The Expression of 11-Hydroxysteroid Dehydrogenase Type I by Lymphocytes Provides a Novel Means for Intracrine Regulation of Glucocorticoid A
http://www.100md.com
免疫学杂志 2005年第2期
Abstract
The 11-hydroxysteroid dehydrogenase (11-HSD) enzymes control the interconversion of active glucocorticoids (GCS) and their inactive 11-keto metabolites, a process commonly referred to as the cortisone/cortisol shuttle. Although the prereceptor metabolism of GCS by 11-HSD is well documented in a variety of cells and tissues, it has not yet been carefully investigated in the major cell types of the immune system. In this study, we demonstrate that 11-HSD1 transcripts, protein, and enzyme activities are actively expressed in murine CD4+, CD8+, and B220+ lymphocytes, as well as CD11c+ dendritic cells. Only reductase activity was observed in living cells, evidenced by the restricted conversion of cortisone to cortisol. Activation of CD4+ T cells increased their 11-HSD1 activity, as did their polarization into Th1 or Th2 cells. CD4+ T cells isolated from aged donors (>16 mo) had increased 11-HSD1 protein and an elevated capacity to convert cortisone to cortisol. The GCS generated in murine CD4+ T cells from their inactive 11-keto metabolites could activate the GCS receptor, demonstrated by an up-regulation of IL-7R and GCS-induced leucine zipper gene expression. The presence of a functional 11-HSD1 provides lymphocytes with a novel intracrine regulatory mechanism that could influence such processes as lymphocyte development, effector function, and susceptibility to apoptosis. Thus, the presence of 11-HSD1 provides an additional means to facilitate GCS influences over lymphocyte activities, uncoupled from the plasma concentration of GCS.
Introduction
As major steroid hormones produced by the adrenal gland, glucocorticoids (GCS)3 are now appreciated to represent essential regulators of the development, homeostasis, and effector functions of the innate and adaptive immune system. GCS can also alter immune cell trafficking through effects on vascular permeability, chemotaxis, and adhesion molecule expression (1, 2, 3, 4, 5). At a cellular level, GCS alter the activation, differentiation, and maturation of most immune cell types, as well as play multiple roles in regulating immune cell sensitivity to apoptosis (6, 7). Some of these effects are accomplished through GCS effects on cytokine and chemokine production, while others are mediated through the ability of GCS to modulate the outcome of numerous cell signaling systems (6, 7).
The potent and pleiotropic effects of GCS on gene expression and the functional properties of numerous transcription factor activities are paralleled by the presence of complex regulatory mechanisms that provide control over their diverse array of cellular influences. One of these controlling mechanisms operates at the level of endogenous GCS production by the adrenal gland, the hypothalamic-pituitary-adrenal axis, which is influenced by the diurnal cycle, organismal stress, and promoters of inflammatory responses (2, 3).
The majority of the bioactive GCS (cortisol and corticosterone in humans, corticosterone only in rodents) in the systemic circulation is found physically associated with albumin or corticosteroid-binding globulin (8, 9). The physical association of GCS with these plasma proteins provides a second means of control over GCS effects by effectively buffering the rate of cellular entry (8, 9). Interestingly, cortisone and 11-dehydrocorticosterone (11DHC), the biologically inactive 11-keto forms of naturally produced GCS, are not found in the plasma in a protein-associated state (8). Consequently, the 11-keto metabolites of GCS are able to freely enter target cells without restrictions.
Intracellularly, the GCS serve as activating ligands for the GCS receptor (GR), a member of the nuclear hormone receptor superfamily of ligand-activated transcription factors (10). Ligand binding of the GR promotes its dissociation from heat shock protein 90, followed by its nuclear translocation, homodimerization, and physical association to specific cis-acting GCS response elements (GREs) of GCS-controlled genes, thereby conferring specific control over their transcription (11, 12). Although GRs are ubiquitously expressed in nucleated cells, a dynamic modulation over their cellular expression has been shown in a variety of cell types (13, 14, 15, 16, 17). The ligand-activated GR can also modulate gene expression indirectly, through protein-protein interactions with numerous transcription factors (e.g., AP-1, NF-B) whose activities are suppressed from such associations (11).
The 11-hydroxysteroid dehydrogenases (11-HSDs) are enzymes that interconvert the functional GCS with their 11-keto metabolites, thereby increasing or decreasing the intracellular levels of bioactive GCS available as activating ligands for the GR. The 11-HSDs are members of the short-chain alcohol dehydrogenase family of enzymes, and two distinct isoforms have been identified and thoroughly investigated to date: 11-HSD type I (11-HSD1) and 11-HSD type II (11-HSD2). Using NAD+ as cofactor to carry out its sole function as a steroid dehydrogenase, 11-HSD2 inactivates bioactive GCS, generating 11-keto metabolites that are incapable of binding or activating the GR (18, 19). In adults, the majority of 11-HSD2 expression is anatomically confined to major mineralcorticoid target tissues (e.g., kidney) (19). During gestation, however, 11-HSD2 is highly expressed in the placenta and fetal tissues, where it functions to inactivate maternal GCS and provide protection of the fetus against the deleterious effects of GCS (20, 21, 22, 23). In contrast, 11-HSD1 exclusively uses NADP(H) as a cofactor and functions predominantly as a reductase in vivo, regenerating bioactive GCS from the circulating 11-keto metabolites (24). Cellular expression of 11-HSD1 is widespread, with the highest levels found in the liver. Thus, at the cellular level, the 11-HSDs provide control over the prereceptor metabolism of GCS, a process commonly referred to as the cortisone/cortisol shuttle.
The physiological importance of prereceptor metabolism of GCS is well documented in many cell and tissue types (25, 26, 27, 28, 29). However, with only a single exception, the prereceptor metabolism of GCS in immune cells has not yet been investigated (30). This publication reported 11-HSD1 expression in monocytes that are induced to mature into macrophages (30).
In the original report that described the cloning and tissue distribution of murine 11-HSD1, the thymus was determined to express low levels of 11-HSD1 mRNA by Northern blot analysis (31). However, this study failed to establish whether the stromal elements of the thymus, thymocytes, or associated tissue macrophages expressed 11-HSD1 transcripts (31).
In this study, we describe the presence of 11-HSD1 in cells of the immune system. Our studies demonstrate that 11-HSD1 mRNA, protein, and enzyme reductase activity are present in murine CD4+ and CD8+ T cells, B220+ B cells, as well as CD11c+ bone marrow-derived dendritic cells (BMDC). Furthermore, the 11-HSD1 in murine lymphocytes was sufficiently active to provide an intracrine mechanism, allowing bioactive GCS produced endogenously to exert control over molecular immune processes. We raise the possibility that an intracellular reactivation of GCS in lymphocytes could provide an autonomous means to modulate individual immune effector functions as well as influence the immediate microenvironment.
Materials and Methods
Experimental animals
BALB/c, C57BL/6, and C3H/HeN mice were purchased from Charles River Laboratories; DO11.10 TCR transgenic mice on a BALB/c background were originally obtained from The Jackson Laboratory and bred in-house. The 11HSD1–/– mice on a C57BL/6 background were obtained from J. Seckl and J. Mullins from the University of Edinburgh (Edinburgh, Scotland), and bred in-house. Female animals, 10–12 wk of age, were used for all experiments described in this work, except in studies in which lymphocytes from 11-HSD1–/– mice were being analyzed. All mice were housed in a specific pathogen-free barrier facility at the University of Utah Animal Resource Center. This facility guarantees strict compliance with regulations established by the Animal Welfare Act. All animals were housed with a 12-h light-dark controlled cycle, and were provided with mouse chow and water ad libitum. At time of sacrifice, mice were anesthetized with halothane, followed by cervical dislocation.
Culture conditions and reagents
Murine primary lymphocytes were isolated from the spleens of normal mice. They were cultured in complete medium (RPMI 1640 (Invitrogen Life Technologies) supplemented with 10% FCS (HyClone), 2 mM L-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, and 50 μM 2-ME), except where otherwise noted. Murine CD11c+ BMDC were generated, as described previously (32). The murine EL-4 T cell line was purchased from American Type Culture Collection. The murine T cell hybridoma, DO11.10, was obtained from J. Woodward (University of Kentucky, Lexington, KY). Glycerrhetinic acid (GA) and all GCS (cortisone, hydrocortisone, corticosterone, and 11DHC) were purchased from Sigma-Aldrich. RU486 was a gift from M. Mitchell (University of Utah). [3H]Cortisone and [3H]cortisol (sp. act., 50 Ci/mmol) were purchased from American Radiolabeled Chemicals. FITC-VAD-FMK was purchased from Promega. Recombinant murine IL-4 and IFN- were purchased from BD Pharmingen; anti-IL-4 (11B11) and anti-IFN- (XMG1.2) were generated in-house. All Abs used for ELISA, activation of T cells, specific cell population enrichments, and FACS analysis were purchased from BD Pharmingen. Coating of tissue culture plates with immobilized anti-CD3 and anti-CD28 Ab was done by incubating plates with anti-CD3 and anti-CD28 Ab at 1–2 μg/ml in Tris-HCl buffer (pH 9.6) at 37°C for 2 h, followed by extensive washing.
Dynabead cell enrichment
For the preparation of CD4+ lymphocytes, freshly isolated spleens were dissociated into a single cell suspension in complete medium. The erythrocytes were lysed by brief treatment with sterile aqueous 0.83% (w/v) ammonium chloride. For enrichment of CD4+ T cells, erythrocyte-depleted splenocytes were incubated with 2 μg/ml each of biotinylated anti-CD45R/B220, anti-CD11b (also known as Mac-1+), and anti-CD8 Abs (BD Pharmingen) for 30 min on ice. Following washes, the cells were resuspended in medium, and streptavidin-coated M-280 magnetic Dynabeads were added (Dynal Biotech) at a bead:cell ratio of 1:1 for a total of 20 min with constant agitation at 4°C. The labeled cells were removed by placing the cells in a magnetic field, and residual cells were separated and washed for use in cell culture or Western blot analysis. The purification of CD8+, B220+, and Mac-1+ cells by negative selection was done in a similar manner. For the enrichment of CD8+ cells, biotinylated anti-CD45R/B220, anti-CD11b, and anti-CD4 Abs were used; for the enrichment of B220+ cells, biotinylated anti-CD4, anti-CD8, and anti-CD11b Abs were used; for the enrichment of Mac-1+ cells, biotinylated anti-CD4, anti-CD8, and anti-B220 Abs were used. Purity of the cells was determined by FACS analysis with FITC anti-mouse CD Abs. The level of purity of negatively selected cells was routinely >90%.
Quantitative PCR
Total RNA was extracted by the method of Chomczynski and Sacchi (33). cDNA was reverse transcribed from RNA, as previously described (34). Quantitative PCR was performed using the Roche Light Cycler, as recommended by the manufacturer. The following primer sets and PCR conditions were used for the amplification of the gene of interest. To initiate the PCR, all samples were denatured at 95°C for 30 s. The murine GAPDH sense 5'-AGT ATG TCG TGG AGT CTA C-3', antisense 5'-CAT ACT TGG CAG GTT TCT C-3', denatured samples were allowed to anneal at 58°C for 10 s, followed by 20 s of extension at 74°C; murine 11-HSD1 sense 5'-AAC CAC ATC ACT CAG ACC-3', antisense 5'-CAC CAT TAG AAC AGA ACT C-3', denatured samples were allow to anneal at 67°C for 5 s, followed by 10 s of extension at 74°C; murine 11-HSD2 sense 5'-GTA TCA AGG TCA GCA TTA TC-3', antisense 5'-GAA GGT GAT TGA TAA AGA AG-3', denatured samples were allowed to anneal at 61°C for 5 s, followed by 14 s of extension at 74°C; murine IL-7R sense 5'-CCA TT TGA GTT TGT TCT CTG-3', antisense 5'-GTC CCT GTG TCT CCA ACT CC-3', denatured samples were allowed to anneal at 62°C for 0 s, followed by 12 s of extension at 74°C; murine GCS-induced leucine zipper (GILZ) sense 5'-AGC AGC CAC TCA AAC CAG C-3', antisense 5' GCC CTA GAC AAC AAG ATT 3', denatured samples were allowed to anneal at 60°C for 0 s, followed by 10 s of extension at 74°C. All PCR were allowed 40 cycles of amplification, which ended with a melting program that spanned the temperature range of 40–95°C with a temperature transition rate of 0.20°C/s. Specificity of the amplification product was determined using the melting temperature (at which all products melted) established during optimization of the gene-specific primer sets.
Luciferase reporter assay
The mouse mammary tumor virus (MMTV)-Luc reporter construct under the control of GREs was obtained from S. Andrews (University of Iowa, Iowa City, IA). Transient transfections were performed, as described in detail previously (35). Briefly, 5 x 106 murine DO11.10 hybridoma T cells were resuspended in 450 μl of RPMI 1640 and placed in 0.4-cm Gene Pulser cuvettes (Bio-Rad) with 2.5 μg of plasmid. Electroporation was conducted at 960 μF and 280 V. Cells were then rested for 5 min and transferred to 100-mm tissue culture plates containing 10 ml of culture medium. Use of the transiently transfected cells in experiments was done within 48 h.
ELISA
Primary murine CD4+ T cells at 5 x 106/ml were placed in multiwell cell culture plates previously coated with either 1 μg/ml anti-CD3 Ab alone, or plates previously coated with anti-CD3 as well as anti-CD28 (both at 1 μg/ml). The cells were cultured at 37°C in an atmosphere of 5% CO2 in air, and all supernatants were collected after 24 h to quantitatively evaluate the levels of IL-2, IFN-, and IL-4 by ELISA, as described previously (36).
The 11-HSD1 Western blot
The DO11.10 T cell hybridomas, the EL-4 T cell thymoma, CD11c+ BMDC, as well as freshly isolated primary murine CD4+, CD8+, and B220+ cells were washed twice with ice-cold PBS and resuspended in lysis buffer (10 mM HEPES, pH 7.8, 0.1 mM EDTA, 10 mM NaCl, 3 mM MgCl2, 300 mM sucrose, 1% Nonidet P-40, and a mixture of protease inhibitors (Roche)). The lysed cells were subjected to one freeze/thaw cycle before centrifugation at 10,600 x g for 10 min to remove nuclear debris. The supernatant was collected and protein content was determined using the bicinchoninic acid Protein Assay kit (Pierce), following the manufacturer’s recommendations. Equal amounts of protein were subjected to electrophoresis on 12% SDS-PAGE gels and transferred to nitrocellulose or polyvinylidene difluoride membranes. After blocking with 5% nonfat milk in TBS, blots were incubated with anti-11-HSD1 Ab (1.5 μg/ml) (American Diagnostica) for 2 h at room temperature. Membranes were then washed and incubated with goat anti-rabbit HRP conjugate (Bio-Rad) for 1 h at room temperature. After extensive washing, bands were visualized using the ECL kit (Amersham Biosciences), according to manufacturer’s instructions.
