Osvaldo Pontiggia

The tumor microenvironment modulates tamoxifen resistance in breast cancer: a role for soluble stromal factors and fibronectin through β1 integrin

Tamoxifen resistance has been largely attributed to genetic alterations in the epithelial tumor cells themselves, such as overexpression of HER-2/Neu. However, in the clinic, only about 15–20% of cases of HER-2/Neu amplification has actually been correlated to the acquisition of endocrine resistance, suggesting that other mechanisms must be involved as well. Using the epithelial LM05-E and the fibroblastic LM05-F cell lines, derived from the estrogen dependent spontaneous M05 mouse mammary tumor, as well as MCF-7 cells, we analyzed whether soluble stromal factors or extracellular matrix components protected against tamoxifen induced cell death. Involvement of signaling pathways was determined by using specific inhibitors and western blot, and phosphorylation of the estrogen receptor alpha by western blot and immunofluorescence. Soluble factors produced by the fibroblastic cells protect the epithelial tumor cells from tamoxifen-induced cell death through a mechanism that involves EGFR and matrix metalloproteinases upstream of PI3K/AKT. Exogenous fibronectin by itself confers endocrine resistance through interaction with β1 integrin and activation of PI3K/AKT and MAPK/ERK 1/2 pathways. The conferred resistance is reversed by blocking β1 integrin. We show also that treatment with both conditioned medium and fibronectin leads to the phosphorylation of the estrogen receptor at serine-118, suggesting stromal factors as modulators of ER activity. Our results show that the tumor microenvironment can modulate tamoxifen resistance, providing an alternative explanation for why patients become refractory to hormone-therapy.

Decreased metastatic phenotype in cells resistant to aminolevulinic acid-photodynamic therapy

Photodynamic therapy (PDT) is a novel cancer treatment utilising a photosensitiser, visible light and oxygen. PDT often leaves a significant number of surviving tumour cells. In a previous work, we isolated and studied two PDT resistant clones derived from the mammary adenocarcinoma LM3 line (Int. J. Oncol. 29 (2006) 397-405). The isolated Clon 4 and Clon 8 exhibited a more fibroblastic, dendritic pattern and were larger than the parentals. In the present work we studied the metastatic potential of the two clones in comparison with LM3. We found that 100% of LM3 invaded Matrigel, whereas only 19+/-6% and 24+/-7% of Clon 4 and Clon 8 cells invaded. In addition, 100% of LM3 cells migrated towards a chemotactic stimulus whereas 38+/-8% and 73+/-10% of Clones 4 and 8, respectively, were able to migrate. In vivo, 100% of the LM3 injected mice developed spontaneous lung metastasis, whereas none of the Clon 8 did, and only one of the mice injected with Clon 4 did. No differences were found in the proteolytic enzyme profiles among the cells. Anchorage-dependent adhesion was also impaired in vivo in the resistant clones, evidenced by the lower tumour take, latency time and growth rates, although both clones showed in vitro higher binding to collagen I without overexpression of beta1 integrin. This is the first work where the metastatic potential of cells surviving to PDT has been studied. PDT strongly affects the invasive phenotype of these cells, probably related to a higher binding to collagen. These findings may be crucial for the outcome of ALA-PDT of metastatic tumours, although further studies are needed to extrapolate the results to the clinic employing another photosensitisers and cell types.

Establishment of an in vitro estrogen-dependent mouse mammary tumor model: a new tool to understand estrogen responsiveness and development of tamoxifen resistance in the context of stromal–epithelial interactions

