Sex steroid-induced neuroplasticity and behavioral activation in birds

Abstract

The brain of adult homeothermic vertebrates exhibits a higher degree of morphological neuroplasticity than previously thought, and this plasticity is especially prominent in birds. In particular, incorporation of new neurons is widespread throughout the adult avian forebrain, and the volumes of specific nuclei vary seasonally in a prominent manner. We review here work on steroid-dependent plasticity in birds, based on two cases: the medial preoptic nucleus (POM) of Japanese quail in relation to male sexual behavior, and nucleus HVC in canaries, which regulates song behavior. In male quail, POM volume changes seasonally, and in castrated subjects testosterone almost doubles POM volume within 2 weeks. Significant volume increases are, however, already observable after 1 day. Steroid receptor coactivator-1 is part of the mechanism mediating these effects. Increases in POM volume reflect changes in cell size or spacing and dendritic branching, but are not associated with an increase in neuron number. In contrast, seasonal changes in HVC volume reflect incorporation of newborn neurons in addition to changes in cell size and spacing. These are induced by treatments with exogenous testosterone or its metabolites. Expression of doublecortin, a microtubule-associated protein, is increased by testosterone in the HVC but not in the adjacent nidopallium, suggesting that neuron production in the subventricular zone, the birthplace of newborn neurons, is not affected. Together, these data illustrate the high degree of plasticity that extends into adulthood and is characteristic of avian brain structures. Many questions still remain concerning the regulation and specific function of this plasticity.

Figure 1

Testosterone-induced plasticity in the medial preoptic nucleus (POM) of the Japanese quail. A. Localization of the POM in the quail brain. The nucleus runs through most of the preoptic area from the level of the tractus septopallio-mesencephalicus (TSM) to the level of the anterior commissure (CA). Gray area denote the presence of dense populations of aromatase cells in the POM and bed nucleus of the stria termialis, medial part (BSTm). E: entopallium; FPL: fasciculus prosencephali lateralis; Hp: hippocampus; N: nidopallium; SL: lateral septal nucleus; SM: medial septal nucleus; VLT: nucleus ventrolateralis thalami. B. Effect of gonadectomy associated or not with a treatment by testosterone on the total volume of the POM (left) and on the average neuronal size in the lateral POM, reflected by the cross sectional area. C. Histological sections through the left POM stained by immunohistochemistry for aromatase illustrating the decrease in number of aromatase-immunoreactive (ARO-ir) cells observed when intact birds are castrated (CX) and the recovery following treatment with exogenous testosterone (CX+T). Magnification bar= 400 μm. D. Effect of gonadectomy associated or not with a treatment by testosterone on the average numbers of ARO-ir cells in the POM (left) and on the specific numbers of these cells located in the medial or lateral parts of the nucleus (right). Data in the two parts of this panel come from different studies in which quantification was performed by different methods so that absolute numbers of ARO-ir cells cannot be directly compared. All quantitative data are means ± SEM. Redrawn from data in (Panzica et al., 1987; Harada et al., 1993; Aste et al., 1994) and (Balthazart & Ball, 2007).

Figure 2

Doublecortin (DCX) expression in the quail brain. DCX-immunoreactive cells are present in large numbers at the level of the lateral ventricle in the telencephalon (A) as well as at the lateral border of the nidopallium where they seem to be migrating in a tangential direction (B). In contrast, no clearly immunoreactive cells are seen in the medial preoptic area where only a few spots of non specific staining can be detected (C). CA: anterior commissure; VL: lateral ventricle; VIII: third ventricle. Magnification bar = 200 μm (Balthazart, J. and Charlier, T.D., unpublished data).

Figure 3

Rapid effects of testosterone on two aspects of male sexual behavior, female neck grab (A) and cloacal contact movements (B) and on the volume of the quail POM defined in Nissl stained sections (C) and based on the dense group of aromatase-immunoreactive cells that identifies the POM (D). Panel E shows photomicrographs illustrating the group of aromatase-immunoreactive cells located in the POM in males that were castrated (CX) or CX and treated with testosterone (CX+T) for one or fourteen days (d). Groups of CX subjects were tested as killed at day 1 or 14 (CX) while CX+T birds were studied at days 1, 2, 7 and 14. All quantitative data are means ± SEM. Redrawn from data in (Charlier et al., 2008).

Figure 4

Schematic representation of the different steps leading from the division of progenitor cells to the incorporation of new neurons and their death in a given brain nucleus (top) and photomicrographs of representative cells at different stages of this process. At any given point in time, various numbers of cells are engaging into a neuronal (as opposed to glial) fate of differentiation (N1), migrate toward their final target (N2), are recruited and differentiate in this target nucleus (N3), are incorporated in this target as mature functional neurons (N4) and finally die by apotosis or necrosis (N5). Only a fraction of neurons at a given stage will progress to the next stage and they will stay in each condition for variable durations (t1, t2, t3). Testosterone could potentially modulate each of these processes and their duration. So far evidence suggests that effects of the steroid take place mainly at the level of the differentiation and survival of neurons (arrows) but actions at other levels cannot be excluded (dotted arrows). The photomicrographs in the bottom row represent cells labeled by BrdU in the ventricle wall (left), an elongated (fusiform) migrating young neuron labeled by immunocytochemistry for DCX (middle left), a slightly older neuron that has started its differentiation but is still immunoreactive for DCX (middle right) and an apoptotic cell identified by its pyknotic nucleus (right).

Figure 5

Effects of testosterone (Testosterone vs. Control; top) or of the social environment (Male-Male [M-M] vs. Male-Female [M-F] housing; bottom) on the volume of HVC (left) and on the number of fusiform (middle) or round (right) doublecortin (DCX)-immunoreactive cells in HVC and in two equivalent areas located laterally (l.HVC) or ventrally (v.HVC) to HVC. The brain drawing in the top-right side schematically illustrates the distribution of DCX-immunoreactive cells in the brain and the location of areas in which systematic quantification was performed. See also text for additional explanations. *= p<0.05 compared to the other experimental group. Redrawn from data in (Balthazart et al., 2008a).