AS and ACG are equal contributors to this study.
Steroid hormone receptor expression and function in microglia
Article first published online: 19 FEB 2008
Copyright © 2008 Wiley-Liss, Inc.
Volume 56, Issue 6, pages 659–674, 15 April 2008
How to Cite
Sierra, A., Gottfried-Blackmore, A., Milner, T. A., McEwen, B. S. and Bulloch, K. (2008), Steroid hormone receptor expression and function in microglia. Glia, 56: 659–674. doi: 10.1002/glia.20644
- Issue published online: 12 MAR 2008
- Article first published online: 19 FEB 2008
- Manuscript Accepted: 24 DEC 2007
- Manuscript Received: 5 JUN 2007
- NIH. Grant Numbers: AG16765, NS007080, HL18974, DA08259
- Moody Foundation
- Ministerio de Educación y Ciencia (Spain)
- steroid hormones;
- brain inflammation
- Top of page
- MATERIALS AND METHODS
Steroid hormones such as glucocorticoids and estrogens are well-known regulators of peripheral immune responses and also show anti-inflammatory properties in the brain. However, the expression of steroid hormone receptors in microglia, the pivotal immune cell that coordinates the brain inflammatory response, is still controversial. Here we use real time RT-PCR to show that microglia, isolated from adult fms-EGFP mice by FACS, express glucocorticoid receptor (GR), mineralocorticoid receptor (MR), and estrogen receptor alpha (ERα). GR was the most abundant steroid hormone receptor transcript in microglia. The presence of GR and ERα immunoreactivity was further confirmed in vivo at the ultrastructural level. To understand the role of steroid hormone receptors during the inflammation process, we evaluated the expression of steroid hormone receptors after inflammatory challenge and found a significant down-regulation of GR, MR, and ERα in microglia. Finally, we tested the immunomodulatory properties of estrogens and glucocorticoids. Estradiol benzoate did not have any significant impact on the inflammatory profile of ex vivo sorted microglia, either in resting conditions or after challenge. Furthermore, corticosterone was a more consistent anti-inflammatory agent than 17β-estradiol in vitro. Our results support the hypothesis that adult microglia are a direct target of steroid hormones and that glucocorticoids, through the predominant expression of GR and MR, are the primary steroid hormone regulators of microglial inflammatory activity. The down-regulation of steroid hormone receptors after LPS challenge may serve as a prerequisite to suppressing the anti-inflammatory actions of endogenous steroid hormones on the immune system, and contribute to a sustained activation of microglia. © 2008 Wiley-Liss, Inc.
- Top of page
- MATERIALS AND METHODS
The neuroendocrine system is a powerful regulator of inflammatory responses. Pharmacological doses of steroid hormones suppress virtually every step of the immune response; however, physiological doses are capable of more subtle modulatory effects (Sternberg, 2001). Steroid hormones are a family of mediators derived from the common precursor cholesterol that includes glucocorticoids, involved in stress responses; mineralocorticoids, which control osmotic balance, androgens, estrogens, and progestagens (Sierra, 2004). Glucocorticoids have been used for more than 50 years in the treatment of several diseases with an inflammatory component, such as asthma, rheumatoid arthritis, and autoimmune diseases (Rosen and Miner, 2005). Sex hormones are also thought to play a significant modulatory role, contributing to sexual dimorphism in the incidence of auto-immune and inflammatory diseases (Sternberg, 2001). Steroid hormone receptors commonly act as nuclear transcription factors, but also participate in activating rapid non-genomic pathways. Steroid hormone receptors are expressed in all the cell types involved in the inflammatory process, including macrophages (Cutolo et al., 1996; Werb et al., 1978), dendritic cells (professional antigen presenting cells) (Mao et al., 2005), T lymphocytes (Cohen et al., 1983; Lippman and Barr, 1977), and endothelial cells (Dietrich, 2004). Nevertheless, the expression of steroid hormone receptors in microglia, the primary immune cell of the brain, is still controversial.
Microglia are critical for proper innate immune responses in the brain because they orchestrate the onset of the inflammatory process after injury or inflammatory challenge, and clean up cell debris and pathogens. Blood-borne cells and resident astrocytes, which express steroid hormone receptors as well (Bohn et al., 1991; Garcia-Ovejero et al., 2002) also contribute to the inflammatory response. Both glucocorticoids (Nadeau and Rivest, 2003) and 17β-estradiol (estradiol) (Vegeto et al., 2003) prevent the activation of microglia by bacterial lipopolysaccharides (LPS) in vivo. Despite some reports showing the expression of glucocorticoid, mineralocorticoid (Tanaka et al., 1997), and estrogen receptors (Baker et al., 2004; Dimayuga et al., 2005; Liu et al., 2005; Mor et al., 1999; Vegeto et al., 2001) in vitro, attempts to identify steroid hormone receptors in resting microglia in vivo by conventional immunohistochemical techniques or in situ hybridization have proved unsuccessful (Garcia-Ovejero et al., 2002, 2005; Patel and Bulloch, 2003; Simerly et al., 1990; Van Eekelen et al., 1988).
To determine if microglia express steroid hormone receptors, we have applied an ex vivo approach using p7.2fms-EGFP mice, in which EGFP (enhanced-green fluorescent protein) is expressed in microglia (Sasmono et al., 2003; Sierra et al., 2007). The expression of glucocorticoid (GR), mineralocorticoid (MR), estrogen alpha and beta (ERα,β), androgen (AR), and progestin (PR) receptors was evaluated by real time RT-PCR, in microglia sorted by fluorescence-activated cell sorting (FACS) from male and female mouse brains, and after inflammatory challenge. Then we confirmed the expression of GR and ERα immunoreactivity by electron microscopy. Finally, we tested the immunomodulatory properties of glucocorticoids and estradiol. We used our ex vivo system to assess the effect of in vivo treatment with estradiol benzoate (EB) upon microglia derived from ovariectomized females; and primary cultures of microglia to compare the anti-inflammatory properties of corticosterone and estradiol at a range of concentrations.
MATERIALS AND METHODS
- Top of page
- MATERIALS AND METHODS
For these studies, the transgenic mouse line p7.2fms-EGFP (C57BL6/6 X CBA background) was used (Sasmono et al., 2003), in which the expression of the enhanced green fluorescent protein (EGFP) is driven by the promoter and the regulatory elements of the c-fms gene, resulting in microglial EGFP expression (Sasmono et al., 2003). Adult animals (2–3 m.o.) were bred in Rockefeller University facilities under 12:12 light: dark cycle and free access to chow and water. All experimental procedures were approved by the Rockefeller University Animal Care and Use Committee and conform to National Institutes of Health guidelines. Two weeks before sacrifice, female mice were bilaterally ovariectomized under Ketamine (100 mg/kg; Ketaset, Fort Dodge Animal Health, Fort Dodge, IO) plus Xylazine (10 mg/kg; AnaSed, Lloyd Laboratories, Shenandoah, IO) anesthesia and then allowed to recover from the surgery for 10 days. The animals were divided randomly into control and experimental groups and injected s.c. either with 0.1 mL of sesame-oil vehicle or 5 μg of estradiol benzoate (EB) in 0.1 mL of sesame oil for two consecutive days (5 μg/day per animal in sesame oil; S-3547 Sigma, St.Louis, MO). Trunk blood was collected at the time of sacrifice and plasma estradiol concentrations were measured by radioimmunoassay using reagents from a Coat-a-Count Estradiol Kit (Diagnostic Products, Los Angeles, CA, USA). To induce inflammation, mice received a single injection with Salmonella typhimurium lipopolysaccharides (LPS; 1–5 mg/kg, i.p.; L2262 Sigma).
