Photoperiod regulates neuronal bromodeoxyuridine labeling in the brain of a seasonally breeding mammal.

Seasonal changes in vertebrate brain function are pervasive, but annual cycles in the rates of neuronal incorporation are established only in songbirds. Although cell division continues in the subependymal and hippocampal subgranular zones of adult rodents, there exists no parallel evidence that seasonal plasticity in mammals extends to changes in neuronal or glial number. We examined the effect of photoperiod on incorporation of new neurons in the brain of the adult golden hamster, a long-day breeder. We administered the cell birth marker 5'-bromodeoxyuridine (BrdU) to males which had either been maintained in long days, transferred to short days for 10 weeks, or moved acutely from long to short or short to long days. The number of cells in specific brain regions immunoreactive (ir) for this thymidine analog was determined 7 weeks later. The number of BrdU-ir cells in the dentate gyrus and subependymal zone increased twofold in short days. Transfer between photoperiods 10 days before the BrdU injections produced intermediate numbers of BrdU-labeled cells in the dentate gyrus, but was as effective as long-term photoperiodic exposure in the subependymal zone. Photoperiod also had similar effects in the hypothalamus and cingulate/retrosplenial cortex, but not in the central gray or preoptic area. Double-label immunocytochemistry indicated that very few of the BrdU-ir cells were glia, but that a majority had neuronal phenotype. In the subependymal zone, short days significantly increased the number of BrdU-labeled neurons. We did not detect significant effects of photoperiod on the volume of either the granule cell layer of the hippocampus or the dentate gyrus as a whole. We conclude that short day lengths increase neuronal birth and/or survival in several brain regions of adult hamsters.

Effects of photoperiod and androgen on pituitary function and neuropeptide staining in Siberian hamsters.

Short photoperiods decrease gonadotropin secretion in Siberian hamsters, but it is unknown whether the negative feedback effects of androgens are amplified under such conditions, as is the case in other species. Photoperiod regulates the synthesis and secretion of gonadotropin-releasing hormone (GnRH), beta-endorphin, and arginine vasopressin (AVP), which influence gonadotropin release and sexual behavior but are themselves regulated by gonadal steroid hormones. To determine the role of androgen in these effects of daylength, immunostaining and gonadotropin concentrations were examined after 8 wk of exposure to long or short days (LD or SD). Animals were either left intact, castrated, or castrated with immediate or delayed replacement of testosterone (T). We also investigated effects of age on photoperiodic influences on brain peptides and serum hormone levels. Serum prolactin concentrations were regulated by photoperiod and by gonadal status in LD hamsters. Effects of T on follicle-stimulating hormone secretion were more pronounced in SD hamsters. Older hamsters were generally less responsive to effects of daylength on pituitary function. Photoperiod and gonadal status regulated the number of AVP-immunoreactive (ir) cells in the bed nucleus of the stria terminalis and the medial amygdala. Androgen treatment yielded more AVP-ir neurons in LD than in SD. Photoperiod influenced the number of GnRH-ir cells only in the medial septum of castrated hamsters. Daylength regulated beta-endorphin-ir neurons in intact hamsters, but not in castrates. Only among old hamsters did photoperiod affect the influence of T on beta-endorphin staining in neurons and fibers. Such fiber staining was unaffected by photoperiod in intact and T-treated castrate hamsters, but was reduced in SD castrates. We conclude that daylength modulates the effects of androgen on gonadotropin secretion and influences the effect of T on neuropeptide staining in regionally specific patterns that depend on the age of the animal and its history of prior steroid exposure.

Distribution of GAD-like immunoreactivity in the retina and central visual system of Rana pipiens.

Immunohistochemistry was used to survey the distribution of glutamic acid decarboxylase (GAD), the synthesizing enzyme for gamma-aminobutyric acid (GABA), throughout the visual system of the frog Rana pipiens. GAD-like immunoreactivity (GAD-LI) was assessed in the retina, in primary retinorecipient targets, and in thalamic nuclear groups postsynaptic to primary retinal terminal zones. Within the retina, the inner plexiform layer displayed intense GAD-LI, but immunoreactivity was absent in the ganglion cell layer (GCL). Putative amacrine, bipolar, and horizontal cell somata were also labeled. Centrally, GAD-LI was observed in all primary visual nuclei. Dense, GAD-like immunoreactive punctate structures (puncta), presumably terminals, were observed in the pretectal nucleus lentiformis mesencephali, posterior thalamic neuropil, and uncinate neuropil. GAD-like immunoreactive puncta were noted in several laminae of the optic tectum, with the highest concentrations located within the 9th and 8th laminae. Moderate numbers of GAD-like immunoreactive puncta were found in the mesencephalic nucleus of the basal optic root, and two thalamic neuropils--corpus geniculatum and neuropil of Bellonci. The ventrolateral area, posterocentral nucleus, and posterolateral nucleus all contained sparse amounts of GAD-LI. These observations suggest that GABA plays an important physiological role in all central visual areas of Rana pipiens.

