The distribution of cells containing estrogen receptor-alpha (ERalpha) and ERbeta messenger ribonucleic acid in the preoptic area and hypothalamus of the sheep: comparison of males and females.

We have used in situ hybridization to compare the distributions of estrogen receptor alpha (ERalpha) and ERbeta messenger RNA (mRNA)-containing cells in the preoptic area and hypothalamus of ewes and rams. Perfusion-fixed brain tissue was collected from luteal phase ewes and intact rams (n = 4) during the breeding season. Matched pairs of sections were hybridized with sheep-specific, 35S-labeled riboprobes, and semiquantitative image analysis was performed on emulsion-dipped slides. A number of sex differences were observed, with females having a greater density of labeled cells than males (P < 0.001) and a greater number of silver grains per cell (P < 0.01) in the ventromedial nucleus for both ER subtypes. In addition, in the retrochiasmatic area, males had a greater (P < 0.05) cell density for ERalpha mRNA-containing cells than females, whereas in the paraventricular nucleus, females had a greater density (P < 0.05) of ERalpha mRNA-containing cells than males. There was a trend (P = 0.068) in the arcuate nucleus for males to have a greater number of silver grains per cell labeled for ERalpha mRNA. In both sexes, there was considerable overlap in the distributions of ERalpha and ERbeta mRNA-containing cells, but the density of labeled cells within each nucleus differed in a number of instances. Nuclei that contained a higher (P < 0.001) density of ERalpha than ERbeta mRNA-containing cells included the preoptic area, bed nucleus of the stria terminalis, and ventromedial nucleus, whereas the subfornical organ (P < 0.001), paraventricular nucleus (males only, P < 0.05), and retrochiasmatic nucleus (females only, P < 0.05) had a greater density of ERalpha than ERbeta mRNA-containing cells. The anterior hypothalamic area and supraoptic nucleus had similar densities of cells containing both ER subtypes. The lateral septum and arcuate nucleus contained only ERalpha, whereas only ERbeta mRNA-containing cells were seen in the zona incerta. The sex differences in the populations of ER mRNA-containing cells in the ventromedial and arcuate nuclei may explain in part the sex differences in the neuroendocrine and behavioral responses to localized estrogen treatment in these nuclei. Within sexes, the differences between the distributions of ERalpha and ERbeta mRNA-containing cells may reflect differential regulation of the actions of estrogen in the sheep hypothalamus. Low levels of ERbeta mRNA in the preoptic area and ventromedial and arcuate nuclei, regions known to be important for the regulation of reproduction, suggest that ERbeta may not be involved in these functions.

Proopiomelanocortin mRNA levels in ovine hypothalamus are not reduced at the time of the preovulatory luteinising hormone surge.

A reduction in inhibition (disinhibition) of gonadotropin-releasing hormone (GnRH) secretion by endogenous opioid systems in the hypothalamus is thought to be permissive of the preovulatory surge of GnRH and luteinising hormone (LH). In rats, proopiomelanocortin (POMC) mRNA levels in the arcuate nucleus are reduced at the time of the LH surge, and this is thought to be part of the neural mechanism of oestrogen positive feedback. There are no studies of POMC mRNA levels at the time of the LH surge in other species. POMC mRNA levels were measured by in-situ hybridization using a 35S-labelled cRNA probe and computer-assisted grain counting in the arcuate nucleus of ovary-intact ewes (n=4) killed on day 10 of the luteal phase or 24 or 48 h into the follicular phase (experiment 1), and ewes killed on day 10 of the luteal phase or during the preovulatory LH surge (experiment 2). Grain counts per cell, the proportion of cellular area covered by grains and the number of labelled cells per section were not significantly different (P > 0.1) between animals killed in the luteal phase and animals killed during the follicular phase or during the LH surge. We conclude that in sheep, POMC mRNA levels are not reduced at the time of the preovulatory LH surge, and reduced POMC gene transcription does not appear to be part of the neural mechanism of oestrogen positive feedback in this species.

Brain 4: a novel mammalian POU domain transcription factor exhibiting restricted brain-specific expression.

The POU domain gene family of transcription factors share a conserved bipartite DNA binding domain, and exhibit distinct temporal and spatial patterns of expression during development, particularly in the forebrain. A cDNA encoding a new member of the POU-III class of the POU domain gene family, referred to as Brn-4, was isolated from a rat hypothalamic cDNA library. Like other mammalian POU-III genes previously characterized (Brn-1, Brn-2, Tst-1), Brn-4 transcripts are initially widely expressed at all levels of the developing neural tube, but in contrast to other previously described POU-III genes, are subsequently restricted to only a few regions of the adult forebrain, including the supraoptic and paraventricular nuclei of the hypothalamus. Brn-4 was shown to bind to DNA sequences containing the octamer motif and to trans-activate promoters containing this DNA binding motif, based on the actions of a unique N-terminal information. This ontogenic pattern of Brn-4 expression in concert with that of Oct-2 and Pit-1, indicates that certain POU domain genes potentially exert their primary functions widely during early neural development, and in a very limited set of neurons in the mature brain.

Tst-1, a member of the POU domain gene family, binds the promoter of the gene encoding the cell surface adhesion molecule P0.

Tst-1, a member of the POU domain gene family, is expressed in specific neurons and in myelinating glia in the mammalian nervous system. Bacterially expressed Tst-1 binds specifically to the promoter of the gene encoding myelin protein P0, a Schwann cell surface adhesion molecule. In cotransfection assays, Tst-1 can specifically repress the P0 promoter. The N-terminal part of Tst-1 protein is highly glycine- and alanine-rich, a structural feature shared by the helix-loop-helix protein TFEB.

Novel developmental specificity in the nervous system of transgenic animals expressing growth hormone fusion genes.

The development of methods for introducing foreign genes into the germ line of mice provides an approach for studying mechanisms underlying inducible and developmental gene regulation. Transgenic animals expressing foreign genes have thus been used to test models of the role played by specific DNA sequences in determining cell-specific expression. Results from these experiments suggest that tissue-specific expression is the consequence of a cis-acting regulatory sequence. However, these results do not exclude the possibility that cell-specific expression of some genes might be 'coded' by combinations of regulatory elements. We have previously described the production of transgenic mice from eggs microinjected with metallothionein-I/growth hormone (MGH) fusion genes, and now demonstrate that the juxtaposition of sequences from two different genes can be deciphered by cells to generate novel tissue specificities. Although expression of the endogenous metallothionein and growth hormone genes has not been detected in neuronal cells, transgenic mice clearly express an MGH fusion gene in a restricted subset of neurones. These results suggest a model in which tissue-specific patterns of expression of certain genes are determined by combinations of cis-acting regulatory sequences.