The 11-HSD1 immunofluorescence
Primary murine CD4+ lymphocytes from C57BL/6 wild-type (WT) and 11-HSD1–/– animals were purified using methods described above. Lymphocytes were then adhered to coverslips coated with poly(L-lysine) solution (0.1% w/v; Sigma-Aldrich) and fixed with 4% paraformaldehyde in TBS. Fixed lymphocytes were then permeabilized with 0.5% Triton X-100 in TBS for 10 min. Coverslips were then blocked with blocking buffer (3% nonfat dry milk, 0.5% Triton X-100 in TBS) at room temperature for 1 h before incubation with anti-11-HSD1 Ab (10 μg/ml in blocking buffer). The coverslips were then washed twice with 0.5% Triton X-100 in TBS, blocked for an additional 30 min, and incubated with Alexa488 (1:500 in blocking buffer) in the dark at room temperature for 1 h. The coverslips were then washed twice, mounted in FluorSave antifade mounting medium (Calbiochem), and dried at room temperature overnight. Samples were analyzed with identical parameters using a FV300 confocal microscope (x60 oil objective with NA 1.4).
The 11-HSD enzymatic activity
The 11-HSD activity was determined in intact cells by measuring the interconversion of cortisone and cortisol. To determine reductase activity, 100 nM cortisone with a tracer amount of [3H]cortisone was added to cell cultures containing cells at a density of 106 cells/ml. At the appropriate time points, steroids were extracted from the supernatant by adding 2 vol of cold ethyl acetate. The organic phase was collected, evaporated to dryness under a stream of nitrogen, and reconstituted with 50 μl of a 4 mg/ml solution containing equal amounts of cortisone and cortisol in chloroform/methanol (9:1). The extract was then spotted onto a TLC plate (Analtech) and developed using chloroform/methanol (9:1) as the mobile phase. The cortisone and cortisol spots were then visualized with iodine vapor, scraped from the TLC plate, and transferred to scintillation vials to determine the ratio of [3H]cortisone and [3H]cortisol. Percent conversion was calculated as (cpm [3H]cortisol/(cpm [3H]cortisol + cpm [3H]cortisone)) x 100.
Th1 and Th2 polarization
CD4+ T cells were isolated from the peripheral lymph nodes (PLN) (axillary, cervical, inguinal, and popliteal) of C3H/HeN mice or the PLN and spleen of DO11.10 TCR transgenic mice by negative selection using the CD4+ T cell selection kit from Miltenyi Biotec, according to the manufacturer’s instructions. Purified populations of CD4+ T cells are then activated by placing them onto tissue culture plates containing coimmobilized anti-CD3 and anti-CD28 (both 2 μg/ml) at a density of 2.5 x 106 - 5.0 x 106/ml in the presence of different polarizing cytokines for a period of 4–5 days to generate Th0, Th1, and Th2 cells. Th0 cultures contained no additional cytokines or Abs, Th1 cultures contained IL-12 (5 ng/ml) and anti-IL-4 (10 μg/ml), and Th2 cultures contained IL-4 (10 ng/ml) and anti-IFN- (10 μg/ml). At the end of the culture period, cells were harvested, washed three times, and resuspended in complete medium at a density of 106 cells/ml for further use. Detection of 11-HSD1 protein and reductase activity was performed, as described above. At the same time, a fraction of the Th0, Th1, and Th2 cells was reactivated with plate-bound anti-CD3 (1 μg/ml) for 24 h, and the supernatant was collected to perform ELISA to determine the pattern of cytokines produced. Upon reactivation, the Th0 cells produced both IFN- and IL-4, Th1 cells produced IFN- and no IL-4, and Th2 cells produced IL-4 and no IFN- (data not shown).
Apoptosis assay
The DO11.10 T cell hybridoma in log-phase growth was cultured at the density of 106 cells/ml. A total of 10–6 M and 10–7 M cortisone or cortisol was added to the cell culture at the predetermined time points backward, starting with the 24-h time points. This manner of treatment allowed all cells to be analyzed by FACS at the same time. At the end of the culture period, all cells were collected by centrifugation at 1500 rpm for 5 min and resuspended in 200 μl of 10 μM FITC-VAD-FMK in PBS, followed by staining for 20 min at 37°C and 5% CO2. After staining, all cells were washed twice with PBS, resuspended in 1 ml of cold PBS, and analyzed by FACS.
Statistical analysis
Statistical analysis was performed with the StatView 5.0 software. The p values were determined by Student’s t test (unpaired).
Results
The 11-HSD1 mRNA and protein are expressed in primary murine immune cells and cell lines
We previously reported that 11-HSD1 is expressed in murine spleen by mRNA analysis (37). With the exception of a single report describing the induction of 11-HSD1 in human monocytes upon differentiation to macrophages (30), the expression of 11-HSD1 in immune cell types has not been characterized. To investigate comprehensively whether 11-HSD1 is expressed in immune cells, we used quantitative RT-PCR to look for the presence of 11-HSD1 transcripts in freshly isolated murine splenic CD4+ and CD8+ T cells, B220+ B cells, as well as Mac-1+ cells. All purified populations analyzed were found to express 11-HSD1 transcripts (Fig. 1A). Mac-1+ cells were found to express the highest level of 11-HSD1 mRNA (Fig. 1A). In lymphocytes, CD4+ and CD8+ T cells consistently expressed higher levels of 11-HSD1 mRNA than B220+ B cells. The murine T cell hybridoma DO11.10 and the EL-4 T cell thymoma were also found to express 11-HSD1 mRNA by quantitative PCR (data not shown), further suggesting that murine lymphocytes, and not a minor cellular contaminant, were responsible for the 11-HSD1 mRNA being observed. In contrast, 11-HSD2 transcripts were undetectable by RT-PCR in these same populations of cells (data not shown). We next questioned whether the presence of 11-HSD1 mRNA correlated with the presence of 11-HSD1 protein. Using a polyclonal Ab specific for murine 11-HSD1, a single band in the expected m.w. region for 11-HSD1 was detected in all of the cell types investigated with the addition of CD11c+ BMDC (Fig. 1B). The single band recognized by the polyclonal Ab was demonstrated to be specific for murine 11-HSD1, as the same band was not present when lysates from purified 11-HSD1–/– CD4+, CD8+, and B220+ lymphocytes were analyzed in parallel. Similarly, the EL-4 thymoma cell line and the DO11.10 T cell hybridoma also demonstrated expression of 11-HSD1 protein by Western blot (data not shown).
FIGURE 1. The 11-HSD1 mRNA and protein are expressed in murine lymphocytes. A, 11-HSD1 transcript levels in primary murine CD4+, CD8+ T cells, B220+ B cells, and Mac-1+ macrophages were analyzed by quantitative RT-PCR. Results were normalized to GAPDH and shown as fold difference relative to levels found in B220+ B cells. B, Western blot analysis of 11-HSD1 protein expression in primary WT and 11-HSD1–/– CD4+, CD8+, and B220+ lymphocytes, as well as CD11c+ BMDC. Bands shown correlate with molecular mass of 36 kDa. C, Cytoplasmic localization of endogenous 11-HSD1 by immunofluorescence. Purified CD4+ T cells from C57BL/6 WT and 11-HSD1–/– mice were fixed, permeabilized, and immunostained for 11-HSD1. All results shown are representative of 3 independent experiments.
The 11-HSD1 has been proposed to be a transmembrane protein associated with the endoplasmic reticulum (38). Using the 11-HSD1 polyclonal Ab for immunostaining, we were able to detect the presence of 11-HSD1 peripheral to the nucleus in primary CD4+ T cells isolated from WT C57BL/6 mice (Fig. 1C). However, the same 11-HSD1-specific staining was not observed in primary CD4+ T cells isolated from 11-HSD1–/– animals (Fig. 1C).
The 11-HSD1 reductase activity is present in primary murine immune cells and T cell lines
To establish the presence of 11-HSD1 reductase activity inside of murine lymphocytes, we used a whole cell assay to analyze the conversion of the biologically inactive 11-keto metabolite of GCS (cortisone) to the active GCS (cortisol) in cultures of purified murine lymphocytes. Freshly isolated CD4+, CD8+, and B220+ lymphocytes were cultured separately in the presence of 100 nM cortisone containing a trace of [3H]cortisone. Supernatants were collected at various time points and analyzed for the conversion of cortisone to cortisol. Reductase activity was detected as determined by the presence of [3H]cortisol in the supernatant of CD4+, CD8+ T cells, as well as the supernatant of the B220+ B cells at all time points examined (Fig. 2A). No conversion of cortisone to cortisol above background levels was detected in medium alone (data not shown). Of the lymphocyte populations, CD4+ and CD8+ T cells showed similar levels of reductase activity, albeit the level of reductase activity was found to decrease significantly (2-fold) over a 24-h time period. In comparison, the B220+ B cell population exhibited a reduced level of reductase activity over a 24-h time period, consistent with the lowered 11-HSD1 transcript level found in B220+ B cells (Fig. 1A). We also found that CD11c+ BMDC possessed higher levels of 11-HSD1 reductase activity when compared with primary lymphocytes (Fig. 2B). The 11-HSD1 reductase activity in BMDC was inhibited by the presence of GA, a known inhibitor of the 11-HSD enzymes (Fig. 2B). Similar to what others have observed (27), we found that 11-HSD1 sp. act. was reduced at the higher concentrations of cortisone used (100 nM) (Fig. 2C). Purified populations of the murine T cell thymoma EL-4 and the murine T cell hybridoma DO11.10 were also found to have the capacity to generate active GCS from their 11-keto metabolites (Fig. 2D), eliminating the possibility of contaminating nonlymphocytes being responsible for the observed enzyme activities.
FIGURE 2. The 11-HSD1 reductase activity is expressed in primary murine immune cells and cell lines. A, Primary CD4+, CD8+, and B220+ T cells were cultured separately in the presence of [3H]cortisone. Supernatants were collected at the indicated time points for steroid extraction, and determination of 11-HSD1 reductase activity evidenced by the conversion of cortisone to cortisol. Results were calculated and shown as fmol of cortisol produced/h/500,000 cells. B, CD11c+ DCs were cultured in the presence of [3H]cortisone in the presence or absence of GA (10 μM). Supernatants were collected to analyze for the conversion of cortisone to cortisol. Results were calculated and shown as fmol of cortisol produced/h/500,000 cells. C, Primary CD4+ T cells were cultured in the presence of increasing concentrations of [3H]cortisone. Supernatants were collected to analyze for the conversion of cortisone to cortisol. Results were calculated and shown as fmol of cortisol produced/h/500,000 cells. D, The murine T cell hybridoma DO11.10 and murine thymoma EL-4 cells were cultured in the presence of [3H]cortisone. Conversion of cortisone to cortisol at the indicated time points was determined by analyzing the supernatants, and is expressed as fmol of cortisol converted/500,000 cells over time. Results shown are mean ± SEM. E, The DO11.10 T cell hybridoma was cultured alone or in the presence of cortisone or cortisol for the indicated time periods. Cells were then stained with FITC-VAD-FMK as an indicator of apoptosis and analyzed by FACS. Results are shown as percentage of cells positive for FITC-VAD-FMK. All experiments were repeated three times with similar results.
It has been previously established that T cell hybridoma lines are susceptible to GCS-induced apoptosis (39), we therefore questioned whether the capacity to generate active GCS (cortisol) from inactive 11-keto metabolites (cortisone) affected the sensitivity of the DO11.10 T cell hybridoma to GCS-induced apoptosis in the presence of cortisone. Log-phase DO11.10 T cell hybridoma cells were cultured in the absence or presence of cortisone or cortisol for various times, followed by staining with FITC-VAD-FMK to determine the percentage of cells undergoing apoptosis. Shown in Fig. 2E, the presence of added cortisol induced significant levels of apoptosis by 24 h in the DO11.10 T cell hybridoma. Similar to what others have observed (39), cortisone addition did not induce a significant level of apoptosis in these same cells by 24 h. These data, coupled with results obtained from Fig. 2D, suggest that even though the DO11.10 T cell hybriboma has the capacity to generate cortisol from cortisone in culture, the amount of active GCS produced over a 24-h time period was insufficient to induce significant levels of apoptosis in this cell line.
An increasing number of reports indicates that 11-HSD1 can function as a dehydrogenase in vitro (25). In intact cells and in vivo, however, 11-HSD1 functions predominantly as an oxidoreductase (40). To determine the directionality of 11-HSD1 enzyme activity in intact murine lymphocytes, [3H]cortisol was added to cultures of CD4+ T cells to question the presence of dehydrogenase activity. As evidenced by the complete lack of conversion of cortisol to cortisone, no dehydrogenase activity above background levels was detected (Fig. 3A). The lack of any dehydrogenase activity further supports the observation that 11-HSD2 transcripts were not detected in CD4+ T cells.
FIGURE 3. The 11-HSD1 is the sole reductase in murine lymphocytes and functions exclusively as a reductase. A, Purified primary CD4+ T cells were cultured in the presence of increasing concentrations of cortisol. No dehydrogenase activity was detected, as determined by the lack of cortisone in the supernatant. B, Thymocytes and purified populations of CD4+ T cells from C57BL/6 and 11-HSD1–/– animals were cultured in the presence of [3H]cortisone. The 11-HSD1 reductase activity is determined by the rate of conversion of cortisone to cortisol. Results were calculated and shown as fmol of cortisol produced/h/500,000 cells. Results shown are representative of 3 independent experiments.
