|
|
||||||||
Mediates Estrogens Immune Protection in Autoimmune Disease

* Department of Neurology, Reed Neurological Research Center, University of California, Los Angeles, CA 90095; and
Department of Biochemistry, Child Health and Animal Sciences, University of Missouri, Columbia, MO 65211
| Abstract |
|---|
|
|
|---|
(ER
). The presence of a functional ER was demonstrated in secondary lymphoid tissues, then ER
expression was shown at both the RNA and protein levels in these tissues. Use of ER
knockout mice revealed that both the estrogen-induced disease protection and the estrogen-induced reduction in proinflammatory cytokines were dependent upon ER
in the prototypic Th1-mediated autoimmune disease experimental autoimmune encephalomyelitis. These findings are central to the design of selective ER modifiers which aim to target biologic responses in specific organ systems. | Introduction |
|---|
|
|
|---|
Gender and hormonal influence play a role in a wide variety of human autoimmune diseases. Multiple sclerosis (MS), rheumatoid arthritis (RA), uveitis, and thyroiditis improve during pregnancy while other diseases such as systemic lupus erythematosis (SLE) tend to worsen (1). An immune shift from Th1 to Th2 that occurs during pregnancy as the maternal immune system attempts to avoid fetal rejection has been proposed as a mechanism underlying the improvement in Th1-mediated autoimmune diseases during pregnancy (1).
Pregnancy is a complex situation characterized by changes in numerous factors including a steady increase in levels of estrogens. Interestingly, increases in estrogens appear to recapitulate the effect of pregnancy on autoimmune diseases, at least in part. Estrogen treatment in mice decreases the severity of experimental autoimmune encephalomyelitis (EAE) and collagen-induced arthritis (CIA), while it worsens SLE (2, 3, 4, 5). Immune shifts from Th1 to Th2 have been observed during estrogen treatment in these animal models (2, 3, 6). There is also evidence in humans with MS and RA that estrogen treatment may be beneficial, while estrogen treatment of SLE worsens the disease (7, 8, 9, 10, 11, 12, 13).
Estrogens may act through genomic or nongenomic mechanisms. Genomic mechanisms were originally described and mediated through ER
, the "classic" ER (14). However, since characterization of ER
, other ERs have been described including ER
(15). Most recently, ER
has been detected, but thus far only in teleosts (16). Nongenomic mechanisms involve membrane effects (17) with a putative membrane associated "ER-X" being recently reported (18). Interestingly, the estrogen-mediated protection in CIA was recently shown to possibly be caused by genomic effects since treatment with ICI 182,780 blocked the protective effect of estrogen on disease when used at a dose known to block ER
and ER
, but not membrane effects (19). However, it was not determined which nuclear receptor was involved in the estrogen induced immune protection in CIA. In this manuscript the prototypic Th1-mediated autoimmune disease EAE is used to determine whether the protective effect of estrogen treatment is mediated through classic ER
.
| Materials and Methods |
|---|
|
|
|---|
Homozygous male ER
knockout mice were generated by backcrossing the knockout (20) onto the C57BL/6 strain for 16 generations. Littermates were used as wild-type (WT) controls. Animals were maintained in accordance with guidelines established by the National Institutes of Health and as mandated by the Univeristy of California, Los Angeles Office for the Protection of Research Subjects. Human ovary was purchased from BioChain Institute (Hayward, CA). Myelin oligodendrocyte glycoprotein (MOG) 3555 peptide was synthesized to >98% purity by Chiron Mimotopes (San Diego, CA).
In vitro estrogen treatment and cytokine analysis
C57BL/6 mice were immunized with MOG3555 peptide (100 µg/mouse) and draining lymph node cells were harvested after 10 days. Cells were stimulated in vitro with autoantigen (25 µg/ml) in the presence or absence of estriol (28 ng/ml). After 6 h of stimulation, TNF-
was detected by intracellular staining with TNF-
-specific Ab (BD PharMingen, San Diego, CA) and subsequent FACS analysis.
EMSA
Cytoplasmic or nuclear proteins from spleen and brain were prepared using NE-PER Nuclear and Cytoplasmic Extraction Reagents based on the manufacturers instruction (Pierce, Rockford, IL). Samples were preincubated with biotin-conjugated estrogen response element (ERE; Panomics, Redwood City, CA) then subjected to electrophoresis. Recombinant ER
and ER
receptors (Affinity, Golden, CO) were used as positive controls.
