No abstract

No abstract

Summary All-trans-retinoic acid (ATRA) is well known to inhibit the proliferation of human breast cancer cells. Much less is known about the antiproliferative activity of the naturally occurring metabolites and isomers of ATRA. In the present study, we investigated the antiproliferative activity of ATRA, its physiological catabolites 4-oxo-ATRA and 5,6-epoxy-ATRA and isomers 9-cis-RA and 1 3-cis-RA in MCF-7 human breast cancer cells by bromodeoxyuridine incorporation. MCF-7 cells were grown in steroid-and retinoid-free medium supplemented with growth factors. Under these culture conditions, ATRA and its naturally occurring catabolites and isomers showed significant antiproliferative activity in MCF-7 cells in a concentration-dependent manner (10-11 M to 10-M). The antiproliferative activity of ATRA catabolites and isomers was equal to that of the parent compound ATRA at concentrations of 10-M and 1 07 M. Only at 10-6 M were the catabolites and the stereoisomer 1 3-cis-RA less potent. The stereoisomer 9-cis-RA was as potent as ATRA at all concentrations tested (1 0-11 M to 10-M). In addition, we show that the catabolites and isomers were formed from ATRA to only a limited extent. Together, our findings suggest that in spite of their high antiproliferative activity the catabolites and isomers of ATRA cannot be responsible for the observed growth inhibition induced by ATRA.

A single-step procedure was devised to separate PRL cells from the rat anterior pituitary gland. After dissociation, cells were centrifuged on a Percoll gradient. Three layers were recovered. The composition of the different layers was evaluated using immunocytochemistry (with antisera to the six pituitary hormones), and in situ hybridization [with DNA complementary to PRL or to GH messenger RNA (mRNA)]. Both methods yielded identical values. PRL cells were recovered in the lower density layer (layer 1) with a good yield (that is 81% of the total PRL cells of the initial cell suspension) and in addition, markedly enriched (indeed 85% of the cells in layer 1 stained for PRL). A second layer (layer 2: intermediate density) contained most of the remaining PRL cells which were, however, heavily contaminated mainly by GH cells and cells that did not stain for any of the known pituitary hormones. A third layer (layer 3: higher density) was enriched in GH cells to 93% (representing, however, only 10% of the initial pituitary GH cells). In addition, PRL and GH were measured by RIA in culture medium and in cell lysates. Hormone biosynthesis was monitored by polyacrylamide gel electrophoresis and autoradiography after culture in the presence of [35S]methionine. These experiments confirmed that layer 1 was enriched in cells containing, and producing, PRL and depleted from GH cells. Cells in layer 2 contained and produced more GH than PRL. PRL cells from layer 1 responded to dopamine and to vasoactive intestinal polypeptide in the same way as PRL cells in the unseparated pituitary cell population. In contrast PRL cells in layer 2 had a lower basal secretion rate but a higher response to vasoactive intestinal polypeptide. Unless this represents a paracrine effect of non-PRL cells, PRL cells in layer 2 exhibit different properties and may therefore form a distinct subpopulation of PRL cells.