• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of pnasPNASInfo for AuthorsSubscriptionsAboutThis Article
Proc Natl Acad Sci U S A. Sep 11, 2007; 104(37): 14813–14818.
Published online Sep 4, 2007. doi:  10.1073/pnas.0703783104
PMCID: PMC1976208
Neuroscience

Differential neuroprotective and antiinflammatory effects of estrogen receptor (ER)α and ERβ ligand treatment

Abstract

Treatment with either estradiol or an estrogen receptor (ER)α ligand has been shown to be both antiinflammatory and neuroprotective in a variety of neurological disease models, but whether neuroprotective effects could be observed in the absence of an antiinflammatory effect has remained unknown. Here, we have contrasted effects of treatment with an ERα vs. an ERβ ligand in experimental autoimmune encephalomyelitis, the multiple sclerosis model with a known pathogenic role for both inflammation and neurodegeneration. Clinically, ERα ligand treatment abrogated disease at the onset and throughout the disease course. In contrast, ERβ ligand treatment had no effect at disease onset but promoted recovery during the chronic phase of the disease. ERα ligand treatment was antiinflammatory in the systemic immune system, whereas ERβ ligand treatment was not. Also, ERα ligand treatment reduced CNS inflammation, whereas ERβ ligand treatment did not. Interestingly, treatment with either the ERα or the ERβ ligand was neuroprotective, as evidenced by reduced demyelination and preservation of axon numbers in white matter, as well as decreased neuronal abnormalities in gray matter. Thus, by using the ERβ selective ligand, we have dissociated the antiinflammatory effect from the neuroprotective effect of estrogen treatment and have shown that neuroprotective effects of estrogen treatment do not necessarily depend on antiinflammatory properties. Together, these findings suggest that ERβ ligand treatment should be explored as a potential neuroprotective strategy in multiple sclerosis and other neurodegenerative diseases, particularly because estrogen-related toxicities such as breast and uterine cancer are mediated through ERα.

Keywords: experimental autoimmune encephalomyelitis, neuroprotection, multiple sclerosis selective estrogen receptor modulators

Estrogen treatment has been effective in numerous neurodegenerative disease models, including multiple sclerosis (MS), Parkinson's disease, spinal cord injury, cerebellar ataxia, Down's syndrome, epilepsy, and some models of stroke and Alzheimer's disease (14), and translational work using estrogen treatment for human neurodegenerative diseases has begun. In general, there has been somewhat of a disparity in results of estrogen treatment of animal models and results in humans, with excellent results in the former and controversial effects in the latter. In reviewing the possible reasons for the disparity, a “healthy cell bias of estrogen action” has been hypothesized (5). Briefly, efficacy of estrogen treatment appears to depend critically on its administration early, as a preventative therapy, before neurodegeneration has occurred (6). Also, early timing of treatment appears to be important, with respect not only to intervention into the neurodegenerative process but also to the need to avoid a period of hypoestrogenicity. In the Women's Health Initiative study, which showed that estrogen treatment afforded no benefit for stroke prevention, women were postmenopausal for many years before initiating estrogen treatment (7). Recently, it has been shown in an ischemic stroke model that estradiol treatment is effective if administered immediately but not 10 weeks after ovariectomy (8). Based on this knowledge, trials are now being designed that will consider the disease duration and menopausal status of the subjects (9).

