| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Rheumatology and Clinical Immunology, Charité University Hospital, Humboldt University,10117 Berlin, Germany;
* Deutsches Rheuma-Forschungszentrum Berlin, 10117 Berlin, Germany; and
Department of Immunology and Biotechnology, University of Pécs, Faculty of Medicine, 7643 Pécs, Hungary
1Correspondence: Department of Rheumatology and Clinical Immunology, Charité University Hospital, Schumannstrasse 20/21, 10117 Berlin, Germany. E-mail: frank.buttgereit{at}charite.de
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
alone is not sufficient to enhance mGCR expression. These in vitro findings are consistent with our clinical observation that in patients with rheumatoid arthritis the frequency of mGCR positive monocytes is increased and positively correlated with disease activity. We conclude that mGCR are 1) indeed physiologically present in healthy blood donors, but remained unidentified by conventional techniques due to their small number per cell and 2) actively up-regulated and transported through the cell after immunostimulation. These receptors may reflect a feedback mechanism of the organism upon immunostimulation and/or play a role in pathogenesis.—Bartholome, B., Spies, C. M., Gaber, T., Schuchmann, S., Berki, T., Kunkel, D., Bienert, M., Radbruch, A., Burmester, G.-R., Lauster, R., Scheffold, A., Buttgereit, F. Membrane glucocorticoid receptors (mGCR) are expressed in normal human peripheral blood mononuclear cells and up-regulated after in vitro stimulation and in patients with rheumatoid arthritis.
Key Words: nongenomic glucocorticoid effects • high-sensitivity immunofluorescent staining • mGCR-positive monocytes • human glucocorticoid receptor alpha
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3
4)
. Recently this classical model was significantly extended through findings on rapid nongenomic effects of the glucocorticoids (5
6
7
8
9)
. The first is the finding that binding of the glucocorticoid molecule to the cGCR is followed secondarily by nongenomic (non-nuclear) actions (e.g., rapid eNOS activation; refs 10
, 11
). Second, at very high but nevertheless clinically relevant glucocorticoid concentrations (e.g., in the case of intra-articular injections or intravenous pulse therapy), unspecific nongenomic effects occur in the form of physicochemical interactions with biological membranes (5
6
7
8)
. Third, glucocorticoids cause specific nongenomic effects that are mediated by membrane-bound glucocorticoid receptors (mGCR). In the past few years, increased evidence has been found of the existence and function of membrane-bound steroid receptors (8
, 12)
. Recently the existence of a plasma membrane receptor for plant steroids was reported (13)
. In humans and animals, the existence of mGCR has been shown for amphibian neuronal membranes (14)
and leukemic/lymphoma cells (15
16
17)
, which raised the question of whether these receptors also exist in humans under physiological and nontumorous pathophysiological conditions. Here we show the presence and induced up-regulation of membrane glucocorticoid receptors (mGCR) in human peripheral blood mononuclear cells (PBMC). |
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
16
17)
, were purchased from American Type Culture Collection (ATCC, Gaithersburg, MD, USA) to serve as positive controls in our experiments (ATCC no. CCL-119). CCRF-CEM is a T lymphoblastoid cell line derived from the peripheral blood buffy coat of a 4-year-old Caucasian female with acute lymphoblastic leukemia. Expression of mGCR is known to depend on the cell cycle (17)
, and our initial experiments with these cells revealed an mGCR expression varying with incubation time. Seventeen hours between warming up the frozen cell sample and measurement of the mGCR expression resulted in the identification of 
5% mGCR-positive cells (40 h/30% and 65 h/20%). We incubated these cells for 40 h (see below) as one parameter of following standardized experimental procedure. Aliquots of 5 x 106 cells each were frozen in DMSO-containing Cryo solution (Art. No. 2040101; Cell-lining GmbH, Berlin, Germany) according to standard procedures shortly after the line was originated. The aliquots were kept frozen (liquid nitrogen, –196°C) until use. For experiments, the cells were warmed up rapidly, washed with medium to minimize the cellular toxic properties of DMSO, then cultured in cell culture medium [RPMI 1640 with 2 mM LL-glutamine (PAA Laboratories, Coelbe, Germany), 100 units/mL penicillin G, and 100 µg/mL streptomycin (PAA Laboratories, Coelbe, Germany), 10% (v/v) heat-inactivated (30 min, 56°C) fetal bovine serum (FBS; Sigma-Aldrich, Steinheim, Germany)] at 37°C in a humidified atmosphere of 5% CO2 for 40 h.
