A clinically relevant bi-cellular murine mammary tumor model as a useful tool for evaluating the effect of retinoic acid signaling on tumor progression

Background The effect of retinoic acid (RA) on breast cancer progression is controversial. Our objective was to obtain information about breast cancer progression, taking advantage of the ER-negative murine mammary adenocarcinoma model LM38 (LM38-LP constituted by luminal (LEP) and myoepithelial-like cells (MEP), LM38-HP mainly composed of spindle-shaped epithelial cells, and LM38-D2 containing only large myoepithelial cells), and to validate the role of the retinoic acid receptors (RARs) in each cell-type compartment. Materials and methods We studied the expression and functionality of the RARs in LM38 cell lines. We analyzed cell growth and cell cycle distribution, apoptosis, the activity of proteases, motility properties, and expression of the molecules involved in these pathways. We also evaluated tumor growth and dissemination in vivo under retinoid treatment. Results LM38 cell lines expressed most retinoic receptor isotypes that were functional. However, only the bi-cellular LM38-LP cells responded to retinoids by increasing RARβ2 and CRBP1 expression. The growth of LM38 cell sublines was inhibited by retinoids, first by inducing arrest in MEP cells, then apoptosis in LEP cells. Retinoids induced inhibitory effects on motility, invasiveness, and activity of proteolytic enzymes, mainly in the LM38-LP cell line. In in-vivo assays with the LM38-LP cell line, RA treatment impaired both primary tumor growth and lung metastases dissemination. Conclusion These in-vivo and in-vitro results show that to achieve maximum effects of RA on tumor progression both the LEP and MEP cell compartments have to be present, suggesting that the interaction between the LEP and MEP cells is crucial to full activation of the RARs.