Decline of vasopressin immunoreactivity and mRNA levels in the bed nucleus of the stria terminalis following castration.

Vasopressinergic (VP) neurons in the bed nucleus of the stria terminalis (BNST) of the rat are regulated by gonadal steroids. Gonadectomy causes the projections of the BNST to lose their VP immunoreactivity gradually over a period lasting more than 2 months. Here we have compared the rate of decline of VP mRNA and VP immunoreactivity in the BNST of adult male rats following castration. In experiment 1, the peak number of VP-immunoreactive cells and the level of VP gene expression were compared in sham-operated controls and at 1, 3, or 8 weeks postcastration. The number of VP-immunoreactive cells was not decreased at 1 week postcastration but was significantly reduced (p less than 0.0001) at 3 and 8 weeks postcastration. VP gene expression declined more rapidly, and both the total number of labeled cells (p less than 0.0001) and the average number of grains per cell (p less than 0.01) were significantly reduced by 1 week postcastration. No VP-expressing cells were detectable at 3 or 8 weeks. The difference in the rate of decline in the number of cells labeled by the two techniques following castration did not appear to be due to colchicine pretreatment. In experiment 2, VP mRNA in the BNST was compared in sham-operated controls or at 1, 3, or 7 d postcastration. A significant decrease (p less than 0.01) in the average number of grains per cell was detectable by just 1 d following castration, and the number of labeled cells was significantly reduced (p less than 0.001) by 3 d postcastration.(ABSTRACT TRUNCATED AT 250 WORDS)

Suprachiasmatic and paraventricular control of photoperiodism in Siberian hamsters.

The effects of lesions of the suprachiasmatic (SCN) and paraventricular nuclei (PVN) of the hypothalamus on photoperiodic responses were examined in adult Siberian hamsters. SCN lesions reduced nocturnal water intake in long days, whereas PVN lesions increased body weight and food intake in both short and long days. SCN or PVN lesions blocked short-day-induced decreases in body, fat pad, and testes weights and in food intake. Serum prolactin (PRL), but not follicle-stimulating hormone, levels were increased. The distribution of immunostained neurons and fibers for gonadotropin-releasing hormone (GnRH), beta-endorphin, arginine vasopressin (AVP), and vasoactive intestinal polypeptide (VIP) resembled that of other rodent species. Short-day exposure reduced AVP staining in lateral septum, medial amygdala, and bed nucleus of the stria terminalis but not in the PVN of the thalamus or the SCN. Short-day-exposed hamsters had fewer beta-endorphin-positive arcuate nucleus cells and tended to have fewer GnRH-positive preoptic cells than long-day controls. VIP staining was unaffected by photoperiod. Most day length effects on immunostaining were eliminated by either lesion. These results establish the importance of the SCN and PVN in the photoperiodic control of several seasonal responses in Siberian hamsters.

The vasopressinergic innervation of the brain in normal and castrated rats.

A detailed description is given of the distribution of vasopressin-immunoreactive structures in the brain of intact adult male rats. By application of a modified immunocytochemical procedure, vasopressin-immunoreactive fibers were detected in many new areas. In adult male rats which were castrated 15 weeks before death, vasopressin-immunoreactive cell bodies had disappeared from the bed nucleus of the stria terminalis and the medial amygdaloid nucleus. No obvious changes were found in vasopressin-immunoreactive cell bodies in other areas. Furthermore, a very strong reduction was seen in the density of vasopressin-immunoreactive fibers in the olfactory tubercle, nucleus of the diagonal band and its immediate surroundings, ventral pallidum, basal nucleus of Meynert, lateral septum, septofimbrial nucleus, ventral hippocampal formation, amygdaloid area, pre- and supramammillary nucleus, supramammillary decussation, (inter)dorsomedial, parafascicular, and ventral aspect of paraventricular thalamic nuclei, zona incerta, lateral habenular nucleus, ventral tegmental area, substantia nigra, periventricular gray, dorsal and median raphe nucleus, and locus coeruleus. No changes were observed in other areas containing vasopressin-immunoreactive fibers. These changes following gonadectomy were not observed in castrated rats which had been treated with testosterone. The results suggest that vasopressin projections from the bed nucleus of the stria terminalis and possibly from the medial amygdaloid nucleus require the presence of gonadal hormones for their normal appearance. This is in contrast to pathways arising from the hypothalamic vasopressin-producing nuclei, which fail to show obvious changes following castration.