Currently, to our knowledge, there are no continuous cell lines derived from estrogen dependent, tamoxifen sensitive spontaneous mouse mammary carcinomas. We describe here the establishment and characterization of a cell line derived from the M05 mouse mammary tumor, LM05-Mix, composed of both an epithelial and a fibroblastic component. From it the respective epithelial LM05-E and fibroblastic LM05-F cell lines were generated by limiting dilution. Immunofluorescence studies confirmed that the epithelial cells were positive for E-cadherin, cytokeratins and vimentin whereas the fibroblastic cells were negative for the epithelial markers and positive for α-smooth muscle actin and vimentin. Both cell types expressed estrogen and progesterone receptors, although only the epithelial LM05-E cells were stimulated by estradiol and inhibited by tamoxifen. In the bicellular LM05-Mix cell line estradiol proved to stimulate cell proliferation whereas the response to tamoxifen was dependent on confluency and the degree of epithelial-fibroblastic interactions. The presence of membrane estrogen receptors in both cell types was suggested by the achievement of non-genomic responses to short treatments with estradiol, leading to the phosphorylation of ERK1/2. Finally, cytogenetic studies suggest that these two cell types represent independent cell populations within the tumor and would not be the result of an epithelial-mesenchymal transition. This model presents itself as a valuable alternative for the study of estrogen responsiveness and tamoxifen resistance in the context of epithelial-stromal interactions.

Non-genomic actions of estradiol and 4-OH-tamoxifen on murine breast cancer cells.

Estrogens and tamoxifen do not only exert their effects at the genomic level, but also play a role at the cell membrane activating downstream signaling pathways. We recently characterized an estrogen receptor-positive epithelial murine breast cancer cell line, LM05-E. Utilizing this cell line and MCF-7 cells, we compared the non-genomic effects of estradiol and 4-OH-tamoxifen. We showed that, similar to estradiol, tamoxifen activated the MAPK/ERK 1/2 pathway; however, we did not find activation of PI3K/AKT by either estradiol or tamoxifen. Short-term treatments with estradiol stimulated, whereas tamoxifen inhibited cell proliferation. Using pharmacological inhibitors we showed that the effect of estradiol was mediated by the MAPK/ERK 1/2 pathway, but that inhibition of this pathway did not affect tamoxifen. Surprisingly, however, blocking of PI3K/AKT signaling interfered with the inhibitory effect of tamoxifen. Analysis of the involvement of the EGFR support previous findings that designate this receptor as a mediator of the non-genomic effects of estradiol; blocking EGFR also reverses the inhibitory effect of tamoxifen. Finally, matrix metalloproteinases (MMPs) were confirmed to be involved in the proliferative effect of estradiol. These results demonstrated the novel non-genomic effects of tamoxifen and revealed that pathways downstream of EGFR and PI3K/AKT are involved in the inhibition of cell proliferation. Caution should be exercised when analyzing strategies that aim at combining endocrine therapy with specific signaling inhibitors.

Abstract 545: Laminin induces tamoxifen resistance through alfa 6 integrin in breast cancer cells

Even though it is now well established that the tumor microenvironment plays a central role in modulating tumor progression and response to therapy, its function as a regulator of endocrine resistance in breast cancer has not yet been thoroughly explored. We previously showed that fibronectin through beta 1 integrin induces tamoxifen resistance in breast cancer cells. The aim of this study was to determine whether other extracellular matrix components induce tamoxifen resistance and what mechanisms are involved. We found that laminin, but not collagen IV, induced tamoxifen resistance in both estrogen receptor positive (ER+) human MCF-7 and mouse LM05-E breast cancer cells (P −8 M) and 4-OH-tamoxifen (10 −6 M) with the addition of either the MAPK/ERK1/2 inhibitor PD98059 (10µM) or the PI3K/AKT inhibitor LY294002 (10µM). Only LY294002 reversed the protective effect of laminin (P −8 M induced a significant stimulation of ER activity (P −6 M 4-OH-tamoxifen (P Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 545. doi:10.1158/1538-7445.AM2011-545

Abstract P4-02-11: Treatment with Fibronectin or Carcinoma Associated Fibroblast Conditioned Media Generate Sustained Tamoxifen Resistance in Breast Cancer Cells

