Insulin-like growth factor 1 regulates the location, stability, and transcriptional activity of beta-catenin

Abstract

The insulin-like growth factor (IGF) type 1 receptor is required for growth, transformation, and protection from apoptosis. IGFs can enhance cell migration, which is known to be influenced via regulation of the E-cadherin/beta-catenin complex. We sought to investigate whether IGF-1 modulated the interaction between E-cadherin and beta-catenin in human colorectal cancer cells. We used the C10 cell line, which we established and have previously shown to lack adenomatous polyposis coli, E-cadherin, or beta-catenin mutations. We found that IGF-1 stimulation enhanced tyrosine phosphorylation of two proteins, beta-catenin and insulin-receptor substrate 1, which formed a complex with E-cadherin. Tyrosine phosphorylation of beta-catenin was accompanied by rapid (<1 min) dissociation from E-cadherin at the plasma membrane, followed by relocation to the cellular cytoplasm. IGF-1 also enhanced the stability of beta-catenin protein. Despite this, we observed no enhancement of transcriptional activity in complex with T-cell factor 4 (Tcf-4) in human embryonic kidney 293 cells treated with IGF-1 or insulin alone. IGF-1 did, however, enhance transcriptional activity in combination with lithium chloride, an inhibitor of glycogen synthase kinase 3 beta, which also stabilizes beta-catenin. In conclusion, we have shown that IGF-1 causes tyrosine phosphorylation and stabilization of beta-catenin. These effects may contribute to transformation, cell migration, and a propensity for metastasis in vivo.

Figure 1

Analysis of IGF1R levels in human colorectal cancer cell lines. Lysates (20 μg) were analyzed for IGF1R and β-catenin levels by Western blotting using an antibody against the β-subunit of the receptor, and filters were reprobed for β-catenin (a). Levels of IGF1R RNA were analyzed by ribonuclease protection assay using the first 286 bases of the IGF1R as an antisense riboprobe (b). Subsequently, detailed analysis of the levels of IGF1R per cell was carried out by Scatchard analysis using ¹²⁵I-IGF-1 (c). Information on the mutational status of APC, β-catenin, and E-cadherin was taken from ref. . n/d, Not done.

Figure 2

Effects of IGF-1 stimulation on tyrosine phosphorylation of proteins that coimmunoprecipitate with E-cadherin. Cells were serum starved for 16 h and treated with recombinant 1.3 nM (10 ng/ml) IGF-1 for 10 min. Cells were lysed in Nonidet P-40 lysis buffer, and proteins were immunoprecipitated using anti-E-cadherin antibody (clone HECD-1). Immunoprecipitates were assessed for tyrosine phosphorylation (a) using phosphotyrosine antibody (PY99). Two proteins of approximate molecular mass of 180 kDa and 90 kDa were identified by Western blotting as insulin receptor substrate-1 (IRS-1) and β-catenin (b).

Figure 3

Effects of IGF-1 stimulation on the interaction between β-catenin and E-cadherin. Cells were serum starved for 16 h, then stimulated with 1.3 nM (10 ng/ml) IGF-1 for 10 min (a and b), pervanadate (0.5 mM sodium orthovanadate and 1.5 mM hydrogen peroxide) for 30 min (c), 1.3 nM IGF-1 for 10 min following preincubation with 100 μM LY294002 or DMSO carrier for 30 min (d), or with 1.3 nM IGF-1 from 1 to 60 min (e), all at 37°C. Cells were lysed in Triton lysis buffer, and the interaction between β-catenin and E-cadherin was examined by immunoprecipitation of β-catenin and Western blotting for E-cadherin.

Figure 4

Effects of IGF-1 stimulation on the localization of β-catenin in c10 cells. (a) Subconfluent c10 cells were serum starved for 16 h and were treated with 1.3 nM IGF-1 for 10 or 30 min. Cells were fractionated into 1% digitonin or RIPA-soluble fractions, and equal amounts of each soluble fraction were separated by SDS/PAGE and immunoblotted for β-catenin. (b) Subconfluent c10 cells were serum starved for 16 h and were treated with 1.3 nM IGF-1 or vehicle control for 30 min at 37°C. Cells were stained with an antibody for β-catenin that was visualized with a fluorescent secondary antibody. DNA was stained with 4′,6-diamidino-2-phenylindole (DAPI). (×400.)

Figure 5

Analysis of effects of IGF-1 signaling on stability of β-catenin protein. Subconfluent C10 cells were pulsed with ³⁵S-Promix and chased with medium containing an excess of cold methionine/cysteine for the indicated times. Cells were lysed and immunoprecipitated for β-catenin. The results were analyzed by densitometry and expressed graphically as a percentage of the value at time 0 h. The figure shows results of a single experiment, which was repeated once with similar results.

Figure 6

(a) Effects of lithium chloride on stability of β-catenin. Subconfluent C10 cells were serum starved for 16 h in the presence or absence of 50 mM LiCl and then pulsed with ³⁵S-Promix for 30 min and chased with medium containing excess of cold methionine/cysteine. Cells were lysed and immunoprecipitated for β-catenin and analyzed as before. The numbers shown indicate a mean of duplicate experiments. (b). Effects of IGF-1 and lithium chloride on transcriptional activity of β-catenin. HEK293 cells were cotransfected with equal amounts of expression plasmids containing full-length wild-type β-catenin, Tcf-4, and either TOPFLASH or FOPFLASH reporter. Total cell lysates were equalized for protein concentration and assayed for luciferase activity. Bars indicate the mean and standard error of triplicate luciferase assays.