Intrigued by the observation that the induction of 11-HSD1 expression was concomitant to the differentiation of immature monocytes into mature macrophages (30), we questioned whether the capacity to generate bioactive GCS from its 11-keto metabolite was restricted to mature lymphocytes in the periphery. When thymocytes were isolated from the thymi of day 3 neonates (which contains >95% CD4–CD8– and CD4+CD8+ immature T cells) or young adult mice, and cultured in the presence of [3H] cortisone, 11-HSD1 reductase activity was indicated by the presence of [3H]cortisol in the supernatant (data not shown). Thus, murine thymocytes as well as mature lymphocytes possess the intrinsic ability to generate bioactive GCS from 11-keto metabolites.
Finally, to establish that 11-HSD1 is responsible for the conversion of cortisone to cortisol by lymphocytes, thymocytes and CD4+ T cells were isolated from both WT and 11-HSD1–/– mice and cultured in the presence of radiolabeled cortisone. Reductase activity in lymphocytes was 13-fold less in cells from the 11-HSD1-deficient animals when compared with WT (Fig. 3B). Similarly, thymocytes isolated from 11-HSD1–/– mice also showed a markedly diminished capacity to generate bioactive GCS from 11-keto metabolites (Fig. 3B). Taken together, this demonstrates 11-HSD1 to be the major GCS reductase in murine CD4+ lymphocytes.
GCS generated from inactive 11-keto metabolites in CD4+ lymphocytes are sufficient to function as activating ligands for the GR
Nuclear hormone receptors, such as the GR, require ligand activation to promote the recruitment of the transcription machinery necessary for trans activation of GRE-containing genes (10). To specifically demonstrate that immunomodulatory quantities of active GCS can be generated by murine CD4+ lymphocytes, an MMTV-Luc reporter construct was transiently transfected into the DO11.10 T cell hybridoma. The MMTV contains well-characterized GREs in its promoter region and is commonly used as a reporter to establish GCS responsiveness (41, 42, 43). The transiently transfected DO11.10 T cell hybridomas were treated with either bioactive GCS (corticosterone) or its 11-keto metabolite (11DHC) for a period of 24 h. These cells were then collected for cytoplasmic protein extraction to quantitatively analyze luciferase activity produced as the end result of the trans activation of the MMTV-Luc reporter construct. As demonstrated in Fig. 4A, exposure to active GCS induced a dose-dependent increase in the relative luciferase activity. Interestingly, exposure to 11-keto metabolites induced a similar dose-dependent increase in relative luciferase activity, confirming the ability of DO11.10 T cell hybridomas to generate active GCS. The coincubation of MMTV-Luc+, DO11.10 cells with 11DHC plus the 11-HSD1 inhibitor GA abrogated the ability of bioactive GCS to be generated, evidenced by a lack of up-regulated luciferase activity (Fig. 4A). GA addition to parallel cultures containing bioactive GCS stimulated the up-regulation of luciferase activity at levels similar to the positive control.
FIGURE 4. GCS generated from inactive 11-keto metabolites by CD4+ lymphocytes function as activating ligands for the GR. A, An MMTV-luciferase reporter construct containing GREs in the promoter region was transiently transfected into DO11.10 T cell hybridomas. The transiently transfected cells were treated with corticosterone (1, 10, and 100 nM) and 11DHC (1, 10, and 100 nM) in the presence or absence of GA (10 μM) (11-HSD1 inhibitor) for 24 h. The cells were then collected for cytoplasmic protein extraction to quantitatively analyze for luciferase activity. Error bars represent mean ± SEM. B and C, Up-regulation of endogenous GCS-responsive genes. Purified primary CD4+ T cells were cultured in the presence of corticosterone (100 nM), 11DHC (100 nM), and ±GA (10 μM) for 5 h. Cells were then collected to perform RNA extraction and quantitative RT-PCR to analyze for an up-regulation of GILZ and IL-7R mRNA. IL-7R, IL 7R subunit. All results shown are representative of 3 separate experiments.
Recent genome-wide microarray experiments revealed that IL-7R -chain (IL-7R) and GILZ are both GCS-inducible genes in human PBMC (44, 45, 46). The GCS-induced expression of IL-7R and GILZ was also demonstrated to play a protective role against activation-induced apoptosis in lymphocytes. To determine whether bioactive GCS produced endogenously by murine CD4+ lymphocytes incubated with 11DHC were sufficient to induce the trans activation of endogenous GCS-responsive genes, we cultured freshly isolated splenic CD4+ lymphocytes with either corticosterone or 11DHC for 5 h, followed by RT-PCR analysis, to determine whether an increased expression in the above mentioned GCS-responsive genes had occurred. The treatment of CD4+ T cells with bioactive GCS resulted in an increase in mRNA expression of both IL-7R and GILZ by 2.5- and 5-fold, respectively (Fig. 4, B and C). Furthermore, treatment of CD4+ T cells with the similar quantities of 11DHC up-regulated IL-7R and GILZ mRNA expression to comparable levels. The addition of GA to cell cultures abolished the ability of 11DHC to induce an increase in mRNA expression without affecting the up-regulation of IL-7R and GILZ by corticosterone (Fig. 4, B and C). Thus, similar to human PBMC, IL-7R and GILZ proved to represent GCS-inducible genes in primary murine CD4+ T cells. Furthermore, the corticosterone produced endogenously by CD4+ T cells was sufficient to up-regulate expression levels of these two GCS-inducible genes.
Ligand activation of the GR by GCS generated by CD4+ T cells modulates immune function
The immunosuppressive and anti-inflammatory effects exerted by GCS are in part due to its ability to modulate the production of inflammatory cytokines by activated CD4+ T cells. The mechanism by which this occurs is partially mediated via the transcriptional interference by the ligand-activated GR of transcription factors such as AP-1 and NF-B (6, 7). To establish that the GR activated by GCS produced endogenously have the capacity to mediate DNA-independent protein-protein interactions, we questioned whether the activation-induced production of certain cytokines by CD4+ T cells could be modulated by the addition of 11DHC to the cell culture system. Freshly isolated splenic CD4+ T cells were activated by placing them onto culture plates coated with immobilized anti-CD3 Ab in the presence or absence of corticosterone or 11DHC. After 24 h, supernatants were collected to quantitatively analyze for the production of IL-2 and IFN- by ELISA. We found that the lymphocytes cultured with 11DHC were inhibited from producing both IL-2 and IFN- (Fig. 5A). As previously demonstrated by us and others (37), the activation-induced production of these cytokines was also inhibited by the added presence of bioactive GCS (Fig. 5A). Exposure of activated CD4+ T cells to GA alone had minor inhibitory effects on both IL-2 and IFN- production, while the coincubation of GA and the bioactive GCS did not affect the inhibited production of either cytokine. However, coincubation of CD4+ T cells with GA plus 11DHC eliminated the inhibition of cytokine production (Fig. 5A), presumably due to the inhibition of 11-HSD1 function by the added GA.
FIGURE 5. Ligand activation of the GR by GCS generated by CD4+ T cells inhibits activation-induced cytokine production. A, Purified splenic CD4+ T cells were activated with immobilized anti-CD3 Ab (1 μg/ml) in the presence or absence of corticosterone or 11DHC with the addition of GA. Supernatants were collected after 24 h to quantitatively analyze for IL-2 and IFN- production by ELISA. B, Purified splenic CD4+ T cells from C57BL/6 WT and 11-HSD1–/– mice were cultured in 1% FBS and activated with immobilized anti-CD3 and anti-CD28 Ab in the presence or absence of corticosterone or 11DHC. Supernatants were collected after 24 h to quantitatively analyze for IL-4 and IFN- production by ELISA. C, Purified primary CD4+ T cells were activated with immobilized anti-CD3 Ab (1 μg/ml) in the presence or absence of bioactive GCS (corticosterone) or the inactive metabolite (11DHC) with the addition of RU486. Supernatants were collected after 24 h to quantitatively analyze for IL-2 and IFN- production by ELISA. RU486 is a GR antagonist, GA, and an 11-HSD1 inhibitor. Results shown are mean ± SEM and are representative of 4 separate experiments.
To further demonstrate that lymphocyte 11-HSD1 is responsible for the intracellular generation of bioactive GCS from 11-keto metabolites, splenic CD4+ T cells were purified from C57BL/6 WT and 11-HSD1–/– mice, cultured in medium containing 1% FBS to minimize influences exerted by any 11-keto GCS present in serum, and activated with anti-CD3 and anti-CD28 Abs (coimmobilized on tissue culture plates) in the presence or absence of corticosterone or 11DHC. Supernatants were collected after 24 h to analyze for activation-induced cytokine production by ELISA. As shown in Fig. 5B, 11DHC suppressed the production of IL-4 and IFN- by WT CD4+ T cells, while this suppressive effect was greatly diminished in cultures of 11-HSD1–/– CD4+ T cells. Interestingly, when identical culture conditions were used, purified CD4+ T cells from 11-HSD1–/– animals possessed an augmented capacity to produce both IL-4 and IFN- when compared with WT CD4+ T cells (Fig. 5B). Collectively, these results demonstrate that murine CD4+ T cells can effectively convert 11DHC to bioactive GCS, providing a means for inactive 11-ketosteroids to function as activating ligands for the GR following prereceptor metabolism by lymphocyte 11-HSD1.
To demonstrate that the endogenous metabolism of 11DHC-mediated inhibition of cytokine production is mediated through the GR, we used RU486, a GR antagonist. Murine splenic CD4+ T cells were activated on culture plates containing immobilized anti-CD3 Ab in the presence or absence of either active corticosterone or 11DHC. Expectedly, RU486 successfully abolished the ability of the active GCS to inhibit both IL-2 and IFN- production (Fig. 5C). We found that the addition of RU486 also abrogated the ability of 11DHC metabolites to suppress cytokine production.
Activation of CD4+ T cells through the TCR and their polarization into Th1 and Th2 effector T cells elevate cellular 11-HSD1 reductase activity
The capacity of CD4+ T cells to convert the 11-keto metabolites of GCS into their bioactive forms, sufficient to suppress the activation-induced production of inflammatory cytokines, was previously unknown. This novel finding led us to further question whether cellular activation through the TCR can induce an increase in the capacity of lymphocyte 11-HSD1 to generate activating ligands for the GR and potentiate the GCS-induced suppression of inflammatory cytokine production. CD4+ T cells were isolated from the PLN as well as the spleen of normal BALB/c mice. The lymphocytes were cultured alone or activated by placing them onto tissue culture plates containing immobilized anti-CD3 Ab or plates containing coimmobilized anti-CD3 and anti-CD28. As seen in Fig. 6A, CD4+ T cells activated through the TCR were able to generate three times the amount of cortisol, when compared with the amount of cortisol generated by an equivalent number of resting CD4+ T cells for the same time period. When CD4+ T cells were activated through the TCR with costimulation, the increased ability of CD4+ T cells to reactivate cortisone into cortisol remained significant, albeit at a slightly reduced level when compared with CD4+ T cells that received an activation signal through the TCR alone.
FIGURE 6. Activation of CD4+ T cells through the TCR and their polarization into Th1 and Th2 effector cells augment their ability to reactivate 11-keto metabolites of GCS. A, Purified splenic CD4+ T cells were cultured either in a resting state or activated by placing them onto tissue culture plates containing immobilized anti-CD3 Ab (1 μg/ml) alone or with immobilized anti-CD3 and anti-CD28 (both 1 μg/ml) in the presence of [3H]cortisone. Supernatants were collected for steroid extraction and the determination of the rate of conversion of cortisone to cortisol. B, CD4+ T cells were purified from the PLN of C3H/HeN mice and activated with immobilized anti-CD3 and anti-CD28 under Th0, Th1, or Th2 polarization conditions. Th0, Th1, and Th2 lineage-committed effector cells were cultured in the presence of [3H]cortisone for an additional 24 h to evaluate their ability to convert cortisone to cortisol. C, The same Th0, Th1, and Th2 lineage-committed effector cells were lysed, and 20 μg of protein was used for Western blot analysis of 11-HSD1. Results shown are mean ± SEM of three independent experiments. *, p < 0.05.
Similar to activation through the TCR, when naive CD4+ T cells were induced to become Th1 and Th2 effector cells, following a 4- to 5-day culture period under the influence of Th1- and Th2-polarizing cytokines, their capacity to generate bioactive GCS from their 11-keto derivatives increased by 2-fold in comparison with the Th0 T cells (Fig. 6B). We were also able to establish that the increase in 11-HSD1 reductase activity correlated well with an observed increase in 11-HSD1 protein expression in Th1 and Th2 T cells (Fig. 6C).
Lymphocyte 11-HSD1 reductase activity is elevated in aged mice
The 11-HSD1 reductase activity has been reported to increase with age in human osteoblasts (47). This increase in the conversion of 11-keto metabolites to bioactive GCS via an intracrine mechanism has been proposed to be linked with the decrease in bone formation and an increased risk in GCS-induced osteoporosis (48, 49). To analyze whether 11-HSD1 activity becomes elevated in the CD4+ T cells of aged mice, purified populations of CD4+ T cells from either young (3-mo) or aged (>16-mo) BALB/c mice were cultured in the presence of [3H]cortisone. Supernatants were analyzed to evaluate whether an observable difference in 11-HSD1 reductase activity occurred as a consequence of aging. Nearly 2-fold higher levels of cortisol were found in the supernatants of CD4+ T cells from aged donors at 24 h when compared with levels of cortisol found in the supernatants of CD4+ T cells from young donors (Fig. 7A). We also found 11-HSD1 protein expression to be elevated in CD4+ T cells from aged donors (Fig. 7B). To establish that the observed increase in 11-HSD1 reductase activity of CD4+ T cells isolated from aged donors is not due to an increase in the ratio of memory/naive T cells in the aged donors, identical experiments were performed using CD4+ T cells isolated from young (3-mo) or aged (>16-mo) DO11.10 TCR transgenic mice. The results obtained were similar (data not shown). Because CD4+ T cells isolated from aged DO11.10 TCR transgenic mice retain a naive phenotype (CD45RBhighCD62Lhigh; our unpublished observation), these findings indicate that, similar to human osteoblasts, 11-HSD1 reductase activity in CD4+ T cells becomes elevated with age.
FIGURE 7. The 11-HSD1 protein expression and reductase activity are elevated in CD4+ T cells from aged donors. A, Primary CD4+ T cells were purified from the spleens of young (3-mo) or aged (>16-mo) mice and cultured in the presence of [3H]cortisone. The 11-HSD1 reductase activity is determined by the rate of conversion of cortisone to cortisol in culture. Results shown are mean ± SEM of five independent experiments. *, p < 0.005. B, Purified primary CD4+ T cells from young or aged donors were lysed to perform protein extraction. A total of 20 μg of cytoplasmic protein was used for 11-HSD1 protein expression by Western blot analysis using a polyclonal 11-HSD1 Ab. Results shown are representative of three independent experiments.
