reports suggest that activated AKT, in response to growth factors such as epidermal growth factor, can directly phosphorylate β-catenin at Ser552in vitro and in vivo causing β-catenin to disassociate from cell-cell contacts and accumulate in the nucleus (36, 37). Alterations of galectin-3 levels by stable transfection, however, did not change the level of β-catenin phosphorylation at Ser552 detected either by Western blotting using anti–phosphorylated β-catenin Ser552 antibody or coimmunoprecipitation in the colon cancer cells we examined (data not shown).

Based on our findings, we propose the following model (Fig. 6). Galectin-3 mediates AKT phosphorylation in colon cancer cells, thereby increasing phosphorylation of GSK-3β and decreasing its activity. Inactivation of GSK-3β leads to reduced degradation of β-catenin, and increased cellular β-catenin levels. β-Catenin then translocates to the nucleus, binds TCF4, and activates the transcription of its specific target genes.

Figure 6.

Proposed model by which galectin-3 mediates Wnt signaling in colon cancer cells. Galectin-3 mediates AKT phosphorylation, thereby increasing phosphorylation of GSK-3β and decreases its activity. Inactivation of GSK-3β leads to a reduction in β-catenin degradation, and increased cellular β-catenin levels. β-Catenin can then translocate to the nucleus, bind to TCF4, and activate the transcription of its specific target genes.

Figure 6.

Proposed model by which galectin-3 mediates Wnt signaling in colon cancer cells. Galectin-3 mediates AKT phosphorylation, thereby increasing phosphorylation of GSK-3β and decreases its activity. Inactivation of GSK-3β leads to a reduction in β-catenin degradation, and increased cellular β-catenin levels. β-Catenin can then translocate to the nucleus, bind to TCF4, and activate the transcription of its specific target genes.

Close modal

In conclusion, our data indicate that galectin-3 regulates β-catenin expression and nuclear accumulation and augments Wnt signaling via posttranslational mechanisms in human colon cancer cells. In addition, we have shown for the first time that galectin-3 modulates β-catenin levels and Wnt signaling by regulating the activity/phosphorylation of GSK-3β via the PI3K/AKT pathway. A more detailed understanding of the mechanisms by which galectin-3 augments Wnt signaling may facilitate the development of chemopreventive and therapeutic strategies for colorectal cancer.

No potential conflicts of interest were disclosed.

