Abstract
Castration of normal male rodents results in significant enlargement of the thymus, and androgen replacement reverses these changes. Androgen-resistant testicular feminization (Tfm) mice also show significant thymus enlargement, which suggests that these changes are mediated by the androgen receptor (AR). The cellular targets of androgen action in the thymus are not known, but may include the lymphoid cells (thymocytes) as well as nonlymphoid epithelial cells, both of which have been believed to express AR. In the present study immunohistochemical analysis and hormone binding assays were used to demonstrate the presence of AR in thymic epithelial cells. The physiological significance of this epithelial cell AR expression was defined by further studies performed in vivo using chimeric mice, produced by bone marrow transplantation, in which AR expression was limited to either lymphoid or epithelial components of the thymus. Chimeric C57 mice engrafted with Tfm bone marrow cells (AR+ epithelium and AR− thymocytes) had thymuses of normal size and showed the normal involutional response to androgens, whereas chimeric Tfm mice engrafted with C57 bone marrow cells (AR− epithelium and AR+ thymocytes) showed thymus enlargement and androgen insensitivity. Furthermore, phenotypic analyses of lymphocytes in mice with AR− thymic epithelium showed abrogation of the normal responses to androgens. These data suggest that AR expressed by thymic epithelium are important modulators of thymocyte development.
THYMUS WEIGHT, cellularity, and cellular composition are very sensitive to changes in androgen status. The thymus gland of male mice is enlarged under conditions of androgen deficiency or in mice with defects in androgen action and shows a shift in the composition of relative numbers of thymocyte phenotypic subpopulations (1, 2). These observations have suggested that androgen effects on the thymus are exerted through conventional receptor-mediated mechanisms.
Many studies have demonstrated the presence of high affinity androgen-binding proteins in thymic tissues (3–5), but localization of androgen receptors (AR) within the different cellular compartments of the thymus has been more controversial. Early reported studies, which used physical separation techniques and ligand binding assays, indicated that the epithelial cell expressed AR, whereas thymocytes were thought to be negative for AR expression (3, 4). In later studies AR expression in purified thymocytes was shown by a variety of methods, including ligand binding assays, flow cytometry, and immunoblotting (6, 7). It is unknown whether the observed effects of androgens on thymic size and cellular composition are mediated by the action of the hormones exerted directly on thymocytes or whether the effects are indirectly mediated by androgen action on thymic epithelial or stromal cells.
In the present study we examined AR expression in nonlymphoid thymic components by ligand binding studies in thymic epithelial cell lines and immunohistochemical techniques on thymic tissue sections. We then tested the functional importance of epithelial AR expression by the use of bone marrow transplantation to create chimeric mice with AR-positive lymphoid and AR-negative stromal-epithelial compartments. Our findings reveal that thymic epithelial expression of AR is required for androgen effects on thymocyte development.
Materials and Methods
Mice and cell preparations
Normal male C57BL/6 Thy 1.1 congenic mice and androgen-resistant Tfm/Y male mice were obtained from The Jackson Laboratory (Bar Harbor, ME). These Tfm/Y mice express an abnormal AR due to a single base deletion that results in a premature stop codon and a truncated protein (8, 9). AR expression (assessed by ligand binding assays) in these mice is estimated to be 10–20% of normal levels (10). Thymuses were harvested and weighed at the time of death. Bone marrow cells for transplantation were obtained from the tibias and femurs of each animal by flushing the marrow cavity with RPMI 1640 medium (Life Technologies, Inc., Grand Island, NY) using a syringe equipped with a 26-gauge needle. Single cell suspensions were prepared by homogenizing the tissues between the frosted ends of microscope slides or by using a ground glass homogenizer.