Unresolved issues in the strategy to use estrogens as neuroprotective agents include whether neuroprotective effects are secondary to antiinflammatory effects of estrogens, and which estrogen receptor mediates each of these protective properties. Although a variety of antiinflammatory mechanisms of estrogen treatment have been described (1012), these are not mutually exclusive of more direct neuroprotective mechanisms, because estrogens are lipophilic, readily traversing the blood–brain barrier (13). Further neuroprotective effects of estrogen treatment in neuronal cultures and other in vitro systems devoid of an inflammatory confound have been described (1416). Regarding estrogen receptors, the actions of estrogen are mediated primarily by nuclear estrogen receptor (ER)α and ERβ, although nongenomic membrane effects have been described (17). ERα and ERβ have partially distinct tissue distributions (18), thereby providing for some tissue selectivity using selective estrogen receptor modifiers. The two receptors act synergistically in some tissues, whereas they act antagonistically in others. These tissue-specific differences in biologic outcomes are thought to be due to tissue-specific differences in transcription factors, which become activated on binding of each ER by ligand (19, 20). Despite the fact ERβ has been shown to be expressed widely in the CNS in adult mice (21, 22), in most neurological disease models, the protective effect of estrogen treatment has been shown to be mediated through ERα and has been associated with antiinflammatory effects (8, 21, 23, 24).

Here, we will contrast effects of treatment with an ERα vs. an ERβ ligand in experimental autoimmune encephalomyelitis (EAE), a MS model with a known pathogenic role for both inflammation and neurodegeneration. Results using the ERβ-selective ligand permit one to dissociate the antiinflammatory from the neuroprotective effect of estrogen treatment and demonstrate that neuroprotective effects of estrogen treatment do not necessarily depend on antiinflammatory properties.

Results

Selected Doses of ERα and ERβ Ligands Induced Known Biological Responses on a Positive Control Tissue, the Uterus.

Before beginning EAE experiments, we used the uterine response to assess whether a known in vivo response would occur during treatment with each of our dosing regimens for the ERα and ERβ ligands. It was known that estrogen treatment increased uterine weight primarily through ERα (25), and it had also been shown that treatment with the ERβ ligand diarylpropionitrile (DPN) could antagonize the ERα-mediated increase in uterine weight (26). Thus, we administered the ERα ligand, propyl pyrazole triol, to ovariectomized C57BL/6 females for 10 days at either an optimal (10 mg/kg per day) or suboptimal (3.3 mg/kg per day) dose and observed a significant increase in uterine weight as compared with vehicle treated mice [supporting information (SI) Fig. 8]. When an ERβ ligand dose (8 mg/kg per day) (27) was given in combination with the ERα ligand, the increase in uterine weight mediated by ERα ligand treatment was significantly reduced. These data demonstrated that our method and dose of delivery of the ERα and ERβ ligands induced a known biological response in vivo on a positive control tissue, the uterus.

Differential Effects of Treatment with ERα and ERβ Ligands on Clinical EAE.

We compared and contrasted effects between ERα and ERβ treatment during EAE. When the ERα ligand was administered 1 week before active EAE induction with myelin oligodendrocyte glycoprotein (MOG) 35–55 peptide in ovariectomized C57BL/6 female mice, clinical disease was completely abrogated (P < 0.0001; Fig. 1Left). This was consistent with our previous findings in this EAE model (23), as well as findings in adoptive EAE in SJL mice by others (28). In contrast, ERβ ligand treatment had no significant effect early in disease (up to day 20 after disease induction) but then demonstrated a significant protective effect later in disease (after day 20), P < 0.001 (Fig. 1 Center).

Fig. 1.
Treatment with ERα- vs. ERβ-selective ligands has differential effects on chronic EAE. Ovariectomized C57BL/6 female mice were given daily s.c. injections of an ER ligand during active EAE and graded for clinical disease severity using ...

Our data showing a protective effect using the ERβ ligand DPN in active EAE in C57BL/6 mice were surprising given that another ERβ ligand (WAY-202041) was shown to have no effect in EAE (28), albeit using a different strain of mice that were followed for a shorter time period. Because WAY-202041 was shown to have a 200-fold selectivity for ERβ, whereas DPN has a 70-fold selectivity (29), it was possible that DPN was not sufficiently selective for ERβ in vivo in our studies. To assess the in vivo selectivity of DPN treatment during EAE, we administered DPN to homozygous ERβ knockout (KO) mice. When DPN was administered to ovariectomized ERβ KO C57BL/6 mice with EAE, the treatment was no longer protective (Fig. 1 Right). These data demonstrated the in vivo selectivity of DPN for ERβ during EAE at the dose used in our studies.