Discussion
Many steroid hormones, including testosterone, vitamin D3, and the retinoids, are produced and released into the systemic circulation as precursor molecules that possess little or no biological activity. These circulating precursors then become end-organ metabolized into their most bioactive forms by tissue- and organ-specific enzymes (50, 51, 52). In contrast, the GCS are produced and released by the adrenal gland in their bioactive forms that can subsequently be metabolized by tissue-specific 11-HSD enzymes. Although >90% of the bioactive GCS are bound to carrier proteins in the plasma, buffering the rate of their cellular entry, the 11-keto metabolites (cortisone and 11DHC) are able to freely diffuse into cells without restrictions (8). Historical reports from the 1950s exist describing the capacity of lymphocytes to generate bioactive GCS from their physiologically inactive 11-keto metabolites (53, 54). Over the next half century, with the exception of a single report describing the inducible expression of 11-HSD1 in human monocytes upon their differentiation into macrophages, the expression of GCS-metabolizing enzymes in immune cells has remained totally unexplored.
In this study, we report that 11-HSD1 mRNA and protein are expressed in murine CD4+, CD8+, and B220+ lymphocytes. The expression of mRNA and protein correlates with 11-HSD1 reductase activity, as evidenced by the exclusive conversion of cortisone to cortisol by cultured lymphocytes. We therefore hypothesized that through its exclusivity in converting inactive 11-keto GCS to their bioactive forms, 11-HSD1 could provide lymphocytes with a novel intracrine mechanism to generate activating ligands for their own intracellular GR.
To support our working hypothesis, we demonstrated that the presence of 11-keto metabolites of GCS not only up-regulated the expression of an MMTV-Luc reporter construct in transiently transfected DO11.10 T cell hybridomas, but also induced the expression of IL-7R and GILZ gene expression in primary murine CD4+ lymphocytes. Both of these genes have been shown to be inducible by dexamethasone in human PBMC (44). The up-regulated expression of IL-7R and GILZ provides protective mechanisms to lymphocytes against activation-induced apoptosis (44, 46). Expression of the IL-7R chain is crucial to the development of both T and B cells because IL-7R–/– mice have profound deficiencies in their T and B lymphocyte repertoire (55, 56). Increased IL-7R expression also exerts antiapoptotic effects on GCS-induced death in T lymphocytes (44), although the molecular mechanism through which this occurs has yet to be established. GILZ has previously been demonstrated to repress activation-induced cell death in lymphocytes via its capacity to mediate protein-protein interactions with AP-1, NF-B, and Raf-1 (57, 58). The ability of murine CD4+ T cells to up-regulate the expression of these prosurvival genes, when cultured in the presence of physiological concentrations of 11-keto GCS, suggests that lymphocytes possess an intrinsic means to control their susceptibility to apoptosis.
To further explore the possibility that intracellular reactivation of GCS provides lymphocytes with a degree of autonomous control over their viability, as well as some of their effector functions, we found that the activation-induced production of IL-2 and IFN- by CD4+ lymphocytes was depressed in cultures containing added 11-keto GCS. Because 11-keto GCS are not capable of directly activating the GR, we presumed they were being converted into the bioactive forms of GCS via intracellular 11-HSD1. In support of this hypothesis, we found that when 11-HSD1 function was inhibited by the addition of GA, or when RU486 was used to antagonize GR signaling, all changes consistent with GCS influences were abrogated. Furthermore, we demonstrated that the ability of 11-keto GCS to suppress IL-4 and IFN- production was greatly diminished in cultures of purified CD4+ T cells from 11-HSD1–/– animals. Together, these observations provide strong evidence that CD4+ lymphocytes have the endogenous capacity to reactivate 11-keto derivatives of GCS into their bioactive forms, providing activating ligands for the GR. Thus, intracellular 11-HSD1 provides CD4+ lymphocytes with an intracrine means to modulate some of their effector functions. Furthermore, the observed inhibition of activation-induced production of IL-2 may be partially due to the up-regulated expression of the GILZ gene and protein, because it has been previously demonstrated that the induction of GILZ mediates the inhibition of activation-induced IL-2 production by dexamethasone, via its transcriptional interference with AP-1 (58).
The cytokines produced by activated T cells impart regulatory influences over the development and nature of the subsequently generated immune responses. Our finding that CD4+ lymphocytes produced less inflammatory cytokines when activated in the presence of 11-keto GCS concurs with the present paradigm that GCS are able to skew the immune response toward a Th2 pathway by inhibiting the production of cytokines necessary to generate an efficient Th1 response (6, 59, 60). Furthermore, we found that the ability to convert 11-keto GCS into their bioactive form is significantly elevated in T cells that have been activated through the TCR or polarized down either a Th1 or Th2 pathway. An augmentation in the ability of Th1 T cells to generate bioactive GCS may assist in controlling the levels of Th1 cytokines, possibly to avoid generating an exaggerated inflammatory response. We and others have previously shown that physiological levels of endogenous GCS can enhance the production of Th2 cytokines by activated lymphocytes in a dose-dependent manner (37). The increase in 11-HSD1 activity in Th2 T cells supports the possibility that elevated levels of endogenous GCS might promote the development of an anti-inflammatory immune response. Thus, the increased capacity of Th1 and Th2 lineage-committed effector T cells to generate bioactive GCS could play a role in directing the progression of the immune response, away from inflammatory Th1 responses. Interestingly, we found that splenic CD4+ T cells from 11HSD1–/– mice possessed an augmented capacity to produce both IL-4 and IFN- in comparison with WT splenic CD4+ T cells in vitro. Whether this phenomenon is due to the differential capacity of activated WT and 11-HSD1–/– CD4+ T cells to produce these cytokines or to developmental differences in the CD4+ T cells between WT and 11-HSD1–/– mice is currently being explored.
Influences on immune functions exerted by GCS are not limited to their ability to shift patterns of cytokine production by activated T lymphocytes. GCS can also extend their immunomodulatory functions to other immune cell types. Increasing evidence now suggests a key role for dendritic cells (DCs) in orchestrating the lineage commitment of naive Th cells (61, 62, 63, 64). Perhaps the capacity of GCS to influence the development of immune effector responses begins with their ability to modulate particular functions of APCs. GCS are known to inhibit the up-regulation of MHC class II molecules, the costimulatory molecules CD80 and CD86, as well as adhesion molecules and chemokine receptors on the DC surface during their maturation (65, 66, 67, 68). Our unique finding that CD11c+ DCs have the capacity to generate bioactive GCS via an intracrine mechanism would suggest a possible role for GCS modulation of DC maturation in vivo.
It has been reported that Ag presentation by DCs in the absence of effective costimulation can lead to an anergy in responding lymphocytes (69, 70). Immature DCs differentially matured under the influence of intracellularly generated bioactive GCS may be unable to properly up-regulate costimulatory molecule expression during Ag presentation to lymphocytes, leading to the development of lymphocyte unresponsiveness. In addition, we have also observed a marked up-regulation in 11-HSD1 mRNA expression in immature DCs following maturation induction with LPS or following treatment with dexamethasone (our unpublished observation). An up-regulation in 11-HSD1 expression could potentially increase the intracellular as well as microenvironmental levels of bioactive GCS during Ag uptake and processing, leading to an altered maturation of the DCs during their migration and trafficking to secondary lymphoid organs.
We have observed that murine thymocytes are capable of generating active GCS from 11-keto GCS substrates in culture. This finding is particularly interesting because there now exists considerable evidence that GCS play essential roles during thymocyte development and differentiation, including the shaping of the Ag-specific T cell repertoire (6, 71). This occurs through the ability of GCS to inhibit activation-induced apoptosis of thymocytes, thus increasing the TCR signaling thresholds required to promote positive and negative selection (72, 73). Thymic epithelial cells have been reported to express steroidogenic enzymes, and the addition of inhibitors of these steroidogenic enzymes has been shown to profoundly affect the viability of the CD4+CD8+ double-positive subpopulation of thymocytes in fetal thymic cultures (74, 75). Our finding that murine neonatal and adult thymocytes readily convert inactive 11-keto GCS into bioactive species of GCS demonstrates an additional intracrine mechanism by which thymocytes themselves might exert control over endogenous GCS levels.
The capacity of thymocytes to generate bioactive GCS may be especially important during fetal development, when placental and fetal tissue 11-HSD2 is functioning to inactivate maternal GCS, preventing any deleterious effects of bioactive GCS on the developing fetus (21, 22, 76). The 11-keto GCS-rich environment present throughout most of gestation would maximize the potential for thymocyte 11-HSD1 activities. Because thymocytes from 11-HSD1–/– animals would exhibit a reduced capacity to generate endogenous bioactive GCS from their 11-keto metabolites, an expanded population of thymocytes from the genetically deficient animals should be susceptible to negative selection by activation-induced apoptosis. This would effectively eliminate developing T cells expressing TCRs having the higher avidities for self peptide/MHC complexes that would normally be spared by natural GCS influences. Hence, animals lacking 11-HSD1 might be less susceptible to autoimmune diseases when compared with WT animals.
The 11-HSD1–/– mice are resistant to age-associated declines in cognitive functions (77). This suggests that changes in 11-HSD1 activity in the brain could be partially responsible for the decline in cognitive functions as age progresses in WT animals. Recent findings with human osteoblasts suggest an increase in 11-HSD1 transcript and reductase activity occurs with aging, a finding that has been linked to the age-associated decline in bone formation as well as GCS-induced osteoporosis (47). Consistent with these findings, we found that 11-HSD1 reductase activity and protein expression are increased in CD4+ T cells from aged donors. This age-associated increase in the capacity to reactivate 11-keto GCS by CD4+ T cells may somehow be linked to the decrease in GCS responsiveness reported to occur in aged animals (78, 79). The possibility also exists that the age-associated increases in circulating levels of proinflammatory cytokines (80, 81, 82, 83, 84) are linked to increases in 11-HSD1 reductase activity because these cytokines (e.g., TNF-, IL-1) are able to increase 11-HSD1 expression as well as reductase activity in nonimmune cells (85). The stimulation of human macrophages by IL-4 or LPS also has the ability to elevate 11-HSD1 mRNA and activity (30).
Recent reports indicate that hexose-6-phosphate dehydrogenase controls the ability of 11-HSD1 to function as a reductase or as a dehydrogenase in murine adipocytes (86, 87). It may be possible that the augmented 11-HSD1 reductase activity in activated/polarized CD4+ T cells and in CD4+ T cells from aged donors is linked to an increase in the activity of hexose-6-phosphate dehydrogenase.
The current paradigm of GCS-mediated influences on lymphocyte development and physiology assumes that the plasma GCS provide the sole source of bioactive steroid. We have discovered and demonstrated the novel capacity of lymphocytes to generate bioactive GCS from circulating inactive and bioavailable 11-keto metabolites, providing a previously unappreciated source of bioactive GCS for modulating immune cell function. A careful examination of T cell development and inducible immune responses by 11-HSD1–/– mice should provide valuable insight into the physiological significance of this intracrine-controlling mechanism in lymphocyte physiology.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grants CA25917 and DK55491, a Browning Foundation grant, and Department of Veterans Affairs medical research funds. T.Y.Z. is supported by National Institutes of Health, Department of Health and Human Services, National Institute of Diabetes and Digestive and Kidney Diseases Hematology Research Training Grant T32 DK07115.
2 Address correspondence and reprint requests to Dr. Raymond A. Daynes, Department of Pathology, University of Utah School of Medicine, 30 North 1900 East, Salt Lake City, UT 84132-2501. E-mail address: daynes.office@path.utah.edu
3 Abbreviations used in this paper: GCS, glucocorticoid; DC, dendritic cell; BMDC, bone marrow-derived DC; 11DHC, 11-dehydrocorticosterone; GA, glycerrhetinic acid; GILZ, GCS-induced leucine zipper; GR, GCS receptor; GRE, GCS response element; 11-HSD, 11-hydroxysteroid dehydrogenase; MMTV, mouse mammary tumor virus; PLN, peripheral lymph node; WT, wild type.
Received for publication May 25, 2004. Accepted for publication November 2, 2004.
References
Rook, G. A.. 1999. Glucocorticoids and immune function. Baillieres Best Pract. Res. Clin. Endocrinol. Metab. 13:567.
Webster, J. I., L. Tonelli, E. M. Sternberg. 2002. Neuroendocrine regulation of immunity. Annu. Rev. Immunol. 20:125
Sternberg, E. M.. 2001. Neuroendocrine regulation of autoimmune/inflammatory disease. J. Endocrinol. 169:429.
Pitzalis, C., N. Pipitone, M. Perretti. 2002. Regulation of leukocyte-endothelial interactions by glucocorticoids. Ann. NY Acad. Sci. 966:108.
Cancedda, C., G. Filaci, F. Puppo, M. Ghio, P. Contini, F. Indiveri. 2002. Immune homeostasis requires several biologic factors including glucocorticoid hormones. Ann. NY Acad. Sci. 966:49.
Ashwell, J. D., F. W. Lu, M. S. Vacchio. 2000. Glucocorticoids in T cell development and function. Annu. Rev. Immunol. 18:309.
Refojo, D., A. C. Liberman, F. Holsboer, E. Arzt. 2001. Transcription factor-mediated molecular mechanisms involved in the functional cross-talk between cytokines and glucocorticoids. Immunol. Cell Biol. 79:385.
Siiteri, P. K., J. T. Murai, G. L. Hammond, J. A. Nisker, W. J. Raymoure, R. W. Kuhn. 1982. The serum transport of steroid hormones. Recent Prog. Horm. Res. 38:457.
Rosner, W.. 1991. Plasma steroid-binding proteins. Endocrinol. Metab. Clin. North Am. 20:697.
Mangelsdorf, D. J., C. Thummel, M. Beato, P. Herrlich, G. Schutz, K. Umesono, B. Blumberg, P. Kastner, M. Mark, P. Chambon, et al 1995. The nuclear receptor superfamily: the second decade. Cell 83:835.