Microglia Isolation by Fluorescence Activated Cell Sorting (FACS)
A single population of microglia was obtained by FACS (Sierra et al., 2007). In brief, adult mice were anaesthetized with pentobarbital (750 mg/kg) and rapidly decapitated. Brains were removed and placed on ice in Hank's balanced salt solution (Gibco, Carlsbad, CA), and meninges, blood vessels, and choroid plexus were carefully removed under a dissecting scope. Brains were minced and incubated with type II-S collagenase (600U; Sigma) and DNAse (450U; Invitrogen, Carlsbad, CA) for 30 min at 37°C in 15 mL HBSS supplemented with 90 mM CaCl2. After incubation, the cell suspension was further refined by repetitive gentle pipetting with fire-polished Pasteur pipettes on ice followed by filtering through a 40-μm cell strainer (BD, Franklin Lakes, NJ).
Cells were pelleted by centrifugation and subjected to the following procedures. (Note: All centrifugations of cell culture suspensions resulting in pellet formation were carried out in a Sorvall RC-3B centrifuge (2500 rpm, 5 min, 4°C) unless otherwise stated). Percoll gradients were used to further enrich for viable microglia. A stock solution of isotonic Percoll (SIP) (Amersham Biosciences, Upsala, Sweden) composed of a 9:1 Percoll to 1X PBS was prepared. All SIP dilutions were made with 1X PBS. Cell pellets were resuspended in 70% SIP, and placed in 14 mL round-bottom polypropylene tubes (1 mL/tube). The cell suspension in each tube was then overlaid with 1.5 mL of 37% SIP, followed by 0.5 mL of PBS. Gradients were centrifuged without brake in a Sorvall RC5C Ultracentrifuge (1200 g, 20 min, 20°C). Cells were collected from the 30/70 interphase, washed with 5% fetal calf serum (FCS, Sigma)-PBS, spun down, and resuspended in 5% FCS-PBS containing 100 ng/mL propidium iodide (PI), before sorting in a FACS Vantage SE Flow Cytometer (BD), with smHighPurity precision. Post-sort analysis was performed to ensure the purity of the collection process.
Primary Microglia Cultures and Microglial Stimulation for Cytokine Production
Microglia cultures were prepared following standard protocols (Hassan et al., 1991). Briefly, 2-day-old mouse pup brains were dissected on ice, and the meninges were carefully removed. The forebrains were minced in 5% FCS-PBS, dissociated using fire polished Pasteur pipettes, and then passed through a 40 μM nylon cell strainer (BD). Cells were washed once in buffer and seeded in culture media at a density of roughly two forebrains per 75 cm2 flask. Culture media, 10%FCS Dulbecco's modified Eagle's medium (DMEM; Gibco) with or without phenol red as a pH indicator, was changed every 5 days and supplemented with 5 ng/mL granulocyte-monocyte colony stimulating factor (GM-CSF; Sigma). After 2 weeks in culture at 37°C, 5%CO2, cells were shaken at 125 rpm for 5 h at 37°C to harvest detached microglia.
Microglia were counted and seeded in 24-well assay plates at a density of 0.25 million cells/well. After plating, microglia were allowed to adhere for 1 h and then rinsed to remove nonadherent glial cells, fed, and incubated as described earlier. The following day, cells were rinsed and incubated with DMEM containing 17β-estradiol (Sigma), corticosterone (Sigma), or vehicle (ethanol, 1:500,000) at different doses (see Results) for 10 min before LPS and interferon gamma (INFγ) were added at a final concentration of 1%FCS, 100 ng/mL LPS and 10 ng/mL INFγ. (In previous experiments carried out with cells grown in charcoal stripped serum, we found similar results to those reported here.). Minimal serum concentrations were used during LPS stimulation (1% FCS), to avoid the interference of hormone binding proteins and other serum factors, which could alter the results. Supernatants were collected 24 h later and assayed for cytokines by ELISA (eBioscience, San Diego, CA) and for nitric oxide by the Greiss assay (Promega) following manufacturers' protocols.
To assay the activity of kinases p38 and p44/42 mitogen-activated protein kinase (MAPK), microglia were plated into 12-well assay plates at a density of 0.5 million cells/well, and rinsed as described earlier. The following day serum was removed and the cells were serum-starved for 24 h; after which they were rinsed and placed in DMEM containing estradiol (10 nM) or corticosterone (1 μM) for 30 min at 37°C. For p44/42 cells were harvested immediately; whereas, for p38MAPK assays, LPS+INFγ was added for an additional 30 min before harvesting of the cells.
Adult microglia were sorted by FACS into RNA lysis buffer [Absolute RNA Microprep kit (Stratagene, La Jolla, CA)], frozen in dry ice, and kept at −80°C until processing. RNA was isolated using the Absolute RNA Microprep kit including a step with DNAse incubation (Qiagen, Valencia, CA) and quantified with RiboGreen RNA Quantitation kit (Molecular Probes), following manufacturer instructions. Ten nanogram of RNA were retrotranscribed with SuperScript II Reverse Transcriptase (Invitrogen) and 3 μl of a 1:3 dilution of the cDNA were amplified by real-time PCR using SYBR Green master mix (AB, Foster City, CA) in a 7900HT SDS thermal cycler (AB), with conventional AB cycling parameters (40 cycles of 95°C, 15 s, 60°C 1 min). After amplification, a denaturing curve was performed to ensure the presence of unique amplification products. The steroid hormone receptor transcripts quantified were AR, ERα, ERβ, GR, MR, and PR, as well as the cytokines TNFα, IL-6, and TGFβ1. Primers sequences are indicated in Table 1; amplicons span at least one intron in order to avoid potential genomic DNA amplification. Each sample was run in triplicate and the average Ct (threshold cycle) was used to calculate the relative amount of product by the Ct-method (AB), using the ribosomal L27A as a housekeeping gene. In each experiment, both positive (brain RNA) and negative (RT minus and water) controls were included.
|Gene||GeneBank||Amplicon size||Sequence (5′ - 3′)||Reference|
For absolute quantification experiments, the GR, MR, ERα, and L27A amplicons were cloned. Brain RNA was isolated using RNeasy Mini kit (Qiagen) and 2 μg were retro-transcribed with SuperScript II Reverse Transcriptase (Invitrogen). The amplicons were generated by PCR with TaqMan polymerase (Invitrogen) and were cloned using the TOPO TA cloning kit with TOP10F' E. coli (Invitrogen), following manufacturer instructions. In short, amplicons were cloned in a pCR4-TOPO vector and competent TOP10F' cells were chemically transformed in the presence of X-galactosidase (X-Gal; Promega, Madison, WI) and isopropyl-β-D-thiogalactopyranoside (IPTG, Promega). Plasmidic DNA was then isolated using QIAprep Miniprep columns (Qiagen) and analyzed by PCR screening. Plasmids containing the desired amplicon were quantified in triplicate by conventional spectrophotometry (absorbance at 260 nm) and with RiboGreen (Molecular Probes; plasmidic DNA of known concentration was used as a reference). Absolute standard curves of plasmidic DNA, containing 300,000-30 plasmid copies per reaction, were run in real-time PCR using SYBR Green master mix (AB) in a 7900HT SDS thermal cycler (AB).
200,000 adult microglia cells were collected by FACS into PBS, centrifuged at 14,000 rpm for 5 min, and frozen at −80°C until processing. Cell pellets were resuspended in protein lysis buffer [6 M Urea, 20 mM Tris-HCl pH7.5, 2% SDS, 10% glycerol, 1% protease inhibitor cocktail (Sigma)] and heated at 70°C for 5 min. Alternatively, cultured microglia were rinsed in cold PBS supplemented with Ca2+Mg2+ and then resuspended in protein lysis buffer supplemented with phosphatase inhibitor (1 M NaVO4; Sigma). Cells were sonicated to homogeneity and then quantified using the BioRad Dc protein assay (BioRad, Hercules, CA). Samples were stored at −20°C until processed by Western blot.
