Ets-1 drives breast cancer cell angiogenic potential and interactions between breast cancer and endothelial cells.

Ets-1 transcription factor overexpression in breast cancers is associated with invasive features and is associated with a poor prognosis. Beyond its role in driving carcinoma cell invasion, in this study, we wished to determine whether Ets-1 overexpression in cancer cells promotes angiogenesis by creating a paracrine pro-invasive environment for endothelial cells as well. To address this question, we set up different co-culture models of cancer cells with endothelial cells. Conditioned media from cancer cells induced endothelial cell proliferation, migration and morphogenesis in matrix models. Of note, co-culture assays in three-dimensional matrix models also revealed the reciprocal induction of cancer cell morphogenesis by endothelial cells, in support of an angiocrine action on tumor cells. Ets-1 emerged as a key regulator of the angiogenic potential of breast cancer cells, favoring their ability to induce, in a paracrine manner, the morphogenesis of endothelial cells and also to physically interact with the latter. Nevertheless, Ets-1 overexpression in cancer cells also restrained their chemoattractive potential for endothelial cells both in Boyden chambers and in ex vivo 3D co-cultures. Finally, Ets-1 modulation in breast cancer cells qualitatively altered the angiogenic pattern of experimental in vivo tumors, with a balance between vessel recruitment and intratumoral small capillaries sprouting. Taken together, our data highlight a critical and intriguing role for Ets-1 in the angiogenic potential of breast cancer cells, and reveal another facet of Ets-1 oncogenic activities.

Ets-1 controls breast cancer cell balance between invasion and growth.

Ets-1 overexpression in human breast cancers is associated with invasiveness and poor prognosis. By overexpressing Ets-1 or a dominant negative mutant in MMT breast cancer cells, we previously highlighted the key role of Ets-1 in coordinating multiple invasive features of these cells. Interestingly, we also noticed that Ets-1 decreased the density of breast cancer cells cultured in three-dimensional extracellular matrix gels. The 3D context was instrumental to this phenomenon, as such downregulation was not observed in cells grown on two-dimensional plastic or matrix-coated dishes. Ets-1 overexpression was deleterious to anchorage-independent growth of MMT cells in soft agar, a standard model for in vitro tumorigenicity. The relevance of this mechanism was confirmed in vivo, during primary tumor growth and in a metastatic assay of lung colonization. In these models, Ets-1 was associated with epithelial-to-mesenchymal transition features and modulated the ratio of Ki67-positive cells, while hardly affecting in vivo apoptotic cell death. Finally, siRNA-mediated knockdown of Ets-1 in human breast cancer cell lines also decreased colony growth, both in anchorage-independent assays and 3D extracellular matrix cultures. These in vitro and in vivo observations shed light on an unsuspected facet of Ets-1 in breast tumorigenesis. They show that while promoting malignancy through the acquisition of invasive features, Ets-1 also attenuates breast tumor cell growth and could therefore repress the growth of primary tumors and metastases. This work also demonstrates that 3D models may reveal mechanisms of tumor biology that are cryptic in standard 2D models.

Ets-1 triggers and orchestrates the malignant phenotype of mammary cancer cells within their matrix environment.

Acquisition of invasive characteristics is a hallmark of breast carcinoma progression. During this phenomenon, Ets-1 transcription factor overexpression is induced and associated with breast cancer invasiveness, and poor prognosis. We hypothesized that Ets-1 transcription factor could be the orchestrator of a genetic program inducing the expression of genes necessary for cell motility, as postulated by the tumor microenvironment invasion model. We aimed at elucidating the role of Ets-1 in the molecular control of mammary cancer cell invasion and aggressiveness within their matrix environment. To that purpose, mouse mammary tumor MMT epithelial cells were engineered to stably overexpress Ets-1, or the dominant negative Ets-1 DNA Binding domain. The biological function of Ets-1 was assessed in three-dimensional extracellular matrix systems recreating a microenvironmental architecture resembling in vivo geometric constraints. Ets-1 overexpression provided MMT cells with a motile and invasive phenotype, leading to cell scattering, and impairing multicellular organization in matrix-mimicking gels. We evidenced that Ets-1 promoted HGF/SF activation, and the expression of its receptor, c-Met. Ets-1 also orchestrated switches in integrin expression pattern, towards a pro-migratory and malignant phenotype. Moreover, Ets-1 concomitantly triggered matrix metalloproteinases (MMP) expression and activation, thus contributing to cell scattering. Functional relevance of these observations was confirmed with blocking antibodies or MMP inhibitors. Our data highlight a critical role for Ets-1 in the orchestration of a network of molecular and phenotypic events, converging to enhance malignant features and invasion by mammary cancer cells of their environment. Ets-1 overexpression hence appears as a probable key step for breast cancer progression.

