The hippocampus of the rat loses neurons with age, a loss which may eventuate in some of the functional impairments typical of senescence. Cumulative exposure to corticosterone (CORT) over the lifespan may be a cause of this neuronal loss, as it is prevented by adrenalectomy at mid-age. In this study, we demonstrate that prolonged exposure to CORT accelerates the process of cell loss. Rats were injected daily with sufficient CORT to produce prolonged elevations of circulating titers within the high physiological range. Animals treated for 3 months (chronic subjects) resembled aged rats in a number of ways. First, both groups had extensive and persistent depletions of CORT receptors in the hippocampus; in the case of chronic rats, no recovery of receptor concentrations occurred 4 months after the end of steroid treatment. Second, autoradiographic analysis revealed that the receptor depletion was due, in part, to a loss of CORT-concentrating cells, especially in the CA3 cell field. Remaining cells bound significantly less [3H]corticosterone than did those of control rats. Finally, analysis of size distributions of hippocampal cell bodies indicated that chronic subjects lost neurons of the same size as those lost in the aged hippocampus. Furthermore, chronic subjects also had increased numbers of small, darkly staining cells of CA3; these corresponded in size to the dark glia whose numbers increase in the aged hippocampus, and which are thought to infiltrate in response to neuronal damage or destruction. Thus, this study supports the hypothesis that cumulative exposure to CORT over the lifespan may contribute to age-related loss of neurons in the hippocampus, and that prolonged stress or exposure to CORT accelerates this process.

The aged rat shows a decline in hippocampal corticosterone receptors and dysfunction in learning and adrenocortical physiology previously linked to glucocorticoid effects upon the hippocampus. The Brattleboro rat, congenitally lacking vasopressin, also has a low number of hippocampal glucocorticoid receptors, as well as learning and endocrine impairments similar to those seen in the aged. Centrally acting vasopressin analogues correct the receptor loss in the hippocampus in the Brattleboro rat but do not influence the hippocampal receptor deficit in the aged rat. Quantitative and high resolution autoradiographic procedures were utilized to characterize the glucocorticoid receptor deficit in the aged and Brattleboro rats. Quantitative autoradiography showed that in both aged and Brattleboro subjects, losses in receptors were most extreme in the pyramidal layer of the CA3a region. High resolution autoradiography revealed striking differences in the cellular basis of the receptor losses. Brattleboro rats had decreased binding of [3H]corticosterone per neuron, whereas aged subjects, in addition, had significant losses in the number of corticosterone-concentrating neurons. Taken together, our findings indicate that the glucocorticoid receptor deficit in the Brattleboro rat probably represents a vasopressin-influenced defect in the synthesis or degradation of the receptor, whereas in the aged rat the deficit originates from loss of both receptor per neuron and the steroid-concentrating neurons themselves, and thus is most likely a permanent and pharmacologically insensitive deficit.