Reichardt, H. M., G. Schutz. 1998. Glucocorticoid signalling: multiple variations of a common theme. Mol. Cell. Endocrinol. 146:1.(Tian Y. Zhang, Xiaohong D)
The 11-hydroxysteroid dehydrogenase (11-HSD) enzymes control the interconversion of active glucocorticoids (GCS) and their inactive 11-keto metabolites, a process commonly referred to as the cortisone/cortisol shuttle. Although the prereceptor metabolism of GCS by 11-HSD is well documented in a variety of cells and tissues, it has not yet been carefully investigated in the major cell types of the immune system. In this study, we demonstrate that 11-HSD1 transcripts, protein, and enzyme activities are actively expressed in murine CD4+, CD8+, and B220+ lymphocytes, as well as CD11c+ dendritic cells. Only reductase activity was observed in living cells, evidenced by the restricted conversion of cortisone to cortisol. Activation of CD4+ T cells increased their 11-HSD1 activity, as did their polarization into Th1 or Th2 cells. CD4+ T cells isolated from aged donors (>16 mo) had increased 11-HSD1 protein and an elevated capacity to convert cortisone to cortisol. The GCS generated in murine CD4+ T cells from their inactive 11-keto metabolites could activate the GCS receptor, demonstrated by an up-regulation of IL-7R and GCS-induced leucine zipper gene expression. The presence of a functional 11-HSD1 provides lymphocytes with a novel intracrine regulatory mechanism that could influence such processes as lymphocyte development, effector function, and susceptibility to apoptosis. Thus, the presence of 11-HSD1 provides an additional means to facilitate GCS influences over lymphocyte activities, uncoupled from the plasma concentration of GCS.
Introduction
As major steroid hormones produced by the adrenal gland, glucocorticoids (GCS)3 are now appreciated to represent essential regulators of the development, homeostasis, and effector functions of the innate and adaptive immune system. GCS can also alter immune cell trafficking through effects on vascular permeability, chemotaxis, and adhesion molecule expression (1, 2, 3, 4, 5). At a cellular level, GCS alter the activation, differentiation, and maturation of most immune cell types, as well as play multiple roles in regulating immune cell sensitivity to apoptosis (6, 7). Some of these effects are accomplished through GCS effects on cytokine and chemokine production, while others are mediated through the ability of GCS to modulate the outcome of numerous cell signaling systems (6, 7).
The potent and pleiotropic effects of GCS on gene expression and the functional properties of numerous transcription factor activities are paralleled by the presence of complex regulatory mechanisms that provide control over their diverse array of cellular influences. One of these controlling mechanisms operates at the level of endogenous GCS production by the adrenal gland, the hypothalamic-pituitary-adrenal axis, which is influenced by the diurnal cycle, organismal stress, and promoters of inflammatory responses (2, 3).
The majority of the bioactive GCS (cortisol and corticosterone in humans, corticosterone only in rodents) in the systemic circulation is found physically associated with albumin or corticosteroid-binding globulin (8, 9). The physical association of GCS with these plasma proteins provides a second means of control over GCS effects by effectively buffering the rate of cellular entry (8, 9). Interestingly, cortisone and 11-dehydrocorticosterone (11DHC), the biologically inactive 11-keto forms of naturally produced GCS, are not found in the plasma in a protein-associated state (8). Consequently, the 11-keto metabolites of GCS are able to freely enter target cells without restrictions.
Intracellularly, the GCS serve as activating ligands for the GCS receptor (GR), a member of the nuclear hormone receptor superfamily of ligand-activated transcription factors (10). Ligand binding of the GR promotes its dissociation from heat shock protein 90, followed by its nuclear translocation, homodimerization, and physical association to specific cis-acting GCS response elements (GREs) of GCS-controlled genes, thereby conferring specific control over their transcription (11, 12). Although GRs are ubiquitously expressed in nucleated cells, a dynamic modulation over their cellular expression has been shown in a variety of cell types (13, 14, 15, 16, 17). The ligand-activated GR can also modulate gene expression indirectly, through protein-protein interactions with numerous transcription factors (e.g., AP-1, NF-B) whose activities are suppressed from such associations (11).
The 11-hydroxysteroid dehydrogenases (11-HSDs) are enzymes that interconvert the functional GCS with their 11-keto metabolites, thereby increasing or decreasing the intracellular levels of bioactive GCS available as activating ligands for the GR. The 11-HSDs are members of the short-chain alcohol dehydrogenase family of enzymes, and two distinct isoforms have been identified and thoroughly investigated to date: 11-HSD type I (11-HSD1) and 11-HSD type II (11-HSD2). Using NAD+ as cofactor to carry out its sole function as a steroid dehydrogenase, 11-HSD2 inactivates bioactive GCS, generating 11-keto metabolites that are incapable of binding or activating the GR (18, 19). In adults, the majority of 11-HSD2 expression is anatomically confined to major mineralcorticoid target tissues (e.g., kidney) (19). During gestation, however, 11-HSD2 is highly expressed in the placenta and fetal tissues, where it functions to inactivate maternal GCS and provide protection of the fetus against the deleterious effects of GCS (20, 21, 22, 23). In contrast, 11-HSD1 exclusively uses NADP(H) as a cofactor and functions predominantly as a reductase in vivo, regenerating bioactive GCS from the circulating 11-keto metabolites (24). Cellular expression of 11-HSD1 is widespread, with the highest levels found in the liver. Thus, at the cellular level, the 11-HSDs provide control over the prereceptor metabolism of GCS, a process commonly referred to as the cortisone/cortisol shuttle.
The physiological importance of prereceptor metabolism of GCS is well documented in many cell and tissue types (25, 26, 27, 28, 29). However, with only a single exception, the prereceptor metabolism of GCS in immune cells has not yet been investigated (30). This publication reported 11-HSD1 expression in monocytes that are induced to mature into macrophages (30).
In the original report that described the cloning and tissue distribution of murine 11-HSD1, the thymus was determined to express low levels of 11-HSD1 mRNA by Northern blot analysis (31). However, this study failed to establish whether the stromal elements of the thymus, thymocytes, or associated tissue macrophages expressed 11-HSD1 transcripts (31).
In this study, we describe the presence of 11-HSD1 in cells of the immune system. Our studies demonstrate that 11-HSD1 mRNA, protein, and enzyme reductase activity are present in murine CD4+ and CD8+ T cells, B220+ B cells, as well as CD11c+ bone marrow-derived dendritic cells (BMDC). Furthermore, the 11-HSD1 in murine lymphocytes was sufficiently active to provide an intracrine mechanism, allowing bioactive GCS produced endogenously to exert control over molecular immune processes. We raise the possibility that an intracellular reactivation of GCS in lymphocytes could provide an autonomous means to modulate individual immune effector functions as well as influence the immediate microenvironment.
Materials and Methods
Experimental animals
BALB/c, C57BL/6, and C3H/HeN mice were purchased from Charles River Laboratories; DO11.10 TCR transgenic mice on a BALB/c background were originally obtained from The Jackson Laboratory and bred in-house. The 11HSD1–/– mice on a C57BL/6 background were obtained from J. Seckl and J. Mullins from the University of Edinburgh (Edinburgh, Scotland), and bred in-house. Female animals, 10–12 wk of age, were used for all experiments described in this work, except in studies in which lymphocytes from 11-HSD1–/– mice were being analyzed. All mice were housed in a specific pathogen-free barrier facility at the University of Utah Animal Resource Center. This facility guarantees strict compliance with regulations established by the Animal Welfare Act. All animals were housed with a 12-h light-dark controlled cycle, and were provided with mouse chow and water ad libitum. At time of sacrifice, mice were anesthetized with halothane, followed by cervical dislocation.
Culture conditions and reagents
Murine primary lymphocytes were isolated from the spleens of normal mice. They were cultured in complete medium (RPMI 1640 (Invitrogen Life Technologies) supplemented with 10% FCS (HyClone), 2 mM L-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, and 50 μM 2-ME), except where otherwise noted. Murine CD11c+ BMDC were generated, as described previously (32). The murine EL-4 T cell line was purchased from American Type Culture Collection. The murine T cell hybridoma, DO11.10, was obtained from J. Woodward (University of Kentucky, Lexington, KY). Glycerrhetinic acid (GA) and all GCS (cortisone, hydrocortisone, corticosterone, and 11DHC) were purchased from Sigma-Aldrich. RU486 was a gift from M. Mitchell (University of Utah). [3H]Cortisone and [3H]cortisol (sp. act., 50 Ci/mmol) were purchased from American Radiolabeled Chemicals. FITC-VAD-FMK was purchased from Promega. Recombinant murine IL-4 and IFN- were purchased from BD Pharmingen; anti-IL-4 (11B11) and anti-IFN- (XMG1.2) were generated in-house. All Abs used for ELISA, activation of T cells, specific cell population enrichments, and FACS analysis were purchased from BD Pharmingen. Coating of tissue culture plates with immobilized anti-CD3 and anti-CD28 Ab was done by incubating plates with anti-CD3 and anti-CD28 Ab at 1–2 μg/ml in Tris-HCl buffer (pH 9.6) at 37°C for 2 h, followed by extensive washing.
Dynabead cell enrichment
For the preparation of CD4+ lymphocytes, freshly isolated spleens were dissociated into a single cell suspension in complete medium. The erythrocytes were lysed by brief treatment with sterile aqueous 0.83% (w/v) ammonium chloride. For enrichment of CD4+ T cells, erythrocyte-depleted splenocytes were incubated with 2 μg/ml each of biotinylated anti-CD45R/B220, anti-CD11b (also known as Mac-1+), and anti-CD8 Abs (BD Pharmingen) for 30 min on ice. Following washes, the cells were resuspended in medium, and streptavidin-coated M-280 magnetic Dynabeads were added (Dynal Biotech) at a bead:cell ratio of 1:1 for a total of 20 min with constant agitation at 4°C. The labeled cells were removed by placing the cells in a magnetic field, and residual cells were separated and washed for use in cell culture or Western blot analysis. The purification of CD8+, B220+, and Mac-1+ cells by negative selection was done in a similar manner. For the enrichment of CD8+ cells, biotinylated anti-CD45R/B220, anti-CD11b, and anti-CD4 Abs were used; for the enrichment of B220+ cells, biotinylated anti-CD4, anti-CD8, and anti-CD11b Abs were used; for the enrichment of Mac-1+ cells, biotinylated anti-CD4, anti-CD8, and anti-B220 Abs were used. Purity of the cells was determined by FACS analysis with FITC anti-mouse CD Abs. The level of purity of negatively selected cells was routinely >90%.
Quantitative PCR
Total RNA was extracted by the method of Chomczynski and Sacchi (33). cDNA was reverse transcribed from RNA, as previously described (34). Quantitative PCR was performed using the Roche Light Cycler, as recommended by the manufacturer. The following primer sets and PCR conditions were used for the amplification of the gene of interest. To initiate the PCR, all samples were denatured at 95°C for 30 s. The murine GAPDH sense 5'-AGT ATG TCG TGG AGT CTA C-3', antisense 5'-CAT ACT TGG CAG GTT TCT C-3', denatured samples were allowed to anneal at 58°C for 10 s, followed by 20 s of extension at 74°C; murine 11-HSD1 sense 5'-AAC CAC ATC ACT CAG ACC-3', antisense 5'-CAC CAT TAG AAC AGA ACT C-3', denatured samples were allow to anneal at 67°C for 5 s, followed by 10 s of extension at 74°C; murine 11-HSD2 sense 5'-GTA TCA AGG TCA GCA TTA TC-3', antisense 5'-GAA GGT GAT TGA TAA AGA AG-3', denatured samples were allowed to anneal at 61°C for 5 s, followed by 14 s of extension at 74°C; murine IL-7R sense 5'-CCA TT TGA GTT TGT TCT CTG-3', antisense 5'-GTC CCT GTG TCT CCA ACT CC-3', denatured samples were allowed to anneal at 62°C for 0 s, followed by 12 s of extension at 74°C; murine GCS-induced leucine zipper (GILZ) sense 5'-AGC AGC CAC TCA AAC CAG C-3', antisense 5' GCC CTA GAC AAC AAG ATT 3', denatured samples were allowed to anneal at 60°C for 0 s, followed by 10 s of extension at 74°C. All PCR were allowed 40 cycles of amplification, which ended with a melting program that spanned the temperature range of 40–95°C with a temperature transition rate of 0.20°C/s. Specificity of the amplification product was determined using the melting temperature (at which all products melted) established during optimization of the gene-specific primer sets.
Luciferase reporter assay
The mouse mammary tumor virus (MMTV)-Luc reporter construct under the control of GREs was obtained from S. Andrews (University of Iowa, Iowa City, IA). Transient transfections were performed, as described in detail previously (35). Briefly, 5 x 106 murine DO11.10 hybridoma T cells were resuspended in 450 μl of RPMI 1640 and placed in 0.4-cm Gene Pulser cuvettes (Bio-Rad) with 2.5 μg of plasmid. Electroporation was conducted at 960 μF and 280 V. Cells were then rested for 5 min and transferred to 100-mm tissue culture plates containing 10 ml of culture medium. Use of the transiently transfected cells in experiments was done within 48 h.
ELISA
Primary murine CD4+ T cells at 5 x 106/ml were placed in multiwell cell culture plates previously coated with either 1 μg/ml anti-CD3 Ab alone, or plates previously coated with anti-CD3 as well as anti-CD28 (both at 1 μg/ml). The cells were cultured at 37°C in an atmosphere of 5% CO2 in air, and all supernatants were collected after 24 h to quantitatively evaluate the levels of IL-2, IFN-, and IL-4 by ELISA, as described previously (36).
The 11-HSD1 Western blot
The DO11.10 T cell hybridomas, the EL-4 T cell thymoma, CD11c+ BMDC, as well as freshly isolated primary murine CD4+, CD8+, and B220+ cells were washed twice with ice-cold PBS and resuspended in lysis buffer (10 mM HEPES, pH 7.8, 0.1 mM EDTA, 10 mM NaCl, 3 mM MgCl2, 300 mM sucrose, 1% Nonidet P-40, and a mixture of protease inhibitors (Roche)). The lysed cells were subjected to one freeze/thaw cycle before centrifugation at 10,600 x g for 10 min to remove nuclear debris. The supernatant was collected and protein content was determined using the bicinchoninic acid Protein Assay kit (Pierce), following the manufacturer’s recommendations. Equal amounts of protein were subjected to electrophoresis on 12% SDS-PAGE gels and transferred to nitrocellulose or polyvinylidene difluoride membranes. After blocking with 5% nonfat milk in TBS, blots were incubated with anti-11-HSD1 Ab (1.5 μg/ml) (American Diagnostica) for 2 h at room temperature. Membranes were then washed and incubated with goat anti-rabbit HRP conjugate (Bio-Rad) for 1 h at room temperature. After extensive washing, bands were visualized using the ECL kit (Amersham Biosciences), according to manufacturer’s instructions.