Bone marrow chimeras
Unseparated bone marrow cells (0.5 × 106) from C57BL/6 or Tfm/Y mice were transferred iv into lethally irradiated (900 rad) recipients of the opposite strain. This radiation dose has been demonstrated to preserve functional thymic epithelial cells while destroying lymphoid components (11). C57 congenic mice that expressed the Thy 1.1 allele on all thymus-derived cells were used to distinguish their cells from those of the Tfm mice, which express the Thy 1.2 allele. This Thy marker discordance permitted assessment of donor cell survival in the irradiated host; 85 ± 5% of thymocytes from chimeric animals were of the donor phenotype. Animals were studied 60 or more days after transplantation. In some experiments, transplant recipients were castrated as described previously (12). Androgen replacement was achieved in some castrated animals using sc pellets of dihydrotestosterone (DHT; Innovative Research, Sarasota, FL). A preliminary series of experiments established that treatment of castrated C57 male mice with a 0.5-mg 21-day release DHT pellet resulted in restoration of thymus size to normal. Animals were killed at the completion of the 21-day release period. Serum testosterone levels measured as previously described (1) in intact and irradiated males (6–10 mice/group) were not significantly different (1.03 ± 0.57 vs. 1.06 ± 0.39 ng/ml; P = 0.97). Tfm/Y animals (n = 6) had somewhat lower levels of serum testosterone (0.72 ± 0.25 ng/ml), but this difference was not significantly different from the other two groups (P = 0.83).
Thymic epithelial cells
The Z210 and TE71.1 cell lines were gifts from Dr. Andrew Farr, University of Washington (Seattle, WA). These cells have been characterized as being of thymic medullary origin (13–15). The 1308.1 and 427 cell lines were obtained from Dr. Barbara Knowles, The Jackson Laboratory. These two cell lines are derived from cortical areas of the thymus (16). Cells were maintained in adherent culture in RPMI 1640 medium (Life Technologies, Inc.) supplemented with 10% FCS (Life Technologies, Inc.) and were passaged by treatment with trypsin/EDTA (Life Technologies, Inc.).
Immunohistochemical procedures
Frozen mouse tissues (thymus and seminal vesicle) were sectioned (5 μm) and fixed in 4% formaldehyde in 0.1 m phosphate buffer, pH 7.4, or in absolute methanol for 10 min at −20 C. Tissues were rinsed with PBS, blocked with BSA, and then incubated with a rabbit polyclonal antibody to a 21-peptide sequence identical to amino acids 1-21 of rat AR (17). Anti-AR antibody was a gift from Dr. Gail Prins (University of Illinois). In one set of experiments, the secondary detection was performed with biotinylated goat antirabbit antibody (Sigma, St. Louis, MO) developed with avidin-biotin complex reagent (Biomeda, Foster City, CA) and detected by reaction with diaminobenzidine. The specificity of this antibody for AR has been shown in previous studies in thymocytes (7). Tissues were counterstained with eosin and examined by brightfield microscopy. In some experiments the biotinylated antibody was detected with streptavidin-conjugated gold particles (Janssen Pharmaceuticals, Piscataway, NJ), and tissue was counterstained with hematoxylin and observed with reflected epiillumination.
Ligand binding assays
Binding of the androgen ligand[ 3H]mibolerone (NEN Life Science Products, Boston, MA) was carried out on monolayers of thymic epithelial cells as previously described (6, 18). The cells were grown to confluence in 6-cm plates, rinsed to remove serum, and incubated for 45 min at 37 C with the radioligand in a range of concentrations (0.05–3.0 nm). Parallel plates received radioligand with an excess of nonradioactive mibolerone to assess nonspecific binding. The cells were washed, harvested by trypsinization, and sonicated in water. Aliquots were removed for measurement of protein and radioactivity. Total and nonspecific binding were analyzed as a function of radioligand concentration, and the method of Scatchard was applied to determine affinity constants.
Flow cytometric analysis
Thymocytes harvested from the transplanted mice were suspended at 1 × 106/0.2 ml FACS buffer (PBS with 2% BSA and 0.1% NaN3) and incubated with saturating concentrations of the conjugated monoclonal antibodies Thy1.2-FITC, Thy1.1-PE, CD4-PE, and CD8-FITC (PharMingen, San Diego, CA) at 4 C for 30 min, washed, and then fixed with 1% paraformaldehyde (EM Sciences, Ft. Washington, PA) before analysis on a FACStar Plus (Becton Dickinson and Co., San Jose, CA).