Equal amounts of protein were separated by SDS-PAGE performed under reducing conditions with 4%–12% acrylamide NuPage gels according to manufacturer's instructions (Invitrogen). Resolved proteins were transferred to PDVF membranes (Invitrogen). Membranes were rinsed in 0.1 M Tris-Buffered Saline with 0.1% Tween-20 (TBS-T) and blocked with a solution of 5% nonfat dry milk in TBS-T for 1 h at room temperature on an orbital shaking platform. Membranes were washed with TBS-T and incubated overnight at 4°C with anti-ERα (NCL-ER-6F11; NovoCastra, Newcastle, UK; 1:1,000) or anti-GR antiserum (M20; Santa Cruz Biotechnology, Santa Cruz, CA; 1:2,000) diluted in 5% BSA in TBS-T solution. Blots with cultured microglia samples were incubated overnight at 4°C with anti-phospho-p38 and anti-p38MAPK antiserum (#9211L, #9102; Cell Signaling, Danvers, MA; 1:2,000) or anti-phospho p44/42 and anti-p44/42MAPK antiserum (#9102, #9101S; Cell Signaling; 1:2,000).
Membranes were washed and incubated with Horseradish Peroxidase-conjugated species-specific anti-antiserum (Pierce, Rockford, IL; 1:20,000) in blocking solution. After washing, membranes were developed with SuperSignal West Pico substrate (Pierce, Rockford, IL), and then exposed to X-Ray film (X-OMAT AR; Kodak, Rochester, NY). To control for protein loading, membranes were incubated in Restore Western Blot Stripping Buffer (Pierce), washed, and immunoblotted as described earlier using an anti-Actin antibody (Sigma; 1:40,000). Developed films were analyzed for densitometry using a computerized image analysis software (MCID-M4; Imaging Research, St. Catherines, ON), and the data was normalized as follows: GR protein band (density × area)/actin (density × area). For MAPK phosphorilation assays, samples were blotted for total MAPK and phospho- MAPK. Densitometry of total MAPK protein (density × area) was used to normalize phospho-MAPK readings (density × area).
Tissue Preparation for Electron Microscopy
Three adult ovariectomized (ovx) fms-EGFP females were deeply anesthetized with sodium pentobarbital (150 mg/kg) and then sequentially perfused with 10–15 mL heparin (1000 U/mL) in normal saline, followed by 40 mL 3.75% acrolein (Polysciences, Washington, PA) in 2% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.4). Brains were removed from the skulls, and postfixed for 30 min in the fixative. Sections (40 μM thick) through the hippocampal formation were cut on a Vibratome (Leica) and collected in PB. Prior to immunochytochemical labeling, sections were treated with 1% sodium borohydride in PB for 30 min to remove active aldehydes. Sections were rinsed in PB, and then in TBS (pH 7.6) before incubating for 30 min in 1% BSA-TBS to minimize nonspecific binding.
Free-floating sections were processed for GFAP, GR, ERα, and ERβ immunolabeling using a pre-embedding peroxidase and immunogold-silver dual labeling method described previously (Towart et al., 2003). Sections were incubated for 24 h at room temperature and for 4 days at 4°C in anti-GR, anti-ERα, or anti-ERβ antisera diluted in 0.1% BSA-TBS. The mouse monoclonal anti-GR-2 antibody (1:1,500; BuGR clone; Affinity BioReagents, Golden, CO) has been previously compared with other anti-GR antisera and may prefer the conformed ligand-bound form of GR (Johnson et al., 2005). The rabbit polyclonal antiserum against the near full-length peptide of the native rat ERα (amino acids 61 through the carboxyl terminus; 1:10,000; clone AS409) was a generous gift of S. Hayashi (Tohoku University, Japan). This antibody has been tested previously for specificity and has been shown to recognize both the ligand-bound and the unbound receptors (Alves et al., 1998; Milner et al., 2001; Okamura et al., 1992). Finally, the anti-ERβ was a polyclonal antibody (clone 485; 1:10,000; Merck Research Laboratories, Rahway, NJ) generated in rabbit against a conserved sequence (rat aa 64-82) of the mouse, human, and rat ERβ that is located within the A/B domain of ERβ (exons 2-3) and is not present in ERα (Mitra et al., 2003). This antibody has been thoroughly tested and its immunoreactivity shows similar cellular and subcellular locations compared to other anti-ERβ antisera (Milner et al., 2005). On day 3 of the incubation period, chicken anti-GFP antiserum (Aves Labs, Tigard, OR, final concentration 1:5,000) was added to the tissue incubating with the GR, ERα, or ERβ antiserum.
For immunoperoxidase labeling of GR, ERα, or ERβ sections were incubated with biotinylated donkey anti-rabbit IgG (Jackson Immunoresearch, West Grove, PA; 1:400) in BSA-TBS for 30 min. Sections were then rinsed in TBS, and incubated with avidin-biotin complex (Vector, Burlingame, CA) for 30 min. After rinsing in TBS, sections were incubated in 3,3′-diaminobenzidine (Aldrich, Milwaukee, WI) and H2O2 in TBS for 6 min. For immunogold labeling of GFP, sections were incubated in a 1:50 dilution of anti-chicken IgG conjugated to 1-nm colloidal gold particles (Electron Microscopy Sciences (EMS), Fort Washington, PA) in 0.001% gelatin and 0.08% BSA in 0.1 M phosphate-buffered saline (PBS) for 2 h at room temperature. Sections were rinsed in PBS, post-fixed in 1.25% glutaraldehyde in PBS, rinsed again in PBS followed by incubation in 0.2 M sodium citrate (pH 7.4) buffer. The conjugated gold particles were enhanced by incubation in silver solution (IntenSE; Ammersham) for 6–7 min.
Dually labeled sections were fixed for 60 min in 2% osmium tetroxide, dehydrated through a graded series of ethanols and propylene oxide, and embedded in EMBed 812 (EMS, Hatfield, PA) between two sheets of Aclar plastic (Milner and Veznedaroglu, 1992). Ultrathin sections (70–75 nm thick) through the dentate gyrus and the CA1 region of the midseptotemporal hippocampal formation were cut on a Leica UCT ultratome. Sections were counterstained with Reynold's lead citrate and uranyl acetate and examined with a Tecnai Biotwin electron microscope equipped with an AMT digital camera. Final microphotographs were generated from digital images in which levels, contrast, and brightness were adjusted using Adobe Photoshop 7.0.
Statistical analysis was performed using Statistica (StatSoft, Tulsa, OK). Experiments involving two goups were compared using a Student's t-test. Experiments involving more than two goups were compared by Analysis of Variance (ANOVA), followed by posthoc analysis with Tukey Honestly Significant Difference (HSD) test. or with Unequal-n HSD test when variances were homogeneous (using Levene's test); or with nonparametric Kruskal- Wallis, followed by Neuman-Keuls test. In experiments measuring the expression of TNFα or IL-6 (but not TGFβ1) mRNA in FACS-sorted microglia, the data histograms did not show a normal distribution nor were the variances homogeneous across groups using Levene's test and therefore, a logarithmic transformation was done to normalize the variables TNFα and IL-6 before ANOVA (22). In experiments measuring TNFα, IL-6, and NO in cell supernatants, experiments were conducted utilizing estradiol and corticosterone in parallel. Results were normalized to 100% from the LPS+INF controls. Throughout the text, we have stated the post-hoc statistic used (Unequal-n HSD, Neuman-Keuls) when a significant difference was found by ANOVA or Kruskal-Wallis respectively. In linear regression analysis, the correlation coefficient R2 and P-value are indicated. Graphs show the mean and the standard error of the mean (SEM).
- Top of page
- MATERIALS AND METHODS
Conventional methodologies such as immunohistochemistry or in situ hybridization have proved to be unsuccessful in identifying GR, ERα, or any other steroid hormone receptors in resting microglia. In this study, we have utilized a more sensitive means to evaluate the presence of steroid hormone receptors by utilizing FACS sorting of live microglia from the adult brain of fms-EGFP mice followed by ex vivo gene expression analysis by real time RT-PCR.
