Search

In human breast cancer, β-catenin localization has been related with disease prognosis. Since HER2-positive patients are an important subgroup, and that in breast cancer cells a direct interaction of β-catenin/HER2 has been reported, in the present study we have explored whether β-catenin location is related with the disease survival. The study was performed in a tumor bank from patients (n = 140) that did not receive specific anti-HER2 therapy. The proteins were detected by immunohistochemistry in serial sections, 47 (33.5%) patients were HER2-positive with a long follow-up. HER2-positive patients that displayed β-catenin at the plasma membrane (completely surrounding the tumour cells) showed a significant better disease-free survival and overall survival than the patients showing the protein on other locations. Then we explored the dynamics of the co-expression of β-catenin and HER2 in human MCF-7 and SKBR3 cells exposed to different stressful situations. In untreated conditions MCF-7 and SKBR3 cells showed very different β-catenin localization. In MCF-7 cells, cadmium administration caused a striking change in β-catenin localization driving it from plasma membrane to cytoplasmic and perinuclear areas and HER2 showed a similar localization patterns. The changes induced by cadmium were compared with heat shock, H2O2 and tamoxifen treatments. In conclusion, this study shows the dynamical associations of HER2 and β-catenin and their changes in subcellular localizations driven by stressful situations. In addition, we report for the first time the correlation between plasma membrane associated β-catenin in HER2-positive breast cancer and survival outcome, and the importance of the protein localization in breast cancer samples.

In a recent study, we have shown that in mammary tumors from mice lacking the Cav-1 gene, there are alterations in specific heat shock proteins as well as in tumor development. With this in mind, we have now investigated other proteins in the same mammary mouse tumor model (Her-2/neu expressing mammary tumors from Cav-1 wild type and Cav-1 null mice), to further comprehend the complex tumor-stroma mechanisms involved in regulating stress responses during tumor development. In this tumor model the cancer cells always lacked of Cav-1, so the KO influenced the Cav-1 in the stroma. By immunohistochemistry, we have found a striking co-expression of β-catenin and Her-2/neu in the tumor cells. The absence of Cav-1 in the tumor stroma had no effect on expression or localization of β-catenin and Her-2/neu. Both proteins appeared co-localized at the cell surface during tumor development and progression. Since Her-2/neu activation induces MTA1, we next evaluated MTA1 in the mouse tumors. Although this protein was found in numerous nuclei, the absence of Cav-1 did not alter its expression level. In contrast, significantly more PTEN protein was noted in the tumors lacking Cav-1 in the stroma, with the protein localized mainly in the nuclei. P-Akt levels were relatively low in tumors from both Cav-1 WT and Cav-1 KO mice. There was also an increase in nuclear NHERF1 expression levels in the tumors arising from Cav-1 KO mice. The data obtained in the MMTV-neu model are consistent with a role for Cav-1 in adjacent breast cancer stromal cells in modulating the expression and localization of important proteins implicated in tumor cell behavior.

The cadherin-catenin proteins have in common with heat shock proteins (HSP) the capacity to bind/interact proteins of other classes. Moreover, there are common molecular pathways that connect the HSP response and the cadherin-catenin protein system. In the present study, we have explored whether in breast cancer the HSP might interact functionally with the cadherin-catenin cell adhesion system. Beta-catenin was immunoprecipitated from breast cancer biopsy samples, and the protein complexes isolated in this way were probed with antibodies against HSP family members. We are thus the first to demonstrate a specific interaction between beta-catenin and Hsp27. However, beta-catenin did not bind Hsp60, Hsp70, Hsp90, gp96, or the endoplasmic reticulum stress response protein CHOP. To confirm the finding of Hsp27-beta-catenin interaction, the 27-kDa immunoprecipitated band was excised from one-dimensional polyacrylamide gel electrophoresis gels and submitted to liquid chromatography-tandem mass spectrometry with electrospray ionization, confirming a role for Hsp27. In addition, beta-catenin interacted with other proteins including heat shock transcription factor 1, P-cadherin, and caveolin-1. In human breast cancer biopsy samples, beta-catenin was coexpressed in the same tumor areas and in the same tumor cells that expressed Hsp27. However, this coexpression was strong when beta-catenin was present in the cytoplasm of the tumor cells and not when beta-catenin was expressed at the cell surface only. Furthermore, murine breast cancer cells transfected with hsp25 showed a redistribution of beta-catenin from the cell membrane to the cytoplasm. When the prognostic significance of cadherin-catenin expression was examined by immunohistochemistry in breast cancer patients (n = 215, follow-up = >10 years), we found that the disease-free survival and overall survival were significantly shorter for patients expressing P-cadherin and for patients showing expression of beta-catenin in the cytoplasm only (not at the cell surface). The interactions of beta-catenin with Hsp27 and with HSF1 may explain some of the molecular pathways that influence tumor cell survival and the clinical significance in the prognosis of the breast cancer patients.