Preparation of PBMC from control subjects and patients
Control subjects
The control group of healthy subjects consisted of 20 women and 34 men (age 18 to 45 years).
Patients with rheumatoid arthritis
Nineteen patients (age 31 to 72 years) with rheumatoid arthritis of variable disease activity admitted to our rheumatology unit were studied. All patients met the 1987 American Rheumatism Association criteria for RA (18)
. Disease activity was estimated according to clinical and laboratory criteria, including measurements of C reactive protein levels and the visual analog score (VAS) for well-being, counting of swollen and tender joints, and calculation of the disease activity score (DAS 28) (19
, 20)
.
Preparation of PBMC
After a complete physical examinations had been performed and informed consent had been obtained from patients and controls, a sample of max. 20 mL of blood was withdrawn into heparinized tubes. PBMC were isolated by density centrifugation using the Ficoll-PaqueTMPLUS technique (Amersham Bioscience AB, Uppsala, Sweden) according to the manufacturer’s instruction. Each sample of venous blood was diluted 1: 2 with phosphate-buffered saline (PBS); density gradient centrifugation was performed at 840 g for 20 min at 20°C. The PBMC-enriched interphase was isolated and washed with PBS. Trypan blue staining revealed the viability of freshly isolated cells to be greater than 95%.
Incubation of PBMC
For in vitro experiments, PBMC were resuspended with cell culture medium [RPMI 1640 with 2 mM LL-glutamine (PAA Laboratories, Coelbe, Germany), 100 units/mL penicillin G and 100 µg/mL streptomycin (PAA Laboratories, Coelbe, Germany), 50 µM 2-mercaptoethanol (GIBCO Brl, Paisley, Scotland), 10% (v/v) heat-inactivated (30 min, 56°C; to inactivate complement protein and prevent an immunological reaction against cultured cells) FBS (Sigma-Aldrich)] at a concentration of 2 x 106 cells/mL. Either 2 µg/mL LPS (lipopolysaccharide to stimulate PBMC from Escherichia coli serotype 0111:B4, Sigma-Aldrich) alone or 2 µg/mL LPS + 5 µg/mL brefeldin A (an inhibitor of the secretory pathway; refs 25
, 26
) from Penicillium brefeldianum (Sigma-Aldrich) was added to the cells and incubated for 24 h at 37°C in a 5% CO2 atmosphere. Control cells were PBMC resuspended and incubated but without LPS or brefeldin A. Untreated Petri dishes (Greiner Bio-One, Frickenhausen, Germany) were used as cell culture vessels. After incubation cells were detached by repeatedly rinsing the Petri dishes with ice-cold PBS, then washed twice.
Antibodies and antibody conjugates
For immunofluorescence analysis and fluorescence imaging, a monoclonal IgG1 antibody anti-GCR 5E4 (21)
directed against the conserved regulatory sequence of human GCR (aa150-176) was used. These antibodies were conjugated to digoxigenin (Dig) using digoxigenin-3-O-methylcarbonyl-
-aminocaproic acid-N-hydroxysuccinimidester (Roche Diagnostics GmbH, Mannheim, Germany) following the instructions of the manufacturer. We also used phycoerythrin (PE) -conjugated anti-humanCD14 (BD PharMingen, San Diego, CA, USA), allophycocyanin(APC) -conjugated anti-humanCD19 (BD PharMingen, San Diego, CA, USA), and peridin-chlorophyll(PerCP) -conjugated anti-humanCD3 (BD Biosciences, San Jose, CA, USA) to identify cell types. Anti-digoxigenin-fluorescein-Fab fragments (Roche, Mannheim, Germany) and/or anti-digoxigenin-Fab fragments (Roche, Mannheim, Germany) coupled to Cy5-NHS-Ester (Amersham Biosciences, Freiburg, Germany) were used to detect the above-mentioned Dig-labeled anti-GCR antibody. The Dig-labeled anti-GCR antibody was used at 5–9 µg/mL. Polyclonal human IgG (Octagam,® Octapharma, Langenfeld, Germany) was used to block unspecific binding (5 mg/mL, 10 min). For staining of CHO cells we used unconjugated anti-GCR, anti-penta-his monoclonal antibody (QIAgen, Hilden, Germany), anti-IFN
-monoclonal antibody (clone 4SB3, DRFZ, Berlin, Germany) as isotype control, and Dig-conjugated anti-IgG1 monoclonal antibody (Milteny Biotec GmbH, Bergisch Gladbach, Germany).
