Negative feedback loop of Wnt signaling through upregulation of conductin/axin2 in colorectal and liver tumors

Abstract

Activation of Wnt signaling through beta-catenin/TCF complexes is a key event in the development of various tumors, in particular colorectal and liver tumors. Wnt signaling is controlled by the negative regulator conductin/axin2/axil, which induces degradation of beta-catenin by functional interaction with the tumor suppressor APC and the serine/threonine kinase GSK3beta. Here we show that conductin is upregulated in human tumors that are induced by beta-catenin/Wnt signaling, i.e., high levels of conductin protein and mRNA were found in colorectal and liver tumors but not in the corresponding normal tissues. In various other tumor types, conductin levels did not differ between tumor and normal tissue. Upregulation of conductin was also observed in the APC-deficient intestinal tumors of Min mice. Inhibition of Wnt signaling by a dominant-negative mutant of TCF downregulated conductin but not the related protein, axin, in DLD1 colorectal tumor cells. Conversely, activation of Wnt signaling by Wnt-1 or dishevelled increased conductin levels in MDA MB 231 and Neuro2A cells, respectively. In time course experiments, stabilization of beta-catenin preceded the upregulation of conductin by Wnt-1. These results demonstrate that conductin is a target of the Wnt signaling pathway. Upregulation of conductin may constitute a negative feedback loop that controls Wnt signaling activity.

FIG. 1.

Western blot analysis of conductin in human tumor cell lines. Western blotting was performed using the C/G7 antibody on protein extracts from the indicated tumor cell lines. Equal amounts of protein were loaded per lane. Conductin was detected as a double band around 100 kDa in most cell lines. In some cell lines, either the upper or the lower band is seen. The same blots were also probed with antitubulin antibodies to demonstrate similar protein loading. Note that the Western blots of conductin in the lower panels (breast, bladder, pancreas, prostate, and melanoma) were exposed five times longer than those in the upper panels (colon, hepatoblastoma, and lung). Numbers at right are molecular masses in kilodaltons.

FIG. 2.

Western blot analysis of conductin in human tumors and normal tissues. Snap-frozen tissues were extracted and processed for Western blotting as described in Materials and Methods. (A) Western blots of normal (N) and corresponding tumor (T) tissue from sporadic colon carcinomas and from adenomatous polyps of FAP patients. The C/G7 antibody detects the major conductin bands at 100 kDa as well as lower-molecular-weight degradation products in the tumor tissues. Cytokeratin 19 and tubulin were also probed to show equal loading. (B) Western blots of normal (N) and tumor (T) tissue from hepatocellular carcinomas (HCC) and hepatoblastoma. Detection of β-catenin on the same samples is also shown. β-Catenin species with reduced size (∗) possibly result from deletions of the regulatory N terminus of the protein (see reference 25). The tumor extractions were performed with detergent-containing buffer. Thus, the β-catenin signal does not necessarily reflect the cytosolic signaling form alone but may include the cadherin-bound fraction. Tubulin was probed to show equal protein loading in the HCC samples; for the hepatoblastoma samples, equal protein loading was verified by Ponceau S staining. Numbers at right are molecular masses in kilodaltons.

FIG. 3.

Expression analysis of conductin and axin in human tumors and normal tissues. (A) ³²P-labeled cDNA probes of conductin and axin were hybridized on a cancer profiling array (Clontech) containing cDNA from 241 human tumor (T) and corresponding normal (N) tissues from individual patients. The outlined groups of dots represent normal, tumor, and metastatic tissues from the same patient. The position of cDNAs from different tumor cell lines is indicated on the right. Note that SW480, which is a colorectal cancer cell line, is strongly positive for conductin. (B) Quantification of the results shown in panel A for a subset of tumor types by PhosphorImager analysis. Results were normalized to the hybridization signals of a ubiquitin probe (see Materials and Methods). The percentage of cases showing tumor/normal tissue expression ratios (T/N) of >2.0, 0.5 to 2.0, and <0.5 is given for each tumor type. Expression changes in the metastases outlined in panel A are not included in the quantification.

FIG. 3.

