WNT/(formula)-Catenin signaling pathway activation in epithelial cancers

Wnt signaling is involved in a large set of cellular and developmental processes, and when mis-regulated can lead to both degenerative diseases and many types of cancer. The involvement of Wnt signaling was already well demonstrated in several types of human cancers such as colorectal cancer. However, in some others such as hepatocellular carcinoma (HCC) and breast cancer, the role of Wnt signaling is not fully understood. To study the role of Wnt pathway in liver cancer, we first classified human hepatoma cell lines into well-differentiated and poorly differentiated groups using hepatocyte-specific biomarkers. Wnt/β-catenin signaling activity was measured using TCF/LEF-dependent reporter assay. Canonical Wnt/β-catenin signaling was constitutively active in 80% of well differentiated and 14% of poorly differentiated cell lines, respectively. Furthermore, ectopic expression mutant of S33Y β-catenin resulted in strong canonical Wnt/β-catenin activity in well differentiated, but not in poorly differentiated HCC cells. Comprehensive analysis of major Wnt signaling components by a rapid RT-PCR assay showed redundant expression of many Wnt ligands, Frizzled receptors, co-receptors and TCF/LEF factors in HCC. In contrast, canonical signaling-inhibitory Wnt5A and Wnt5B ligands were selectively expressed in poorly differentiated HCC cell lines. Our observations indicate that canonical Wnt/β-catenin signaling is active in well differentiated, but repressed in poorly differentiated HCC cells. Thus, canonical Wnt/β-catenin signaling plays a dual role in HCC. To study the role of Wnt pathway in breast cancer, we performed a comprehensive expression analysis, by RT-PCR, of Wnt signaling molecules, including 19 Wnt ligands, ten Frizzled receptors, two LRP co-receptors and four Lef/TCF transcription factors in immortalized normal human mammary epithelial cells (HMECs), six breast cancer cell lines (BCCL) and 14 primary breast tumors (PBT). BCCL expressed/over-expressed all Frizzleds except FZD10, LRP5/6 and Lef/TCFs. They also overexpressed WNT4, WNT7B, WNT8B, WNT9A and WNT10B, but the expression of WNT1, WNT2B, WNT3, WNT5A, WNT5B and WNT16 was lost or decreased in most BCCL. Wnt expression correlated with nuclear β-catenin accumulation and cyclin D1 induction in BCCL, compared to HMECs, indicating a reactivation of the canonical Wnt signaling in malignant cells. Furthermore, the expression of FZD1, WNT-4, WNT7B, WNT8B, WNT9A and WNT10B, all implicated in canonical Wnt signaling, was upregulated in PBT, whereas the non-canonical WNT5A expression was down-regulated. Our study gave strong evidences for the differential involvement of Wnt pathway in liver and breast cancers. In liver cancer, Wnt pathway activity seems to be linked to the differentiation status of HCC cell lines. Furthermore, our data showed that the canonical Wnt pathway was active in well-differentiated HCC cell lines and repressed in poorly differentiated ones. In contrast, the study of Wnt pathway in breast cancer cell lines showed similarities rather than differences. Indeed, our study revealed a significant correlation between Wnt ligands mRNA expression profile and the induction of Cyclin D and nuclear β-catenin protein accumulation in all breast cancer cell lines studied. We concluded that, although involved in both types of cancers, Wnt signaling is acting differently in liver and breast cancers. More interestingly, in the same type of cancer such as HCC, Wnt signaling displayed differential activity depending on the cell differentiation status.