RT-PCR
Total RNA (2 µg) was reverse-transcribed to cDNA using the Reverse Transcription System (Applied Bioscience, Branchburg, NJ). For amplification, the outer primer set corresponded to nucleotides 461480 (sense) and 19181937 (anti-sense) of the ER
gene, while with the inner primer set corresponded to 659679 (sense) and 981-1003 (anti-sense) as previously numbered (21). The primer sequences are as follows: mER
outer set sense 461480 5'-GCC GCC TTC AGT GCC AAC AG-3'; mER
inner set sense 659679 5'-AAT TCT GAC AAT CGA CGC CAG-3'; mER
inner set anti-sense 1003981 5'-GTG CTT CAA CAT TCT CCC TCC TC-3'; mER
outer set anti-sense 19371918 5'-AGG AAT GTG CTG AAG TGG AG-3'.
Western blots
Intracellular proteins were prepared using NE-PER Nuclear and Cytoplasmic Extraction Reagents (as above) and Western blots run as described previously (22). To detect ER
(67 kDa) a rabbit ER
-specific primary Ab (Santa Cruz Biotechnology, Santa Cruz, CA) was used with a biotin-labeled goat anti-rabbit IgG secondary Ab. Alternatively, westerns were probed with E2-BSA (Sigma-Aldrich, St. Louis, MO) labeled with biotin using the EZ-link Sulfo-NHS-LC-Biotinylation kit (Pierce).
In vivo hormone treatment, induction of active EAE, and cytokine analysis
C57BL/6 male mice were treated with estriol (5 mg s.c. 90-day release pellets) or with placebo pellets (Innovative Research of America, Sarasota, FL) as described (2). One week later, active EAE was induced by immunizing with MOG3555 peptide in complete Freunds adjuvant as described (23). Mice were monitored daily for signs of EAE on a standard EAE grading scale of 05 (0, unaffected; 5, moribund). Mice in each group were sacrificed at days 1624 during EAE, splenocytes were harvested and stimulated with autoantigen (MOG3555 peptide at 25 µg/ml) as described (2). Supernatants were collected after 48 h and levels of TNF-
, IFN-
, IL-2, and IL-5 were determined by cytometric bead array (BD PharMingen).
| Results |
|---|
|
|
|---|
We first examined immune cells from secondary lymphoid tissues of mice for the presence of a functional ER. Lymph node cells from MOG peptide 3555-immunized C57BL/6 mice were stimulated with the autoantigen in the presence or absence of estriol in vitro, and a significant decrease in TNF
(TNF-
) was observed, p = 0.036 (Fig. 1A). This reduction in TNF-
production during an autoantigen-specific immune response had been previously reported during in vivo estrogen treatment of mice with EAE (3). These in vitro data indicated that estrogens effect can be mediated by a direct effect of an estrogen on murine immune cells during an autoantigen-specific immune response. Next, we assessed murine splenic tissue proteins for their ability to bind to the ERE. In the nuclear fraction, binding was evidenced by retardation of migration of bands in EMSA (Fig. 1B). The lack of well delineated bands in lanes containing tissue proteins, with clear single bands in lanes containing recombinant proteins, is likely due to heterogeneity in ER types within tissues. This is consistent with the fact that either ER
or ER
could bind the ERE and that variants of ERs have been demonstrated in nonimmune tissues previously (24, 25). Together the alteration in cytokine production during in vitro estriol treatment of immune cells and the binding of splenic protein to ERE indicated that functional ERs are expressed in murine secondary lymphoid tissues.
