Breast cancer progression in MCF10A series of cell lines is associated with alterations in retinoic acid and retinoid X receptors and with differential response to retinoids.

In most breast carcinomas and in breast cancer cell lines, retinoic acid receptor beta (RARbeta) is lost or down-regulated, whereas retinoic acid receptor alpha and gamma (RARalpha, gamma) and retinoid X receptors (RXRalpha, beta, gamma) are variably expressed. Little is known about alterations of the above receptors in hyperplastic and premalignant stages of breast cancer development. In this study, we employed the MCF10A series of breast epithelial cell lines (the parental and benign MCF10A, premalignant MCF10AT, and malignant MCF10CA1a) to assess whether in the course of their malignant transformation specific alterations in RARalpha, beta, gamma and RXRalpha, beta, gamma expression occur and whether they may affect the sensitivity of cells to retinoids. Malignant properties of the above cell lines were estimated by the nude mice xenograft transplantation assay. Among the above receptors most significant alterations occurred in RARbeta2, which was detected in the normal breast epithelial cells both, at mRNA and protein levels, but expressed in the MCF10A cell lines at mRNA level only. The transformation of benign MCF10A cells into premalignant MCF10AT and malignant MCF10CA1a was also associated with increase in RARalpha, RARgamma, RXRalpha, and RXRbeta proteins expression. All-trans retinoic acid (atRA), 9-cis retinoic acid (9cRA), and 4-(hydroxyphenyl) retinamide (4-HPR) induced RARbeta2 protein expression exclusively in the benign MCF10A cells and the former two retinoids, mRNA expression in MCF10A and MCF10AT cells, but not in malignant, MCF10CA1a cells, suggesting that the loss of inducible RARbeta expression is associated with the progression and malignant transformation of MCF10A cells. Retinoids also variable decreased the RARalpha, RARgamma and RXRalpha protein expression preferentially in the premalignant and malignant, but not in benign MCF10A cells. Among the above retinoids, 4-HPR was most efficacious in inhibiting the growth of the three cell lines and this apparently was not dependent on the levels of the RARbeta2 transcriptional activation. Thus, our data support the hypothesis that breast epithelial cells in the course of their progression and malignant transformation may differentially respond to retinoids and that not only RARbeta, but RARalpha, gamma and/or RXRalpha, beta may also serve as potential targets for retinoids in breast cancer prevention and therapy trials.