Together, these results indicated that treatment with an ERα ligand is protective at the acute onset and throughout the course of EAE, whereas treatment with an ERβ ligand is protective during the later phase of the disease, after the acute initial phase.

Differential Effects of Treatment with ERα and ERβ Ligands on Autoantigen-Specific Cytokine Production.

Because estrogen treatments in EAE had previously been associated with either a down-regulation of proinflammatory cytokines or an up-regulation of antiinflammatory cytokines (10, 11), we next assessed autoantigen-specific cytokine production by systemic immune cells during ERα vs. ERβ ligand-treated EAE. ERα ligand treatment significantly reduced levels of cytokines (TNFα, IFN-γ, and IL-6) known to be proinflammatory in EAE, whereas it increased the antiinflammatory cytokine IL-5, during both early (Fig. 2A) and later (Fig. 2C) stages of EAE. In contrast, ERβ ligand treatment was no different from vehicle treatment in all measured cytokines (TNFα, IFN-γ, IL-6, and IL-5) at either the early (Fig. 2B) or later (Fig. 2D) time points. These results indicated that, whereas ERα ligand treatment was antiinflammatory in the systemic immune system, ERβ ligand treatment was not.

Fig. 2.
Treatment with ERα- vs. ERβ-selective ligands has differential effects on the systemic immune response. At day 19 (A and B) or day 40 (C and D) after disease induction, mice were killed, and cytokine production by autoantigen-stimulated ...

Treatment with an ERα Ligand, but Not an ERβ Ligand, Reduces CNS Inflammation.

We then addressed whether treatment with ERα vs. ERβ ligands resulted in differences in inflammation within the CNS. At both early (day 19) and later (day 40) stages of EAE, spinal cord sections from mice treated with either vehicle, ERα or ERβ ligand were assessed for inflammation by using anti-CD45 antibody to stain inflammatory cells. ERα ligand-treated EAE compared with vehicle-treated EAE mice had less CD45 staining in white matter. This reduction in CD45 staining was present at both the early (Fig. 3A) and later (Fig. 3B) timepoints in EAE. In contrast, ERβ ligand-treated EAE mice did not have reduced CD45 staining in white matter, at either time point. Quantification of CD45+ cells revealed that ERα ligand-treated mice at the early stage of EAE had a reduction in inflammation, such that levels were no different as compared with those in normal control mice, whereas CD45+ cell numbers in ERβ ligand-treated EAE mice remained significantly increased and comparable to those in vehicle-treated EAE mice (Fig. 3C). At the later time point, quantification detected some inflammation in ERα ligand-treated EAE mice, whereas inflammation in ERβ ligand-treated remained very high and similar to vehicle-treated EAE mice (Fig. 3D).

Fig. 3.
Treatment with an ERα ligand, not an ERβ ligand, reduced inflammation in spinal cords of mice with EAE. Consecutive thoracic spinal cord sections coimmunostained with NF200 (green) and CD45 (red) at ×10 magnification are shown ...

Additionally, CD45 staining of cells in gray matter of vehicle-treated EAE mice was observed at both the early and later time points, with these cells demonstrating a morphology suggestive of activated microglia (Fig. 3 Insets). This gray matter inflammation was also decreased with ERα ligand but not ERβ ligand treatment.

H&E staining also revealed that vehicle-treated EAE mice had extensive white matter inflammation at both the early (SI Fig. 9A) and later (SI Fig. 9D) time points, and that this inflammation was reduced by treatment with the ERα but not the ERβ ligand. Further, when anti-CD3 antibody was used to stain T lymphocytes, and anti-Mac 3 antibody was used to stain cells of the macrophage lineage, the infiltrate was shown to be composed of both T cells and macrophage lineage cells. Treatment with the ERα ligand but not the ERβ ligand reduced this T cell and macrophage lineage cell staining at both the early (SI Fig. 9 B and C) and later (SI Fig. 9 E and F) time points.