GR, MR, and ERα Messages are Expressed in Ex Vivo Microglia
In the brain parenchyma of p7.2fms-EGFP mice, microglia selectively expressed EGFP (Fig. 1A) (Sierra et al., 2007). In tissue sections, fms-EGFP cells were identified as microglia based on morphology and expression of Iba1 and CD11b [Fig. 1A and (Sierra et al., 2007)]. Pure populations of live microglia were obtained by FACS (Fig. 1B) (Sierra et al., 2007) and the expression of steroid hormone receptors was assessed using real time RT-PCR. To avoid detection of potential contaminant genomic DNA, primers were designed to span at least one intron. Microglial mRNA samples were run in parallel with an RT-control (microglial mRNA not retro-transcribed into cDNA) and a positive control (brain mRNA). The expression of AR, ERα, ERβ, GR, MR, and PR, and two ribosomal housekeeping genes, L27A and 28S (see Fig. 2) was assessed. We detected the expression of ERα, GR, and MR in microglia, but not that of AR, ERβ, or PR, as shown by the amplification curves (Fig. 2A). In quantification experiments, we chose to use only L27A to normalize the data, because its expression levels are in the same range as the steroid hormone receptors and therefore it has better quantification accuracy. The dissociation plots show unique peaks corresponding to unique amplification products of the expected Tm (melting temperature) (Fig. 2B).
To establish whether there are sex differences in the expression of mRNAs for the steroid hormone receptors, we compared males, and ovariectomized (ovx; 2weeks) females replaced with either vehicle or estradiol benzoate (EB, 5 μg/day per animal for 2 consecutive days) (n = 6). GR expression in ovx females was (2.0 ± 0.3)-fold of the male expression (P = 0.0357, Tukey HSD, Fig. 3A); the EB replacement decreased the expression of GR to (1.8 ± 0.2)-fold (n.s. from both males and ovx females). The expression of MR (Fig. 3B) and ERα (Fig. 3C) was not significantly different between males and females. ERβ, AR, and PR were not found in female microglia, either.
GR is the Most Abundant Steroid Hormone Receptor Transcript in Ex Vivo Microglia
Real time PCR consistently demonstrated throughout the experiments (n = 6) that the expression of GR mRNA had a smaller threshold cycle value (Ct = 28.9 ± 0.4) than MR (Ct = 30.8 ± 0.3) and ERα (33.3 ± 0.2). The Ct represents the cycle at which the fluorescence reaches the threshold intensity; thus, it is inversely proportional to the initial amount of target mRNA. This initial observation was further evaluated by absolute quantification of the mRNA expression using standard curves with known copies of plasmids containing the GR, MR, ERα, and L27a amplicons (see Fig. 4). These results support our original PCR data and showed that fms-EGFP sorted microglia expressed 14.1 ± 3.5 copies of GR, 0.30 ± 0.09 copies of MR and 0.17 ± 0.04 copies of ERα per copy of the housekeeping gene L27A.
GR and ERα Immunoreactivity in Microglia
The expression of GR and ERα protein in hippocampal microglia was confirmed by electron microscopic immunocytochemistry (see Fig. 5). In hippocampal sections of fms-EGFP female mice, microglia were visualized with an anti-GFP antiserum.
Single labeling with an anti-GR antibody revealed a strong immunostaining in the granule cell nuclei, which was visible at the light level, as expected (Lawson et al., 1991). In double-labeling experiments, GR immunoreactivity was seen in microglial profiles, located in the cytoplasm (Fig. 5A) and in the membrane (Figs. 5A,B). Although typically expressed in cell nuclei, GR labeling has been observed before in neuronal and glial membranes (Johnson et al., 2005).
Single labeling with an anti-ERα antibody (Okamura et al., 1992) revealed an extra-nuclear staining pattern, in axons, dendrites, and glial processes (data not shown) consistent with reports in the rat (Milner et al., 2001). Double labeling experiments revealed most ERα in neuronal profiles and rarely in microglial profiles (Figs. 5C,D). In some instances, we did observe microglia labeled with ERα in the cytoplasm of cell processes (Figs. 5E,F), and near the cell nucleus (data not shown). In some occasions, ER-immunoreactivity was detected in parenchymal microglia nearby blood vessels. No double-labeled processes were detected in sections dually labeled for GFP and ERβ (data not shown).
Acute Inflammation Down-Regulates the Expression of Steroid Hormone Receptors
To understand the role of steroid hormones during the inflammation process, we evaluated the expression of steroid hormone receptors after LPS challenge. Male fms-EGFP mice were treated with LPS (5 mg/Kg, n = 3–6) and fms-EGFP microglia were sorted 0, 3, 12, or 24 h later (Fig. 6A). Sorted microglia showed an increased dispersion in the forward scatter (FSC, related to cell surface area) axis along the time course (Fig. 6A), corresponding to the LPS-induced hypertrophy that is observed in tissue sections (Fig. 6B) (Sierra et al., 2007).
Along the time course after LPS challenge, GR, MR, and ERα were down-regulated (Figs. 6C–E). By 24 h, GR expression was down-regulated to 40.5% ± 7.9% (P = 0.0316, Unequal-n HSD) compared to control. The expression of MR was down-regulated to 42.5% ± 4.2% at 3 h (P < 0.0001, Neuman-Keuls) and down-regulated further to 11.1% ± 1.0% by 24 h (P = 0.0002, Neuman-Keuls). ERα expression was also down-regulated to 68.3% ± 13.4% at 3 h (P = 0.0025, Unequal-n HSD) and to 33.6% ± 6.5% by 24 h (P < 0.0017, Unequal-n HSD) compared to control. Furthermore, the expression of ERβ, AR, and PR remained undetectable (data not shown).
To ensure that LPS induced a microglial inflammatory state, the mRNA of three major cytokines, pro-inflammatory tumor necrosis factor alpha (TNFα), interleukin 6 (IL-6), and anti-inflammatory transforming growth factor beta1 (TGFβ1), was quantified (Figs. 6F–H). TNFα expression was rapidly and transiently up-regulated to (20.5 ± 1.3)-fold compared to control at 3 h (P = 0.0002, unequal-n HSD), decreased to (3.4 ± 0.6)-fold at 12 h (P = 0.0147 respect to control, P = 0.0008 respect to 3 h, unequal-n HSD) and decreased further to (2.3 ± 0.6)-fold at 24 h (P = 0.0399 respect to control, P = 0.0002 respect to 3 h, unequal-n HSD). IL-6 expression was up-regulated to (12.8 ± 1.2)-fold at 3 h (P = 0.0022, unequal-n HSD), was maintained elevated to 8.1 ± 1.9 fold at 12 h (P = 0.0111 respect to control, nonsignificant respect to 3 h, unequal-n HSD) and recovered basal levels at 24 h. TGFβ1 expression was down-regulated to 18.6% ± 1.1% by 12 h (P = 0.0002, Unequal-n HSD) and to 41.0% ± 5.9% by 24 h (P = 0.0002, Unequal-n HSD) compared to control. Interestingly, the expression of TGFβ1 along the time course correlated significantly with the expression of ERα (R2 = 0.8715, P < 0.0001) and that of MR (R2 = 0.5691, P < 0.0003) using linear regression analysis; whereas, GR mRNA expression was not significantly correlated with any cytokine.
LPS-induced down-regulation of GR protein was confirmed through Western blot of the sorted microglia from adult male fms-EGFP mice (Fig. 7A). A single band of a molecular weight of 87 kDa was present in microglia protein samples (Fig. 7A) and a 77.1% ± 5.2% down-regulation (P = 0.03791, t-test; Fig. 7B) was observed after LPS challenge (24 h). Immunoblotting with an anti-ERα antiserum did not show any band at the specific molecular weight of 67 kDa in microglia protein samples derived from either control or LPS-treated animals (Fig. 7A), confirming the low expression of ERα message observed before (see Fig. 4).
