I. Introduction II. Mechanisms of Estrogen Action A. “Genomic” and “nongenomic” mechanisms B. Steroid hormone actions on gene expression C. Subtypes of estrogen receptors D. Steroid hormone actions on putative receptors on membranes E. Rapid actions of steroids on neuronal excitability F. Steroid hormone actions via second messengers G. Neuroprotective effects of estrogens H. Summary III. Areas of the Brain Affected Outside of the Hypothalamus A. Studies of hypothalamic and extrahypothalamic actions of estrogens B. Estrogens and the cholinergic system C. Estrogens and the serotonergic system D. Catecholaminergic neurons E. Spinal cord F. Hippocampus G. Glial cells, endothelial cells, and the blood-brain barrier H. Summary IV. Effects of Estrogens on Learning and Memory V. Estrogens, Neuroprotection, and Alzheimer’s Disease VI. Conclusions

Estrogen actions in the central nervous system.

According to the traditional model, steroid hormones bind to intracellular receptors and subsequently modulate transcription and protein synthesis, thus triggering genomic events finally responsible for delayed effects. Based upon similarities in molecular structure, specific receptors for steroids, vitamin D3 derivatives, thyroid hormone, retinoids, and a variety of orphan receptors are considered to represent a superfamily of steroid receptors. In addition, very rapid effects of steroids mainly affecting intracellular signaling have been widely recognized that are clearly incompatible with the genomic model. These rapid, nongenomic steroid actions are likely to be transmitted via specific membrane receptors. Evidence for nongenomic steroid effects and distinct receptors involved is presented for all steroid groups including related compounds like vitamin D3 and thyroid hormones. The physiological and clinical relevance of these rapid effects is still largely unclear, but their existence in vivo has been clearly shown in various settings including human studies. Drugs that specifically affect nongenomic steroid action may find applications in various clinical areas such as cardiovascular and central nervous disorders, electrolyte homeostasis, and infertility. In addition to a short description of genomic steroid action, this review pays particular attention to the current knowledge and important results on the mechanisms of nongenomic steroid action. The modes of action are discussed in relation to their potential physiological or pathophysiological relevance and with regard to a cross-talk between genomic and nongenomic responses.

Multiple Actions of Steroid Hormones—A Focus on Rapid, Nongenomic Effects

Non-genomic and genomic effects of steroids on neural activity

Traditionally, steroid hormone action has been described as the modulation of nuclear transcription, thus triggering genomic events that are responsible for physiological effects. Despite early observations of rapid steroid effects that were incompatible with this theory, nongenomic steroid action has been widely recognized only recently. Evidence for these rapid effects is available for steroids of all clones and for a multitude of species and tissues. Examples of nongenomic steroid action include rapid aldosterone effects in lymphocytes and vascular smooth muscle cells, vitamin D3 effects in epithelial cells, progesterone action in human sperm, neurosteroid effects on neuronal function, and vascular effects of estrogens. Mechanisms of action are being studied with regard to signal perception and transduction, and researchers have developed a patchy sketch of a membrane receptor-second messenger cascade similar to those involved in catecholamine and peptide hormone action. Many of these effects appear to involve phospholipase C, phosphoinositide turnover, intracellular pH and calcium, protein kinase C, and tyrosine kinases. The physiological and pathophysiological relevance of these effects is unclear, but rapid steroid effects on cardiovascular, central nervous, and reproductive functions may occur in vivo. The cloning of the cDNA for the first membrane receptor for steroids should be achieved in the near future, and the physiological and clinical relevance of these rapid steroid effects can then be established.

Specific, nongenomic actions of steroid hormones

In female rats the gonadal hormones estrogen and progesterone modulate dopamine (DA) activity in the striatum and nucleus accumbens. For example, there is estrous cycle-dependent variation in basal extracellular concentration of striatal DA, in amphetamine (AMPH)-stimulated DA release, and in striatal DA-mediated behaviors. Ovariectomy attenuates basal extracellular DA, AMPH-induced striatal DA release, and behaviors mediated by the striatal DA system. Estrogen rapidly and directly acts on the striatum and accumbens, via a G-protein-coupled external membrane receptor, to enhance DA release and DA-mediated behaviors. In male rats, estrogen does not affect striatal DA release, and removal of testicular hormones is without effect. These effects of estrogen also result in gender differences in sensitization to psychomotor stimulants. The effects of the gonadal hormones on the striatum and ascending DA systems projecting to the striatum and nucleus accumbens are hypothesized to occur as follows: estrogen induces a rapid change in neuronal excitability by acting on membrane receptors located in intrinsic striatal GABAergic neurons and on DA terminals. The effect of these two actions results in enhanced stimulated DA release through modulation of terminal excitability. These effects of gonadal hormones are postulated to have important implications for gender differences in susceptibility to addiction to the psychomotor stimulants. It is suggested that hormonal modulation of the striatum may have evolved to facilitate reproductive success in female rats by enhancing pacing behavior.

Gender differences in dopaminergic function in striatum and nucleus accumbens.

Radioiodinated bovine serum albumin conjugated to progesterone was used as a probe to examine binding parameters of steroids to membrane preparations from rat brain tissue. The binding of 11 alpha-hydroxyprogesterone-11-hemisuccinate-125I-bovine serum albumin conjugate reached saturation after 30 min of incubation at 5 degrees C. Several bovine serum albumin-conjugated steroids were then tested for competition displacement studies. Among these steroid conjugates, the bovine serum albumin conjugate at position 3 of progesterone had the highest affinity, with an estimated inhibition constant of 28.5 +/- 2.1 nM (n = 3), whereas bovine serum albumin itself and the 17 beta-estradiol 6-(O-carboxy-methyl)oxime-bovine serum albumin conjugate showed no specific displacement. In addition, the binding sites were localized in an axolemma-enriched fraction of rat brainstem. Specific binding was obtained in tissues from cerebral cortex, brainstem, cerebellum, corpus striatum, and hypothalamus, but little or no binding occurred in uterus, ovary, liver, and spleen. The present data indicate that progesterone-125I-bovine serum albumin conjugate can be used as a ligand to study progesterone-membrane receptor interactions.

Ferng‐Chun Ke

Papers

Binding of progesterone to nerve cell membranes of rat brain using progesterone conjugated to 125I-bovine serum albumin as a ligand.