Together, these data indicated that ERα but not ERβ ligand treatment reduced inflammation in the CNS of mice with EAE.

Treatment with Both ERα and ERβ Ligands Reduces Demyelination in White Matter.

The degree of myelin loss was then assessed by myelin basic protein (MBP) immunostaining in the dorsal columns of thoracic cords. Extensive demyelination occurred at the sites of inflammatory cell infiltrates in vehicle-treated EAE mice, whereas less demyelination occurred in ERα and ERβ ligand treated mice (Fig. 4 A and B). Quantification of demyelination by density analysis of MBP immunostained spinal cord sections revealed a 32% (P < 0.01) and 34% (P < 0.005) decrease in myelin density in vehicle-treated EAE mice, at the early and later time points, respectively, as compared with healthy controls (Fig. 4 C and D). In contrast, myelin staining was somewhat decreased but relatively preserved in both ERα and ERβ ligand-treated mice with no significant difference as compared with healthy controls. Double immunostaining with antibodies to MBP and to 200-kDa neurofilament (NF200) revealed relatively intact red myelin rings around green axons in the ERα and ERβ ligand-treated EAE mice (SI Fig. 10).

Fig. 4.
Treatment with ERα and ERβ ligands, each preserved MBP immunoreactivity in white matter of spinal cords of mice with EAE. At days 19 (A) and 40 (B) after disease induction, vehicle-treated EAE mice had reduced MBP immunoreactivity as compared ...

Treatment with Both ERα and ERβ Ligands Reduces Axonal Loss in White Matter.

Staining with anti-NF200 antibody revealed axonal loss in white matter of vehicle-treated EAE mice at both early and later time points of disease as compared with healthy controls, whereas both ERα and ERβ ligand-treated EAE mice had less axonal loss (Fig. 5 A and B). Quantification of NF200 staining in the anterior funiculus revealed a 49 ± 12% (P < 0.01) and 40 ± 8% (P < 0.005) reduction in vehicle-treated EAE, at the early and later time points, respectively, as compared with healthy controls (Fig. 5 C and D), whereas axon numbers in ERα and ERβ ligand-treated EAE mice were not significantly reduced as compared with those in healthy controls.

Fig. 5.
Treatment with ERα and ERβ ligands each preserved axonal densities in white matter of spinal cords of mice with EAE. Part of the anterior funiculus of thoracic spinal cord sections was imaged at ×40 after coimmunostaining with ...

Treatment with Both ERα and ERβ Ligands Reduces Neuronal Pathology in Gray Matter.

Recently, we demonstrated neuronal abnormalities surprisingly early during EAE (day 15), which were prevented by treatment with either estradiol or ERα ligand (23). Here, we asked whether ERβ ligand treatment might preserve neuronal integrity. We used a combination of Nissl stain histology and NeuN/β3-tubulin immunolabeling to identify and semiquantify neurons in gray matter. A decrease in neuronal staining in gray matter occurred at both time points in vehicle-treated EAE mice as compared with healthy controls, whereas neuronal staining in gray matter was preserved in EAE mice treated with either the ERα or the ERβ ligand at the early and the later time points (Fig. 6 A and B). Quantification of NeuN+ cells in gray matter demonstrated a 41 ± 13% (P < 0.05) and 31 ± 8% (P < 0.05) reduction at the early and later time points, respectively, in vehicle-treated EAE mice as compared with normal controls, whereas ERα and ERβ ligand-treated mice had NeuN+ cell numbers that were fewer but not significantly different from those in healthy controls (Fig. 6 C and D).

Fig. 6.
Treatment with ERα and ERβ ligands each preserved neuronal staining in gray matter of spinal cords of mice with EAE. Split images of thoracic spinal cord sections stained with NeuN (red, i) and Nissl (ii) at ×4 magnification, derived ...

Protection from Neuropathology Is Mediated by ERβ.