|
expression at the RNA and protein levels
ER
expression has been reported in bone marrow and thymus (26, 27, 28), but direct evidence of expression in murine peripheral immune tissues is lacking. While RT-PCR data detecting ER
transcripts in peripheral immune tissues have been reported in mice, there has been no sequencing of these products (29, 30). Furthermore, expression of ER
at the protein level in the peripheral immune system has remained poorly characterized since Western blot data were negative for specific bands of the correct size (31) while flow cytometry of immune cells was positive (29, 32, 33). Together these previous reports raised the question of whether the classic ER
, vs another ER variant (24, 25), was detected in immune cells by flow cytometry. Thus, we characterized ER
expression in secondary lymphoid tissues of mice at the RNA level by both nested RT-PCR and sequencing and at the protein level by Western blot using both Ab and ligand binding. Nested RT-PCR using ER
-specific primers revealed bands of the expected size in both spleen (Fig. 2A) and lymph node (data not shown). The product from the inner primer set was then sequenced and confirmed to be ER
. Western blots of splenic tissue using an ER
-specific Ab detected expression at the protein level at the expected size of 67 kDa (Fig. 2B). ER
expression at the protein level was then confirmed by probing Western blots with 17
estradiol-BSA whereby a band was identified in spleen which colocalized to the size of the band detected with ER
-specific Ab probing (Fig. 2C). Together these data are conclusive evidence of the expression of classic ER
in murine peripheral immune tissues.
|

While an estrogen-mediated protective effect had been previously reported in EAE and in CIA, the mechanism of the immune protection has remained unknown. Therefore, we used the EAE model to determine whether the estrogen induced immune protection in this prototypic Th1 autoimmune disease was mediated through ER
. Since mechanisms of action of estrogens have been shown to differ depending upon whether physiologic or pharmacologic doses are used (34), we used a dose of estriol for EAE treatment which had previously been shown to induce estriol levels in sera of mice which were physiologic with natural murine pregnancy (2). Homozygous ER
KO mice were not protected from disease during estriol treatment, while WT mice were protected as evidenced by a significant decrease in disease severity in estriol vs placebo treated, p < 0.0001 (Fig. 3). These data demonstrate that the protective effect of estriol in EAE is dependent upon the presence of an intact ER
.
|

Significant reductions in proinflammatory cytokine production (TNF-
, p < 0.005; IFN-
, p < 0.005; and IL-2, p < 0.05) were observed during autoantigen-specific immune responses of WT mice treated with estriol as compared with placebo (Fig. 4). Importantly, these reductions did not occur in ER
knockout mice treated with estriol. Since not all cytokines were reduced by estriol treatment, apoptosis was an unlikely explanation for the decrease in the above cytokines in estriol-treated WT mice. Levels of the Th2 cytokine IL-5 were increased in estriol-treated WT mice as compared with placebo-treated WT mice. This estriol-induced increase in IL-5 production was less dramatic but still present in ER
KO mice. Together these data demonstrate that both the disease protection and the reduction in proinflammatory cytokine production induced by estriol treatment in EAE are mediated through ER
.
|
| Discussion |
|---|
|
|
|---|
was shown to be critical in estrogen-induced protection against brain injury during stroke (34). Direct estrogen-induced, ER
-mediated protective effects on the CNS remain possible in EAE as well. However, the fact that estrogen treatment influences a wide variety of autoimmune diseases which do not have the CNS as a target organ indicates that the protective effects of estrogen are clearly not limited to direct effects on the CNS. This conclusion is consistent with the finding in this report that the estrogen induced alterations in cytokine production during autoantigen-specific immune responses were mediated through ER
.
Regarding the effects of estrogen treatment on cytokine production, it is interesting to contrast our findings with those in a previous report in which estradiol was given in vivo and primary T cell responses to OVA were shown to be enhanced for Th1 cytokine production, with this effect dependent upon ER
(35). While both studies agree that ER
is important for immune efffects, the studies differ in that Th1 cytokine production was decreased by estrogen treatment in our paper while being increased in the previous paper. The differences between the two papers could be that estriol was used in our studies while estradiol was used in the previous study. Estriol and estradiol are clearly not the same. Estradiol has been shown to lead to hypertension, increased coagulation and to protect from osteoporosis more than estriol (36, 37, 38, 39, 40, 41). They can even have opposing effects. For example, estriol can protect from estradiol-mediated breast cancer (42, 43). However, since reports by three groups have shown that both estradiol and estriol are protective in both EAE and CIA (2, 3, 44, 45), there do not appear to be major differences between the two estrogen types in immune effects on disease. Indeed, the finding that estradiol treatment increases Th1 cytokine production during OVA-specific immune responses (35) is in direct contrast to previous reports by several groups that estradiol treatment ameliorates the prototypic Th1-mediated disease EAE. Perhaps differences between the effect estradiol treatment on OVA-specific responses in healthy mice and autoantigen-specific responses in EAE mice relates to either the strain or the age of the mice treated. F5 generation C57BL/6 mice were used in the former study while F16 C57BL/6 mice were used in our study. However, such strain differences also appear to be an unlikely explanation since estradiol has been shown to ameliorate EAE in numerous strains (C57BL/6, SJL, B10.PL, B10RIII). The age of the mice treated remains a consideration since the former report involved treatment of 4-wk-old mice, while all the EAE studies involved treatment of adult mice at least 8 wk of age. Differences in the nature of the immune response to OVA vs autoantigen is also a consideration.
