Redundant expression of canonical Wnt ligands in human breast cancer cell lines

Human breast cancer displays nuclear accumulation of β-catenin and induction of cyclin D1 expression, which suggests that canonical Wnt/β-catenin signaling is activated. In other cancers, the activation of canonical wnt/β-catenin signaling is associated with APC, CTNNB1 or AXIN1 mutations. However, these mutations are rare or absent in breast cancer. In search of alternative mechanisms, we performed comprehensive expression analysis of Wnt signaling molecules, including 19 Wnt ligands, ten Frizzled receptors, two co-receptors and four Lef/TCF transcription factors in immortalized normal human mammary epithelial cells (HMEC) and six breast cancer cell lines. HMEC expressed all Frizzled receptors except FZD9 and FZD10. They also expressed LRP5 and LRP6 co-receptors, as well as four Lef/TCF transcription factors. HMEC cells also expressed many Wnt ligands, including WNT1, WNT2B, WNT3, WNT5A, WNT5B, WNT7B, WNT9A, WNT10B and WNT16. Redundant expression of Wnt ligands, Frizzled receptors, co-receptors and Lef/TCF transcription factors was maintained in breast cancer cell lines with some exceptions. The most important changes in cancer cell lines concerned Wnt ligand expression. We noticed that most breast cancer cell lines overexpressed WNT3A, WNT4, WNT6, WNT8B, WNT9A and WNT10B. In contrast, the expression of WNT5A, WNT5B and WNT16 was usually down-regulated. It is noteworthy that all six Wnt ligands that are overexpressed in malignant cell lines are known to signal through the canonical Wnt/β-catenin signaling pathway, whereas down-regulated WNT5A and WNT5B ligands signal via the non-canonical pathway. The expression of both canonical Wnt ligands and most Frizzled receptors in breast cancer cell lines suggests that canonical Wnt/β-catenin signaling is activated in these cell lines by an autocrine/paracrine mechanism. In support of this prediction, we observed nuclear β-catenin accumulation and cyclin D1 induction in breast cancer cell lines, but not in HMEC. These results imply that ligand-dependent canonical Wnt/β-catenin signaling is active in human breast cancer.