Flow cytometric analysis
Detection of mGCR
mGCR were detected by high-sensitivity immunofluorescent staining using anti-GCR-Dig, followed by anti-Dig magnetofluorescent liposomes (22)
. All stainings were performed on ice using PBS containing 0.5% (w/v) bovine serum albumin (PAA Laboratories, Coelbe, Germany), 0.02% (v/v) sodium azide (NaN3) (Sigma-Aldrich), and 5 mM EDTA (Sigma-Aldrich) as the staining buffer (PBA/EDTA). Antibody staining was done for 10 min and staining with liposomes for 30–35 min, with gentle agitation of the cells. To block unspecific binding the cells were incubated for 10 min with 5 mg/mL polyclonal human IgG. Propidium iodide (Sigma Aldrich) was added directly before data acquisition at 1 µg/mL for identification of dead cells. To control specificity, cells were incubated without hapten-labeled mGCR antibody or with a 50- to 100-fold excess of unlabeled mGCR antibody before staining with the anti-GCR-Dig conjugate. The frequency of positive cells was calculated from the positive sample by subtracting the background signals obtained by cold blocking.
Intracellular staining
Intracellular staining was performed at room temperature using PBA with 0.5% (w/v) saponin (Sigma-Aldrich) as the staining buffer. Prior to intracellular staining, cells were fixed using 2% (v/v) formaldehyde (Merck, Darmstadt, Germany).
Data were acquired using FACS-Calibur (equipped with a 488 nm argon laser and a 635 nm red diode laser) and CellQuest software (Becton Dickinson, San Jose, CA, USA). The setup of FACS-Calibur was performed according to the manufacturer’s instructions using unstained and single stained cells. Data analysis was performed using FCS Express (De Novo Software, Thornhill, Ontario, Canada). For exclusion of debris and dead cells, gates were set according to FSC/SSC and PI staining.
Fluorescence microscopy
LPS-treated PBMC were stained for mGCR as described above. Being mGCR-positive, monocytes were isolated using magnetic cell separation (MACS) columns and anti-PE microbeads (Miltenyi Biotec, Bergisch-Gladbach, Germany). Isolated monocytes were diluted to a final concentration of 103 to 104 cells/mL. Images were taken using an imaging fluorescence system based on an inverse microscope (IX50, Olympus, Hamburg, Germany) equipped with 40x and 60x objectives (numerical aperture 0.3 and 0.6, respectively; Olympus), a xenon light source, and a charge-coupled device camera (Till Photonics, Planegg, Germany). Cells were monitored by excitation at 470 nm and the emitted fluorescence signal was detected above 520 nm using a 490 nm dichroic mirror and a 520 nm long-pass filter.
Cloning of amino-terminal his-tagged human GCR , CHO cell culture, and transfection
The coding sequence of human glucocorticoid receptor (GCR) (NCBI accession no. M10901) was derived from cDNA of total RNA of Ficoll prepared PBMC (QIAgen RNeasy Minikit, Roche TaqMAN cDNA kit according to the manufacturer’s instructions) using PCR with the following primer set: 5'-GATCGCTAGC ATGGACCACC ATCACCATCA CCATGACTCC AAAGAATCAT TAAC-3', 5'-GATCTCTAGA TCACTTTTGA TGAAACAGAA GTTTT-3' (purchased from TIB MolBiol, Berlin, Germany). PCR was performed in a 20 µL reaction volume with an initial denaturation step of 95°C for 2 min, followed by 95°C for 30 s, 60°C for 30 s, and 72°C for 4 min (35 cycles). The PCR product was inserted into the NheI site of pIRES2-eGFP (Clontech, Palo Alto, CA, USA) leading to pI2EGR and sequenced using the following sequencing primers: 5'-CATTATGCCC AGTACATGACC-3', 5'-AAGTGATGGG AAATGACCTG-3', 5'-TTGGAGTTTT CTTCTGGGTC-3', 5'-TGGGCACAGT TTACTGTCAG-3', 5'-AGCATGAGAC CAGATGTAAG-3', 5'-AGGAATTCAG CAGGCCACTA-3', 5'-CTCCTGGATG TTTCTTATGG-3', 5'-GAATGACCTA CATCAAAGAG-3'.
