Retinoic acid receptor antagonist BMS453 inhibits the growth of normal andmalignant breast cells without activating RAR-dependent gene expression

To elucidate the role of RAR-dependent gene transcription in inhibiting bre ast cell growth, we have investigated the ability of retinoids to suppress growth of normal, immortal, and malignant breast cells. We compared the abi lity of all trans retinoic acid (atRA) to activate retinoid receptors in no rmal, immortal, and malignant breast cells, with its ability to inhibit the growth of these cells. Our studies demonstrate that normal breast cells ar e more sensitive to the growth inhibitory effect of atRA than are immortal nonmalignant breast cells and breast cancer cells. atRA activated RAR-depen dent gene transcription in both atRA-sensitive and -resistant breast cells as determined by transfection of a RARE-containing reporter gene. These res ults demonstrate that activation of RAR-dependent gene transcription by atR A is not sufficient to inhibit growth in atRA-resistant breast cancer cells . To determine whether activation of RAR-dependent gene transcription by at RA is necessary for growth inhibition, we tested the growth suppressive eff ect of a retinoid (BMS453) which binds RAR receptors and transrepresses AP- 1 but does not activate RAR-dependent gene expression. This retinoid inhibi ted the growth of normal breast cells (HMEC and 184) and T47D breast cancer cells. Breast cancer cells which were resistant to atRA, were also resista nt to BMS453. Normal human breast cells were most sensitive to the anti-pro liferative effects of BMS453. These results indicate that in some breast ce lls RAR-dependent transactivation is not necessary for retinoids to inhibit growth. Instead, retinoids may suppress growth by inhibiting transcription factors such as AP-1 through transcription factor crosstalk.