While both estadiol and estriol have been shown to ameliorate EAE, we have focused considerable attention on estriol treatment of EAE for two reasons. First, unlike estradiol which is present during the menstrual cycle of the nonpregnant state, estriol is not present in appreciable amounts in the nonpregnant state. The nonpregnant female state is not a time of protection from autoimmune diseases such as MS and RA, since nonpregnant females are known to be highly susceptible. Estriol becomes detectable only during the first trimester of pregnancy and increases progressively throughout gestation, and pregnancy is a state of relative protection from MS and RA. Thus, there a correlation between disease protection and the presence of estriol, while the relationship between disease protection and the presence of estradiol appears more complex. Second, we focused on estriol treatment of EAE as a translational research strategy for MS in light of the fact that estriol is known to have less toxicity in humans than estradiol (36, 37, 38). The ultimate goal of the findings in this manuscript are to define the precise mechanism of action of the therapeutic efficacy of estriol treatment in EAE. These findings will then be contrasted to findings of the precise mechanism of action of estrogen related toxicities such as increased risk of breast cancer, endometrial cancer, and vascular disease. Our findings suggest that therapeutic efficacy may be achieved through stimulation of ER
. If it is ultimately shown that some or all toxicities are mediated through ER
or nongenomic effects, then a window of selectivity exists whereby one could treat disease with an ER
agonist or other SERM to achieve efficacy and minimize toxicity. In contrast, if it is ultimately shown that all toxicities are mediated through ER
, then selectivity will not be possible at the level of the ER. However, selectivity may still be achievable through modulation of tissue-specific cofactors involved in ER
action. To pursue the latter, one must first define the ER used in a given tissue for a given biologic effect. Then one may begin to identify tissue-specific differences in cofactors for each of the biologic effects.
In conclusion, the protective effect of estrogen treatment on autoantigen-specific immune responses in a prototypic Th1-mediated autoimmune disease is mediated through ER
. This does not however rule out additional nongenomic (membrane, ER-X) or genomic (ER
, ER
) effects on estrogen-mediated immune protection. The finding of an important role for ER
carries far reaching implications for the design of SERMs for the treatment of a wide variety of autoimmune diseases which aim toward optimizing efficacy and minimizing toxicity. They are equally as relevant to the design of SERMs for treatment of nonautoimmune diseases with different target organs (the reproductive system, the cardiovascular system, bone, breast, colon, and prostate) which aim to avoid effects on the immune system.
| Footnotes |
|---|
1 Address correspondence and reprint requests to Dr. Rhonda Voskuhl, Reed Neurological Research Center, Room A145, 710 Westwood Plaza, Los Angeles, CA 90095. E-mail address: rvoskuhl{at}ucla.edu ![]()
2 Abbreviations used in this paper: ER, estrogen receptor; EAE, experimental autoimmune encephalomyelitis; SERM, selective ER modifier; ERE, estrogen response element; SLE, systemic lupus erythematosis; CIA, collagen-induced arthritis; MOG, myelin oligodendrocyte glycoprotein; WT, wild type; MS, multiple sclerosis; RA, rheumatoid arthritis. ![]()
Received for publication May 29, 2003. Accepted for publication October 10, 2003.
| References |
|---|
|
|
|---|
production and reduces the severity of experimental autoimmune encephalomyelitis in cytokine knockout mice. J. Immunol. 167:542.
is expressed in the estrogen receptor
knockout mouse. J. Mol. Endocrinol. 29:281.[Abstract]
in mouse splenic lymphocytes: possible involvement in immunity. Toxicol. Lett. 133:221.[Medline]
gene responsible for the transient transcription of a variant receptor. Endocrinology 142:4106.