DNA sequencing was carried out by the dideoxy chain termination method using the dye terminator cycle sequencing kit (Perkin-Elmer, Norwalk, CT, USA) and the Applied Biosystems 310 Genetic Analyzer according to the manufacturer’s instructions. The CHO-K1 cell line was obtained from ATCC and grown in RPMI 1640 plus 10% FCS (GibcoBRL) in 5% CO2, saturated humidity, 37°C. Cells were transiently transfected with pI2EGR (1 µg/106 cells) or as a control with pIRES2-eGFP (1 µg/106 cells; Clontech) using Cell Line NucleofectorTM Kit T (Amaxa Biosystems, Cologne, Germany) incubated for 48 h under the recommended media. Transient transfected CHO-K1 cells were collected after 48 h of incubation.
For flow cytometric analysis, cells were incubated with the monoclonal antibody anti-GCR 5E4 (2 µg/mL), anti-penta-his monoclonal antibody (2 µg/mL; QIAgen, Hilden, Germany), or anti-IFN monoclonal antibody (2 µg/mL) as an isotype control or without a primary monoclonal antibody as a second control. This was followed by incubation with a Dig-conjugated anti-IgG1 monoclonal antibody (0.99 µg/mL; Miltenyi Biotec GmbH, Bergisch Gladbach, Germany) and finally anti-Dig magnetofluorescent Cy5 liposomes (22)
or anti-Dig-Cy5 (1 µg/mL) for intracellular staining. Intracellular staining of GCR was as described above.
Statistical methods
All data are expressed as mean ± SD. If not stated otherwise, differences between the groups were verified using a Mann-Whitney test or Wilcoxon test. Probability values of P < 0.05 were considered to be statistically significant. Spearman test was used for correlation analysis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. This was achieved by Western/autoradiographic analysis and by FACS and immunocytochemical analyses in lymphocytes of leukemic patients and in the CCRF-CEM cell line (CCL-119 cells) (15)
. We cultivated CCRF-CEM cells and examined them for the presence of cGCR and mGCR with the aid of a monoclonal antibody against the cytosolic glucocorticoid receptor (cGCR). The antibody used (5E4 clone-IgG1) was raised against a conserved sequence (150–176 amino acids) of the regulatory part of the cGCR (21)
. Using immunofluorescence with conventional antibody staining, the cGCR could be identified after fixation and permeabilization of the cells, but it was not possible at first to confirm the existence of mGCR on the surface of intact cells (Fig. 1
A, B). We suspected that so few mGCR were expressed that immunofluorescence with conventional antibody staining was not sensitive enough to detect them. For this reason, we applied high-sensitivity immunofluorescent staining (liposome staining) (22)
(Fig. 1C
). This method uses antibody-conjugated magnetofluorescent liposomes, which can increase fluorescence signal intensity up to 1000-fold compared with conventional methods and allows the detection of 50–100 target molecules per cell (Fig. 2
). Using this technique, it was possible to confirm the existence of the mGCR (described by Gametchu et al.) on 7–17% of the CCRF-CEM cells. Figure 1C
shows a typical measurement.
|
|
Identification of mGCR on PBMC of healthy donors
We began to systematically search for mGCR expression on PBMC. Here, too, it was found that the antibody used detects cGCR by immunofluorescence with conventional antibody staining, but not mGCR (Fig. 3
). Using the high-sensitivity immunofluorescence with liposome staining, however, a significant mGCR expression could be detected specifically. A representative measurement for monocytes, B lymphocytes, and T lymphocytes is shown in Fig. 4
. Specific binding of the 5E4 monoclonal antibody to the GCR was confirmed 1) by competition with molar excess of unlabeled antibodies ("block" in Fig. 4
) and 2) with the synthetic GCR peptide fragment (called APTEK-26) used for immunization and screening of the hybridomas (21)
. Fifty-four healthy subjects with no known chronic or acute diseases in the previous 8 wk were investigated. mGCR expression on monocytes varied among individuals, ranging from 0 to 9.2% with a median of 2.5% (mean±SD: 3.0±2.4%). For B lymphocytes, 0–12.3% of the cells were also mGCR positive, with a median value of 2.9% (mean±SD: 3.8±3.3%). T cells showed no significant mGCR expression (min 0.0%, max 1.0%, median 0.1%, mean±SD: 0.2±0.2%). The data are presented in detail in Fig. 5
.