To confirm whether the effect of DPN treatment in vivo on CNS neuropathology was indeed mediated through ERβ, we next assessed neuropathology in DPN-treated EAE mice deficient in ERβ. At day 38 after disease induction, inflammation, demyelination, and reductions in axon numbers were present in white matter, whereas neuronal staining was decreased in gray matter of vehicle-treated EAE mice (SI Fig. 11). In contrast to the neuroprotection observed during DPN treatment of WT mice (Figs. 446), DPN treatment of ERβ KO mice failed to prevent this white and gray matter pathology (SI Fig. 11). These data demonstrated that neuroprotective effects mediated by DPN treatment in vivo during EAE are mediated through ERβ.

Treatment with an ERβ Ligand Induces Recovery of Motor Performance.

Because treatment with an ERβ ligand was found to be neuroprotective in EAE, we then assessed the clinical significance of this neuroprotective effect using an outcome frequently used in spinal cord injury, rotarod performance. Vehicle-treated EAE mice demonstrated an abrupt and consistent decrease in the number of seconds they were able to remain on the rotarod, beginning at day 12 after disease induction, and this disability remained throughout the observation period. ERβ ligand-treated mice also had an abrupt decrease in the number of seconds they could remain on the rotarod. However, later, at days 30–40, they had significant recovery (Fig. 7Left). These data demonstrated that ERβ ligand treatment induced functional recovery in motor performance at later time points during EAE. Finally, the improvement in rotarod performance with DPN treatment was no longer observed in the ERβ KO (Fig. 7 Right), demonstrating that the DPN-induced recovery in motor performance later in disease was indeed mediated through ERβ.

Fig. 7.
Treatment with an ERβ ligand results in recovery of motor function late during EAE. (Left) Ovariectomized C57BL/6 female mice with EAE were treated with ERβ ligand and assessed for motor performance on a rotarod apparatus. Although mean ...

Discussion

Previously, it had been shown that treatment with either estradiol or an ERα ligand was antiinflammatory and neuroprotective in EAE, stroke, and other disease models (8, 21, 23, 24). Whether neuroprotective effects could be observed in the absence of an antiinflammatory effect remained unknown, with a recent study suggesting that an antiinflammatory effect was necessary to observe neuroprotection in stroke (8). In this study, we have contrasted effects of treatment with ERα vs. ERβ ligands in EAE, the MS model with a known pathogenic role for both inflammation and neurodegeneration. We found that treatment with the ERβ ligand was neuroprotective, with no evidence of an antiinflammatory effect, when assessing both systemic immune responses and CNS inflammation. Thus, by using the ERβ-selective ligand, we have dissociated the antiinflammatory effect from the neuroprotective effect of estrogen treatment and have shown that neuroprotective effects of estrogen treatment do not necessarily depend on antiinflammatory properties. What remains unknown is whether the neuroprotective effect of ERβ ligand treatment is mediated through binding to ERβ receptors on neurons, oligodendrocytes, or astrocytes, because all of these CNS cell types have been shown to express ERβ in vivo (30).

It was interesting to contrast the effects of ERα vs. ERβ ligand treatment. Treatment with the ERα ligand, which was antiinflammatory, resulted in complete abrogation of the onset of EAE and thereafter. In contrast, treatment with the ERβ ligand, which was not antiinflammatory, resulted in no significant effect at the onset of disease but induced clinical protection later in disease. On rotarod testing, motor performance was severely impaired initially but then underwent recovery in the ERβ ligand-treated mice. Finally, ERβ ligand-treated mice had extensive inflammation yet demonstrated significant preservation of axons and neurons throughout the disease course. Together, these data suggested that early treatment targeted toward preserving axonal and neuronal integrity changed the ultimate result of a given inflammatory attack, from permanent disability in untreated, to temporary disability in treated.

