EUROPROTECTIVE effects of estrogen have been demonstrated against a variety of cytotoxic insults. We present data here addressing a possible mechanism of estrogen neuroprotection in the human teratocarcinoma cell line NT2 terminally differentiated to a neuronal phenotype. Cell death induced by H2O2 or glutamate results in a dose-dependent cell death of NT2 neurons, while 24 h of estrogen pretreatment significantly enhances neuronal viability. Bcl-2 expression has been shown to reduce oxidative stress and prevent cell death. In NT2 neurons, Bcl-2 levels are dramatically elevated upon differentiation and are further enhanced with estrogen treatment. These results suggest that neuroprotective effects of estrogen may be related to increases in Bcl-2 expression.

Early postnatal expression of VP mRNA in the bed nucleus of the stria terminalis (BNST) and medial amygdala (MA) of the male rat has been shown to be testosterone (T) dependent, However, it remains unclear whether these effects of T are mediated exclusively through androgen or estrogen receptors, or both. This study was performed to determine which gonadal hormone receptor(s) is (are) responsible for the neonatal expression of VP and mRNA expression in the BNST and MA. Male and female:rats were given early postnatal (day 1-5) treatment with estrogen or androgen receptor agonists or antagonists and VP mRNA was then detected by in situ hybridization in the BNST and MA. In male rats, tamoxifen (estrogen antagonist) blocked T effects on VP mRNA expression in postnatal day six animals; while the estrogen agonist diethylstilbestrol mimicked the effects of T in the female rat. The data clearly indicate a primary role of the estrogen receptor in mediating the initial expression of VP mRNA in the BNST and MA of male and female rats.

Variations in levels of estrogen receptor mRNA were investigated in the medial preoptic nucleus, arcuate nucleus, and ventromedial nucleus of the hypothalamus throughout the phases of the female estrous cycle and compared with those in ovariectomized female and intact male rats. Female Wistar rats were killed during estrus, metestrus, diestrus, or proestrus or 72 h after ovariectomy as were a group of intact male rats. Brains were removed and frozen, and 20-microns cryostat sections were thaw-mounted onto slides and hybridized with a 35S-labeled antisense estrogen receptor probe. Section-mounted slides were processed, apposed to x-ray film, then dipped in liquid emulsion, and quantified. After exposure, estrogen receptor mRNA was detected in several brain regions, including the medial preoptic nucleus, arcuate nucleus, and ventromedial nucleus of the hypothalamus. Estrogen receptor mRNA levels in the medial preoptic nucleus were highest during estrus and metestrus, attenuated at diestrus, and low during proestrus. In contrast, the hybridization signal in the arcuate and ventromedial nuclei was low during estrus and then gradually increased throughout the cycle until it peaked during proestrus. Ovariectomized females exhibited an elevated level of estrogen receptor mRNA in all brain regions investigated. Hybridization signal in male medial preoptic nucleus and ventromedial nucleus was reduced compared with those in both intact and ovariectomized females. Estrogen receptor mRNA levels in the arcuate nucleus were similar to those in intact females, but less than those in ovariectomized animals. The results of these studies demonstrate that estrogen receptor mRNA levels are sexually dimorphic, vary during the estrous cycle, and increase after ovariectomy. Furthermore, these results indicate that the magnitude and direction of change observed during the estrous cycle are region specific and suggest that factors other than endogenous estrogen levels differentially modulate estrogen receptor mRNA expression in the hypothalamus.

In situ hybridization was performed in Long-Evans (LE) and heterozygous (HET) and homozygous (HOM) Brattleboro rats using a 48-base oligonucleotide complementary to the last 16 amino acids of the vasopressin messenger RNA (VP mRNA). The number of cells expressing the VP mRNA in the supraoptic (SON) and paraventicular (PVN) nuclei was not significantly different between LE, HET, and HOM rats; however, the relative amount of mutant VP mRNA expressed in neurons of the PVN and SON of the HOM rat was significantly lower than that in the other two genotypes. In contrast, the suprachiasmatic nucleus (SCN), the bed nucleus of the stria terminalis (BNST), and the medial amygdala (MA) of the HOM rat showed a significant reduction in both the number of neurons and the level of mutant VP mRNA expression per cell compared to those in either the LE or the HET rat. To determine whether the reduced level of mutant VP mRNA in the HOM rat was due to decreased transcription, in situ hybridization to detect the VP primary transcript was performed. The number of neurons and the amount of nuclear VP RNA expressed per cell for the PVN, SCN, BNST, and MA in the HET and HOM rats were not significantly different from those in the LE rat. However, in the SON, the HOM rat exhibited a significant increase (P < 0.10) in the amount of nuclear VP RNA expressed per cell compared to the LE and HET rats, but the number of positively labeled cells was not significantly different. Therefore, these data suggest that the mutant gene in the HOM rat is transcribed at rates comparable to those in the LE and HET rats, while cytoplasmic mutant VP mRNA is significantly reduced in all nuclei of the HOM rat. The reduced cytoplasmic VP mRNA in the HOM rat may be attributed to instability (degradation) of the mutant message.