, not
, is a critical link in estradiol-mediated protection against brain injury. Proc. Natl. Acad. Sci. USA 98:1952.
expression in hematopoietic cells. Eur. J. Immunol. 33:512.[Medline]
This article has been cited by other articles:
![]() |
D. Fairweather, S. Frisancho-Kiss, and N. R. Rose Sex Differences in Autoimmune Disease from a Pathological Perspective Am. J. Pathol., September 1, 2008; 173(3): 600 - 609. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Tiwari-Woodruff, L. B. J. Morales, R. Lee, and R. R. Voskuhl Differential neuroprotective and antiinflammatory effects of estrogen receptor (ER){alpha} and ERbeta ligand treatment PNAS, September 11, 2007; 104(37): 14813 - 14818. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. H. Straub The Complex Role of Estrogens in Inflammation Endocr. Rev., August 1, 2007; 28(5): 521 - 574. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Dahlman-Wright, V. Cavailles, S. A. Fuqua, V. C. Jordan, J. A. Katzenellenbogen, K. S. Korach, A. Maggi, M. Muramatsu, M. G. Parker, and J.-A. Gustafsson International Union of Pharmacology. LXIV. Estrogen Receptors Pharmacol. Rev., December 1, 2006; 58(4): 773 - 781. [Full Text] [PDF] |
||||
![]() |
L. B. J. Morales, K. K. Loo, H.-b. Liu, C. Peterson, S. Tiwari-Woodruff, and R. R. Voskuhl Treatment with an estrogen receptor alpha ligand is neuroprotective in experimental autoimmune encephalomyelitis. J. Neurosci., June 21, 2006; 26(25): 6823 - 6833. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Hildebrand, W. J. Hubbard, M. A. Choudhry, B. M. Thobe, H.-C. Pape, and I. H. Chaudry Are the protective effects of 17{beta}-estradiol on splenic macrophages and splenocytes after trauma-hemorrhage mediated via estrogen-receptor (ER)-{alpha} or ER-{beta}? J. Leukoc. Biol., June 1, 2006; 79(6): 1173 - 1180. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. C. Lambert, E. M. Curran, B. M. Judy, G. N. Milligan, D. B. Lubahn, and D. M. Estes Estrogen Receptor {alpha} (ER{alpha}) Deficiency in Macrophages Results in Increased Stimulation of CD4+ T Cells while 17{beta}-Estradiol Acts through ER{alpha} to Increase IL-4 and GATA-3 Expression in CD4+ T Cells Independent of Antigen Presentation J. Immunol., November 1, 2005; 175(9): 5716 - 5723. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Mao, V. Paharkova-Vatchkova, J. Hardy, M. M. Miller, and S. Kovats Estrogen Selectively Promotes the Differentiation of Dendritic Cells with Characteristics of Langerhans Cells J. Immunol., October 15, 2005; 175(8): 5146 - 5151. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. M. Palaszynski, D. L. Smith, S. Kamrava, P. S. Burgoyne, A. P. Arnold, and R. R. Voskuhl A Yin-Yang Effect between Sex Chromosome Complement and Sex Hormones on the Immune Response Endocrinology, August 1, 2005; 146(8): 3280 - 3285. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. H. van den Broek, J. G. Damoiseaux, M. H De Baets, and R. M. Hupperts The influence of sex hormones on cytokines in multiple sclerosis and experimental autoimmune encephalomyelitis: a review Multiple Sclerosis, June 1, 2005; 11(3): 349 - 359. [Abstract] [PDF] |
||||
![]() |
M M. Elloso, K. Phiel, R. A Henderson, H. A Harris, and S. J Adelman Suppression of experimental autoimmune encephalomyelitis using estrogen receptor-selective ligands J. Endocrinol., May 1, 2005; 185(2): 243 - 252. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Garidou, S. Laffont, V. Douin-Echinard, C. Coureau, A. Krust, P. Chambon, and J.-C. Guery Estrogen Receptor {alpha} Signaling in Inflammatory Leukocytes Is Dispensable for 17{beta}-Estradiol-Mediated Inhibition of Experimental Autoimmune Encephalomyelitis J. Immunol., August 15, 2004; 173(4): 2435 - 2442. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |