Activating just the β isoform of the estrogen receptor (ERβ) is enough to improve learning and memory in mice, according to a paper published in the February 24 Nature Neuroscience. Researchers led by Feng Liu, Mark Day, and Nicholas Brandon at Wyeth Pharmaceuticals, Monmouth Junction, New Jersey, and Collegeville, Pennsylvania, report that selectively activating ERβ increases synaptic strength in the hippocampus, an area of the brain that plays a key role in learning and memory. ERβ agonists also improve hippocampus-dependent cognition, according to the paper. The finding supports the idea that selective estrogen receptor modulators, or SERMs, are worth pursuing as therapeutics for AD and other cognitive disorders, and that they might yield better results than natural estrogens, which have so far proved disappointing in clinical trials.

Potential side effects notwithstanding, estrogen replacement therapy was once touted as a treatment for dementia, and with good reason. Epidemiological evidence suggested that hormone replacement therapy (HRT) in postmenopausal women lowered the risk of getting AD (see, e.g., Zandi et al., 2002). Molecular evidence offered up a plausible explanation for this protection: in the mammalian brain, estrogen promotes dendritic spines and synaptic activity, increases long-term potentiation, elevates expression of glutamatergic NMDA receptors, and even protects against Aβ deposition in a triple transgenic mouse model of AD (see Carroll et al., 2007). But results from the Women’s Health Initiative Memory Study (WHIMS) of HRT proved disappointing. Not only did combination estrogen/progesterone offer no protection against cognitive decline, but it actually increased the risk of getting AD (see Shumaker et al., 2003). Estrogen alone was no better (see Shumaker et al., 2004).

This setback curtailed clinical research into estrogen and cognition, and sent researchers back to the drawing board. Some posited that timing is the key. Administering HRT 15 years after menopause (as in many of the WHIMS participants) may do more harm than good, the theory goes, while taking it around the time of menopause may help (see ARF related news story). This is the premise of KEEPS (Kronos Early Estrogen Prevention Study), which plans to examine if HRT, given within 3 years of menopause, prevents cardiovascular disease. Cognition is a secondary endpoint of that trial.

An alternative approach to reaping a neurologic benefit from estrogen signaling is to selectively target estrogen receptors, something estradiol itself doesn't do. ERβ, for example, seems to have a better neuroprotective profile than ERα (see Tiwari-Woodruff et al., 2007), and the new data from the Wyeth group support this idea.

Joint first authors Feng Liu, Mark Day, and colleagues report that estradiol improves learning and memory when given to ovariectomized mice, but not when given to Esr2 knockout animals (the Esr2 gene encodes ERβ protein). They also report that ERβ-selective compounds developed at Wyeth (WAY-200070 and WAY-202779) mimic the effect of estrogen on the hippocampus in ways that the ERα agonist PPT (4,4',4''-[4-propyl-(1H)-pyrazole-1,3,5-triyl]trisphenol) does not. For example, when given to ovariectomized animals, the Wyeth compounds cause nuclear estrogen receptor levels in the hippocampus to rise—this indicates receptor activation. They also cause hippocampal phospho-CREB to rise—this is an early event during learning and memory encoding. WAY-200070, the more brain-soluble of the two, also increased levels of the glutamate receptor GluR1 and postsynaptic density 95, two proteins important in synaptic activity. These proteins stayed unchanged when the compound was given to Esr2-/- mice or when PPT was used instead.

These findings suggest that the ERβ compounds may accentuate synaptic transmission. In fact, the researchers found that dousing hippocampal slices with WAY-200070 enhanced long-term potentiation. Furthermore, they report that synaptic architecture changed when ovariectomized rats were given WAY-20070. Dendritic interactions among CA1 pyramidal cells increased, and in the dentate gyrus the number of dendritic branches and mushroom-shaped spines grew (these are stable spines believed to be necessary for memory storage). Finally, the researchers report that their compound improved spatial memory: in a radial arm maze, ovariectomized rats on WAY-200070 made fewer errors than untreated animals and performed on par with estradiol-treated rats. The ERα agonist had no effect.

