Introduction

Sex hormones play an important role in the plasticity of the hippocampus and provide insights into how males and female differ in their response to experiences. A recent study on cognitive decline in aging men and women found that increasing levels of cortisol over a three year interval predicted cognitive decline in women, whereas results for men were not statistically significant.(1)

The finding that women's brains are sensitive to the conditions associated with increased cortisol and show a negative impact on cognitive function may, however, be due, at least in part, to the absence of estrogens, since the subjects in this study were postmenopausal. What are the interactions between stress and sex hormones in hippocampal plasticity? We need to look at the important role of protective factors in the hippocampus -- how they mitigate the degenerative effects of excitatory amino acids and stress hormones.

As we will see in this Cyberounds^®, our new appreciation of the plasticity of the hippocampus has opened the way to possible treatment strategies to reverse hippocampal atrophy that may be able to retard the onset and progression of Alzheimer's disease. As noted in the previous conference, one of the effects of estrogens is to induce new synaptic connections between nerve cells in the hippocampus, a region of the brain that is important in declarative and spatial memory and a brain structure that degenerates in Alzheimer's disease.

Importance of Sex Differences in the Hippocampus

One factor that may contribute to individual differences in aging is gender -- both the effects of sex hormones and the process of sexual differentiation itself that take place early in development. Sex differences in brain structure and function occur in other brain regions besides the hypothalamus, such as the hippocampus, and they appear to be involved in aspects of cognitive function and other processes that go beyond the reproductive process itself. The higher incidence of depression among women and of substance abuse among men is one example.(2) Male and female rats use different strategies to solve spatial problems and this seems to be programmed by testosterone early in postnatal life (and these are most likely to reside in the hippocampus which is responsible for spatial learning and memory).(3) Moreover, structural sex differences in the hippocampus have been reported which reinforce the notion that the hippocampal formation undergoes sexual differentiation.(4)(5)

There are also sex differences seen in the severity of brain damage resulting from transient ischemia,(6) in the response of the brain to lesions(7) and in the response to severe, chronic stress.(8),(9) A recent study has shown that the stress-induced atrophy of apical dendrites of CA3 pyramidal neurons occurs in male rats but not in female rats.(10) What is not yet clear in this case is whether this atrophy, which is reversible, increases or decreases the vulnerability of the male and female brain to permanent damage.

Effects of Estrogens on Cognitive Function

Estrogens appear to have protective effects on the brain in relation to aging, as well as acute effects on verbal memory and other cognitive functions linked to the hippocampus. We note again that one action of estrogens is to induce new synapses between nerve cells in the hippocampus. In animal experiments, it has been difficult to detect cyclicity of performance in spatial tasks -- some studies have reported no effect;(11) some have demonstrated differences in motivational or attentional parameters;(12) while others found an impairment in performance on the day of proestrus, when ovulation occurs.(13) This lack of agreement and paucity of effects may be a reflection of the relative insensitivity of the measures used to detect behaviors that female rats actually use in their natural environments at the time of mating.

Some success, however, has resulted from studying longer-term effects of ovariectomy and estrogen replacement on hippocampal-dependent learning and memory. Short-term estrogen treatment of ovariectomized female rats helps them do better on two different spatial memory tasks(14),(15) and long-term estrogen replacement improves performance in a working memory task(16) as well as in one of the spatial memory tasks.(17) Moreover, estrogen treatment actually changes the animal's behavior, in that it promotes a shift in the strategy that female rats use to solve a task in which they have to find and discriminate between two different foods, with estrogen treatment increasing the probability of the rat's using a response strategy as opposed to a spatial strategy.(18) Furthermore, estrogens are involved in promoting cognitive function in aging rats, since aging female rats that have low plasma estradiol levels in the 'estropause' are reported to perform significantly worse in solving a spatial maze than female rats with high estradiol levels.(19)

The effects of estrogen replacement in rats are reminiscent of the effects of estrogen treatment in women whose ovarian function has been eliminated (by surgical menopause) or modified (by a gonadotrophin releasing hormone antagonist used to shrink the size of fibroids prior to surgery).(20),(21),(22) In general, these effects are seen within a number of weeks and are reversible.

Are Estrogens Neuroprotective?

Another aspect of estrogen action in the aging brain is that estrogen treatment of post-menopausal women appears to have a protective effect on the brain with respect to Alzheimer's disease.(23),(24),(25),(26),(27),(28) It is likely that these protective actions involve a number of other actions of estrogens besides inducing synapses and, indeed, there are many aspects of brain function that are degenerating in Alzheimer's disease and involve the death of neurons rather than simply the loss of synapses between nerve cells. The neuroprotective actions of estrogens include suppression of the production of the toxic form of beta amyloid protein(29),(30) and inhibition of free-radical induced toxicity.(31),(32) The hippocampus is not the only target that may be involved in these cognitive changes. Estrogens affect many regions of the brain, including the basal forebrain cholinergic systems, the serotonergic system, the dopaminergic system and the noradrenergic system (for review, see (33),(34)). The basal forebrain cholinergic system promotes attention while the serotonergic system is very important in the control of mood and the noradrenergic system is involved in arousal, while the dopamine systems is very important in reward and in motor activity.