The 11-HSD1 immunofluorescence
Primary murine CD4+ lymphocytes from C57BL/6 wild-type (WT) and 11-HSD1–/– animals were purified using methods described above. Lymphocytes were then adhered to coverslips coated with poly(L-lysine) solution (0.1% w/v; Sigma-Aldrich) and fixed with 4% paraformaldehyde in TBS. Fixed lymphocytes were then permeabilized with 0.5% Triton X-100 in TBS for 10 min. Coverslips were then blocked with blocking buffer (3% nonfat dry milk, 0.5% Triton X-100 in TBS) at room temperature for 1 h before incubation with anti-11-HSD1 Ab (10 μg/ml in blocking buffer). The coverslips were then washed twice with 0.5% Triton X-100 in TBS, blocked for an additional 30 min, and incubated with Alexa488 (1:500 in blocking buffer) in the dark at room temperature for 1 h. The coverslips were then washed twice, mounted in FluorSave antifade mounting medium (Calbiochem), and dried at room temperature overnight. Samples were analyzed with identical parameters using a FV300 confocal microscope (x60 oil objective with NA 1.4).
The 11-HSD enzymatic activity
The 11-HSD activity was determined in intact cells by measuring the interconversion of cortisone and cortisol. To determine reductase activity, 100 nM cortisone with a tracer amount of [3H]cortisone was added to cell cultures containing cells at a density of 106 cells/ml. At the appropriate time points, steroids were extracted from the supernatant by adding 2 vol of cold ethyl acetate. The organic phase was collected, evaporated to dryness under a stream of nitrogen, and reconstituted with 50 μl of a 4 mg/ml solution containing equal amounts of cortisone and cortisol in chloroform/methanol (9:1). The extract was then spotted onto a TLC plate (Analtech) and developed using chloroform/methanol (9:1) as the mobile phase. The cortisone and cortisol spots were then visualized with iodine vapor, scraped from the TLC plate, and transferred to scintillation vials to determine the ratio of [3H]cortisone and [3H]cortisol. Percent conversion was calculated as (cpm [3H]cortisol/(cpm [3H]cortisol + cpm [3H]cortisone)) x 100.
Th1 and Th2 polarization
CD4+ T cells were isolated from the peripheral lymph nodes (PLN) (axillary, cervical, inguinal, and popliteal) of C3H/HeN mice or the PLN and spleen of DO11.10 TCR transgenic mice by negative selection using the CD4+ T cell selection kit from Miltenyi Biotec, according to the manufacturer’s instructions. Purified populations of CD4+ T cells are then activated by placing them onto tissue culture plates containing coimmobilized anti-CD3 and anti-CD28 (both 2 μg/ml) at a density of 2.5 x 106 - 5.0 x 106/ml in the presence of different polarizing cytokines for a period of 4–5 days to generate Th0, Th1, and Th2 cells. Th0 cultures contained no additional cytokines or Abs, Th1 cultures contained IL-12 (5 ng/ml) and anti-IL-4 (10 μg/ml), and Th2 cultures contained IL-4 (10 ng/ml) and anti-IFN- (10 μg/ml). At the end of the culture period, cells were harvested, washed three times, and resuspended in complete medium at a density of 106 cells/ml for further use. Detection of 11-HSD1 protein and reductase activity was performed, as described above. At the same time, a fraction of the Th0, Th1, and Th2 cells was reactivated with plate-bound anti-CD3 (1 μg/ml) for 24 h, and the supernatant was collected to perform ELISA to determine the pattern of cytokines produced. Upon reactivation, the Th0 cells produced both IFN- and IL-4, Th1 cells produced IFN- and no IL-4, and Th2 cells produced IL-4 and no IFN- (data not shown).
Apoptosis assay
The DO11.10 T cell hybridoma in log-phase growth was cultured at the density of 106 cells/ml. A total of 10–6 M and 10–7 M cortisone or cortisol was added to the cell culture at the predetermined time points backward, starting with the 24-h time points. This manner of treatment allowed all cells to be analyzed by FACS at the same time. At the end of the culture period, all cells were collected by centrifugation at 1500 rpm for 5 min and resuspended in 200 μl of 10 μM FITC-VAD-FMK in PBS, followed by staining for 20 min at 37°C and 5% CO2. After staining, all cells were washed twice with PBS, resuspended in 1 ml of cold PBS, and analyzed by FACS.
Statistical analysis
Statistical analysis was performed with the StatView 5.0 software. The p values were determined by Student’s t test (unpaired).
Results
The 11-HSD1 mRNA and protein are expressed in primary murine immune cells and cell lines
We previously reported that 11-HSD1 is expressed in murine spleen by mRNA analysis (37). With the exception of a single report describing the induction of 11-HSD1 in human monocytes upon differentiation to macrophages (30), the expression of 11-HSD1 in immune cell types has not been characterized. To investigate comprehensively whether 11-HSD1 is expressed in immune cells, we used quantitative RT-PCR to look for the presence of 11-HSD1 transcripts in freshly isolated murine splenic CD4+ and CD8+ T cells, B220+ B cells, as well as Mac-1+ cells. All purified populations analyzed were found to express 11-HSD1 transcripts (Fig. 1A). Mac-1+ cells were found to express the highest level of 11-HSD1 mRNA (Fig. 1A). In lymphocytes, CD4+ and CD8+ T cells consistently expressed higher levels of 11-HSD1 mRNA than B220+ B cells. The murine T cell hybridoma DO11.10 and the EL-4 T cell thymoma were also found to express 11-HSD1 mRNA by quantitative PCR (data not shown), further suggesting that murine lymphocytes, and not a minor cellular contaminant, were responsible for the 11-HSD1 mRNA being observed. In contrast, 11-HSD2 transcripts were undetectable by RT-PCR in these same populations of cells (data not shown). We next questioned whether the presence of 11-HSD1 mRNA correlated with the presence of 11-HSD1 protein. Using a polyclonal Ab specific for murine 11-HSD1, a single band in the expected m.w. region for 11-HSD1 was detected in all of the cell types investigated with the addition of CD11c+ BMDC (Fig. 1B). The single band recognized by the polyclonal Ab was demonstrated to be specific for murine 11-HSD1, as the same band was not present when lysates from purified 11-HSD1–/– CD4+, CD8+, and B220+ lymphocytes were analyzed in parallel. Similarly, the EL-4 thymoma cell line and the DO11.10 T cell hybridoma also demonstrated expression of 11-HSD1 protein by Western blot (data not shown).
FIGURE 1. The 11-HSD1 mRNA and protein are expressed in murine lymphocytes. A, 11-HSD1 transcript levels in primary murine CD4+, CD8+ T cells, B220+ B cells, and Mac-1+ macrophages were analyzed by quantitative RT-PCR. Results were normalized to GAPDH and shown as fold difference relative to levels found in B220+ B cells. B, Western blot analysis of 11-HSD1 protein expression in primary WT and 11-HSD1–/– CD4+, CD8+, and B220+ lymphocytes, as well as CD11c+ BMDC. Bands shown correlate with molecular mass of 36 kDa. C, Cytoplasmic localization of endogenous 11-HSD1 by immunofluorescence. Purified CD4+ T cells from C57BL/6 WT and 11-HSD1–/– mice were fixed, permeabilized, and immunostained for 11-HSD1. All results shown are representative of 3 independent experiments.
The 11-HSD1 has been proposed to be a transmembrane protein associated with the endoplasmic reticulum (38). Using the 11-HSD1 polyclonal Ab for immunostaining, we were able to detect the presence of 11-HSD1 peripheral to the nucleus in primary CD4+ T cells isolated from WT C57BL/6 mice (Fig. 1C). However, the same 11-HSD1-specific staining was not observed in primary CD4+ T cells isolated from 11-HSD1–/– animals (Fig. 1C).
The 11-HSD1 reductase activity is present in primary murine immune cells and T cell lines
To establish the presence of 11-HSD1 reductase activity inside of murine lymphocytes, we used a whole cell assay to analyze the conversion of the biologically inactive 11-keto metabolite of GCS (cortisone) to the active GCS (cortisol) in cultures of purified murine lymphocytes. Freshly isolated CD4+, CD8+, and B220+ lymphocytes were cultured separately in the presence of 100 nM cortisone containing a trace of [3H]cortisone. Supernatants were collected at various time points and analyzed for the conversion of cortisone to cortisol. Reductase activity was detected as determined by the presence of [3H]cortisol in the supernatant of CD4+, CD8+ T cells, as well as the supernatant of the B220+ B cells at all time points examined (Fig. 2A). No conversion of cortisone to cortisol above background levels was detected in medium alone (data not shown). Of the lymphocyte populations, CD4+ and CD8+ T cells showed similar levels of reductase activity, albeit the level of reductase activity was found to decrease significantly (2-fold) over a 24-h time period. In comparison, the B220+ B cell population exhibited a reduced level of reductase activity over a 24-h time period, consistent with the lowered 11-HSD1 transcript level found in B220+ B cells (Fig. 1A). We also found that CD11c+ BMDC possessed higher levels of 11-HSD1 reductase activity when compared with primary lymphocytes (Fig. 2B). The 11-HSD1 reductase activity in BMDC was inhibited by the presence of GA, a known inhibitor of the 11-HSD enzymes (Fig. 2B). Similar to what others have observed (27), we found that 11-HSD1 sp. act. was reduced at the higher concentrations of cortisone used (100 nM) (Fig. 2C). Purified populations of the murine T cell thymoma EL-4 and the murine T cell hybridoma DO11.10 were also found to have the capacity to generate active GCS from their 11-keto metabolites (Fig. 2D), eliminating the possibility of contaminating nonlymphocytes being responsible for the observed enzyme activities.
FIGURE 2. The 11-HSD1 reductase activity is expressed in primary murine immune cells and cell lines. A, Primary CD4+, CD8+, and B220+ T cells were cultured separately in the presence of [3H]cortisone. Supernatants were collected at the indicated time points for steroid extraction, and determination of 11-HSD1 reductase activity evidenced by the conversion of cortisone to cortisol. Results were calculated and shown as fmol of cortisol produced/h/500,000 cells. B, CD11c+ DCs were cultured in the presence of [3H]cortisone in the presence or absence of GA (10 μM). Supernatants were collected to analyze for the conversion of cortisone to cortisol. Results were calculated and shown as fmol of cortisol produced/h/500,000 cells. C, Primary CD4+ T cells were cultured in the presence of increasing concentrations of [3H]cortisone. Supernatants were collected to analyze for the conversion of cortisone to cortisol. Results were calculated and shown as fmol of cortisol produced/h/500,000 cells. D, The murine T cell hybridoma DO11.10 and murine thymoma EL-4 cells were cultured in the presence of [3H]cortisone. Conversion of cortisone to cortisol at the indicated time points was determined by analyzing the supernatants, and is expressed as fmol of cortisol converted/500,000 cells over time. Results shown are mean ± SEM. E, The DO11.10 T cell hybridoma was cultured alone or in the presence of cortisone or cortisol for the indicated time periods. Cells were then stained with FITC-VAD-FMK as an indicator of apoptosis and analyzed by FACS. Results are shown as percentage of cells positive for FITC-VAD-FMK. All experiments were repeated three times with similar results.
It has been previously established that T cell hybridoma lines are susceptible to GCS-induced apoptosis (39), we therefore questioned whether the capacity to generate active GCS (cortisol) from inactive 11-keto metabolites (cortisone) affected the sensitivity of the DO11.10 T cell hybridoma to GCS-induced apoptosis in the presence of cortisone. Log-phase DO11.10 T cell hybridoma cells were cultured in the absence or presence of cortisone or cortisol for various times, followed by staining with FITC-VAD-FMK to determine the percentage of cells undergoing apoptosis. Shown in Fig. 2E, the presence of added cortisol induced significant levels of apoptosis by 24 h in the DO11.10 T cell hybridoma. Similar to what others have observed (39), cortisone addition did not induce a significant level of apoptosis in these same cells by 24 h. These data, coupled with results obtained from Fig. 2D, suggest that even though the DO11.10 T cell hybriboma has the capacity to generate cortisol from cortisone in culture, the amount of active GCS produced over a 24-h time period was insufficient to induce significant levels of apoptosis in this cell line.
An increasing number of reports indicates that 11-HSD1 can function as a dehydrogenase in vitro (25). In intact cells and in vivo, however, 11-HSD1 functions predominantly as an oxidoreductase (40). To determine the directionality of 11-HSD1 enzyme activity in intact murine lymphocytes, [3H]cortisol was added to cultures of CD4+ T cells to question the presence of dehydrogenase activity. As evidenced by the complete lack of conversion of cortisol to cortisone, no dehydrogenase activity above background levels was detected (Fig. 3A). The lack of any dehydrogenase activity further supports the observation that 11-HSD2 transcripts were not detected in CD4+ T cells.
FIGURE 3. The 11-HSD1 is the sole reductase in murine lymphocytes and functions exclusively as a reductase. A, Purified primary CD4+ T cells were cultured in the presence of increasing concentrations of cortisol. No dehydrogenase activity was detected, as determined by the lack of cortisone in the supernatant. B, Thymocytes and purified populations of CD4+ T cells from C57BL/6 and 11-HSD1–/– animals were cultured in the presence of [3H]cortisone. The 11-HSD1 reductase activity is determined by the rate of conversion of cortisone to cortisol. Results were calculated and shown as fmol of cortisol produced/h/500,000 cells. Results shown are representative of 3 independent experiments.
Intrigued by the observation that the induction of 11-HSD1 expression was concomitant to the differentiation of immature monocytes into mature macrophages (30), we questioned whether the capacity to generate bioactive GCS from its 11-keto metabolite was restricted to mature lymphocytes in the periphery. When thymocytes were isolated from the thymi of day 3 neonates (which contains >95% CD4–CD8– and CD4+CD8+ immature T cells) or young adult mice, and cultured in the presence of [3H] cortisone, 11-HSD1 reductase activity was indicated by the presence of [3H]cortisol in the supernatant (data not shown). Thus, murine thymocytes as well as mature lymphocytes possess the intrinsic ability to generate bioactive GCS from 11-keto metabolites.