Estradiol Benzoate Does Not Appear to be Anti-inflammatory in Ex Vivo Microglia
A comparison of the immunomodulatory efficacy of glucocorticoids and estrogen in animals is not easily achieved because it involves multiple surgeries (adrenalectomy plus gonadectomy). Furthermore, in vivo challenging with LPS induces an activation of the hypothalamo-pituitary-adrenal axis that results in increased circulating levels of protective, anti-inflammatory glucocorticoids (Nadeau and Rivest, 2003; Nakano et al., 1987), and adrenalectomy produces a 40-fold increase in the lethal effects of LPS (Koniaris et al., 2001). Thus, we decided to test the modulatory effects of estrogens alone on microglial activity.
We compared the expression of cytokines in microglia sorted from ovariectomized females replaced with vehicle or with EB (see Fig. 8) that were injected with vehicle or with a lower dose of LPS (1 mg/kg, 3 h). EB replacement induced a significant increase in the circulating levels of estrogen (36.3 ± 2.1 pg/mL in ovx vs. 126.7 ± 55.6 pg/mL), but produced no significant changes in the mRNA expression of pro-inflammatory cytokines (IL-6, TNFα) or the anti-inflammatory cytokine, TGFβ1, either in resting conditions or 3 h after LPS stimulation (see Fig. 8).
Corticosterone, But Not Estradiol, Consistently Attenuates Anti-inflammatory Mediator Production by Activated Cultured Microglia
Since GR, MR, and ERα were the only steroid hormone receptors found in our system, we compared the immunomodulatory effects of glucocorticoids and estradiol in primary cultures of microglia. Cells were pretreated with corticosterone or estradiol and then challenged with LPS + interferon gamma (LPS+IFNγ) to elicit the production of inflammatory cytokines and NO (see Fig. 9). We tested the effects of corticosterone and estradiol over a range of concentrations (10 pM, 10 nM, and 10 μM), in line with other studies (Bruce-Keller et al., 2000; Ghisletti et al., 2005; Jacobsson et al., 2006). The production of the proinflammatory cytokines TNFα, IL-6 as well as NO, was assessed 24 h after treatment with LPS+INFγ either in conventional media (containing phenol red as a pH indicator dye; Figs. 9A,C,E) or in phenol red-free media (Figs. 9B,D,F), to ascertain potential pro-estrogenic effects of lipophilic impurities with estrogenic properties sometimes present in commercial preparations of phenol red (Bindal et al., 1988). TNFα, IL-6, and NO were not detectable in the absence of LPS+INFγ or with hormone treatment alone (data not shown).
Corticosterone (1 μM) significantly reduced TNFα secretion by 53.7% ± 4.6% in phenol red-containing media (P < 0.001, Neuman-Keuls test; Fig. 9A), and by 47.1% ± 4.9% in phenol red-free media (P < 0.001, Neuman-Keuls test; Fig. 9B) while it was ineffective at lower concentrations (10 nM). In contrast, estradiol did not significantly modulate microglial TNFα over the entire range of concentrations tested, in either phenol red-containing or phenol-free media (Figs. 9A,B). Similarly, the high dose of corticosterone (1 μM) significantly reduced the secretion of IL-6 by 45.0% ± 3.8% in phenol red-containing media (P < 0.001, Neuman-Keuls test; Fig. 9C), and by 47.6% ± 2.0% in phenol red-free media (P < 0.001, Neuman-Keuks test; Fig. 9D). In contrast, estradiol effects on IL-6 secretion were highly dependent on the concentration when tested in phenol red-containing media: the lowest concentration tested (10 pM) produced a small decrease in IL-6 production of 14.6% ± 6.3% (P = 0.01, Neuman-Keuls test). At a higher dose (10 μM), estradiol had no significant effect; yet at the highest dose (1 μM) it produced a small but significant increase of 16.9% ± 5.0% (P = 0.004, Neuman-Keuls test) (Fig. 9C). On the contrary, the highest dose (1 μM) produced a significant reduction in IL-6 production of 30.0% ± 8.5% in phenol red-free media (P < 0.05, Neuman-Keuls test; Fig. 9D). The production of NO was reduced by corticosterone (1 μM) by 25.5% ± 3.0% in phenol red-containing media (P < 0.001, Neuman-Keuls test; Fig. 9E) and by 8.4% ± 1.1% in phenol red-free media, (P < 0.05, Neuman Keuls test; Fig. 9F) whereas estradiol was ineffective in modulating NO production at any of the concentrations tested in either phenol red-containing or phenol-free media (Figs. 9E,F).
A key signaling molecule activated during the inflammatory response is the p38 mitogen-activated protein kinase (p38MAPK) (Han et al., 2005). LPS-induced phosphorylation of p38MAPK was prevented by a 30 min pre-treatment with 1 μM corticosterone [52.7% ± 1.9%, P = 0.05, Neuman-Keuls test), but a similar pretreatment with estradiol (10 nM) did not have a significant effect (Fig. 10A) in phenol red-containing media. Similarly, estradiol did not modulate the basal phosphorylation levels of the ERK1/2 (Fig. 10B), a MAPK whose phosphorylation is rapidly increased in response to estradiol in a variety of brain cells (Belcher and Zsarnovszky, 2001).
- Top of page
- MATERIALS AND METHODS
Expression of Steroid Hormone Receptors in Microglia
Using a novel, sensitive approach, based on a combination of FACS, real time RT-PCR, and Western blot (Sierra et al., 2007) we have detected the expression of GR, MR, and ERα in microglia from both male and female mice. Previously, the expression of steroid hormone receptors in microglia was a controversial issue, since their detection by conventional immunohistochemistry or in situ hybridization in the adult brain remained inconclusive (Garcia-Ovejero et al., 2002, Garcia-Ovejero et al., 2005; Patel and Bulloch, 2003; Simerly et al., 1990; Van Eekelen et al., 1988). In addition, we have corroborated the expression of GR and ERα in microglia at the ultrastructural level, showing that the pattern of immunoreactivity for these receptors in microglia is primarily extranuclear.
Several reasons can be proposed to explain why steroid hormone receptors had not readily been detected in microglia by conventional techniques. For instance, low expression levels of steroid hormone receptor transcripts, non-nuclear localization of the receptors, and low level of staining in thin microglial processes that can easily be regarded as background. Thus, the anti-inflammatory effects of steroid hormones in the brain have been suggested to be mediated through other cell types that express steroid hormone receptors (Soucy et al., 2005), such as astrocytes (Bohn et al., 1991; Garcia-Ovejero et al., 2002), the endothelial cells forming part of the blood brain barrier (Dietrich, 2004), or the circumventricular organs (de Kloet et al., 1986; Sibug et al., 1991). Our results showing the presence of GR, MR, and ERα in microglia indicates for the first time that in the adult rodent brain, microglia are a direct target of steroid hormones.
On the other hand, we did not detect the expression of ERβ, AR and PR in microglia under any of the conditions tested, although it is possible that their expression level is below the sensitivity threshold of our techniques. To the best of our knowledge, there is no evidence in the literature of PR expression in microglia (see ERβ and AR later). Interestingly, PR was not induced by an estradiol benzoate replacement treatment, as occurs in other types of estrogen-responsive cells (Mahesh and Muldoon, 1987), suggesting a low or selective effectiveness of estradiol in microglia, possibly as a consequence of the low expression levels of ERα. Furthermore, the pattern of steroid hormone receptors expression was essentially independent of the sex, although a statistically significant increased expression of GR in ovariectomized females compared to males was found. Sexual dimorphism has been found in incidence, symptom severity, and progression of several inflammatory and autoimmune diseases (Czlonkowska et al., 2005; Obendorf and Patchev, 2004) as well as neurodegenerative diseases including Parkinson's disease, Alzheimer's disease, multiple sclerosis, or stroke. Therefore, it is possible that differences in the effectiveness of steroid hormone-based treatments in anti-inflammatory therapies may exist.