The results suggest that ERβ activation may be one way of treating cognitive disorders, according to the authors. “Activation of this pathway may confer some of the CNS-mediated benefits of estrogen without the feminizing side effects and may offer a new therapeutic approach for diseases with cognitive deficits such as Alzheimer's disease and schizophrenia,” they write. However, there may be some work to do yet to fully understand the ramifications of selectively activating estrogen receptors. For example, Christian Pike at the University of Southern California, Los Angeles, just reported that other ERα and ERβ agonists have differential effects on amyloid deposits in the triple transgenic mice. In those experiments, the ERα agonist PPT reduced Aβ in the hippocampus, subiculum, and amygdala but less so in the frontal cortex. The ERβ agonist DPN, on the other hand, reduced Aβ in the amygdala, but not in the hippocampus and subiculum. Unlike the ERα agonists, DPN had no effect on memory, whereas PPT significantly improved hippocampal-dependent working memory in these animals, as judged by a spontaneous alternation task (see Carroll and Pike, 2008). Whether that memory improvement is related to the reduction in Aβ is not clear. Nevertheless, the differential effects of ERα and ERβ agonists on neurons and Aβ pathology suggest that, as with HRT, timing SERM treatment might be important for successful prevention or treatment of AD.—Tom Fagan


  1. This study investigated cellular and molecular mechanisms that underlie changes in estrogen-mediated effects on synaptic plasticity and hippocampal-dependent spatial memory. The particular focus is on whether Wyeth compounds, specifically estrogen receptor β (ERβ) agonists, can mediate these beneficial effects of estrogen treatment. Much data are shown that are consistent with a beneficial action of their ERβ agonist on these cellular and molecular mechanisms, as well as on behavioral tests of memory.

    It is somewhat problematic, however, that the authors switch back and forth between rats and mice when addressing the different mechanisms. It is known that ERα and ERβ expression in the brain differ between mice and rats. Thus, for consistency, it would have been desirable to have all mechanisms studied in their entirety in at least one species. Notably, extrapolation of findings to humans should be made with caution for the same reason.

    In addition, some experiments were done in females and others in males. Further, some had undergone gonadectomy and others were gonadally intact. Since sex hormones can affect ER expression, different results may occur in each of these scenarios. Indeed, there are inconsistencies in the results in the paper between these variables that are not addressed.

    Finally, it has previously been shown that states of disease, such as stroke, can alter ER expression in the brain. This study was done in normal, healthy rodents. Therefore, one cannot assume that observing ERβ-mediated beneficial effects on a variety of mechanisms during health will also be the case in states of diseases like Alzheimer’s.

    View all comments by Marina Ziehn
  2. There is evidence that estradiol (E2) modulates many advanced brain functions, including hippocampus-dependent learning and memory. Several independent studies have demonstrated beneficial effects of estrogen replacement on cognition in normally aging women and in women suffering from dementia associated with Alzheimer disease. Estrogen affects brain function through estrogen receptors (ERα and ERβ), which have been found in the hippocampus and cortex. We have demonstrated that ERβ is the predominant estrogen receptor expressed in the hippocampus and cortex during embryogenesis; loss of ERβ causes brain malformation (Fan et al., 2006; Wang et al., 2003). All of these data indicate that estrogen can affect hippocampus function, but the underlying mechanism is not clear.

    Liu et al. in this paper have explored estrogen effects on hippocampal synaptic plasticity and memory through estrogen receptor β. An in vivo study showed that key synaptic proteins including PSD-95, synaptophysin, and the AMPA-receptor subunit GluR1 were increased by selective ERβ agonists. In hippocampal slices, ERβ activation enhanced long-term potentiation. All of these effects were absent in ERβ knockout mice. This is the first demonstration of ERβ receptor activation of hippocampal plasticity.

    Another study previously found that estradiol stimulates glutamatergic synapse formation in the developing hippocampus through an ERα-dependent mechanism (Jelks et al., 2007). It is likely that ERα and ERβ act on separate pathways with different mechanisms in the hippocampus. Several studies have reported that loss of ERβ induces dysfunction of hippocampus-dependent learning and memory (Rissman et al., 2002; Day et al., 2005).