Statistics
Data are presented as the mean and sem. Comparisons between two groups were made using Student’s t test. P < 0.05 was considered significant.
Results
Immunohistochemical examination and ligand binding assays reveal AR expression in thymic epithelial cells
Immunohistochemical staining of thymus sections with anti-AR antibodies revealed scattered positive cells with staining intensity comparable to that seen in a positive control tissue (seminal vesicle; Fig. 1). Immunolocalization techniques using reflected epiillumination showed the positive cells to be nonlymphoid cells with epithelioid morphology (Fig. 1, C and D). AR expression in thymic epithelium was further confirmed by ligand binding assays in established thymic epithelial cell lines. Three of four such cell lines tested were found to have saturable high affinity binding of the synthetic androgen receptor ligand mibolerone (Fig. 2). Two of the AR+ cell lines (TE711 and Z210) were derived from thymic medulla, and one was from thymic cortex (427). Only the 1308 cortical cell line was found to lack AR expression. Scatchard analyses were carried out for the two medullary lines, yielding corresponding Kd values of less than 0.1 nm.
Immunohistochemical detection of AR using an anti-AR peptide antibody. A seminal vesicle (A), stained using peroxidase, illustrates intranuclear localization of AR in this classic target tissue. A thymus (B) demonstrates scattered positive cells. Thymus tissue stained using gold particle-conjugated second antibody and examined in a brightfield (C) shows large epithelioid cells, which are revealed as positive for AR staining using reflected epiillumination (D).
Ligand binding assays for detection of AR in four thymic epithelial cell lines. Total (•) and nonspecific (○) binding curves are shown for each. The medullary lines Z210 and TE71 show high levels of specific hormone binding, as does one of the cortical lines, 427.1. The cortical line 1308.1 is negative for AR.
AR expression in thymic epithelium is necessary for androgen-induced thymic involution
Bone marrow transplantation between normal and AR-defective mice was used to create chimeric animals with AR expression confined to either thymic epithelium or thymocytes. These chimeric animals were compared with normal C57BL/6 male mice or to androgen-insensitive Tfm/Y mice. In the normal male mouse (expressing AR in both thymocytes and thymic epithelium), castration results in an approximate doubling in thymic size that is reversed by androgen replacement (Fig. 3A). The Tfm/Y mouse (AR negative in both thymocytes and thymic epithelium) shows significant thymic enlargement, comparable to that in androgen-deficient animals; this thymic expansion is not affected by androgen administration (Fig. 3B). Chimeric animals with AR (+) thymocytes transplanted into the AR− thymic microenvironment had thymuses that were 34% larger than intact C57 controls (58.8 ± 3.7 compared with 43.7 ± 3.5 mg; P = 0.026) and were insensitive to the involutional effects of androgens (Fig. 3C). In chimeric mice with AR expression in the thymic epithelium, but with thymocytes derived from AR- bone marrow, castration resulted in only modest thymic enlargement that did not reach statistical significance, but the involutional effects of androgens were preserved. After androgen administration, thymic weight in the chimeras was decreased from 51.33 ± 7.2 to 23.85 ± 2.83 mg (P = 0.02; Fig. 3D). The lack of thymic expansion upon androgen withdrawal in these chimeric mice could not be attributed to the transplantation procedure itself, as control transplants between congenic C57 mice showed statistically significant castration-induced thymic enlargement compared with corresponding intact animals (data not shown). Castration experiments were not carried out in chimeric Tfm recipients of C57 bone marrow because of the technical difficulty of gonadectomy in these mice, in which the testes are intraabdominal and about 10% of normal size (19).