The Drosophila learning mutant, rutabaga, is deficient in the calmodulin-sensitive adenylate cyclase, and studies of associative learning in Aplysia have implicated this enzyme in neuroplasticity. Therefore, the distribution of mRNA encoding the calmodulin-sensitive adenylate cyclase in rat brain was examined by in situ hybridization. mRNA for this enzyme is expressed in specific areas of brain that have been implicated in learning and memory, including the neocortex, the hippocampus, and the olfactory system. The presence of mRNA for this enzyme in the pyramidal and granule cells of the hippocampal formation provides evidence that it is found in neurons. These data are consistent with the proposal that the calmodulin-sensitive adenylate cyclase plays an important role in learning and memory.

Vasopressin (AVP)-immunoreactive cells have been previously reported in the medial amygdala (AME) and in the locus coeruleus (LC). The present study was designed to verify the presence of AVP-synthesizing neurons in these areas using in situ hybridization histochemistry. A 35S-labelled oligonucleotide probe, complementary to the glycopeptide portion of the vasopressin-encoding mRNA, was used to label cells expressing the AVP gene in brain sections from male Wistar rats. AVP mRNA-positive cells were identified in the AME and were located throughout the anterodorsal and posterodorsal aspect of the nucleus. Cells in the LC, however, did not exhibit labelling for the glycopeptide portion of the AVP gene. The highest density of labelled cells in the medial amygdala occurred 2.30 to 2.80 mm caudal to bregma. The labelling intensity of the cells averaged 53.8 +/- 3.9 grains/cells and was constant throughout the rostro-caudal extent of the AME. These data demonstrate the presence of AVP-synthesizing cells in the AME and provide a method for quantifying their activity. In addition, these data suggest that the cells in the LC may not synthesize vasopressin.

Recent immunocytochemical studies have suggested that vasopressin (VP) neurons in the bed nucleus of the stria terminalis (BNST) of the rat are gonadal steroid sensitive. In this paper we have used in situ hybridization and quantitative autoradiography to determine whether testosterone (T) and/or its metabolites modulate the biosynthetic capacity of VP neurons in the BNST of adult male rats. In Exp 1 the number of labeled cells and the average number of grains per cell were compared in sections sampled through the BNST of intact, castrated, and castrated male rats treated with physiological levels of T (1.6 +/- 0.1 ng/ml plasma). Castration dramatically reduced the number of labeled cells (P less than 0.01) and the intensity of labeling (P less than 0.05) of cells in the BNST. T, treatment of castrated animals reversed the effect of castration on both cell number and grains per cell. In Exp 2 treatment of castrated rats with supraphysiological levels of T (7.6 +/- 0.7 ng/ml plasma) increased the number of labeled BNST cells (P less than 0.05) and the intensity of labeling (P less than 0.05) over those in castrates treated with physiological levels of T or intact rats. These results indicate that T and/or its metabolites modulate expression of the VP gene by neurons in the BNST of adult male rats.

The β-lipotropin fragments, [des-Tyr1]-γ-endorphin (DTγE, β-LPH62–77) and α-endorphin (β-LPH61–76) affect self-stimulating behavior associated with electrical stimulation of neurons of the ventral tegmentum area of rats in an opposite way. Subcutaneous administration of DTγE (5 and 25 μg) attenuated and that of α-endorphin (5 and 25 μg) facilitated this behavior. Similar opposite effects were observed after subcutaneous treatment with respectively the neuroleptic haloperidol (5 μg) and the psychostimulant amphetamine (100 μg). By using a biphasic testparadigm of decreasing and subsequent increasing the stimulating current intensity it was noted that the neuropeptides predominantly exerted their effect on responding at current intensities in the neighbourhood of the threshold for eliciting the behavior, whereas the neuroleptic and psychostimulant drug appeared to affect responding at currents associated with maximal performance as well. In contrast to haloperidol, the effectiveness of DTγE was of a long term nature, in that performance of the rat was still affected 24 hr after peptide treatment. The results support the hypothesis that DTγE in some aspects interacts with brain substrates in a way comparable to that of neuroleptics. The data further suggest that closely related fragments of β-lipotropin modulate on-going activity of in particular dopaminergic neuronal systems.