    The present study demonstrates that an ERβ agonist can improve performance in hippocampus-dependent memory tasks. All of these data support a role of ERβ for hippocampus-dependent cognition. Another study showed that estradiol via estrogen receptor β improves cerebellar memory formation (Andreescu et al., 2007).

    Alzheimer disease (AD) is the primary cause of dementia and memory loss in the elderly, and is characterized by an insidious development of hippocampal pathology. The present study shows that activation of ERβ may offer a new therapeutic approach for diseases with cognitive deficits such as Alzheimer’s and schizophrenia. Clinical trials have shown that estrogen cannot slow the progression of already diagnosed AD, and estrogen replacement can prevent rather than reverse neurodegeneration. More research is needed to find out the function of estrogen in the brain under normal and pathological conditions, and to determine if estrogen may play some role in AD. Since this study clearly has demonstrated that estrogen affects hippocampus-dependent cognition through estrogen receptor β, ERβ agonists may be a better choice than estrogen for improving cognition.


    . Estrogen receptor beta expression in the embryonic brain regulates development of calretinin-immunoreactive GABAergic interneurons. Proc Natl Acad Sci U S A. 2006 Dec 19;103(51):19338-43. Epub 2006 Dec 11 PubMed.

    . Estrogen receptor (ER)beta knockout mice reveal a role for ERbeta in migration of cortical neurons in the developing brain. Proc Natl Acad Sci U S A. 2003 Jan 21;100(2):703-8. PubMed.

    . Estradiol targets synaptic proteins to induce glutamatergic synapse formation in cultured hippocampal neurons: critical role of estrogen receptor-alpha. J Neurosci. 2007 Jun 27;27(26):6903-13. PubMed.

    . Disruption of estrogen receptor beta gene impairs spatial learning in female mice. Proc Natl Acad Sci U S A. 2002 Mar 19;99(6):3996-4001. PubMed.

    . Beta estrogen receptor knockout (BERKO) mice present attenuated hippocampal CA1 long-term potentiation and related memory deficits in contextual fear conditioning. Behav Brain Res. 2005 Oct 14;164(1):128-31. PubMed.

    . Estradiol improves cerebellar memory formation by activating estrogen receptor beta. J Neurosci. 2007 Oct 3;27(40):10832-9. PubMed.

    View all comments by Margaret Warner
  3. In this manuscript, Liu et al. convincingly demonstrate that activation of ERβ positively regulates synaptic plasticity and enhances long-term potentiation in the adult rodent hippocampus, providing the basis for structural and physiological changes underlying learning and memory. Consistently, the authors demonstrate that the ERβ agonist WAY-200070 improves hippocampus-dependent spatial memory. Thus, this study elucidates a novel ERβ-dependent mechanism by which estrogen enhances hippocampal synaptic plasticity and improves memory, and supports a study demonstrating that ERβ is a prerequisite for optimal spatial learning in mice (Rissman et al., 2002). In addition, the present study reveals that WAY-200070 increases phospho-CREB via a rapid, “non-genomic” mechanism. Milner et al. (2005) has recently demonstrated that ERβ is extensively expressed at extranuclear locations within the hippocampal neurons, suggesting that ERβ may serve primarily as a non-genomic transducer of estrogen actions in the hippocampus. In humans, ERβ seems to the predominant ER subtype found in the adult hippocampus (González et al., 2007).

    Recently, our laboratory has demonstrated that activation of ERα induces glutamatergic synapse formation in the developing hippocampus within a time frame during which hippocampal ERα expression is transiently increased in neonates. Furthermore, we have previously demonstrated that ERα, but not ERβ, mediates the ability of estrogen to protect the brain against ischemic injury (Dubal et al., 2001). Consistently, ischemic stroke injury upregulates the expression of ERα in the injured brain (Dubal et al., 1999). We hypothesize that the dramatic re-expression of ERα following injury is a recapitulation of its expression during development when E2 plays pivotal trophic roles in the brain. Together, these findings indicate that the hippocampus is capable of responding to both ERα and ERβ to potentially improve cognitive functions, depending on the ER subtypes that are available in specific physiological as well as neurodegenerative conditions.

