NIA Home > Research > Research Programs (Extramural) > Division of Neuroscience > Related Conferences, Workshops, and Meetings
Print this page E-mail this page

Bench to Bedside: Basic Science

I. Basic Science: Effects of Estrogen and Progesterone on CNS Cellular Systems and in Animal Models

A. Neurobiology of Estrogen and Progesterone

Estrogen Mechanisms of Action in Neurons: Factors That Determine Outcome In Vitro as Predictors of Efficacy In Vivo—Dr. Roberta Brinton

Presentation Outline

Dr. Brinton's presentation highlighted the following:

  1. Overview of mechanisms of estrogen action in neurons
  2. Timing of estrogen exposure matters: Estrogen therapy prior to—versus following, versus simultaneous—exposure to beta amyloid toxicity. Healthy cell bias of estrogen action
  3. Dose matters: dose response of 17 beta-estradiol versus conjugated equine estrogens (CEE); more 17 beta-estradiol is not better
  4. Formulation matters: estrogenic components; progestin components
  5. Strategies to optimize estrogen therapy for the brain

Specific points included:

  • Hippocampal neurons contain multiple splice variants of the ERß receptor, including missing the nuclear receptor or the DNA binding domain.
  • 17-ß estradiol has an inverted U-shaped dose response curve for hippocampal neuronal branching.
  • In cell culture, with amyloid present, low dose 17 ß-estradiol (10 ng/ml) leads to protection, while high dose (200 ng/ml) does not. Also important is whether the estradiol is given continuously.
  • In cells without amyloid, all doses and modes of administration are protective.
  • In hippocampal neurons, estrogens reduce tau hyperphosphorylation; addition of MPA reverses this effect.
  • For protection from amyloid in cell culture, there is less protection from estrogen if given after the cells are exposed to amyloid.
  • Hypothesis: hormone therapy might work at an earlier age when cells are healthier.

In Defense of Estradiol and Progesterone—Dr. Dominique Toran-Allerand

Dr. Toran-Allerand's presentation highlighted the following:

  1. The brain is a major target of the gonadal steroids.
  2. The biology of the gonadal steroids and of their receptors must be considered in any experimental design.
  3. All gonadal steroids are not equal: the type of hormone used is crucial.
  4. The method (timing) and route (oral versus transdermal) of replacement is crucial.

Specific points included:

  • Women's Health Initiative Memory Study (WHIMS) was not physiological.
  • Studies are needed using pure estradiol 17ß, not Premarin.
  • Studies on cyclical therapy are needed.
Discussion of Basic Science
  • The healthy cell bias predicts healthy neurons are protected by estrogen, but unhealthy cells are not. The healthy cell bias is consistent with epidemiological data.
  • CEEs in culture are associated with low dendritic sprouting, while 17ß estradiol produces abundant dendritic spines.
  • Provera is not neuroprotective in culture.
  • Women who started with cognitive health in WHIMS may have maintained it. Women who may have had pre-existing underlying and unidentified neuropathology may have been more vulnerable to neuron damage by the therapy provided in the study.

B. Cognitive Function: Age-Related Changes in Estrogen/Progesterone Levels and Cognitive Function

Hormone Therapy and Cognitive Performance: Reconciling Animal Studies With Clinical Data—Dr. Robert Gibbs

Dr. Gibbs's presentation highlighted the following:

  1. Therapy for the brain must be tailored to the brain.
  2. Results depend on: (1) dose and regimen, and (2) timing of hormone therapy with respect to age and loss of ovarian function.
  3. Basal forebrain cholinergic projections play a very important role in mediating effects of estrogen replacement on cognitive performance.

Specific points included:

  • In rat models, 17-ß estradiol enhances learning and memory performance on multiple tasks. This effect can be blocked by immunotoxic lesions of septal cholinergic cells. The effect is dependent on these cells.
  • Some effects of 17-ß estradiol on learning and memory correlate with effects on hippocampal plasticity, which are dependent on basal forebrain cholinergic inputs.
  • Hormone treatment can help prevent age-related cognitive impairment in rats; however, studies suggest there may be a timeframe (less than 10 months) following ovariectomy during which hormone treatment must be initiated to be effective.
  • Simultaneous and sustained E and P treatment has negative effects on basal forebrain cholinergic neurons in rats and in cynomologous monkeys, contrary to other treatment regimens.

Estrogen Influences on Cognitive Aging in Monkeys—Dr. Peter Rapp

Dr. Rapp's presentation addressed the following:

  1. Are dose, schedule, and timing critical for mediating the cognitive and neurobiological effects of estrogen?
  2. Are the effects of estrogen replacement age dependent?
  3. Is research outside the hippocampus needed to understand the neurobiological basis of estrogen effects on cognitive function?