Expression analysis of conductin and axin in human tumors and normal tissues. (A) ³²P-labeled cDNA probes of conductin and axin were hybridized on a cancer profiling array (Clontech) containing cDNA from 241 human tumor (T) and corresponding normal (N) tissues from individual patients. The outlined groups of dots represent normal, tumor, and metastatic tissues from the same patient. The position of cDNAs from different tumor cell lines is indicated on the right. Note that SW480, which is a colorectal cancer cell line, is strongly positive for conductin. (B) Quantification of the results shown in panel A for a subset of tumor types by PhosphorImager analysis. Results were normalized to the hybridization signals of a ubiquitin probe (see Materials and Methods). The percentage of cases showing tumor/normal tissue expression ratios (T/N) of >2.0, 0.5 to 2.0, and <0.5 is given for each tumor type. Expression changes in the metastases outlined in panel A are not included in the quantification.

FIG.4.

Downregulation of conductin by dominant-negative TCF. (A) Downregulation of conductin upon expression of dominant-negative TCF1 in DLD1 colon carcinoma cells. The parental DLD1 cell clone (TR7), containing the Tet repressor, and two cell clones containing tetracycline-inducible dominant-negative TCF1 (ΔNTCF1 clones 1 and 2) were treated with doxycycline as indicated for 12 h. Conductin, axin, and tubulin (as a loading control) were detected by Western blotting on extracts of these cells. (B) Time course of downregulation of conductin by dominant-negative TCF determined by Western blotting. (C) Northern blot analysis of conductin mRNA in TR7 and ΔNTCF1 clone 1 after treatment with doxycycline for 14 h.

FIG. 5.

Upregulation of conductin by Wnt signaling. (A) Expression of conductin in Rat2/Wnt-1 cells and Rat2/MV7 control cells demonstrated by Western blotting using the C/G7 antibody. Note increased levels of conductin in the Wnt-1-expressing cells. (B) Upregulation of conductin in MDA MB 231 breast carcinoma cells by Wnt-1. Cells were incubated with media conditioned by Rat2/Wnt-1 (Wnt) or Rat2/MV7 cells (MV7) for 4 and 18 h as indicated above the lanes. Conductin levels were determined by Western blotting as described for panel A. Western blotting for β-catenin was performed on cytosolic extracts prepared from parallel cultures. Note that both conductin and β-catenin levels increase after 4 h of stimulation by Wnt-1-conditioned medium and decrease to baseline after 18 h. Control conditioned medium had no effect. (C) Increase in conductin mRNA after treatment of MDA MB 231 cells with Wnt-1-conditioned medium for 4 h as determined by Northern blotting. GAPDH, glyceraldehyde-3-phosphate dehydrogenase. (D) Time course of upregulation of conductin and β-catenin by Wnt-1 in MDA MB 231 cells as determined by Western blotting. In this experiment, both conductin and β-catenin were detected from the same cytosolic extracts. Note that β-catenin peaks between 2 and 6 h while conductin peaks between 4 and 8 h after stimulation with Wnt-1. Tubulin was probed to demonstrate protein loading. (E) Upregulation of conductin after transient expression of dishevelled-2 (Dvl-2) in Neuro2A cells. Conductin and tubulin were detected by Western blotting.

FIG. 6.

Upregulation of conductin expression in Min tumors. Min mice were crossed with heterozygous conductin^+/lacZ mice. Expression of lacZ and thus activation of the conductin promoter were detected by β-galactosidase staining of the small intestine. (A) Schematic representation of the conductin wild-type (wt) locus and the targeted allele in the vicinity of exon 2. In the conductin lacZ allele, a lacZ gene containing a nuclear localization signal (NLS) was introduced in frame to the endogenous conductin start codon by the homologous recombination technique, thereby replacing most of exon 2. (B) Prominent whole-mount β-galactosidase staining of adenomas (arrows) in the intestine. Note that the intestine was opened longitudinally to expose the luminal site. Bar, 1 mm. (C) Paraffin section of a tumor showing strong β-galactosidase staining in the adenoma (bracket) and light staining of crypt epithelium adjacent to the muscularis layer (arrow). Counterstaining was performed with nuclear fast red. Bar, 100 μm. (D) Whole-mount β-galactosidase staining of early dysplastic lesions. Bar, 100 μm. (E) Magnification of the crypt epithelium layer in panel C to reveal β-galactosidase staining in the epithelial cells. Bar, 50 μm. (F) In situ hybridization for TCF4 mRNA on tissue sections of mouse intestinal epithelium. Bar, 50 μm. (G) In situ hybridization of conductin mRNA on tissue sections of mouse intestinal epithelium. Bar, 50 μm. Note similar staining patterns for TCF4 and conductin in the basal crypt epithelium in panels F and G.