    Recent results from the Woman’s Health Initiative Memory Study (WHIMS) have been reported that hormone therapy (HT) improves cognitive performance when given within five years after the onset of menopause, but not after a prolonged period of hypoestrogenicity. The duration from loss of ovarian hormone production to the initiation of HT also seems to dictate the beneficial actions of estrogen in stroke injury. Despite numerous observational and retrospective studies demonstrating cerebrovascular benefits of HT among postmenopausal women, the WHI reported that HT increased the risk for stroke, or afforded no benefit. The vast majority of women in the WHI were on average 12 years postmenopausal prior to the initiation of HT, whereas in observational studies that reported beneficial effects of HT, the majority of subjects initiated hormone replacement in their perimenopausal period. We have recently demonstrated that a long duration from surgically induced menopause to the initiation of estrogen reintroduction disrupts the beneficial effects of estrogen on the outcomes of experimentally induced ischemic stroke, as well as the ability of the injured brain to upregulate ERα expression (Suzuki et al., 2007). It is intriguing to speculate that a prolonged period of hypoestrogenicity may also disrupt ER expression in the hippocampus, causing diminished responsiveness of the hippocampal neurons to HT to produce positive effects on cognitive function. A therapeutic approach to maintain or to enhance the expression of specific ER subtypes, in addition to HT in the form of 17β-estradiol, ERα or ERβ agonists, may prove to confer beneficial actions against a variety of neurodegenerative conditions including cerebrovascular stroke and Alzheimer disease.


    . Disruption of estrogen receptor beta gene impairs spatial learning in female mice. Proc Natl Acad Sci U S A. 2002 Mar 19;99(6):3996-4001. PubMed.

    . Ultrastructural localization of estrogen receptor beta immunoreactivity in the rat hippocampal formation. J Comp Neurol. 2005 Oct 17;491(2):81-95. PubMed.

    . Distribution patterns of estrogen receptor alpha and beta in the human cortex and hippocampus during development and adulthood. J Comp Neurol. 2007 Aug 20;503(6):790-802. PubMed.

    . Estrogen receptor alpha, not beta, is a critical link in estradiol-mediated protection against brain injury. Proc Natl Acad Sci U S A. 2001 Feb 13;98(4):1952-7. PubMed.

    . Estradiol modulates bcl-2 in cerebral ischemia: a potential role for estrogen receptors. J Neurosci. 1999 Aug 1;19(15):6385-93. PubMed.

    . Timing of estrogen therapy after ovariectomy dictates the efficacy of its neuroprotective and antiinflammatory actions. Proc Natl Acad Sci U S A. 2007 Apr 3;104(14):6013-8. PubMed.

    View all comments by Phyllis M. Wise

Make a Comment

To make a comment you must login or register.


News Citations

  1. Estrogen and Neuroprotection: A Matter of Time?

Paper Citations

  1. . Hormone replacement therapy and incidence of Alzheimer disease in older women: the Cache County Study. JAMA. 2002 Nov 6;288(17):2123-9. PubMed.
  2. . Progesterone and estrogen regulate Alzheimer-like neuropathology in female 3xTg-AD mice. J Neurosci. 2007 Nov 28;27(48):13357-65. PubMed.
  3. . Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA. 2003 May 28;289(20):2651-62. PubMed.
  4. . Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women's Health Initiative Memory Study. JAMA. 2004 Jun 23;291(24):2947-58. PubMed.
  5. . Differential neuroprotective and antiinflammatory effects of estrogen receptor (ER)alpha and ERbeta ligand treatment. Proc Natl Acad Sci U S A. 2007 Sep 11;104(37):14813-8. PubMed.
  6. . Selective estrogen receptor modulators differentially regulate Alzheimer-like changes in female 3xTg-AD mice. Endocrinology. 2008 May;149(5):2607-11. PubMed.

External Citations

  1. Kronos Early Estrogen Prevention Study

Further Reading

Primary Papers

  1. . Activation of estrogen receptor-beta regulates hippocampal synaptic plasticity and improves memory. Nat Neurosci. 2008 Mar;11(3):334-43. PubMed.