Glucocorticoid Receptors in Microglia
Of all the steroid hormone receptors we examined in microglia, we find that GR is the most robustly expressed at both mRNA and protein level. GR was present in the membrane and the cytosol of microglial processes at the ultrastructural level in the hippocampus. Our results agree with previous reports showing a higher number of binding sites for GR than for MR in microglia (Tanaka et al., 1997). In our current study, using purified microglial cultures stimulated with LPS+IFNγ, we have showed that corticosterone acted consistently as an anti-inflammatory hormone, attenuating the production of TNFα, IL-6 and NO, while estradiol was not effective over an identical range of concentrations. Interestingly, corticosterone decreased the LPS+IFNγ-induced phosphorylation of p38 MAPK. Although GRs are typically ligand-activated transcription factors, they also mediate some rapid effects, some of which have been attributed to membrane, G protein-coupled receptors (Tasker et al., 2006). Our results support a model where glucocorticoids may serve through the activation of GRs as the main steroid hormone involved in down-regulating the microglial inflammatory response in mice.
Glucocorticoids have well-known immunosuppressive properties in the peripheral immune system which make them clinically useful for the treatment of autoimmune diseases, chronic inflammation, or to avoid transplant rejection (Dinkel et al., 2002). The rapid increase of glucocorticoid circulating levels after systemic inflammatory challenge has been interpreted as a protective, anti-inflammatory response (Nadeau and Rivest, 2003; Nakano et al., 1987); in agreement, adrenalectomy results in a 40-fold increase in the lethal effects of LPS (Koniaris et al., 2001). In the brain, high concentrations of pro-inflammatory cytokines such as TNFα promote glutamate mediated-excitotoxicity (Pickering et al., 2005), and therefore, a sustained activation of the brain innate immune responses induces neurodegeneration (Aloisi, 2005). Although glucocorticoids can exacerbate kainic acid-induced neurodegeneration (Dinkel et al., 2003), they can also mediate anti-inflammatory and neuroprotective responses. The GR/PR antagonist RU486 induces a dramatic neurodegeneration after intraparenchymal injection of LPS in the the substantia nigra (Castano et al., 2002) and the cerebral cortex (Nadeau and Rivest, 2003), by increasing and prolonging the expression of pro-inflammatory cytokines (Nadeau and Rivest, 2003). On the contrary, the GR agonist dexamethasone increases the survival of dopaminergic neurons after intranigral injection of LPS (Castano et al., 2002), suggesting that GRs play an essential role in protecting the brain against an inflammatory challenge.
Estrogen Receptors in Microglia
ERα showed the lowest levels of steroid hormone receptor transcript expression, a fact consistent with the absence of distinct ERα immuno-specific bands by Western blotting. Despite this low level of expression, we were able to confirm the presence of some ERα immunoreactivity in microglia at the ultrastructural level. ERα expressing microglia could be either resident cells or have been originated in the bone marrow. However, we did not observe any traces of ERβ immunoreactivity in mouse adult microglia. A series of studies have shown that primary cultures of microglia derived from the postnatal rat brain (Liu et al., 2005; Mor et al., 1999) express both ERα and ERβ, like the microglial cell line N9 (Baker et al., 2004; Dimayuga et al., 2005), but unlike the BV2 microglial line, which expresses only ERβ (Dimayuga et al., 2005). The discrepancy regarding the ERβ expression in microglia may be attributed to differences between species (mouse vs. rat), developmental stages (adult vs. postnatal) or cell immortalization. Nevertheless, in vivo experiments support a predominant role of ERα in regulating brain innate immune responses. For instance, the effects of estradiol on the number of microglial cells expressing activation markers such as the matrix metalloproteinase 9 (MMP9) (Vegeto et al., 2003), or the brain expression of pro-inflammatory cytokines (Soucy et al., 2005), induced by intracerebral injection of LPS inflammatory challenge are prevented in ERα- but not ERβ-KnockOut animals.
We have further shown that estradiol did not act consistently, if at all, as an anti-inflammatory agent in regulating the microglial expression of IL-6, TNFα, TGFβ1 mRNAs in vivo after peripheral administration of LPS. Mice were treated with an estradiol conjugate, estradiol benzoate, whose anti-inflammatory properties in vivo have been described before in peripheral models of inflammation (Begon et al., 2002; Ozveri et al., 2001). A possible explanation for this discrepancy with the literature cited above is that in our experiments LPS was administered systemically. Since LPS does not cross the blood brain barrier (Singh and Jiang, 2004), the activation of microglia we observe is likely to be secondary to peripheral pro-inflammatory cytokines. Additionally, it is possible that estradiol is affecting the microglial morphology or expression of activation markers, an effect that can not be ruled out from our experiments.
However, we did not observe any effects of estradiol in the production of TNFα and NO in cultured microglia either. The effects of estradiol on IL-6 production in cultured microglia were dependent on the presence of phenol red in the culture media. Phenol red is a common pH indicator dye that has been reported to show some pro-estrogenic effects due to lipophilic impurities with estrogenic properties sometimes present in commercial preparations of phenol red (Bindal et al., 1988). The production of IL-6 was increased by the highest dose of estradiol (1 μM) in phenol-containing media, but reduced in phenol-free containing media. In agreement, a number of studies have found no effects of estradiol on microglial activity when compared to other steroids (Johnson and Sohrabji, 2005; Lieb et al., 2003; Tanaka et al., 1997).
Our findings of extra-nuclear ERα immunoreactivity in microglia suggested that estradiol may exert nongenomic effects in these cells as is reported for neurons (Milner et al., 2005; Towart et al., 2003). Yet, acute estradiol treatment was ineffective in blocking the LPS-induced phosphorylation of p38 MAPK or modulating the basal levels of ERK1/2 MAPK. As we show here that ERα is expressed in microglia at very low levels, we hypothesize that other cell types expressing estrogen receptors, such as astrocytes Garcia-Ovejero et al., 2002) or endothelial cells (Dietrich, 2004), are contributing to the regulatory effects of estradiol on inflammatory processes observed in the brain in vivo.
In fact, the effects of estrogens on microglial activity are still controversial, although the differences reported by different authors may be due to different culture condicions, such as LPS concentration, serum concentration, or use of media containing phenol red. Suuronen and coworkers report anti-inflammatory effects of various selective estrogen receptor modulators (SERMs), but not of estradiol, and further suggest that SERM-induced modulation of LPS-activated pro-inflammatory signaling cascades is not estrogen receptor-mediated (Suuronen et al., 2005). For instance, the SERM ICI-182,780, a pure antagonist of ERs, has strong effects preventing microglial activation, whereas estradiol has no effects under the same culture conditions (Suuronen et al., 2005; Tanaka et al., 1997). Yet, several reports support a pure anti-inflammatory role of estradiol (Baker et al., 2004; Bruce-Keller et al., 2000; Dimayuga et al., 2005; Ghisletti et al., 2005; Liu et al., 2005; Mor et al., 1999; Vegeto et al., 2003, 2006). Some recent papers have shown otherwise, namely, that ovary-derived estradiol is essential for mounting a proper inflammatory response to bacterial LPS, suggesting that circulating estradiol is exerting a permissive, pro-inflammatory effect (Soucy et al., 2005). In addition, it has been shown that estradiol has pro-inflammatory properties in cultured microglia derived from aging females, while it has no significant effects in microglia derived from young females (Johnson and Sohrabji, 2005). Furthermore, the potential pro-inflammatory effects of estradiol have been proposed to contribute to the failure of clinical trials to prove the benefits of hormone replacement therapies (Bushnell, 2005; Stork et al., 2004). Thus, steroid hormones should be defined as immuno-modulatory agents, exerting permissive, stimulatory, or blocking effects depending on hormone concentration, endpoints measured, cell types studied, and inflammation models used, as observed for glucocorticoids (Dinkel et al., 2002; Sternberg, 2001; Yeager et al., 2004).