Background: breast cancer is the most common neoplasm in woman and the second cancer-related cause of death. About 70% of breast cancers are positive for estrogen receptor (ER) and progesterone receptors. Tamoxifen is the main hormonal therapy for ER+ patients. However, one third of the patients that are initially responsive eventually develop resistance to this treatment. We previously showed that treatment with fibronectin or conditioned media from carcinoma associated fibroblasts conferred tamoxifen resistance to otherwise sensitive ER+ breast cancer cells (in particular MCF-7 and LM05-E cells). On the other hand we also found that in the case of conditioned media, the PI3K/AKT pathway was involved downstream of EGFR and ≥1 integrin, whereas for fibronectin both the MAPK/ERK1/2 and the PI3K/AKT pathways were involved downstream of ≥1. The aim of this study was to investigate whether the protective effect conferred by the stromal elements was reversible, or, if on the contrary, it was sustained over time even after they were removed from the culture. In order to do so we pre-treated LM05-E cells with DMEM/F12 containing 1% charcoal stripped serum (CSS) as a control or with the addition of fibronectin (10μg/ml) or conditioned media from the LM05-F carcinoma associated fibroblast cell line. Subsequently cells were trypsinized, washed and allowed to attach ON in 8 well chamber slides in the absence of the stromal factors. They were then treated for 48 hours with estradiol 10-8 M, or estradiol plus 4-OH-tamoxifen 10-6 M in the presence of 1% CSS. Cell death was measured by propidium iodide incorporation. The 48 hour pre-treatment with either conditioned media or fibronectin conferred resistance to tamoxifen induced cell death. This resistance was sustained for at least a week in both cases, as determined by culturing the cells for this period of time in the absence of the stromal factors before testing the effect of 4-OH-tamoxifen. Shorter pretreatments of 6 or 12 hours weren9t capable of conferring endocrine resistance. Inhibitors of the MAPK/ERK, PI3K/AKT or EGFR were added to the pre-treatments in order to study involvement of these pathways. Neither addition of the PI3K inhibitor LY294002 (10 μM), the MAPK inhibitor PD98059 (10 μM) or the EGFR inhibitor AG1478 (6,4 μM) were able to reverse the sustained protection induced by either fibronectin or conditioned media. Conclusion: our results suggest that both of these stromal elements are able to induce sustained changes in the breast cancer cells that lead to tamoxifen resistance even after they have been removed from the culture. At least a 48 hour period is required to accomplish this protection and this effect is sustained for at least a week. These results have important clinical implications on the lasting effects that the tumor microenvironment may have on the long term response to endocrine treatment in breast cancer. This work is supported by a grant from the Susan G. Komen for the Cure Foundation to MS (BCTR0600341). Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P4-02-11.

Microenvironmental Factors Modulate Tamoxifen Resistance in Breast Cancer.

The tumor microenvironment plays a critical role in tumor growth and progression. However, its impact on response to endocrine therapy has not been fully addressed in the literature. From the estrogen-dependent M05 mouse mammary tumor we established a bicellular cell line, LM05 MIX, which is composed of an epithelial and a fibroblastic cell population which were separated to generate the LM05-E and LM05-F cell lines respectively. Previous work demonstrated that the LM05-E cells, which are positive for estrogen and progesterone receptors, are sensitive to 4-OH-tamoxifen (TAM; 10 -6 M) being this effect reverted when cells are co-cultured with the fibroblastic LM05-F cell population. Our working hypothesis is that extracellular matrix components and soluble factors produced by the fibroblastic cells may induce tamoxifen resistance on the otherwise sensitive epithelial LM05-E tumor cells. Immunofluorescence staining for the extracellular matrix component laminin (LN) was found to be associated to the fibroblastic cells in mixed cultures. To investigate whether LN conferred TAM resistance, LM05-E cells were cultured on recombinant LN and treated for 48 hours with TAM. We found that LN conferred resistance (P≤ 0.001) to TAM treatment. We used the specific inhibitors PD98059 (ERK ½ inhibitor, 10μM) and LY294002 (PI3K inhibitor, 10μM) to determine which signaling pathways were involved. Only LY294002 reverted TAM resistance, implicating the participation of the PI3K/AKT but not the ERK pathway. Next, to study whether soluble factors secreted by LM05-F cells were mediating TAM resistance we tested the effect of their conditioned media on LM05-E cells in the presence of TAM. We found a decreased number of apoptotic cells in conditioned media-treated cells, as compared to cells treated with non-conditioned media. Treatments with AG1478 (EGFR inhibitor, 6.4 μM), GM6001 (MMP inhibitor, 10 μM), PD98059, LY294002 and AIIB2 (β1-integrin blocking antibody) were used to elucidate which pathways were involved in the protective effect conferred by the conditioned media. The resistance to TAM treatment was reverted with AG1478, GM6001, AIIB2 and LY294002 (P≤ 0.001) but not with PD98059. MMPs or membrane-associated MMP activities have been reported to activate EGFR by shedding EGFR ligands from plasma membrane. Moreover activation of β1-integrin by fibroblast-conditioned media, as suggested by the detection of FAK phosphorylation by western blot, might collaborate with the EGFR signaling pathway. Our results suggest that TAM resistance may be the result of the modulation of epithelial cell responses by tumor fibroblasts via microenvironmental cues such as extracellular matrix components and soluble factors. Targeting the interaction of breast cancer cells with the microenvironment could be a novel and future alternative for patients with breast cancer whose tumors progress to TAM resistance. This work is supported by a grant from the Susan G. Komen for the Cure Foundation to MS (BCTR0600341). Citation Information: Cancer Res 2009;69(24 Suppl):Abstract nr 5133.