Specific points included:

  • Estrogen replacement regimen used in the behavioral studies consisted of estradiol cypionate, 100µg/1ml sterile peanut oil (i.m., given in a single injection once every 3 weeks).
  • Estrogen administration substantially improved delayed response performance—a frontally mediated task—in aged ovariectomized monkeys relative to age-matched, vehicle-treated subjects.
  • The effects of estrogen on recognition memory mediated by the medial temporal lobe were modest, and hormone treatment had no effect on simple object discrimination learning in aged ovariectomized monkeys.
  • Preliminary data indicate the cognitive effects of estrogen are qualitatively different in young monkeys.
  • Conclusion: estrogen influences on cognitive function are task and age dependent. The findings also establish a primate model for defining the neurobiological basis of ovarian hormone effects of cognitive aging (see Dr. Morrison's presentation, below).

Estrogen and the Aging Cortical Synapse: Implications for Cognitive Effects in Aged Monkeys—Dr. John Morrison

Dr. Morrison's presentation addressed the following research question: what is the behavioral impact of the observed synaptic effects of estrogen?

  1. In primates, the effects of estrogen on prefrontal cortex may be more important than those on the hippocampus with respect to the cognitive effects.
  2. In primates, how long does the aged cortical synapse continue to be responsive to estrogen? Is its responsiveness dependent on variables such as:
    • Length of time without estrogen prior to replacement?
    • Age of initiation of replacement?
    • Schedule of replacement (e.g., continuous or cyclic)?
  3. Aged cortical synapses may react differently to estrogen replacement than young synapses. However, such differences may not be consistent across rats and nonhuman primates.
  4. What are the molecular mechanisms of estrogen-induced spine/synapse enhancement, and how are they affected by aging?  

Specific points included:

  • Unopposed E—administered cyclically for 2.5 years—causes an enhancement of cognitive performance and an increase in spine counts in the prefrontal cortex of aged female rhesus monkeys.
  • The behavioral effect was particularly pronounced on a Delayed Response task, a task dependent on prefrontal cortex.

Wrapup—Dr. MaryLou Voytko

The wrap-up discussion about multiple processes involved changes in cognitive function and appropriateness of the ovariectomized animal model to human menopause. The following outline presents the topics included in this discussion:

Important Issues for Cognitive Function

  • Age of the animal is essential
  • Treatment timing: earlier the better in rats and monkeys
  • Route: Parenteral versus oral
    Regimen
    • Continuous versus cyclical
    • E alone versus E and P
    • Types of E and P
    • Dose: rats tested only; additional monkey studies are needed

I. Examples of Cognitive Testing in Memory and Attention With Hormone Therapy

Rats

  • Morris Water Maze—spatial reference memory
  • Radial Arm Maze—spatial working and reference memory
  • Delayed Matching to Position in a T-maze—spatial working memory

Monkeys

  • Delayed Response—spatial working memory
  • Delayed Nonmatching to Sample—visual working memory
  • Delayed Recognition Span—visual or spatial working memory
  • Discriminations and Reversals—associative learning and cognitive flexibility
  • Visuospatial Cued Reaction Time—visuospatial attention

II. Ovarian Hormones Affect Brain Neurobiology

  • Neurochemical systems: cholinergic, dopaminergic, serotonergic, noradrenergic, gabaergic, glutamatergic, neurotrophic
    • Neuronal excitability
    • Synaptogenesis/spines
    • Glial cells
    • Cerebral blood flow
    • Glucose uptake
    • Neurotoxicity
    • Oxidative Stress

III. Sites of Ovarian Hormone Actions in the Brain Relevant to Cognition

  • Hippocampus
  • Cerebral Cortex: frontal lobe, parietal lobe, entorhinal cortex

IV. Differences Between Natural Versus Surgical Menopause

    1. Natural   

      • Gradual drop in hormones
      • Androgen production by ovaries
      • Mean age of 51 years
    2. Surgical

      • Abrupt drop in hormones
      • Ovarian androgens absent
      • Mean age of 45 years
Next Steps

It is necessary to have better integrated communication between basic science and clinical investigators working on ovarian hormone therapy issues and its affects on cognition and the brain.

  • Clinical to Basic: basic scientists need to communicate with clinicians regarding the animal and cell culture studies needed to best inform them about critical issues related to treating women for their cognitive and brain health.
  • Basic to Clinical: clinicians need to use the information from basic science studies to develop strategies and make decisions on when and how to treat women.
Discussion—Cognitive Function

Earlier is better: waiting a long time before administering estrogen (10 years after loss of ovarian function) may result in the selective loss of function. This may be subtle and difficult to measure.