|
|
|
LPS stimulation drastically increases the proportion of mGCR-expressing monocytes
We suspected that the expression of mGCR might be connected to activity levels of immune cells, and so we investigated PBMC that had been stimulated with 2 µg/mL LPS over 24 h (for details, see Materials and Methods). It is known that, like various cytokines, LPS causes an increase in cGCR expression (23
, 24)
. We were able to confirm this observation in our own experiments, which showed a significant MFI (mean fluorescence intensity) increase in LPS-treated monocytes by 39.1 ± 16.1% (n=5) compared with untreated cells (Wilcoxon test, P<0.05). In addition, we found that LPS causes a marked increase in the proportion of mGCR-positive monocytes from 4.1 ± 3.0% to 33.8 ± 14.0% (n=12) (Fig. 6
). In contrast, 24 h incubation of the cells without LPS did not significantly alter mGCR expression at 4.7 ± 2.1%. To clarify whether mGCR expression is dependent upon transcellular transport processes, we also conducted these investigations with brefeldin A. We found that in the presence of 5 µg/mL brefeldin A, LPS stimulation does not lead to a significant change in baseline value: only 4.7 ± 3.6% (n=6) of the monocytes were mGCR positive after 24 h incubation. This observation demonstrates the ability of brefeldin A to abrogate the induced up-regulation of mGCR (Fig. 6)
. Figure 7
shows a representative measurement.
|
|
Confirmation of the existence of mGCR on monocytes by fluorescence microscopy
We confirmed the existence of mGCR on monocytes in the model of LPS stimulation using fluorescence microscopy. PBMC were incubated according to the above-mentioned protocol without and with LPS for 24 h. The monocytes were then isolated with the aid of the MACS technique (anti-PE antibodies labeled with magnetic beads) and analyzed under a fluorescence microscope. Figure 8
shows that antibody-conjugated magnetofluorescent liposomes do not penetrate the cell interior and so do not detect cGCR. However, identification of mGCR on the cell surface can be clearly seen and mGCR-positive distinguished from mGCR-negative monocytes. The proportion of mGCR-positive cells is much higher than in the FACS measurements, because the MACS sorting leads to a positive selection (accumulation) of the mGCR-positive monocytes due to the magnetic properties of the magnetofluorescent liposomes.
|
Overexpression of human GCR gene does not increase mGCR expression
LPS stimulation increases both cGCR and mGCR expression in monocytes, leading us to wonder whether there may be a dynamic equilibrium between the two types of receptors or that mGCR expression may represent an "overflow expression" of cGCR. Alternatively, transcription and translation, post-transcriptional/post-translational processing of the gene product in the direction of mGCR (27)
, as well as transcellular transport may be stimulated by LPS. Both theories are based on the assumption that only one gene exists for coding, because the amino acid sequence detectable by our antibody occurs only in the GCR receptor. To differentiate between these two possibilities, we examined the ability of overexpressed GCR to effect increased mGCR expression as observed for LPS stimulation. We used transient transfection in CHO cells to overexpress the coding sequence of human cytosolic GCR. This sequence was his-tagged to allow correct differentiation between endogenously expressed GCR and overexpressed GCR. We found that this procedure leads to a significant increase in cGCR protein expression as confirmed by a parallel increase of penta-his-positive protein (Fig. 9
A, C). We found that CHO cells express mGCR to a certain extent per se, but that after transfection neither an increase of mGCR expression nor penta-his-positive membrane protein could be observed [GFP+ vs. GFP–, mGCR+: 5.4±4.2% vs. 4.1±3.4% (n=4), His+: 0.7±3.5% vs. 0.1±3.5% (n=4)] (Fig. 9B, D
). This demonstrates that, unlike LPS stimulation, overexpression of cGCR does not increase mGCR expression.