Finally, it is important to note that virtually nothing is known about protective effects towards Alzheimer's disease in men, either by estrogens or by testosterone. This is an area deserving of investigation -- a future Cyberounds^® will discuss our knowledge of what androgens do to the brain.

How to Define Protective Factors in the Brain

It is evident that both men and women are vulnerable to atrophy of the hippocampus and associated memory deficits, particularly later in life. Thus, besides the possible neuroprotection afforded by sex hormones, it is important to understand the factors that contribute to the resilience of the brain to stress and stress hormones.

There are at least four approaches to identify protective factors. The first is to manipulate genes that are likely to provide protection, such as the neurotrophins or superoxide dismutase. Neurotrophins and their receptors are involved in promoting survival of brain cells and in regulating their plasticity, while superoxide dismutase is an enzyme that helps to quench free radicals that can cause cell damage and death. Mice with deficiencies in these genes should be more vulnerable to stress-induced damage of hippocampal structure and function. Studies are underway to test the validity of this strategy.

The second approach is to manipulate metabolic factors, e.g., by making rats diabetic or by stressing animals that have genetic risk for either Type I or Type II diabetes. Some initial results indicate that diabetes may accelerate stress-induced dendritic atrophy in the hippocampus and promote stress-induced neuronal damage.(35),(36)

A third approach is to use hippocampal cell culture models and study the interaction of androgens, estrogens, glucocorticoids and excitatory amino acids in producing damage through the excitation of nerve cells to the point of damage, referred to as "excitotoxicity."(37),(38),(39) As noted above, the vulnerability to excitotoxicity in hippocampal neurons has been related to increased calcium channel activity that develops with increasing age in culture.(38)

The cell culture approach has been extended recently to demonstrate the protective effects of another steroid that declines with age in humans, namely, dehydroepiandrosterone (DHEA), towards NMDA-induced neurotoxicity. NMDA refers to N-methyl-D-aspartate, an amino acid that acts specifically on a set of receptors (NMDA).

A fourth strategy is to use targeted delivery of genetic material, attached to viral vectors, in order to overcome the restrictions of energy supply in the face of excitotoxic challenge; this can be accomplished by using the viral vectors to supply glucose transporters to cells, thus increasing their ability to obtain glucose while under challenge by over-excitation from NMDA receptor activation.(41)

Are Age-Related Memory Impairments Treatable?

In the previous Cyberounds^® of this series, we have pointed out that adrenal steroids affect the structure and function of the hippocampus in a variety of ways. We have also seen that, in human subjects, there is evidence for both cognitive impairment as well as hippocampal atrophy associated with altered levels of adrenal steroids and traumatic stress (e.g., in Cushing's disease, recurrent depressive illness, post-traumatic stress disorder, as well as individual differences in aging).(42) Besides permanent damage and loss of neurons, what are the possible mechanisms for the atrophy and are they treatable?

Adrenal steroids do not act alone. In the case of dendritic atrophy in the hippocampus, both excitatory amino acids and serotonin release, possibly facilitated by circulating glucocorticoids, play a key role. In fact, the final common path for CA3 dendritic atrophy in rats, treated with either corticosterone or by restraint stress, involves blocking glutamate release or actions, using phenytoin or an NMDA antagonist, respectively, or facilitating serotonin reuptake, using tianeptine. The efficacy of these agents raises the attractive possibility of treating individuals - perhaps the depressed elderly - with agents, such as phenytoin or tianeptine, that can improve cognitive function.(42) If such studies are carried out, it will be important to determine the degree to which hippocampal volume may be increased by such treatments and whether long-term treatment protects these same individuals from dementia.

Conclusions

The glucocorticoid cascade hypothesis of hippocampal aging, published in 1986,(43) has served extremely well to promote research into the factors that cause individual differences in rates of brain aging. This hypothesis emphasized that glucocorticoids participate in a feed-forward cascade of effects on the brain and body, in which progressive glucocorticoid-induced damage to the hippocampus promotes progressive elevation of adrenal steroids and dysregulation of the hypothalamo-pituitary-adrenal (HPA) axis.(43) This research, together with advances in many other aspects of neuroscience in the past 12 years, has led to a far more sophisticated view of the interactions between genes, early development and environmental influences on brain aging.

We also now have a greater appreciation of events at the cellular and molecular level that can either protect, damage or destroy neurons. In addition to stress hormones, we have begun to understand the role of sex hormones and sex differences as factors that regulate hippocampal function, influence cognitive processes and exert neuroprotective effects. While documenting the ultimate vulnerability of the brain to stressful challenges and to the aging process, the research has shown us the magnificent resilience of the brain and thus renews our optimism that treatment strategies can be developed to maintain this resilience in the aging brain.

Acknowledgments

Research in the author's laboratory on some of the topics discussed in this article is supported by NIH Grants NS07080 and MH41256 and by the Health Foundation (New York), Servier (France) and UCB (Belgium).