Finally, to establish that 11-HSD1 is responsible for the conversion of cortisone to cortisol by lymphocytes, thymocytes and CD4+ T cells were isolated from both WT and 11-HSD1–/– mice and cultured in the presence of radiolabeled cortisone. Reductase activity in lymphocytes was 13-fold less in cells from the 11-HSD1-deficient animals when compared with WT (Fig. 3B). Similarly, thymocytes isolated from 11-HSD1–/– mice also showed a markedly diminished capacity to generate bioactive GCS from 11-keto metabolites (Fig. 3B). Taken together, this demonstrates 11-HSD1 to be the major GCS reductase in murine CD4+ lymphocytes.
GCS generated from inactive 11-keto metabolites in CD4+ lymphocytes are sufficient to function as activating ligands for the GR
Nuclear hormone receptors, such as the GR, require ligand activation to promote the recruitment of the transcription machinery necessary for trans activation of GRE-containing genes (10). To specifically demonstrate that immunomodulatory quantities of active GCS can be generated by murine CD4+ lymphocytes, an MMTV-Luc reporter construct was transiently transfected into the DO11.10 T cell hybridoma. The MMTV contains well-characterized GREs in its promoter region and is commonly used as a reporter to establish GCS responsiveness (41, 42, 43). The transiently transfected DO11.10 T cell hybridomas were treated with either bioactive GCS (corticosterone) or its 11-keto metabolite (11DHC) for a period of 24 h. These cells were then collected for cytoplasmic protein extraction to quantitatively analyze luciferase activity produced as the end result of the trans activation of the MMTV-Luc reporter construct. As demonstrated in Fig. 4A, exposure to active GCS induced a dose-dependent increase in the relative luciferase activity. Interestingly, exposure to 11-keto metabolites induced a similar dose-dependent increase in relative luciferase activity, confirming the ability of DO11.10 T cell hybridomas to generate active GCS. The coincubation of MMTV-Luc+, DO11.10 cells with 11DHC plus the 11-HSD1 inhibitor GA abrogated the ability of bioactive GCS to be generated, evidenced by a lack of up-regulated luciferase activity (Fig. 4A). GA addition to parallel cultures containing bioactive GCS stimulated the up-regulation of luciferase activity at levels similar to the positive control.
FIGURE 4. GCS generated from inactive 11-keto metabolites by CD4+ lymphocytes function as activating ligands for the GR. A, An MMTV-luciferase reporter construct containing GREs in the promoter region was transiently transfected into DO11.10 T cell hybridomas. The transiently transfected cells were treated with corticosterone (1, 10, and 100 nM) and 11DHC (1, 10, and 100 nM) in the presence or absence of GA (10 μM) (11-HSD1 inhibitor) for 24 h. The cells were then collected for cytoplasmic protein extraction to quantitatively analyze for luciferase activity. Error bars represent mean ± SEM. B and C, Up-regulation of endogenous GCS-responsive genes. Purified primary CD4+ T cells were cultured in the presence of corticosterone (100 nM), 11DHC (100 nM), and ±GA (10 μM) for 5 h. Cells were then collected to perform RNA extraction and quantitative RT-PCR to analyze for an up-regulation of GILZ and IL-7R mRNA. IL-7R, IL 7R subunit. All results shown are representative of 3 separate experiments.
Recent genome-wide microarray experiments revealed that IL-7R -chain (IL-7R) and GILZ are both GCS-inducible genes in human PBMC (44, 45, 46). The GCS-induced expression of IL-7R and GILZ was also demonstrated to play a protective role against activation-induced apoptosis in lymphocytes. To determine whether bioactive GCS produced endogenously by murine CD4+ lymphocytes incubated with 11DHC were sufficient to induce the trans activation of endogenous GCS-responsive genes, we cultured freshly isolated splenic CD4+ lymphocytes with either corticosterone or 11DHC for 5 h, followed by RT-PCR analysis, to determine whether an increased expression in the above mentioned GCS-responsive genes had occurred. The treatment of CD4+ T cells with bioactive GCS resulted in an increase in mRNA expression of both IL-7R and GILZ by 2.5- and 5-fold, respectively (Fig. 4, B and C). Furthermore, treatment of CD4+ T cells with the similar quantities of 11DHC up-regulated IL-7R and GILZ mRNA expression to comparable levels. The addition of GA to cell cultures abolished the ability of 11DHC to induce an increase in mRNA expression without affecting the up-regulation of IL-7R and GILZ by corticosterone (Fig. 4, B and C). Thus, similar to human PBMC, IL-7R and GILZ proved to represent GCS-inducible genes in primary murine CD4+ T cells. Furthermore, the corticosterone produced endogenously by CD4+ T cells was sufficient to up-regulate expression levels of these two GCS-inducible genes.
Ligand activation of the GR by GCS generated by CD4+ T cells modulates immune function
The immunosuppressive and anti-inflammatory effects exerted by GCS are in part due to its ability to modulate the production of inflammatory cytokines by activated CD4+ T cells. The mechanism by which this occurs is partially mediated via the transcriptional interference by the ligand-activated GR of transcription factors such as AP-1 and NF-B (6, 7). To establish that the GR activated by GCS produced endogenously have the capacity to mediate DNA-independent protein-protein interactions, we questioned whether the activation-induced production of certain cytokines by CD4+ T cells could be modulated by the addition of 11DHC to the cell culture system. Freshly isolated splenic CD4+ T cells were activated by placing them onto culture plates coated with immobilized anti-CD3 Ab in the presence or absence of corticosterone or 11DHC. After 24 h, supernatants were collected to quantitatively analyze for the production of IL-2 and IFN- by ELISA. We found that the lymphocytes cultured with 11DHC were inhibited from producing both IL-2 and IFN- (Fig. 5A). As previously demonstrated by us and others (37), the activation-induced production of these cytokines was also inhibited by the added presence of bioactive GCS (Fig. 5A). Exposure of activated CD4+ T cells to GA alone had minor inhibitory effects on both IL-2 and IFN- production, while the coincubation of GA and the bioactive GCS did not affect the inhibited production of either cytokine. However, coincubation of CD4+ T cells with GA plus 11DHC eliminated the inhibition of cytokine production (Fig. 5A), presumably due to the inhibition of 11-HSD1 function by the added GA.
FIGURE 5. Ligand activation of the GR by GCS generated by CD4+ T cells inhibits activation-induced cytokine production. A, Purified splenic CD4+ T cells were activated with immobilized anti-CD3 Ab (1 μg/ml) in the presence or absence of corticosterone or 11DHC with the addition of GA. Supernatants were collected after 24 h to quantitatively analyze for IL-2 and IFN- production by ELISA. B, Purified splenic CD4+ T cells from C57BL/6 WT and 11-HSD1–/– mice were cultured in 1% FBS and activated with immobilized anti-CD3 and anti-CD28 Ab in the presence or absence of corticosterone or 11DHC. Supernatants were collected after 24 h to quantitatively analyze for IL-4 and IFN- production by ELISA. C, Purified primary CD4+ T cells were activated with immobilized anti-CD3 Ab (1 μg/ml) in the presence or absence of bioactive GCS (corticosterone) or the inactive metabolite (11DHC) with the addition of RU486. Supernatants were collected after 24 h to quantitatively analyze for IL-2 and IFN- production by ELISA. RU486 is a GR antagonist, GA, and an 11-HSD1 inhibitor. Results shown are mean ± SEM and are representative of 4 separate experiments.
To further demonstrate that lymphocyte 11-HSD1 is responsible for the intracellular generation of bioactive GCS from 11-keto metabolites, splenic CD4+ T cells were purified from C57BL/6 WT and 11-HSD1–/– mice, cultured in medium containing 1% FBS to minimize influences exerted by any 11-keto GCS present in serum, and activated with anti-CD3 and anti-CD28 Abs (coimmobilized on tissue culture plates) in the presence or absence of corticosterone or 11DHC. Supernatants were collected after 24 h to analyze for activation-induced cytokine production by ELISA. As shown in Fig. 5B, 11DHC suppressed the production of IL-4 and IFN- by WT CD4+ T cells, while this suppressive effect was greatly diminished in cultures of 11-HSD1–/– CD4+ T cells. Interestingly, when identical culture conditions were used, purified CD4+ T cells from 11-HSD1–/– animals possessed an augmented capacity to produce both IL-4 and IFN- when compared with WT CD4+ T cells (Fig. 5B). Collectively, these results demonstrate that murine CD4+ T cells can effectively convert 11DHC to bioactive GCS, providing a means for inactive 11-ketosteroids to function as activating ligands for the GR following prereceptor metabolism by lymphocyte 11-HSD1.
To demonstrate that the endogenous metabolism of 11DHC-mediated inhibition of cytokine production is mediated through the GR, we used RU486, a GR antagonist. Murine splenic CD4+ T cells were activated on culture plates containing immobilized anti-CD3 Ab in the presence or absence of either active corticosterone or 11DHC. Expectedly, RU486 successfully abolished the ability of the active GCS to inhibit both IL-2 and IFN- production (Fig. 5C). We found that the addition of RU486 also abrogated the ability of 11DHC metabolites to suppress cytokine production.
Activation of CD4+ T cells through the TCR and their polarization into Th1 and Th2 effector T cells elevate cellular 11-HSD1 reductase activity
The capacity of CD4+ T cells to convert the 11-keto metabolites of GCS into their bioactive forms, sufficient to suppress the activation-induced production of inflammatory cytokines, was previously unknown. This novel finding led us to further question whether cellular activation through the TCR can induce an increase in the capacity of lymphocyte 11-HSD1 to generate activating ligands for the GR and potentiate the GCS-induced suppression of inflammatory cytokine production. CD4+ T cells were isolated from the PLN as well as the spleen of normal BALB/c mice. The lymphocytes were cultured alone or activated by placing them onto tissue culture plates containing immobilized anti-CD3 Ab or plates containing coimmobilized anti-CD3 and anti-CD28. As seen in Fig. 6A, CD4+ T cells activated through the TCR were able to generate three times the amount of cortisol, when compared with the amount of cortisol generated by an equivalent number of resting CD4+ T cells for the same time period. When CD4+ T cells were activated through the TCR with costimulation, the increased ability of CD4+ T cells to reactivate cortisone into cortisol remained significant, albeit at a slightly reduced level when compared with CD4+ T cells that received an activation signal through the TCR alone.
FIGURE 6. Activation of CD4+ T cells through the TCR and their polarization into Th1 and Th2 effector cells augment their ability to reactivate 11-keto metabolites of GCS. A, Purified splenic CD4+ T cells were cultured either in a resting state or activated by placing them onto tissue culture plates containing immobilized anti-CD3 Ab (1 μg/ml) alone or with immobilized anti-CD3 and anti-CD28 (both 1 μg/ml) in the presence of [3H]cortisone. Supernatants were collected for steroid extraction and the determination of the rate of conversion of cortisone to cortisol. B, CD4+ T cells were purified from the PLN of C3H/HeN mice and activated with immobilized anti-CD3 and anti-CD28 under Th0, Th1, or Th2 polarization conditions. Th0, Th1, and Th2 lineage-committed effector cells were cultured in the presence of [3H]cortisone for an additional 24 h to evaluate their ability to convert cortisone to cortisol. C, The same Th0, Th1, and Th2 lineage-committed effector cells were lysed, and 20 μg of protein was used for Western blot analysis of 11-HSD1. Results shown are mean ± SEM of three independent experiments. *, p < 0.05.
Similar to activation through the TCR, when naive CD4+ T cells were induced to become Th1 and Th2 effector cells, following a 4- to 5-day culture period under the influence of Th1- and Th2-polarizing cytokines, their capacity to generate bioactive GCS from their 11-keto derivatives increased by 2-fold in comparison with the Th0 T cells (Fig. 6B). We were also able to establish that the increase in 11-HSD1 reductase activity correlated well with an observed increase in 11-HSD1 protein expression in Th1 and Th2 T cells (Fig. 6C).
Lymphocyte 11-HSD1 reductase activity is elevated in aged mice
The 11-HSD1 reductase activity has been reported to increase with age in human osteoblasts (47). This increase in the conversion of 11-keto metabolites to bioactive GCS via an intracrine mechanism has been proposed to be linked with the decrease in bone formation and an increased risk in GCS-induced osteoporosis (48, 49). To analyze whether 11-HSD1 activity becomes elevated in the CD4+ T cells of aged mice, purified populations of CD4+ T cells from either young (3-mo) or aged (>16-mo) BALB/c mice were cultured in the presence of [3H]cortisone. Supernatants were analyzed to evaluate whether an observable difference in 11-HSD1 reductase activity occurred as a consequence of aging. Nearly 2-fold higher levels of cortisol were found in the supernatants of CD4+ T cells from aged donors at 24 h when compared with levels of cortisol found in the supernatants of CD4+ T cells from young donors (Fig. 7A). We also found 11-HSD1 protein expression to be elevated in CD4+ T cells from aged donors (Fig. 7B). To establish that the observed increase in 11-HSD1 reductase activity of CD4+ T cells isolated from aged donors is not due to an increase in the ratio of memory/naive T cells in the aged donors, identical experiments were performed using CD4+ T cells isolated from young (3-mo) or aged (>16-mo) DO11.10 TCR transgenic mice. The results obtained were similar (data not shown). Because CD4+ T cells isolated from aged DO11.10 TCR transgenic mice retain a naive phenotype (CD45RBhighCD62Lhigh; our unpublished observation), these findings indicate that, similar to human osteoblasts, 11-HSD1 reductase activity in CD4+ T cells becomes elevated with age.
FIGURE 7. The 11-HSD1 protein expression and reductase activity are elevated in CD4+ T cells from aged donors. A, Primary CD4+ T cells were purified from the spleens of young (3-mo) or aged (>16-mo) mice and cultured in the presence of [3H]cortisone. The 11-HSD1 reductase activity is determined by the rate of conversion of cortisone to cortisol in culture. Results shown are mean ± SEM of five independent experiments. *, p < 0.005. B, Purified primary CD4+ T cells from young or aged donors were lysed to perform protein extraction. A total of 20 μg of cytoplasmic protein was used for 11-HSD1 protein expression by Western blot analysis using a polyclonal 11-HSD1 Ab. Results shown are representative of three independent experiments.