|
The proportion of mGCR-positive monocytes correlates positively with disease activity of patients with rheumatoid arthritis
The results of in vitro stimulation with LPS led us to suspect that the proportion of mGCR-positive monocytes correlates with the activity of the immune system in vivo. To test this hypothesis and the clinical relevance of our results, we chose rheumatoid arthritis as our disease model for the reason that the disease activity can be defined very precisely by parameters such as C reactive protein, swollen/tender joints, VAS score for well-being, DAS 28 (19
, 20)
. Nineteen consecutive patients with established rheumatoid arthritis of variable disease activity were studied. Our results are summarized in Fig. 10
. Statistical analysis of the data revealed a strong positive correlation between the number of mGCR-positive monocytes and disease activity.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 12
, 28
29
30)
. For rapid glucocorticoid effects, different mechanisms (including specific interactions with membrane-bound glucocorticoid receptors, or mGCR) are currently under consideration as being responsible (5
6
7
8
9
10
11
, 28
29
30
31)
. Binding sites in membranes have been characterized that display binding features compatible with an involvement in rapid steroid signaling. Evidence of nongenomic steroid effects and distinct receptors involved is available for glucocorticoids, mineralocorticoids, gonadal hormones, vitamin D, and thyroid hormones (8
, 30
31
32
33
34
35
36
37)
. For glucocorticoids, however, mGCR have so far been detected only in amphibian brain (14)
and on leukemic/lymphoma cells (15
16
17)
. There has been no direct evidence of the physiological existence of mGCR in humans, although reports are available on rapid effects in humans. Hafezi-Moghadam et al. recently reported cardiovascular protective effects of dexamethasone and suspected the mechanism to be a binding of the glucocorticoid to the cGCR, leading to nontranscriptional activation of PI3K, Akt, and eNOS (11)
. Alternatively, mediation of such effects by mGCR must be discussed. But why has it not been possible to demonstrate the physiological existence of these receptors to date? We suspected that, like other important molecules (22)
, mGCR are expressed on the cell surface in such small numbers they are not detected by conventional methods. Our experiments indeed showed that a monoclonal anti-GCR antibody identifies cGCR through binding but is negative for identification of mGCR in the conventional flow cytometry method using antibody staining. These data were initially obtained on the lymphoma cells we used as a positive control (CCRF-CEM cells), whose mGCR positivity had been detected by other methods (15
16
17)
. However, in this paper we have used high-sensitivity immunofluorescent staining with magnetofluorescent liposomes (22)
not only to confirm the existence of mGCR on CCRF-CEM cells but, even more important, to detect mGCR on human PBMC (Figs. 4
, 5
, 8)
. Using this technique we were able to demonstrate by immunofluorescence and a fluorescence imaging technique for the first time significant expression of mGCR on human PBMC from healthy controls that had been undetectable by conventional technologies. Up to 9.2% of monocytes and up to 12.3% of B lymphocytes, but not T lymphocytes, were found to be positive for mGCR. The latter observation is of particular interest, since precisely the reverse has been described for the existence on lymphocytes on other membrane-bound steroid receptors. Benten et al. found recently that T lymphocytes and macrophages (but not B lymphocytes) express membrane-based estrogen and testosterone receptors (38)
.
Immunofluorescence-based analysis with conventional antibody staining is widely used for characterizing cells according to their expression profile of specific proteins. However, the sensitivity of this method is limited by the fact that several thousand antigens are necessary on the cell surface in order to identify these cells as positive for the respective protein (22)
. This means that functionally important molecules go undetected if they are expressed in numbers that are too low. This includes, for example, many cytokine or hormone receptors (39)
and pre-B and pre-T cell receptors (40
, 41)
. mGCR must also be classified in this category, which may explain why detection of these receptors on nonmalignant human cells was until now unsuccessful.