To our knowledge, the only previously described neuroprotective agents for EAE, which did not decrease CNS inflammation, were blockers of glutamate receptors (31, 32). These treatments resulted in a modest reduction in neurologic impairment, and the effect was lost after cessation of treatment (32, 33). Glutamate blockers are currently used in amyotrophic lateral sclerosis and Alzheimer's disease with modest success. In MS, brain atrophy on MRI has been detected at the early stages of disease (34), thus a neuroprotective agent would need to be started relatively early, generally at ages 20–40 years, and continued for decades. Because glutamate is needed for normal neuronal plasticity and memory (35), treatment of relatively young individuals with glutamate blockers for decades may be associated with significant toxicity (36). Hence, the identification of an alternative neuroprotective agent represents an important advance in preclinical drug development in MS and other chronic neurodegenerative diseases.

One must consider the risk/benefit ratio of any estrogen treatment when considering its use in neurodegenerative diseases. The goal is to optimize efficacy and minimize toxicity. Hence, determining which estrogen receptor mediates the neuroprotective effect of estrogen treatment is of central importance. Our data demonstrating that treatment with an ERβ ligand is neuroprotective are of clinical relevance, because breast and uterine endometrial cancer are both mediated through ERα, not ERβ. For neurodegenerative diseases with only a minimum inflammatory component, treatment with an ERβ ligand may suffice. For diseases such as MS with a significant inflammatory component, a standard antiinflammatory treatment could be used in combination with ERβ ligand treatment. In each of these scenarios, the neuroprotective properties of estrogen treatment could be maintained while avoiding the increased risk of cancer in the breast and uterus.

Materials and Methods

Animals.

Female WT C57BL/6 mice and ERβ homozygous KO mice on the C57BL/6 background, age 8 weeks, were obtained from Taconic Farms (Germantown, NY). Animals were maintained in accordance with guidelines set by the National Institutes of Health and as mandated by the University of California Los Angeles Office for the Protection of Research Subjects and the Chancellor's Animal Research Committee.

Reagents.

Propyl pyrazole triol and diarylpropionitrile (DPN), ERα and ERβ agonists, respectively, were purchased from Tocris Bioscience (Ellisville, MO). Estradiol was purchased from Sigma–Aldrich (St. Louis, MO). Miglyol 812 N liquid oil was obtained from Sasol North America (Houston, TX). MOG peptide, amino acids 35–55, was synthesized to >98% purity by Mimotopes (Clayton, Victoria, Australia).

Hormone Manipulations and EAE Induction.

Ovariectomized mice were treated with daily s.c. injections of estradiol at 0.04 mg/kg per day (37), DPN at 8 mg/kg per day (27), propyl pyrazole triol at 10 mg/kg per day (25), or vehicle beginning 7 days before EAE induction and throughout the entire disease duration. Active EAE was induced by immunizing with 300 μg of MOG peptide, amino acids 35–55, and 500 μg of Mycobacterium tuberculosis in complete Freund's adjuvant as described (10), and mice were monitored daily for clinical signs as described in SI Text. Some mice were followed clinically for up to 50 days after disease induction, whereas others were killed earlier for mechanistic studies at day 19 after disease induction, corresponding to days 4–6 after the onset of clinical signs in the vehicle-treated group. Uterine weights to assess the biological response to dosing were as described in SI Text.

Rotarod Testing.

Motor behavior was tested up to two times per week for each mouse using a rotarod as described in SI Text.

Immune Responses.

Splenocytes were stimulated with autoantigen at 25 μg/ml, supernatants were collected after 48 and 72 h, and levels of TNFα, IFN-γ, IL-6, and IL-5 were determined by cytometric bead array (BD Biosciences, San Diego, CA), as described (10), and IL-17 was measured by ELISA (R&D Systems, Minneapolis, MN)

Histologic Preparation and Immunohistochemistry.

Perfusion and spinal cord collections were carried out as described in SI Text. Serial sections were stained with H&E or Nissl.

