Cellular variations in estrogen receptor mRNA translation in the developing brain: evidence from combined [ 125I]estrogen autoradiography and non-isotopic in situ hybridization histochemistry

Abstract

The spatial distribution of cells in the adult rodent forebrain which express estrogen receptor mRNA, as shown by in situ hybridization histochemistry with isotopically-labeled probes, has been reported to overlap with regions that are known targets of estrogen and which bind estrogen. The extent to which detection of estrogen receptor mRNA within developing forebrain neurons of the postnatal day 10–12 female rat is accompanied by translation into estrogen binding sites was investigated by combining [ 125I]estrogen autoradiography with non-isotopic (digoxigenin) in situ hybridization, using a 48-base oligodeoxyribonucleotide probe encoding a sequence of the estrogen-binding domain of rat uterine estrogen receptor cDNA. Estrogen receptor mRNA and estrogen binding sites appeared to be restricted to neurons. No mRNA or binding was seen in ependymal cells. Cells expressing estrogen receptor mRNA were widely distributed in the developing rat forebrain and were found in brain regions generally corresponding to those previously shown in the adult, with the addition of some regions not previously described, such as the medial habenula and dorsal endopiriform nucleus. Although there was widespread overlapping of estrogen receptor mRNA expression with known estrogen binding sites, there were regional and cellular variations in the extent of receptor mRNA translation. This pattern was true for developing forebrain regions previously defined as estrogen receptor-containing (hypothalamus, preoptic area, medial and lateral septum, vertical and horizontal nuclei of the diagonal band, cerebral cortex, hippocampus and amygdala) as well as for regions heretofore not considered estrogen targets (the thalamus, dorsal endopiriform nucleus, claustrum, ventral pallidum/substantia innominata and the basal nucleus of Meynert) or characterized as estrogen-responsive in the adult without previously documented estrogen binding [caudate-putamen (striatum)]. While estrogen binding and receptor mRNA expression always co-localized, neurons expressing estrogen receptor mRNA did not always exhibit ligand binding and there was no clear-cut relationship between the intensity of the hybridization signal and estrogen binding. Little, however, is known about translational control of estrogen receptor expression in the brain. Localization of estrogen binding sites to regions not generally considered targets of estrogen would appear to reflect the greater sensitivity of the iodinated ligand than the tritiated estrogens more commonly used for autoradiography. Non-isotopic in situ hybridization histochemistry combined with [ 125I]estrogen autoradiography represents a very powerful tool with which to study regulation of estrogen receptor gene expression at the single cell level with an exceptional degree of cellular and anatomical resolution. These findings suggest that the extent of estrogen receptor mRNA translation in the developing brain has been underestimated. Differential patterns of translation may be related to regional differences in the ontogeny of the estrogen receptor and to developmentally-regulated differences in post-translational processing as well as to the autoradiographic method itself, particularly with respect to regional differences in the length of exposure required.