The immunostimulation of PBMC leads to the proportion of mGCR-positive monocytes being dramatically increased (Fig. 6)
. Although up to 57.9% of monocytes (33.8±14.0%, n=12; Fig. 6
) were mGCR positive, there was no significant effect on the mGCR expression of lymphocytes. These results could be achieved only by high-sensitivity immunofluorescent liposome staining, whereas tests for mGCR by conventional immunofluorescence were negative under these conditions (data not shown). Thus, although an LPS-induced increase in the number of mGCR per monocyte is possible, receptor density clearly is still so low (>100 but <2000/per cell) that conventional detection is unsuccessful.
That this effect can be completely blocked by the inhibitor of the secretory pathway brefeldin A suggests an active transport of the mGCR from the cytosol to the membrane surface. Just as difficult to answer about the membrane-bound estrogen receptor (12)
is the question of how the protein is inserted into the membrane. If cGCR and mGCR are derived from the same transcript, then there is only a single mRNA producing two similar but functionally distinct proteins that move either into the cytosol (cGCR) or to the membrane (mGCR). To be inserted in the plasma membrane, a transmembrane domain within the protein should exist. However, a genuine transmembrane domain within the original protein has not been described yet and there is no evidence now that any post-translational modification occurs that would facilitate membrane insertion. This problem is unsolved for the membrane-bound estrogen receptor (12)
. This raises the question of whether cGCR and mGCR might be coded by two different genes. These genes would have to display a high sequence homology; otherwise the monoclonal antibody we used would not have been able to identify both proteins specifically. The group led by Gametchu used various monoclonal and polyclonal antibodies directed against different epitopes in the cGCR molecule to identify mGCR on leukemic/lymphoma cells (15
, 42)
. However, a search in the NCBI protein database only yields an amino acid sequence match with a single protein, the human GCR. We therefore overexpressed human GCR to determine whether this would lead to increased mGCR expression. Our working hypothesis was that GCR overexpression would lead to a new equilibrium between cGCR and mGCR by more cGCR or post-translationally edited cGCR protein finding its way to the cell surface. But our experiments on the overexpression of his-tagged GCR did not show any increased mGCR expression on the cell surface. This leads us to the conclusion that the mGCR is not just a cGCR that is transported to the surface in unchanged form. As suspected by Diba et al., the mGCR is probably a variant of the cGCR produced by differential splicing and/or by promotor switching (27)
. Alternatively, the mGCR is produced from the cGCR by post-translational editing. According to our data, we suggest that LPS stimulation causes both increased production of the mGCR protein and a stimulation of active transport to the cell surface.
Our observation of LPS-induced mGCR expression gave rise to the suspicion that more monocytes may be mGCR positive in the case of disease-related increased activity of the immune system. We investigated this hypothesis in a clinically characterized cohort of patients with rheumatoid arthritis. Our data show a strongly positive correlation of the frequency of mGCR-positive monocytes and different parameters of disease activity (Fig. 10)
. These observations can be interpreted to mean that mGCR may play a role in the etiopathogenesis of this disease. According to the latest findings, however, it is more probable they cause negative feedback regulation. An excessive immune reaction might thus be limited by the glucocorticoid–mGCR interaction leading to apoptosis of the cell (15
, 41
, 42)
. Over-excessive immune activation would thus be prevented by increased mGCR expression with an appropriately increased rate of apoptosis. This supposition was deduced from the observation that the presence of mGCR in different leukemic and lymphoma cells and cell lines correlates with the ability of these cells to respond in vitro to the lymphocytolytic activity of glucocorticoids (15
, 42
, 43)
.
How could a glucocorticoid cause these effects by binding to the mGCR? The most probable assumption is that the binding triggers a signaling through cell surface receptors. It has been shown for estradiol that binding to membrane receptors leads e.g., to a rapid rise in the intracellular Ca2+ concentration, which is suggested to be mediated through a novel G-protein-coupled membrane estrogen receptor (36)
. Similar observations have been made for testosterone (37)
and other steroids (9)
. For glucocorticoids there are, on the one hand, many reports on rapid actions (44)
and, on the other, the reports mentioned on the existence of mGCR in amphibian brain and on lymphoma cells and membrane binding sites for several glucocorticoids in various tissues or cells (44)
. The causal connection between mGCR and these rapid effects is not yet clear, however. According to our results, the cause is probably to be found in the fact that the small number of mGCR per cell and the dependency of their expression frequency on external factors make suitable investigations very complicated.
