Estrogen replacement increases both the number of dendritic spines and the density of axospinous synapses in the hippocampal CA1 region in young rats, yet this is attenuated in aged rats. The estrogen receptor-␣ (ER-␣) is localized within select spines of CA1 pyramidal cells in young animals and thus may be involved locally in this process. The present study investigated the effects of estrogen on the ultrastructural distribution of ER-␣ in the CA1 of young (3-4 months) and aged (22-23 months) Sprague Dawley rats using postembedding immunogold electron microscopy. Within dendritic spines, most ER-␣ immunoreactivity (IR) was seen in plasmalemmal and cytoplasmic regions of spine heads, with a smaller proportion within 60 nm of the postsynaptic density. In presynaptic terminals, ER-␣-IR was clustered and often associated with synaptic vesicles. Significant effects of both aging and estrogen were observed. Quantitative analysis revealed that nonsynaptic pools of ER-␣-IR within the presynaptic and postsynaptic compartments were decreased (35 and 27%, respectively) in the young estrogen-replaced animals compared with those that received vehicle. Such localized regulation of ER-␣ in response to circulating estrogen levels might directly affect synaptic signaling in CA1 pyramidal cells. No estrogen treatment-related differences were observed in the aged animals. However, 50% fewer spines contained ER-␣ in the aged compared with young hippocampus. These data suggest that the decreased responsiveness of hippocampal synapses to estrogen in aged animals may result from age-related decrements in ER-␣ levels and its subcellular localization vis-à -vis the synapse. Such a role for spinous ER-␣ has important implications for age-related attenuation of estrogen-induced hippocampal plasticity.

Estrogen interacts with N-methyl-D-aspartate (NMDA) receptors to regulate multiple aspects of morphological and functional plasticity. In hippocampus, estrogen increases both dendritic spine density and synapse number, and NMDA antagonists block these effects. Thus, estrogen-mediated hippocampal plasticity may be of particular importance in the context of age-related changes in endocrine status and cognitive performance. NR1 levels per synapse are increased in CA1 by estrogen in aged rats but not young rats, although no information is available on estrogen-induced synaptic alterations in other NMDA receptor subunits that might impact function. Therefore, the present study was designed to investigate the effect of estrogen on the synaptic and subsynaptic distributions of the NMDA receptor subunits, NR2A and NR2B in CA1 pyramidal cells, within the context of aging. Our results demonstrated that the overall synaptic levels of NR2A and NR2B are similar in young and aged female rats, regardless of estrogen treatment. However, in the aged CA1, estrogen restores NR2B levels back to young levels in the lateral portions of the active synaptic zone. Thus, estrogen may impact the mobility of NMDA receptors across the synapse and, in the process, restore a more youthful synaptic profile. These findings have important implications for the mechanism of estrogen-induced alterations in NMDA receptor-mediated processes, particularly in the context of aging.

No abstract