Discussion
Many steroid hormones, including testosterone, vitamin D3, and the retinoids, are produced and released into the systemic circulation as precursor molecules that possess little or no biological activity. These circulating precursors then become end-organ metabolized into their most bioactive forms by tissue- and organ-specific enzymes (50, 51, 52). In contrast, the GCS are produced and released by the adrenal gland in their bioactive forms that can subsequently be metabolized by tissue-specific 11-HSD enzymes. Although >90% of the bioactive GCS are bound to carrier proteins in the plasma, buffering the rate of their cellular entry, the 11-keto metabolites (cortisone and 11DHC) are able to freely diffuse into cells without restrictions (8). Historical reports from the 1950s exist describing the capacity of lymphocytes to generate bioactive GCS from their physiologically inactive 11-keto metabolites (53, 54). Over the next half century, with the exception of a single report describing the inducible expression of 11-HSD1 in human monocytes upon their differentiation into macrophages, the expression of GCS-metabolizing enzymes in immune cells has remained totally unexplored.
In this study, we report that 11-HSD1 mRNA and protein are expressed in murine CD4+, CD8+, and B220+ lymphocytes. The expression of mRNA and protein correlates with 11-HSD1 reductase activity, as evidenced by the exclusive conversion of cortisone to cortisol by cultured lymphocytes. We therefore hypothesized that through its exclusivity in converting inactive 11-keto GCS to their bioactive forms, 11-HSD1 could provide lymphocytes with a novel intracrine mechanism to generate activating ligands for their own intracellular GR.
To support our working hypothesis, we demonstrated that the presence of 11-keto metabolites of GCS not only up-regulated the expression of an MMTV-Luc reporter construct in transiently transfected DO11.10 T cell hybridomas, but also induced the expression of IL-7R and GILZ gene expression in primary murine CD4+ lymphocytes. Both of these genes have been shown to be inducible by dexamethasone in human PBMC (44). The up-regulated expression of IL-7R and GILZ provides protective mechanisms to lymphocytes against activation-induced apoptosis (44, 46). Expression of the IL-7R chain is crucial to the development of both T and B cells because IL-7R–/– mice have profound deficiencies in their T and B lymphocyte repertoire (55, 56). Increased IL-7R expression also exerts antiapoptotic effects on GCS-induced death in T lymphocytes (44), although the molecular mechanism through which this occurs has yet to be established. GILZ has previously been demonstrated to repress activation-induced cell death in lymphocytes via its capacity to mediate protein-protein interactions with AP-1, NF-B, and Raf-1 (57, 58). The ability of murine CD4+ T cells to up-regulate the expression of these prosurvival genes, when cultured in the presence of physiological concentrations of 11-keto GCS, suggests that lymphocytes possess an intrinsic means to control their susceptibility to apoptosis.
To further explore the possibility that intracellular reactivation of GCS provides lymphocytes with a degree of autonomous control over their viability, as well as some of their effector functions, we found that the activation-induced production of IL-2 and IFN- by CD4+ lymphocytes was depressed in cultures containing added 11-keto GCS. Because 11-keto GCS are not capable of directly activating the GR, we presumed they were being converted into the bioactive forms of GCS via intracellular 11-HSD1. In support of this hypothesis, we found that when 11-HSD1 function was inhibited by the addition of GA, or when RU486 was used to antagonize GR signaling, all changes consistent with GCS influences were abrogated. Furthermore, we demonstrated that the ability of 11-keto GCS to suppress IL-4 and IFN- production was greatly diminished in cultures of purified CD4+ T cells from 11-HSD1–/– animals. Together, these observations provide strong evidence that CD4+ lymphocytes have the endogenous capacity to reactivate 11-keto derivatives of GCS into their bioactive forms, providing activating ligands for the GR. Thus, intracellular 11-HSD1 provides CD4+ lymphocytes with an intracrine means to modulate some of their effector functions. Furthermore, the observed inhibition of activation-induced production of IL-2 may be partially due to the up-regulated expression of the GILZ gene and protein, because it has been previously demonstrated that the induction of GILZ mediates the inhibition of activation-induced IL-2 production by dexamethasone, via its transcriptional interference with AP-1 (58).
The cytokines produced by activated T cells impart regulatory influences over the development and nature of the subsequently generated immune responses. Our finding that CD4+ lymphocytes produced less inflammatory cytokines when activated in the presence of 11-keto GCS concurs with the present paradigm that GCS are able to skew the immune response toward a Th2 pathway by inhibiting the production of cytokines necessary to generate an efficient Th1 response (6, 59, 60). Furthermore, we found that the ability to convert 11-keto GCS into their bioactive form is significantly elevated in T cells that have been activated through the TCR or polarized down either a Th1 or Th2 pathway. An augmentation in the ability of Th1 T cells to generate bioactive GCS may assist in controlling the levels of Th1 cytokines, possibly to avoid generating an exaggerated inflammatory response. We and others have previously shown that physiological levels of endogenous GCS can enhance the production of Th2 cytokines by activated lymphocytes in a dose-dependent manner (37). The increase in 11-HSD1 activity in Th2 T cells supports the possibility that elevated levels of endogenous GCS might promote the development of an anti-inflammatory immune response. Thus, the increased capacity of Th1 and Th2 lineage-committed effector T cells to generate bioactive GCS could play a role in directing the progression of the immune response, away from inflammatory Th1 responses. Interestingly, we found that splenic CD4+ T cells from 11HSD1–/– mice possessed an augmented capacity to produce both IL-4 and IFN- in comparison with WT splenic CD4+ T cells in vitro. Whether this phenomenon is due to the differential capacity of activated WT and 11-HSD1–/– CD4+ T cells to produce these cytokines or to developmental differences in the CD4+ T cells between WT and 11-HSD1–/– mice is currently being explored.
Influences on immune functions exerted by GCS are not limited to their ability to shift patterns of cytokine production by activated T lymphocytes. GCS can also extend their immunomodulatory functions to other immune cell types. Increasing evidence now suggests a key role for dendritic cells (DCs) in orchestrating the lineage commitment of naive Th cells (61, 62, 63, 64). Perhaps the capacity of GCS to influence the development of immune effector responses begins with their ability to modulate particular functions of APCs. GCS are known to inhibit the up-regulation of MHC class II molecules, the costimulatory molecules CD80 and CD86, as well as adhesion molecules and chemokine receptors on the DC surface during their maturation (65, 66, 67, 68). Our unique finding that CD11c+ DCs have the capacity to generate bioactive GCS via an intracrine mechanism would suggest a possible role for GCS modulation of DC maturation in vivo.
It has been reported that Ag presentation by DCs in the absence of effective costimulation can lead to an anergy in responding lymphocytes (69, 70). Immature DCs differentially matured under the influence of intracellularly generated bioactive GCS may be unable to properly up-regulate costimulatory molecule expression during Ag presentation to lymphocytes, leading to the development of lymphocyte unresponsiveness. In addition, we have also observed a marked up-regulation in 11-HSD1 mRNA expression in immature DCs following maturation induction with LPS or following treatment with dexamethasone (our unpublished observation). An up-regulation in 11-HSD1 expression could potentially increase the intracellular as well as microenvironmental levels of bioactive GCS during Ag uptake and processing, leading to an altered maturation of the DCs during their migration and trafficking to secondary lymphoid organs.
We have observed that murine thymocytes are capable of generating active GCS from 11-keto GCS substrates in culture. This finding is particularly interesting because there now exists considerable evidence that GCS play essential roles during thymocyte development and differentiation, including the shaping of the Ag-specific T cell repertoire (6, 71). This occurs through the ability of GCS to inhibit activation-induced apoptosis of thymocytes, thus increasing the TCR signaling thresholds required to promote positive and negative selection (72, 73). Thymic epithelial cells have been reported to express steroidogenic enzymes, and the addition of inhibitors of these steroidogenic enzymes has been shown to profoundly affect the viability of the CD4+CD8+ double-positive subpopulation of thymocytes in fetal thymic cultures (74, 75). Our finding that murine neonatal and adult thymocytes readily convert inactive 11-keto GCS into bioactive species of GCS demonstrates an additional intracrine mechanism by which thymocytes themselves might exert control over endogenous GCS levels.
The capacity of thymocytes to generate bioactive GCS may be especially important during fetal development, when placental and fetal tissue 11-HSD2 is functioning to inactivate maternal GCS, preventing any deleterious effects of bioactive GCS on the developing fetus (21, 22, 76). The 11-keto GCS-rich environment present throughout most of gestation would maximize the potential for thymocyte 11-HSD1 activities. Because thymocytes from 11-HSD1–/– animals would exhibit a reduced capacity to generate endogenous bioactive GCS from their 11-keto metabolites, an expanded population of thymocytes from the genetically deficient animals should be susceptible to negative selection by activation-induced apoptosis. This would effectively eliminate developing T cells expressing TCRs having the higher avidities for self peptide/MHC complexes that would normally be spared by natural GCS influences. Hence, animals lacking 11-HSD1 might be less susceptible to autoimmune diseases when compared with WT animals.
The 11-HSD1–/– mice are resistant to age-associated declines in cognitive functions (77). This suggests that changes in 11-HSD1 activity in the brain could be partially responsible for the decline in cognitive functions as age progresses in WT animals. Recent findings with human osteoblasts suggest an increase in 11-HSD1 transcript and reductase activity occurs with aging, a finding that has been linked to the age-associated decline in bone formation as well as GCS-induced osteoporosis (47). Consistent with these findings, we found that 11-HSD1 reductase activity and protein expression are increased in CD4+ T cells from aged donors. This age-associated increase in the capacity to reactivate 11-keto GCS by CD4+ T cells may somehow be linked to the decrease in GCS responsiveness reported to occur in aged animals (78, 79). The possibility also exists that the age-associated increases in circulating levels of proinflammatory cytokines (80, 81, 82, 83, 84) are linked to increases in 11-HSD1 reductase activity because these cytokines (e.g., TNF-, IL-1) are able to increase 11-HSD1 expression as well as reductase activity in nonimmune cells (85). The stimulation of human macrophages by IL-4 or LPS also has the ability to elevate 11-HSD1 mRNA and activity (30).
Recent reports indicate that hexose-6-phosphate dehydrogenase controls the ability of 11-HSD1 to function as a reductase or as a dehydrogenase in murine adipocytes (86, 87). It may be possible that the augmented 11-HSD1 reductase activity in activated/polarized CD4+ T cells and in CD4+ T cells from aged donors is linked to an increase in the activity of hexose-6-phosphate dehydrogenase.
The current paradigm of GCS-mediated influences on lymphocyte development and physiology assumes that the plasma GCS provide the sole source of bioactive steroid. We have discovered and demonstrated the novel capacity of lymphocytes to generate bioactive GCS from circulating inactive and bioavailable 11-keto metabolites, providing a previously unappreciated source of bioactive GCS for modulating immune cell function. A careful examination of T cell development and inducible immune responses by 11-HSD1–/– mice should provide valuable insight into the physiological significance of this intracrine-controlling mechanism in lymphocyte physiology.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by National Institutes of Health Grants CA25917 and DK55491, a Browning Foundation grant, and Department of Veterans Affairs medical research funds. T.Y.Z. is supported by National Institutes of Health, Department of Health and Human Services, National Institute of Diabetes and Digestive and Kidney Diseases Hematology Research Training Grant T32 DK07115.
2 Address correspondence and reprint requests to Dr. Raymond A. Daynes, Department of Pathology, University of Utah School of Medicine, 30 North 1900 East, Salt Lake City, UT 84132-2501. E-mail address: daynes.office@path.utah.edu
3 Abbreviations used in this paper: GCS, glucocorticoid; DC, dendritic cell; BMDC, bone marrow-derived DC; 11DHC, 11-dehydrocorticosterone; GA, glycerrhetinic acid; GILZ, GCS-induced leucine zipper; GR, GCS receptor; GRE, GCS response element; 11-HSD, 11-hydroxysteroid dehydrogenase; MMTV, mouse mammary tumor virus; PLN, peripheral lymph node; WT, wild type.
Received for publication May 25, 2004. Accepted for publication November 2, 2004.
References
Rook, G. A.. 1999. Glucocorticoids and immune function. Baillieres Best Pract. Res. Clin. Endocrinol. Metab. 13:567.
Webster, J. I., L. Tonelli, E. M. Sternberg. 2002. Neuroendocrine regulation of immunity. Annu. Rev. Immunol. 20:125
Sternberg, E. M.. 2001. Neuroendocrine regulation of autoimmune/inflammatory disease. J. Endocrinol. 169:429.
Pitzalis, C., N. Pipitone, M. Perretti. 2002. Regulation of leukocyte-endothelial interactions by glucocorticoids. Ann. NY Acad. Sci. 966:108.
Cancedda, C., G. Filaci, F. Puppo, M. Ghio, P. Contini, F. Indiveri. 2002. Immune homeostasis requires several biologic factors including glucocorticoid hormones. Ann. NY Acad. Sci. 966:49.
Ashwell, J. D., F. W. Lu, M. S. Vacchio. 2000. Glucocorticoids in T cell development and function. Annu. Rev. Immunol. 18:309.
Refojo, D., A. C. Liberman, F. Holsboer, E. Arzt. 2001. Transcription factor-mediated molecular mechanisms involved in the functional cross-talk between cytokines and glucocorticoids. Immunol. Cell Biol. 79:385.
Siiteri, P. K., J. T. Murai, G. L. Hammond, J. A. Nisker, W. J. Raymoure, R. W. Kuhn. 1982. The serum transport of steroid hormones. Recent Prog. Horm. Res. 38:457.
Rosner, W.. 1991. Plasma steroid-binding proteins. Endocrinol. Metab. Clin. North Am. 20:697.
Mangelsdorf, D. J., C. Thummel, M. Beato, P. Herrlich, G. Schutz, K. Umesono, B. Blumberg, P. Kastner, M. Mark, P. Chambon, et al 1995. The nuclear receptor superfamily: the second decade. Cell 83:835.
Reichardt, H. M., G. Schutz. 1998. Glucocorticoid signalling: multiple variations of a common theme. Mol. Cell. Endocrinol. 146:1.(Tian Y. Zhang, Xiaohong D)