Why should steroid hormones interact with receptors both at the cell surface and within the cell? It has been suggested that the synthesis of macromolecules as a result of genomic actions of steroid hormones must be preceded by rapid changes in the cellular environment to "prime" the cell and support such activity. These preparative changes may include altered ion influxes, import of amino acids and sugars, or phosphorylation of key enzymes, all of which could be triggered by interactions of steroid hormones with cell surface receptors (28)
.
Another theory is that membrane-based steroid receptors cause receptor-mediated endocytosis (28)
. Bound to its plasma carrier protein, the steroid hormone is brought into the cell via a cell surface receptor. The complex is broken down inside the lysosome; free steroid hormone diffuses into the cell, where it subsequently exerts its action at the genomic level or undergoes metabolism. However, we consider this theory to be improbable for the mGCR in view of their small number.
We conclude that mGCR 1) are indeed physiologically present in healthy blood donors, but remained unidentified by conventional techniques due to their small number per cell and 2) are actively up-regulated and transported through the cell after immunostimulation. These receptors may reflect a feedback mechanism of the organism upon immunostimulation and/or play a role in pathogenesis. If these suppositions are confirmed, it would indicate a need for the development of drugs to specifically modulate specific nongenomic action via selective activation or inhibition of mGCR. The next steps to be taken are 1) experimental work to investigate the signaling as mediated by mGCR occupation and 2) to carry out affinity assays in order to explore mGCR binding characteristics.
|
ACKNOWLEDGMENTS |
|---|
Received for publication May 29, 2003. Accepted for publication September 22, 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. M. Spies, B. Bartholome, T. Berki, G-.R. Burmester, A. Radbruch, A. Scheffold, and F. Buttgereit Membrane glucocorticoid receptors (mGCR) on monocytes are up-regulated after vaccination Rheumatology, February 1, 2007; 46(2): 364 - 365. [Full Text] [PDF]
|
|
|
|
 |
|
 S Capellino, B Riepl, L Rauch, P Angele, M Cutolo, and R H Straub Quantitative determination of steroid hormone receptor positive cells in the synovium of patients with rheumatoid arthritis and osteoarthritis: is there a link to inflammation? Ann Rheum Dis, January 1, 2007; 66(1): 53 - 58. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. W. Leung, Y. L. Pon, R. N. S. Wong, and A. S. T. Wong Ginsenoside-Rg1 Induces Vascular Endothelial Growth Factor Expression through the Glucocorticoid Receptor-related Phosphatidylinositol 3-Kinase/Akt and beta-Catenin/T-cell Factor-dependent Pathway in Human Endothelial Cells J. Biol. Chem., November 24, 2006; 281(47): 36280 - 36288. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. ten Hove, L. A. Houben, J. A. M. Raaijmakers, L. Koenderman, and M. Bracke Rapid Selective Priming of Fc{alpha}R on Eosinophils by Corticosteroids J. Immunol., November 1, 2006; 177(9): 6108 - 6114. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C M Spies, D H S Schaumann, T Berki, K Mayer, M Jakstadt, D Huscher, C Wunder, G-R Burmester, A Radbruch, R Lauster, et al. Membrane glucocorticoid receptors are down regulated by glucocorticoids in patients with systemic lupus erythematosus and use a caveolin-1-independent expression pathway Ann Rheum Dis, September 1, 2006; 65(9): 1139 - 1146. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. R. Pondugula, N. N. Raveendran, Z. Ergonul, Y. Deng, J. Chen, J. D. Sanneman, L. G. Palmer, and D. C. Marcus Glucocorticoid regulation of genes in the amiloride-sensitive sodium transport pathway by semicircular canal duct epithelium of neonatal rat Physiol Genomics, January 12, 2006; 24(2): 114 - 123. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Horvath and A. Wanner Inhaled corticosteroids: effects on the airway vasculature in bronchial asthma Eur. Respir. J., January 1, 2006; 27(1): 172 - 187. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Maier, D. Runzler, J. Schindelar, G. Grabner, W. Waldhausl, G. Kohler, and A. Luger G-protein-coupled glucocorticoid receptors on the pituitary cell membrane J. Cell Sci., August 1, 2005; 118(15): 3353 - 3361. [Abstract] [Full Text] [PDF]
|
|
1.0%) mGCR expression.