Down-Regulation of Steroid Hormone Receptors in Acute Inflammation
We have further shown that there is a dramatic down-regulation of GR, MR and ERα in microglia that takes place shortly after inflammation is induced by systemic injection of LPS. Multiple factors can contribute to the LPS-induced down-regulation of steroid hormone receptors. For instance, LPS induces the secretion of glucocorticoids (Nakano et al., 1987), which decrease the expression of GR (Ahima et al., 1992). Furthermore, the down-regulation of MR and ERα was correlated with the expression of TGFβ1, a very well known regulator of microglial activity (Buckwalter and Wyss-Coray, 2004). Whether this correlation means that there is a cause-effect relationship or whether the two events are consequence of the same process remains unanswered. We speculate that the down-regulation of steroid hormone receptors induced by LPS will allow microglia to reach a full inflammatory state, relieving the cell from the modulatory control of endogenous anti-inflammatory modulators such as steroid hormones. Interestingly, LPS also down-regulates the expression of the cannabinoid receptor CB2 (Carlisle et al., 2002), another major endogenous immunomodulatory system (Klein, 2005). Being desensitized to the anti-inflammatory effects of steroids and cannabinoids, microglia would have a greater success rate of eliminating pathogens.
The bidirectional regulation of the immune and the neuroendocrine systems occurs at multiple levels (Sternberg, 2001); hormones are potent immunomodulators of every step of the inflammatory process and, in turn, inflammatory cytokines modulate the hypothalamic-pituitary-adrenal and hypothalamic-pituitary-gonadal axes (Foster et al., 2003). An altered expression of steroid hormone receptors in microglia during the initiation of the inflammatory response adds another step of regulation to the neuroimmune-endocrine interaction. For instance, in rats AR is induced in microglia and ERβ in astrocytes after deep wound and kainic acid-induced injury (DonCarlos et al., 2006; Garcia-Ovejero et al., 2002, 2005), ERβ expression increases in astrocytes and GR and MR are over-expressed in gerbil microglia after ischemia (Hwang et al., 2006), ERβ is induced in microglia after ischemia in macaque monkeys (Takahashi et al., 2004), and several components of the brain steroidogenic machinery are up-regulated in neurons after brain injury in the rat brain (Azcoitia et al., 2001; Sierra et al., 2003). Since steroid hormones have well-described neuroprotective properties, it has been suggested that this response is as an attempt to cope with neurodegeneration (Garcia-Ovejero et al., 2005; Sierra, 2004).
It is possible that a differential regulation of the expression of steroid hormone receptors in different states of activation of microglia (i.e., inflammation vs. neurodegeneration) may serve different purposes. Although neurodegenerative processes have an inflammatory component, they are more complicated than pure inflammatory processes, involving a coordinated activation of microglia and astrocytes in response to neuronal death. Nevertheless, because inflammation exacerbates neurodegeneration, an up-regulation of steroid hormone receptors in microglia would restore homeostasis and prevent further neuronal damage.
- Top of page
- MATERIALS AND METHODS
We have utilized a powerful ex vivo approach to analyze reliably and in a quantitative manner the activation of microglia and to study the expression of low-abundance genes such as steroid hormone receptors that cannot be easily detected by more conventional techniques. Although this approach does not allow the evaluation of regional changes in the microglia, the advantage of real time RT-PCR for quantification purposes is widely accepted (Wong and Medrano, 2005). Using this methodology, we have demonstrated the presence of GR, MR and ERα in adult microglia, implying they are a direct target of steroid hormones. GR was the most abundant steroid hormone receptor expressed in microglia and, in agreement, corticosterone was a consistent anti-inflammatory while estradiol at the same concentration range had no consistent effect. Finally, the down-regulation of steroid hormone receptors after LPS challenge could serve as a mechanism to relieve microglia from the control of steroid hormones. As they are a pivotal brain cell-type controlling inflammation, microglia should be taken into account when designing immunomodulatory hormone-based therapies.
- Top of page
- MATERIALS AND METHODS
The authors thank Dr. D. Hume (University of Queensland, Australia) and Dr. J. Pollard (Albert Einstein College of Medicine, NY) for the kind gift of the 7.2fms-EGFP mice; the staff of the Rockefeller Flow Cytometry Resource Center for their expertise with flow cytometry procedures; and Dr. J.M. Encinas (CSHL, NY) for intellectually encouraging discussions; and Mr. Scott Herrick for technical assistance.
- Top of page
- MATERIALS AND METHODS
- 1992. Sexual dimorphism in regulation of type II corticosteroid receptor immunoreactivity in the rat hippocampus. Endocrinology 131: 1409–1416. , , , .
- 2005. Cytokine production. In: KettenmannHR,RansomB, editors. Neuroglia. New York: Oxford UP. pp 285–301. .
- 1998. Immunocytochemical localization of nuclear estrogen receptors and progestin receptors within the rat dorsal raphe nucleus. J Comput Neurol 391: 322–334. , , , .
- 2001. Brain aromatase is neuroprotective. J Neurobiol 47: 318–329. , , , , , .
- 2004. Estrogen modulates microglial inflammatory mediator production via interactions with estrogen receptor beta. Endocrinology 145: 5021–5032. , , .
- 2002. Assessment of the relationship between hyperalgesia and peripheral inflammation in magnesium-deficient rats. Life Sci 70: 1053–1063. , , , , , .
- 2001. Estrogenic actions in the brain: estrogen, phytoestrogens, and rapid intracellular signaling mechanisms. J Pharmacol Exp Ther 299: 408–414. , .
- 1988. Lipophilic impurities, not phenolsulfonphthalein, account for the estrogenic activity in commercial preparations of phenol red. J Steroid Biochem 31: 287–293. , , , .
- 1991. Glial cells express both mineralocorticoid and glucocorticoid receptors. J Steroid Biochem Mol Biol 40: 105–111. , , , .
- 2000. Antiinflammatory effects of estrogen on microglial activation. Endocrinology 141: 3646–3656. , , , , , .
- 2004. Modelling neuroinflammatory phenotypes in vivo. J Neuroinflammation 1: 10. , .
- 2005. Oestrogen and stroke in women: Assessment of risk. Lancet Neurol 4: 743–751. .
- 2002. Differential expression of the CB2 cannabinoid receptor by rodent macrophages and macrophage-like cells in relation to cell activation. Int Immunopharmacol 2: 69–82. , , , , .
- 2002. The degenerative effect of a single intranigral injection of LPS on the dopaminergic system is prevented by dexamethasone, and not mimicked by rh-TNF-alpha, IL-1beta and IFN-gamma. J Neurochem 81: 150–157. , , , .
- 1983. Sex steroid receptors in peripheral T cells: Absence of androgen receptors and restriction of estrogen receptors to OKT8-positive cells. J Immunol 131: 2767–2771. , , , , .
- 1996. Androgen and estrogen receptors are present in primary cultures of human synovial macrophages. J Clin Endocrinol Metab 81: 820–827. , , , , , , , , , , , .
- 2005. Estrogen and cytokines production—The possible cause of gender differences in neurological diseases. Curr Pharm Des 11: 1017–1030. , , , .
- 1986. Function and plasticity of brain corticosteroid receptor systems: Action of neuropeptides. J Steroid Biochem 25: 723–731. , , , , .
- 2004. Endothelial cells of the blood-brain barrier: A target for glucocorticoids and estrogens? Front Biosci 9: 684–693. .
- 2005. Estrogen and brain inflammation: Effects on microglial expression of MHC, costimulatory molecules and cytokines. J Neuroimmunol 161: 123–136. , , , , , , , , , .
- 2003. Novel glucocorticoid effects on acute inflammation in the CNS. J Neurochem 84: 705–716. , , .
- 2002. Glucocorticoids and central nervous system inflammation. J Neurovirol 8: 513–528. , , .
- 2006. Novel cellular phenotypes and subcellular sites for androgen action in the forebrain. Neuroscience 138: 801–807. , , , , , , .
- 2003. Dysregulation of the hypothalamic-pituitary-gonadal axis in experimental autoimmune encephalomyelitis and multiple sclerosis. J Neuroimmunol 140: 78–87. , , , .