Grant support: NIH/National Cancer Institute grant R01CA69480 (R.S. Bresalier), an American Gastroenterological Association Research Scholar Award (S. Song), and Public Health Service grant DF56338, which supports the Texas Medical Center Digestive Diseases Center.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1
Kolligs FT, Bommer G, Goke B. Wnt/β-catenin/tcf signaling: a critical pathway in gastrointestinal tumorigenesis.
Digestion
2002
;
66
:
131
–44.
2
Giles RH, van Es JH, Clevers H. Caught up in a Wnt storm: Wnt signaling in cancer.
Biochim Biophys Acta
2003
;
1653
:
1
–24.
3
Behrens J, Jerchow BA, Wurtele M, et al. Functional interaction of an axin homolog, conductin, with β-catenin, APC, and GSK3β.
Science
1998
;
280
:
596
–9.
4
Sparks AB, Morin PJ, Vogelstein B, Kinzler KW. Mutational analysis of the APC/β-catenin/Tcf pathway in colorectal cancer.
Cancer Res
1998
;
58
:
1130
–4.
5
Rubinfeld B, Albert I, Porfiri E, Fiol C, Munemitsu S, Polakis P. Binding of GSK3β to the APC-β-catenin complex and regulation of complex assembly.
Science
1996
;
272
:
1023
–6.
6
Li J, Mizukami Y, Zhang X, Jo WS, Chung DC. Oncogenic K-ras stimulates Wnt signaling in colon cancer through inhibition of GSK-3β.
Gastroenterology
2005
;
128
:
1907
–18.
7
Desbois-Mouthon C, Cadoret A, Blivet-Van Eggelpoel MJ, et al. Insulin and IGF-1 stimulate the β-catenin pathway through two signalling cascades involving GSK-3β inhibition and Ras activation.
Oncogene
2001
;
20
:
252
–9.
8
Liu FT, Rabinovich GA. Galectins as modulators of tumour progression.
Nat Rev Cancer
2005
;
5
:
29
–41.
9
Schoeppner HL, Raz A, Ho SB, Bresalier RS. Expression of an endogenous galactose-binding lectin correlates with neoplastic progression in the colon.
Cancer
1995
;
75
:
2818
–26.
10
Bresalier RS, Yan PS, Byrd JC, Lotan R, Raz A. Expression of the endogenous galactose-binding protein galectin-3 correlates with the malignant potential of tumors in the central nervous system.
Cancer
1997
;
80
:
776
–87.
11
Bresalier RS, Mazurek N, Sternberg LR, et al. Metastasis of human colon cancer is altered by modifying expression of the β-galactoside-binding protein galectin 3.
Gastroenterology
1998
;
115
:
287
–96.
12
Yoshii T, Inohara H, Takenaka Y, et al. Galectin-3 maintains the transformed phenotype of thyroid papillary carcinoma cells.
Int J Oncol
2001
;
18
:
787
–92.
13
Elad-Sfadia G, Haklai R, Balan E, Kloog Y. Galectin-3 augments K-Ras activation and triggers a Ras signal that attenuates ERK but not phosphoinositide 3-kinase activity.
J Biol Chem
2004
;
279
:
34922
–30.
14
Shimura T, Takenaka Y, Tsutsumi S, Hogan V, Kikuchi A, Raz A. Galectin-3, a novel binding partner of β-catenin.
Cancer Res
2004
;
64
:
6363
–7.
15
Shimura T, Takenaka Y, Fukumori T, et al. Implication of galectin-3 in Wnt signaling.
Cancer Res
2005
;
65
:
3535
–7.
16
Shi Y, He B, Kuchenbecker KM, et al. Inhibition of Wnt-2 and galectin-3 synergistically destabilizes β-catenin and induces apoptosis in human colorectal cancer cells.
Int J Cancer
2007
;
121
:
1175
–81.
17
Song S, Byrd JC, Mazurek N, Liu K, Koo JS, Bresalier RS. Galectin-3 modulates MUC2 mucin expression in human colon cancer cells at the level of transcription via AP-1 activation.
Gastroenterology
2005
;
129
:
1581
–91.
18
Shishodia S, Koul D, Aggarwal BB. Cyclooxygenase (COX)-2 inhibitor celecoxib abrogates TNF-induced NF-κB activation through inhibition of activation of IκBα kinase and Akt in human non-small cell lung carcinoma: correlation with suppression of COX-2 synthesis.
J Immunol
2004
;
173
:
2011
–22.
19
Ding Q, He X, Hsu JM, et al. Degradation of Mcl-1 by β-TrCP mediates glycogen synthase kinase 3-induced tumor suppression and chemosensitization.
Mol Cell Biol
2007
;
27
:
4006
–17.
20
Dumic J, Dabelic S, Flogel M. Galectin-3: an open-ended story.
Biochim Biophys Acta
2006
;
1760
:
616
–35.
21
Lin HM, Pestell RG, Raz A, Kim HR. Galectin-3 enhances cyclin D(1) promoter activity through SP1 and a cAMP-responsive element in human breast epithelial cells.
Oncogene
2002
;
21
:
8001
–10.
22
Nakahara S, Raz A. Regulation of cancer-related gene expression by galectin-3 and the molecular mechanism of its nuclear import pathway.
Cancer Metastasis Rev
2007
;
26
:
605
–10.
23
He TC, Sparks AB, Rago C, et al. Identification of c-MYC as a target of the APC pathway.
Science
1998
;
281
:
1509
–12.
24
Shtutman M, Zhurinsky J, Simcha I, et al. The cyclin D1 gene is a target of the β-catenin/LEF-1 pathway.
Proc Natl Acad Sci U S A
1999
;
96
:
5522
–7.
25
Crawford HC, Fingleton BM, Rudolph-Owen LA, et al. The metalloproteinase matrilysin is a target of β-catenin transactivation in intestinal tumors.
Oncogene
1999
;
18
:
2883
–91.
26
Koh TJ, Bulitta CJ, Fleming JV, Dockray GJ, Varro A, Wang TC. Gastrin is a target of the β-catenin/TCF-4 growth-signaling pathway in a model of intestinal polyposis.
J Clin Invest
2000
;
106
:
533
–9.
27
Kolligs FT, Nieman MT, Winer I, et al. ITF-2, a downstream target of the Wnt/TCF pathway, is activated in human cancers with β-catenin defects and promotes neoplastic transformation.
Cancer Cell
2002
;
1
:
145
–55.
28
Araki Y, Okamura S, Hussain SP, et al. Regulation of cyclooxygenase-2 expression by the Wnt and ras pathways.
Cancer Res
2003
;
63
:
728
–34.
29
Lu Z, Hunter T. Wnt-independent β-catenin transactivation in tumor development.
Cell Cycle
2004
;
3
:
571
–3.
30
Liu C, Li Y, Semenov M, et al. Control of β-catenin phosphorylation/degradation by a dual-kinase mechanism.
Cell
2002
;
108
:
837
–47.
31
Lu Z, Ghosh S, Wang Z, Hunter T. Downregulation of caveolin-1 function by EGF leads to the loss of E-cadherin, increased transcriptional activity of β-catenin, and enhanced tumor cell invasion.
Cancer Cell
2003
;
4
:
499
–515.
32
Cross DA, Alessi DR, Cohen P, Andjelkovich M, Hemmings BA. Inhibition of glycogen synthase kinase-3 by insulin mediated by protein kinase B.
Nature
1995
;
378
:
785
–9.
33
Weston CR, Davis RJ. Signal transduction: signaling specificity—a complex affair.
Science
2001
;
292
:
2439
–40.
34
Oka N, Nakahara S, Takenaka Y, et al. Galectin-3 inhibits tumor necrosis factor-related apoptosis-inducing ligand-induced apoptosis by activating Akt in human bladder carcinoma cells.
Cancer Res
2005
;
65
:
7546
–53.
35
Oloumi A, Syam S, Dedhar S. Modulation of Wnt3a-mediated nuclear β-catenin accumulation and activation by integrin-linked kinase in mammalian cells.
Oncogene
2006
;
25
:
7747
–57.
36
He XC, Yin T, Grindley JC, et al. PTEN-deficient intestinal stem cells initiate intestinal polyposis.
Nat Genet
2007
;
39
:
189
–98.
37
Fang D, Hawke D, Zheng Y, et al. Phosphorylation of β-catenin by AKT promotes β-catenin transcriptional activity.
J Biol Chem
2007
;
282
:
11221
–9.