Consecutive sections were also examined by immunohistochemistry (23) by using primary antibodies: anti-β3 tubulin and anti-neurofilament-NF200, anti-neuronal specific nuclear protein (NeuN), anti-CD45, and anti-MBP [Chemicon (Temecula, CA) and Sigma], as described in SI Text.

Microscopy and Quantification.

Sections from spinal cord levels T1–T5 were examined, six from each mouse, with n = 3 mice per treatment group, for a total of 18 sections per treatment group. Images were captured under microscope (×4, ×10, or ×40) by using the DP70 Image software and a DP70 camera (both from Olympus, Melville, NY). All images were converted to grayscale and then analyzed by density measurement with ImageJ, ver. 1.29 (http://rsb.info.nih.gov/ij). Increase in total number of infiltrating cells was measured by density measurements of DAPI+ nuclei in the whole white matter. Neuronal cells were quantified by counting the NeuN+/β3-tubulin+/DAPI+ cells per mm2 in the whole gray matter. Laser-scanning confocal microscopic scans were performed on MBP+/NF200+ and CD45+/NF200+ immunostained spinal cord sections, each as described in SI Text.

Statistical Analysis.

EAE clinical disease severity was compared between treatment groups by using the Friedman test; histopathological changes were assessed by using 1 × 4 ANOVAs; uterine weights, proliferative responses, and cytokine levels were compared between treatment groups using Student's t test, and time on rotorod was compared between treatment groups by using ANOVA.

Supplementary Material

Supporting Information:

Acknowledgments

We thank Cory Peterson for technical laboratory assistance. Support for this work was provided by National Institutes of Health Grant NS45443 and National Multiple Sclerosis Society Grants RD3407 and CA1028 (to R.R.V.).

Abbreviations

EAE
experimental autoimmune encephalomyelitis
MOG
myelin oligodendrocyte glycoprotein
DPN
diarylpropionitrile
MBP
myelin basic protein
MS
multiple sclerosis
KO
knockout.

Footnotes

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.

This article contains supporting information online at www.pnas.org/cgi/content/full/0703783104/DC1.