Pea3 transcription factor cooperates with USF-1 in regulation of the murine bax transcription without binding to an Ets-binding site.

The Pea3 transcription factor (which belongs to the PEA3 group) from the Ets family has been shown to be involved in mammary embryogenesis and oncogenesis. However, except for proteinases, only few of its target genes have been reported. In the present report, we identified bax as a Pea3 up-regulated gene. We provide evidence of this regulation by using Pea3 overexpression and Pea3 silencing in a mammary cell line. Both Pea3 and Erm, another member of the PEA3 group, are able to transactivate bax promoter fragments. Although the minimal Pea3-regulated bax promoter does not contain an Ets-binding site, two functional upstream stimulatory factor-regulated E boxes are present. We further demonstrate the ability of Pea3 and USF-1 to cooperate for the transactivation of the bax promoter, mutation of the E boxes dramatically reducing the Pea3 transactivation potential. Although Pea3 did not directly bind to the minimal bax promoter, we provide evidence that USF-1 could form a ternary complex with Pea3 and DNA. Taken together, our results suggest that Pea3 may regulate bax transcription via the interaction with USF-1 but without binding to DNA.

Prognostic value of ERM gene expression in human primary breast cancers.

We measured the expression of ERM gene, a nuclear transcription factor belonging to the ets family, in a series of 364 unselected primary breast cancers from patients who underwent locoregional surgery in the Centre Oscar Lambret between May 1989 and December 1991. The expression of ERM was quantified with a real-time one-step reverse transcription-PCR assay based on the 5'-nuclease activity of the TaqDNA polymerase and with an Abi Prism 7700 Sequence Detector System (Applied Biosystems, Courtaboeuf, France). ERM was positively correlated (Spearman test) to epidermal growth factor receptor (EGFR; P < 0.001, r = 0.296) and to histoprognostic grading (P = 0.044, r = 0.112), whereas it was negatively correlated to estradiol receptors (P = 0.019, r = -0.124), HER3 (c-erbB-3; P = 0.01, r = -0.135), and HER4 (c-erbB-4; P = 0.003, r = -0.154). Using the chi2 test, a positive relationship was found between the expression of ERM and EGFR (chi2 = 7.795, P = 0.007). In overall survival studies, Cox univariate analyses demonstrated a prognostic value of ERM (P = 0.006; risk ratio, 2.95) besides the classical prognostic factors histoprognostic grading, node involvement, tumor size, estradiol receptors, progesterone receptors, EGFR, HER3, and HER4. In multivariate analyses, ERM preserved its prognostic value (P = 0.004; risk ratio, 3.779) together with histoprognostic grading, tumor size, estradiol receptors, and progesterone receptors. In relapse-free survival studies, univariate analyses demonstrated that histoprognostic grading, node involvement, tumor size, and HER4 were prognostic factors. These parameters, except histoprognostic grading, retained their prognostic value in multivariate analyses. This study demonstrates for the first time that ERM gene expression is an independent adverse prognostic factor for overall survival in breast cancer patients.

PEA3 transcription factors are expressed in tissues undergoing branching morphogenesis and promote formation of duct-like structures by mammary epithelial cells in vitro.

The genetic program that controls reciprocal tissue interactions during epithelial organogenesis is still poorly understood. Erm, Er81 and Pea3 are three highly related transcription factors belonging to the Ets family, within which they form the PEA3 group. Little information is yet available regarding the function of these transcription factors. We have previously used in situ hybridization to compare their expression pattern during critical stages of murine embryogenesis [Oncogene 15 (1997), 937; Mech. Dev. 108 (2001), 191]. In this study, we have examined the expression of PEA3 group members during organogenesis of the lung, salivary gland, kidney, and mammary gland. In all of these developmental settings, we observed a tight correlation between branching morphogenesis and the expression of specific members of the PEA3 group. To assess the functional relevance of these findings, Erm and Pea3 were overexpressed in the TAC-2.1 mammary epithelial cell line, which has the ability to form branching duct-like structures when grown in collagen gels. We found that overexpression of Erm and Pea3 markedly enhances branching tubulogenesis of TAC-2.1 cells and also promotes their invasion into a collagen matrix. Collectively, these findings suggest that the differential expression of PEA3 group transcription factors has an important role in the regulation of branching morphogenesis and raise the question of their implication in branching signaling.