Thymus weights in C57, Tfm/Y, and chimeric mice created by bone marrow transplantation after androgen depletion and androgen replacement. C57BL/6 male mice (A) show significant thymus enlargement after castration, with return to normal size after androgen replacement. Tfm/Y mice (B) have large thymuses and insensitivity to DHT. Chimeras made between Tfm AR− recipients and C57 AR+ donors (C) show enlarged thymuses with no response to DHT. Chimeras with C57 AR+ recipients and Tfm AR− donors (D) have normal thymus weights, no significant enlargement with castration, and significant involution after DHT replacement. Groups A and D were castrated before DHT replacement; groups B and C were not castrated before DHT treatment.
Depletion of the CD4+CD8+ thymocyte subpopulation after androgen administration requires thymic epithelial AR expression
When castrated C57BL/6 male mice were treated with DHT, the fraction of CD4+CD8+ double positive thymocytes was reduced by nearly 35% (Fig. 4A), with a corresponding increase in the CD4−CD8− double negative population. This reduction in CD4+CD8+ cells was not seen after androgen administration to androgen-resistant Tfm/Y mice (Fig. 4B). Chimeric mice with androgen-responsive AR+ thymic epithelium but with AR− thymocytes also exhibit androgen-mediated depletion of CD4+CD8+ cells (Fig. 4C) whereas no significant changes were observed in the CD4+CD8+ population of chimeric mice with AR+ thymocytes but AR− thymic epithelium (Fig. 4D).
Thymocyte subsets defined by markers CD4 and CD8 in C57, Tfm/Y, and chimeric mice created by bone marrow transplantation with and without DHT treatment. C57BL/6 mice (A) show a significant decrease in the double positive (DP) subset of thymocytes with DHT treatment. Tfm/Y mice show no significant thymocyte changes with DHT (B). Chimeric mice with AR− thymocytes and AR+ epithelium (C) display the normal DP decrease with DHT, whereas chimeras with AR+ thymocytes and AR- epithelium show no shifts in thymocyte subpopulations (D).
Discussion
The thymus has a central role in the maturation and development of T lymphocytes. Interactions between thymic epithelial cells and developing thymocytes determine which cells undergo negative selection by apoptosis within the gland and which ones are positively selected to mature and form the peripheral T cell repertoire. Precise mechanisms by which positive and negative thymocyte selection take place are not completely understood, but involve interaction between the T cell receptor (TCR) molecules expressed by thymocytes and major histocompatibility complex molecules on thymic epithelium (11, 20, 21).
Factors outside the immune system exert important effects on thymus function. Among these are hormones such as estrogens, androgens, glucocorticoids, progesterone, and somatostatin (22–26). Receptors for each of these hormones have been found to be expressed in thymus tissues and cells (6, 25, 27–30), and both thymocytes and thymic epithelial cells have been implicated as targets of hormone action based on localization of specific receptors in these two major thymic compartments. High affinity, specific receptors for androgen were demonstrated in early reports using ligand binding assays in homogenates of whole thymus tissue (3, 4) and steroid autoradiography in tissue sections (31). In other studies using both human and murine tissues, AR were detected in thymocytes using ligand binding assays and flow cytometry (6, 7). A recent study using quantitative RT-PCR for detection of AR messenger RNA demonstrated significantly more abundant messenger RNA for the receptor in the thymic epithelial cells than in the thymocytes (23). However, separation techniques used in these studies most likely did not lead to pure preparations of epithelial cells.
The present studies confirm that AR is present in thymic epithelial cells using two different techniques, ligand binding and immunohistochemistry. No significant staining for AR was noted in the thymocytes despite previous observations of AR expression in thymocytes detected by ligand binding, immunoblotting, and flow cytometry (6, 7). The immunohistochemical findings may differ because AR is less abundant in thymocytes than in epithelial cells (23) [although our ligand binding data (7) suggest similar levels] or, more likely, because fixation procedures required for the immunohistochemical studies interfered with detection of thymocyte AR. Both cortical and medullary thymic epithelial cell lines were positive for AR in ligand binding assays, and the immunohistochemical studies also suggest that the AR-expressing epithelial cells are distributed in both of these major areas of the thymus.