  • Mode of administration: for continuous administration, more studies needed.
  • Integration of basic science findings into clinical studies and clinical outcomes to drive new basic science studies.  
  • Impaired glucose homeostasis is correlated with cognitive function decline.
  • Women who started with cognitive health in WHIMS may have maintained it. Women who may have had pre-existing underlying and unidentified neuropathology may have been more vulnerable to neuron damage by the therapy provided in the study.
  • The ovariectomized aging rodent animal model is relevant and somewhat comparable to the studies conducted by Barbara Sherwin. There are no other similar studies in humans.
  • Differential sensitivity:
    1. Sensitivity of brain to steroids is region dependent (e.g., prefrontal cortex is steroid sensitive, but medial temporal cortex is not). Some brain regions do not respond in either rodent or nonhuman primate.
    2. Some brain regions are less sensitive to age-related changes than others.
    3. Changes in memory function depend on the modality tested. Some modalities are not sensitive to aging.
  • Primate studies:
    1. Hippocampal dendritic spines in the primate are responsive to estrogen.
    2. Providing estrogen to ovx primate results in the induction of approximately 1 billion dendritic spines in specific brain regions (prefrontal cortex), but not others (visual cortex).
    3. There is no effect of estrogen on dendritic arbors, dendritic complexity, or neuron death.
  • Reorganization of function with age: changes in neural networks underlying changes in cognitive function.
  • Stroke and prefrontal tasks.
  • Metabolic syndrome, diabetes, CHS; E in combination with other treatments.
  • Interactive effects of E across systems and across brain regions.

C. Neuroprotection: Summary of Data Indicating Estrogen/Progesterone Are or Are Not Neuroprotective

Role of Nonfeminizing Estrogens in Brain Protection From Cerebral Ischemia: An Animal Model of Alzheimer's Disease Neuropathology—Dr. James Simpkins

Dr. Simpkins was unable to present. Dr. McEwen summarized the work.

In advance of the meeting, Dr. Simpkins provided the following discussion points:

  1. The WHI reported side effects of continuous oral equine estrogens are primarily mediated by continuous stimulation of estrogen receptor alpha in reproductive tissues and the liver.
  2. Review of the approaches to eliminate these effects of estrogens.
  3. Focus on nonfeminizing estrogen studies and their potent neuroprotective and anti-Alzheimer's neuropathological effects in rodents.

Dr. McEwen summarized Dr. Simpkin's work and addressed the following:

  • Effects are mediated by known ERs (Alpha, beta, X). Nonfeminizing Es do not bind to ERs, are protective in ischemic models, and are antioxidative and free radical quenchers.
  • Wise model-estrogen receptor knock out (ERKO): ER alpha has important neuroprotective effects (genomic or nongenomic), but ER beta is less important
  • Aromatization in the brain is activated by excitatory neurotransmitters; cholesterol side chain cleavage to pregnenolone (first step of estrogen synthesis) is activated by excitatory neurotransmission.
  • Endogenous E production is not sufficient for replacement, but may be a limited servo mechanism for protection.
  • Aromatase KO is susceptible to ischemic damage; add back E, then is protected.
  • Calcium homeostasis is essential for neuroprotection.

Ovarian Steroids, Neuroinflammatory Responses, and Aging—Dr. Caleb Finch

Dr. Finch's presentation highlighted the following:

  1. Systemic estrogens and serum inflammatory markers: effects of mode of administration
  2. Sex steroid effects on glia
  3. Aging, glial activation, and neuronal functions

Specific points included:

  • Oral estrogens increase interleukin 6 (IL6) and C-reactive protein (CRP) at higher body mass index.
  • CRP elevations may predict vascular events and dementia decades in advance of pathologic events.
  • In a wound model in cell culture, progesterone—but not medroxyprogesterone acetate (MPA)—stimulates neuritic outgrowth.
Discussion—Neuroprotection
  • Inflammatory proteins are found in plaques; complement deposits are associated with amyloid beta. Estrogen supports sprouting in vitro (wounding-in-a-dish model). However, sensitivity to estrogens changes with age, because astrocytes from old rat cortex lose responsiveness to estrogen. The age increase in astrocytic GFAP is implicated as a primary factor. GFAP levels can be downregulated with siRNA, and this restores responses to estrogen in aging astrocytes.
  • For anti-inflammatory treatment—in epidemiological studies—low dose is as effective as high dose.
  • In young astrocytes, progesterone can cause lengthening of processes, but this does not occur with MPA. MPA blocks the support function of E2.
  • Progesterone alone does not have much effect in the uterus, but it can be neuroprotective in the brain.
  • CEE treatment is associated with increased CRP.
  • Capillary density is increased in plaques where there appear to be local angiogenesis.
  • Among rodents, there are species differences such that ER alpha is present (i.e., three times more in rats).
  • The Women's Estrogen for Stroke Trial, The Heart and Estrogen/Progestin Replacement Study, and WHI had poor vascular outcomes. Consequently, vascular markers need to be examined.
  • Targets for therapy must include early sets of events.
  • Synaptic changes may be observed in the absence of neuron death.
  • Changes in small vessels may be different than those occurring in large vessels.
  • There were no data on the effects of E on the coagulation cascade.
  • The diet for the monkey consisted of pellets laced with soy protein (consider the effect on neuroanatomy as compared to humans).

Page last updated Feb 16, 2008