- 2005. Glia-neuron crosstalk in the neuroprotective mechanisms of sex steroid hormones. Brain Res Brain Res Rev 48: 273–286. , , , , .
- 2002. Glial expression of estrogen and androgen receptors after rat brain injury. J Comput Neurol 450: 256–271. , , , .
- 2005. 17beta-estradiol inhibits inflammatory gene expression by controlling NF-kappaB intracellular localization. Mol Cell Biol 25: 2957–2968. , , , .
- 2005. TNFalpha-induced long pentraxin PTX3 expression in human lung epithelial cells via JNK. J Immunol 175: 8303–8311. , , , , , , , , .
- 1991. Isolation and flow cytometric characterization of newborn mouse brain-derived microglia maintained in vitro. J Leukoc Biol 50: 86–92. , , , , .
- 2006. Mineralocorticoid and glucocorticoid receptor expressions in astrocytes and microglia in the gerbil hippocampal CA1 region after ischemic insult. Neurosci Res 54: 319–327. , , , , , , , , .
- 2006. Corticosterone inhibits expression of the microglial glutamate transporter GLT-1 in vitro. Neuroscience 139: 475–483. , , , .
- 2005. Estrogen's effects on central and circulating immune cells vary with reproductive age. Neurobiol Aging 26: 1365–1374. , .
- 2005. Localization of glucocorticoid receptors at postsynaptic membranes in the lateral amygdala. Neuroscience 136: 289–299. , , , , .
- 2005. Cannabinoid-based drugs as anti-inflammatory therapeutics. Nat Rev Immunol 5: 400–411. .
- 2001. TNF mediates a murine model of Addison's crisis. Shock 15: 29–34. , , .
- 1991. Postnatal development of corticosteroid receptor immunoreactivity in the rat hippocampus. Brain Res Dev Brain Res 62: 69–79. , , , .
- 2003. Inhibition of LPS-induced iNOS and NO synthesis in primary rat microglial cells. Neurochem Int 42: 131–137. , , .
- 1977. Glucocorticoid receptors in purified subpopulations of human peripheral blood lymphocytes. J Immunol 118: 1977–1981. , .
- 2005. Estrogen provides neuroprotection against activated microglia-induced dopaminergic neuronal injury through both estrogen receptor-alpha and estrogen receptor-beta in microglia. J Neurosci Res 81: 653–665. , , , , , , .
- 1987. Integration of the effects of estradiol and progesterone in the modulation of gonadotropin secretion. J Steroid Biochem 27: 665–675. , .
- 2005. Estrogen selectively promotes the differentiation of dendritic cells with characteristics of Langerhans cells. J Immunol 175: 5146–5151. , , , , .
- 2005. Ultrastructural localization of estrogen receptor beta immunoreactivity in the rat hippocampal formation. J Comput Neurol 491: 81–95. , , , , , , , .
- 2001. Ultrastructural evidence that hippocampal alpha estrogen receptors are located at extranuclear sites. J Comput Neurol 429: 355–371. , , , , , .
- 1992. Ultrastructural localization of neuropeptide Y-like immunoreactivity in the rat hippocampal formation. Hippocampus 2: 107–125. , .
- 2003. Immunolocalization of estrogen receptor beta in the mouse brain: Comparison with estrogen receptor alpha. Endocrinology 144: 2055–2067. , , , , , , , , , , others.
- 1999. Estrogen and microglia: A regulatory system that affects the brain. J Neurobiol 40: 484–496. , , , , , , .
- 2003. Glucocorticoids play a fundamental role in protecting the brain during innate immune response. J Neurosci 23: 5536–5544. , .
- 1987. Significance of increased secretion of glucocorticoids in mice and rats injected with bacterial endotoxin. Brain Behav Immun 1: 159–172. , , .
- 2004. Interactions of sex steroids with mechanisms of inflammation. Curr Drug Targets Inflamm Allergy 3: 425–433. , .
- 1992. A polyclonal antibody to the rat oestrogen receptor expressed in Escherichia coli: Characterization and application to immunohistochemistry. J Endocrinol 135: 333–341. , , , , .
- 2001. Estrogens ameliorate remote organ inflammation induced by burn injury in rats. Inflamm Res 50: 585–591. , , , , , , .
- 2003. Type II glucocorticoid receptor immunoreactivity in the mossy cells of the rat and the mouse hippocampus. Hippocampus 13: 59–66. , .
- 2005. Actions of TNF-alpha on glutamatergic synaptic transmission in the central nervous system. Exp Physiol 90: 663–670. , , .
- 2005. The search for safer glucocorticoid receptor ligands. Endocr Rev 26: 452–464. , .
- 2003. A macrophage colony-stimulating factor receptor-green fluorescent protein transgene is expressed throughout the mononuclear phagocyte system of the mouse. Blood 101: 1155–1163. , , , , , , , , .
- 1991. Distribution of estrogen target sites in the 2-day-old mouse forebrain and pituitary gland during the ‘critical period’ of sexual differentiation. Brain Res Dev Brain Res 61: 11–22. , , , , .
- 2004. How does peripheral lipopolysaccharide induce gene expresion in the brain of rats? Toxicology 201: 197–207. , .
- 2004. Neurosteroids: the StAR protein in the brain. J Neuroendocrinol 16: 787–793. .
- 2007. Microglia derived from aging mice exhibit an altered inflammatory profile. Glia 55: 412–424. , , , .
- 2003. Steroidogenic acute regulatory protein in the rat brain: cellular distribution, developmental regulation and overexpression after injury. Eur J Neurosci 18: 1458–1467. , , , , , .
- 1990. Distribution of androgen and estrogen receptor mRNA-containing cells in the rat brain: an in situ hybridization study. J Comput Neurol 294: 76–95. , , , .
- 2005. Estradiol is required for a proper immune response to bacterial and viral pathogens in the female brain. J Immunol 174: 6391–6398. , , , .
- 2001. Neuroendocrine regulation of autoimmune/inflammatory disease. J Endocrinol 169: 429–435. .
- 2004. Estrogen, inflammation and cardiovascular risk in women: A critical appraisal. Trends Endocrinol Metab 15: 66–72. , , , .
- 2005. Anti-inflammatory effect of selective estrogen receptor modulators (SERMs) in microglial cells. Inflamm Res 54: 194–203. , , , , , .
- 2004. Expression of estrogen receptor-beta in the postischemic monkey hippocampus. Neurosci Lett 369: 9–13. , , , , , , .
- 1997. Glucocorticoid- and mineralocorticoid receptors in microglial cells: the two receptors mediate differential effects of corticosteroids. Glia 20: 23–37. , , , , , .
- 2006. Minireview: Rapid glucocorticoid signaling via membrane-associated receptors. Endocrinology 147: 5549–5556. , , .
- 2003. Subcellular relationships between cholinergic terminals and estrogen receptor-alpha in the dorsal hippocampus. J Comput Neurol 463: 390–401. , , , , , .
- 1988. Distribution of the mineralocorticoid and the glucocorticoid receptor mRNAs in the rat hippocampus. J Neurosci Res 21: 88–94. , , , .
- 2003. Estrogen receptor-alpha mediates the brain antiinflammatory activity of estradiol. Proc Natl Acad Sci USA 100: 9614–9619. , , , , , , , , , , et al.
- 2006. The endogenous estrogen status regulates microglia reactivity in animal models of neuroinflammation. Endocrinology 147: 2263–2272. , , , , , .
- 2001. Estrogen prevents the lipopolysaccharide-induced inflammatory response in microglia. J Neurosci 21: 1809–1818. , , , , , , , , , .
- 1978. Interaction of glucocorticoids with macrophages. Identification of glucocorticoid receptors in monocytes and macrophages. J Exp Med 147: 1684–1694. , , .
- 2005. Real-time PCR for mRNA quantitation. Biotechniques 39: 75–85. , .
- 2004. Glucocorticoid regulation of the inflammatory response to injury. Acta Anaesthesiol Scand 48: 799–813. , , .