β1 integrin mediates tamoxifen resistance in breast cancer cells.

Abstract #1055 Even though it is now well established that the microenvironment plays a central role in modulating tumor progression and response to therapy, its function as a regulator of endocrine resistance in breast cancer has not yet been explored. To do so we have established a bicellular in vitro cell line, called LM05-Mix, from the estrogen dependent M05 mouse mammary tumor (BCRT, in press). The epithelial and fibroblastic cell populations of the LM05-Mix cell line were further separated to generate the LM05-E and LM05-F cell lines respectively, both of which are estrogen receptor (ER) and progesterone receptor positive. We have previously shown that treatment with 10-6M 4-OH-tamoxifen (TAM) induces apoptosis of the LM05-E cells, and that this effect is inhibited when the epithelial cells are co-cultured with the fibroblastic cell population. Immunofluorescence staining to the extracellular matrix component fibronectin (FN) was found to be especially associated to the fibroblastic cells in the co-culture. To investigate whether FN conferred TAM resistance, LM05-E cells and human MCF-7 cells were cultured on recombinant FN and treated for 48 hours with TAM. In both cases FN conferred resistance (P≤0.001). To understand what signaling pathways were involved, the experiments were repeated in the presence of specific inhibitors and both PD98059 and LY294002 reverted the protective effect, implicating the participation of both the MAPK/ERK and PI3K/AKT pathways. Next, to corroborate the activation of these pathways by FN, starved LM05-E cells were treated with FN (30 μg/ml). Phosphorylation of FAK, MAPK/ERK1/2 and AKT was detected by western blot between 3 and 15 minutes of treatment. To find out whether the activation of these pathways by FN could lead to ligand independent activation of the ER and thus account for the induction of TAM resistance, immunofluorescence staining was carried out to serine 118-phosphorylated ER. An increase in the phosphorylation of ER at this site was evident at 15 minutes of treatment both in LM05-E and MCF-7 cells. Finally, to confirm that the protective effect was mediated by the binding of β1 integrin to FN, cells were pre-incubated with either a specific β1 blocking antibody (AIIB2) or mouse IgG and seeded on FN. The protective effect of FN over TAM induced cell death was only reverted in the AIIB2 treated group (P≤0.01). Our results show that FN through its binding to β1 integrin confers TAM resistance and suggest that the mechanism involved would be through the modulation of the activity of ER. Targeting the interaction of breast cancer cells with the microenvironment could be a novel and future alternative for patients with breast cancer whose tumors progress to TAM resistance. This work is supported by a grant from the Susan G. Komen for the Cure Foundation to MS (BCTR0600341). Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 1055.