References

1. Heikkinen T, Kalesnykas G, Rissanen A, Tapiola T, Iivonen S, Wang J, Chaudhuri J, Tanila H, Miettinen R, Puolivali J. Exp Neurol. 2004;187:105–117. [PubMed]
2. Leranth C, Roth RH, Elswoth JD, Naftolin F, Horvath TL, Redmond DE., Jr J Neurosci. 2000;20:8604–8609. [PubMed]
3. Sierra A, Azcoitia I, Garcia-Segura L. Endocrine. 2003;21:43–51. [PubMed]
4. Sribnick EA, Wingrave JM, Matzelle DD, Ray SK, Banik NL. Ann NY Acad Sci. 2003;993:125–133. discussion 159–160. [PubMed]
5. Brinton RD. Ann NY Acad Sci. 2005;1052:57–74. [PubMed]
6. Mulnard RA, Corrada MM, Kawas CH. Curr Neurol Neurosci Rep. 2004;4:368–373. [PubMed]
7. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, et al. J Am Med Assoc. 2002;288:321–333. [PubMed]
8. Suzuki S, Brown CM, Dela Cruz CD, Yang E, Bridwell DA, Wise PM. Proc Natl Acad Sci USA. 2007;104:6013–6018. [PMC free article] [PubMed]
9. Manson JE, Allison MA, Rossouw JE, Carr JJ, Langer RD, Hsia J, Kuller LH, Cochrane BB, Hunt JR, Ludlam SE, et al. N Engl J Med. 2007;356:2591–2602. [PubMed]
10. Liu HB, Loo KK, Palaszynski K, Ashouri J, Lubahn DB, Voskuhl RR. J Immunol. 2003;171:6936–6940. [PubMed]
11. Polanczyk M, Zamora A, Subramanian S, Matejuk A, Hess DL, Blankenhorn EP, Teuscher C, Vandenbark AA, Offner H. Am J Pathol. 2003;163:1599–1605. [PMC free article] [PubMed]
12. Vegeto E, Belcredito S, Etteri S, Ghisletti S, Brusadelli A, Meda C, Krust A, Dupont S, Ciana P, Chambon P, Maggi A. Proc Natl Acad Sci USA. 2003;100:9614–9619. [PMC free article] [PubMed]
13. Wise PM, Dubal DB, Wilson ME, Rau SW, Bottner M. Endocrinology. 2001;142:969–973. [PubMed]
14. McEwen B. Rec Prog Horm Res. 2002;57:357–384. [PubMed]
15. Nilsen J, Diaz Brinton R. Proc Natl Acad Sci USA. 2003;100:2842–2847. [PMC free article] [PubMed]
16. Woolley CS. Curr Opin Neurobiol. 1999;9:349–354. [PubMed]
17. Weiss DJ, Gurpide E. J Steroid Biochem. 1988;31:671–676. [PubMed]
18. Enmark E, Gustafsson JA. J Intern Med. 1999;246:133–138. [PubMed]
19. Paech K, Webb P, Kuiper GG, Nilsson S, Gustafsson J, Kushner PJ, Scanlan TS. Science. 1997;277:1508–1510. [PubMed]
20. Shang Y, Brown M. Science. 2002;295:2465–2468. [PubMed]
21. Merchenthaler I, Lane MV, Numan S, Dellovade TL. J Comp Neurol. 2004;473:270–291. [PubMed]
22. Mitra SW, Hoskin E, Yudkovitz J, Pear L, Wilkinson HA, Hayashi S, Pfaff DW, Ogawa S, Rohrer SP, Schaeffer JM, et al. Endocrinology. 2003;144:2055–2067. [PubMed]
23. Morales LB, Loo KK, Liu HB, Peterson C, Tiwari-Woodruff S, Voskuhl RR. J Neurosci. 2006;26:6823–6833. [PubMed]
24. Suzuki S, Brown CM, Wise PM. Endocrine. 2006;29:209–215. [PubMed]
25. Harris HA, Katzenellenbogen JA, Katzenellenbogen BS. Endocrinology. 2002;143:4172–4177. [PubMed]
26. Frasor J, Barnett DH, Danes JM, Hess R, Parlow AF, Katzenellenbogen BS. Endocrinology. 2003;144:3159–3166. [PubMed]
27. Carswell HV, Macrae IM, Gallagher L, Harrop E, Horsburgh KJ. Am J Physiol. 2004;287:H1501–H1504. [PubMed]
28. Elloso MM, Phiel K, Henderson RA, Harris HA, Adelman SJ. J Endocrinol. 2005;185:243–252. [PubMed]
29. Meyers MJ, Sun J, Carlson KE, Marriner GA, Katzenellenbogen BS, Katzenellenbogen JA. J Med Chem. 2001;44:4230–4251. [PubMed]
30. Platania P, Laureanti F, Bellomo M, Giuffrida R, Giuffrida-Stella AM, Catania MV, Sortino MA. Neuroendocrinology. 2003;77:334–340. [PubMed]
31. Pitt D, Werner P, Raine CS. Nat Med. 2000;6:67–70. [PubMed]
32. Smith T, Groom A, Zhu B, Turski L. Nat Med. 2000;6:62–66. [PubMed]
33. Kanwar JR, Kanwar RK, Krissansen GW. Brain. 2004;127:1313–1331. [PubMed]
34. Chard DT, Griffin CM, Parker GJ, Kapoor R, Thompson AJ, Miller DH. Brain. 2002;125:327–337. [PubMed]
35. Miyamoto E. J Pharmacol Sci. 2006;100:433–442. [PubMed]
36. Gardoni F, Di Luca M. Eur J Pharmacol. 2006;545:2–10. [PubMed]
37. Jansson L, Olsson T, Holmdahl R. J Neuroimmunol. 1994;53:203–207. [PubMed]

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

    Your browsing activity is empty.

    Activity recording is turned off.

    Turn recording back on

    See more...