In this issue of BioEssays, Remage-Healey describes an elegant body of work demonstrating that estradiol, traditionally considered an ovarian hormone, is synthesized within the brain and affects local neuronal activity 1. In zebra finches, hearing conspecific song induces estradiol release in the auditory forebrain, which rapidly alters firing properties both in that area and in downstream sensorimotor regions.
The use of songbirds for this work magnifies its impact; historically, they have been among the most important models for studying the regulation of brain and behavior by steroids of gonadal origin. One of the first demonstrations of large-scale plasticity in the adult brain was the discovery in 1981 of seasonal changes in the circuits controlling courtship vocalizations in canaries 2. That work gave rise to an explosion of research on how gonadal steroids affect communication pathways – not only in songbirds but also in fish, amphibians, and mammals. The prevailing view, which still makes a lot of sense, was that the rise in plasma gonadal steroids at the onset of the breeding season coordinates courtship vocalizations with reproductive readiness.
Over the next 20 years, researchers studying birds began to question that view. Gonadectomy failed to reduce plasma estradiol 3, and it was ultimately discovered that the avian brain can produce estrogens de novo, without peripheral androgen precursors 4. Membrane steroid receptors, as well as rapid actions of steroids on both behavior and neuronal activity, began to be reported (reviewed in ref. 5). Today, evidence is overwhelming that the avian brain makes its own estradiol, and that estradiol has rapid effects on neural responses and behavior.
That estradiol is a neuromodulator should not have come as a surprise. Estrogen-mediated signaling mechanisms are among the most ancient forms of intercellular communication, having evolved more than 600 million years ago 6. They are likely to function in all kinds of sensory processes, not just those involved in vocal communication. The phenomenon has already been demonstrated in visual cortex 7 and will likely prove ubiquitous in other sensory areas not previously thought to be regulated by steroids. We should expect that estradiol, like other neuromodulators, plays a fundamental role in learning and attention. In songbirds, the auditory forebrain is well-understood to respond selectively to behaviorally relevant sounds such as conspecific song, and this response includes estradiol release. It will be interesting to see whether other kinds of behaviorally relevant sounds, for example non-song sounds that have acquired meaning via associative learning, also engage estrogen signaling pathways.
Estradiol is not the first hormone-turned-neuromodulator. Remage-Healey points out a parallel between estradiol and pituitary hormones such as oxytocin that are also synthesized and released in the brain. Oxytocin's neural effects on affiliation, which are limited to a few brain areas and clearly distinct from its peripheral effects, have led to mass confusion both in the scientific literature and in the popular press; the neuromodulator is now widely regarded as a “love hormone” that modulates social responses by crossing into the brain from the circulation. In this case, the roles of hormone and neuromodulator have been unfortunately lumped. Those of us who study the neural effects of estradiol, which unlike oxytocin can easily cross into and out of the brain, must be careful not to muddy our own waters by considering just one or the other source. Ongoing work with seasonally breeding species is showing that the ability to synthesize neurosteroids may be heavily influenced by endocrine state, in part defined by plasma levels of gonadal hormones 8. The actions of estradiol and other steroids as neuromodulators are best interpreted in light of the whole picture – which includes both brain and gonads.