If AR is present in both thymocytes and thymic epithelium, then signaling pathways for androgens may involve either or both cell types. Results in the chimeric mice of the present study show that the apparent restraining effect of endogenous androgens on thymocyte proliferation is dependent at least in part upon the expression of AR by epithelial cells. This is suggested by the finding that chimeras in which androgen-resistant thymocytes develop in the context of a normal epithelium show no thymus enlargement, whereas chimeras with AR− epithelium showed significantly enlarged thymuses. However, a contribution of the thymocyte AR is also suggested by the observation that the thymuses in these chimeras were not as large as in the Tfm/Y mice. As these animals were not castrated, circulating androgens could potentially be exerting a negative effect on the AR+ thymocytes, thus contributing to the decreased size. This hypothesis will be tested by treating these animals with AR-blocking agents.
Two epithelial-dependent steps in the T cell developmental pathway might be affected by the presence of a defective AR. The first is migration of stem cells from the bone marrow to the thymus, and the second is expansion and selection of immature thymocyte precursors within the thymus gland itself. Migration of precursors from bone marrow to the thymus is most likely dependent on both chemotactic factors and adhesion molecules, both of which are expressed by thymic epithelial cells (32). Expansion and maturation of precursor cells within the gland are in part dependent on rearrangement of the αβ TCR locus, interaction between these TCRs and epithelial major histocompatibility complex molecules, and soluble factors produced by the thymic epithelial cells (33). Other epithelial-mediated effects, notably apoptosis, may be independent of the TCR (34–36). Mechanisms by which androgens might alter either or both of these epithelial cell-dependent steps are unknown, but effects on the production of cytokines or the expression of cell surface molecules are possible.
The importance of thymic epithelial hormone receptors for thymus development has been demonstrated for two other hormone-receptor systems, progesterone and estrogen. Studies in the progesterone receptor null mouse model have demonstrated that functional epithelial cell progesterone receptor is required for normal pregnancy and for the thymic involution that normally accompanies pregnancy (25). Progesterone action mediated via the epithelial progesterone receptor also blocks maturation of very early thymocytes within the double negative population, presumably by a paracrine mechanism (25). Epithelial expression of estrogen receptorα by thymic epithelium also appears to be required for normal thymus development and for mediating estrogen-induced thymic atrophy (37). The current findings suggest that thymic epithelial AR is also required for normal thymic development and thymocyte selection.
The finding that castration of C57 chimeric recipients of Tfm/Y bone marrow did not result in significant thymic enlargement was unexpected. The mechanism responsible for castration-induced thymus enlargement includes induction of cell cycling by the thymocytes (12). How androgen insensitivity might interfere with the ability of Tfm/Y thymocytes to respond to the proliferative stimulus that follows androgen deprivation is currently under investigation. In contrast, the signal for involution appears not to require androgen-sensitive thymocytes, but only androgen-responsive thymic epithelium.
In summary, the present study confirms previous findings (3) that receptors for androgens are expressed in thymic epithelial cells and demonstrates for the first time that these receptors appear to have an important functional role in modulating thymus size and normal thymocyte development. These studies may have relevance to understanding human immune dysfunction, as a marked decline in thymic function after the fourth to fifth decades of life has been implicated in the difficulty of regenerating adequate immune responses in older patients with HIV infection or allogeneic transplants (38). Elucidation of the molecular and cellular pathways of androgen signaling in thymic epithelial cells may suggest approaches to enhancing thymus function in such patients.
Acknowledgments
The expert technical assistance of Maxine Turney, V. A. Winfrey, Wendell Nicholson, Adam Swallows, Yuxin Dong, and Andrew Strang is appreciated. David MacFarland of the Vanderbilt Flow Cytometry Laboratory provided assistance with flow cytometric analyses.
References
Author notes
This work was supported by NIH Grants DK-41053 and AI-41575 and grants from the Lupus Foundation of America and its Nashville Chapter.
Recipient of an NIH postdoctoral fellowship under Training Grant HD-07043. Current address: Department of Biochemistry, Midwestern University, Downers Grove, Illinois 60515.
Recipient of a Career Development Award (Clinical Investigator) from the Department of Veteran’s Affairs.
