- Email: [email protected]
Apoptosis and the Thymic Microenvironment in Murine Lupus
Article No. jaut.1999.0325, available online at http://www.idealibrary.com on
Journal of Autoimmunity (1999) 13, 325–334
Apoptosis and the Thymic Microenvironment in Murine Lupus Yuichi Takeoka1,2, Nobuhisa Taguchi1,2, Leonard Shultz3, Richard L. Boyd4, Mitsuru Naiki2, Aftab A. Ansari5 and M. Eric Gershwin1 1
Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, School of Medicine, Davis, California 95616, USA 2 Institute of Bio-Active Science, Nippon Zoki Pharmaceutical Co. Ltd., Yashiro, Hyogo 673–1461, Japan 3 Jackson Laboratories, Bar Harbor, ME 04609, USA 4 Department of Pathology and Immunology, Monash University, Prahran, Victoria 3181, Australia 5 Department of Pathology, Emory University, School of Medicine, Atlanta, GA 30322, USA Received 18 January 1999 Accepted 14 June 1999 Key words: apoptosis, autoimmunity, murine lupus, thymic microenvironment
The thymus of New Zealand black (NZB) mice undergoes premature involution. In addition, cultured thymic epithelial cells from NZB mice undergo accelerated preprogrammed degeneration. NZB mice also have distinctive and well-defined abnormalities of thymic architecture involving stromal cells, defined by staining with monoclonal antibodies specific for the thymic microenvironment. We took advantage of these findings, as well as our large panel of monoclonal antibodies which recognize thymic stroma, to study the induction of apoptosis in the thymus of murine lupus and including changes of epithelial architecture. We studied NZB, MRL/lpr, BXSB/Yaa, C3H/gld mice and BALB/c and C57BL/6 as control mice. Apoptosis was studied both at basal levels and following induction with either dexamethasone or lipopolysaccharide (LPS). The apoptotic cells were primarily found in the thymic cortex, and the frequency of apoptosis in murine lupus was less than 20% of controls. Moreover, all strains of murine lupus had severe abnormalities of the cortical network. These changes were not accentuated by dexamethasone treatment in cultured thymocytes. However, the thymus in murine lupus was less susceptible to LPS-induced apoptosis than control mice. Finally we note that the number of thymic nurse cells (TNC) was lowest in NZB mice. Our findings demonstrate significant abnormalities in the induction of apoptosis and the formation of TNC-like epithelial cells in SLE mice, and suggest that the abnormalities of the thymic microenvironment have an important role in the pathogenesis of murine lupus. © 1999 Academic Press
Introduction
positive and negative selection; the latter being more pronounced in thymic medulla. The process involved in negative selection, known as apoptosis, has generated considerable interest. Apoptosis has been shown to occur not only in immature thymocytes secondary to ligation of the TCR with appropriate peptide bearing MHC molecules, but also to be initiated by glucocorticoid hormones. The role of glucocorticoids is highlighted by the finding that in vivo administration of LPS increases glucocorticoid secretion by the adrenal gland, leading to a marked increase in numbers of apoptotic cells within the murine thymus within 24 h [5–8]. Investigation of LPS-mediated apoptosis has permitted evaluation of the relative frequency of cells that are susceptible to apoptosis. Our laboratory has previously documented severe abnormalities in the thymic tissue of murine lupus compared with several control strains [9–16]. In this study, we have extended our observations by focusing on changes of thymic architecture following the induction of LPS-mediated apoptosis in murine lupus.
Bone marrow-derived precursor T cells traffic through the thymus and participate in a phenomenon termed negative and positive selection. This process of positive/negative selection involves the interaction of precursor T cells with various cell lineages, including thymic stromal cells, which form the framework of the thymic tissue [1–4]. The precise precursor T cell and thymic stromal cell lineages, the receptor–ligand interaction and the intracellular molecules that are involved in processes that distinguish negative from positive selection are not well understood. It is generally accepted that precursor T cells upon interaction with cell lineages, including thymic epithelial cells and thymic nurse cells (TNC), undergo both Correspondence to: Dr M. Eric Gershwin, Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, School of Medicine, One Shields Avenue, TB192, Davis, CA 95616–8660, USA. Fax: 530–752–4669. E-mail: [email protected] 325 0896–8411/99/070325+10 $30.00/0
© 1999 Academic Press
326
Materials and Methods Mice Female NZB-Bln, MRL/MP-Faslpr (MRL/lpr), C3H/ HeJ-Fasgld (C3H/gld), BALB/c, C57BL/6 and male BXSB/MpJ Yaa (BXSB/Yaa) mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA). All mice were studied beginning at 4–5 weeks of age.
Y. Takeoka et al.
in a humidified chamber for 30 min at room temperature (RT). For detection, a 0.05% 3,3′-diaminobenzidine tetrahydrochloride substrate solution was applied to the sections. Finally, after washing in PBS, the sections were stained by immunohistochemical techniques as described below. Apoptotic cells were counted using a BH-2 Olympus Microscope at ×400. The data are expressed as mean±SE of apoptotic cells visualized in a total of five distinct fields. Immunohistochemistry
Administration of LPS An LPS preparation, extracted by Westphal method from Salmonella typhosa 0901, was obtained from Difco Laboratories (Detroit, MI, USA). All mice, except controls, were injected intraperitoneally (ip) with LPS suspended in saline at 2.5 mg/kg body weight. Control mice received saline. Groups of eight mice in each experimental cohort were killed 18 h following injection.
Tissue sections Thymii were snap-frozen in dry ice-cold 2-methyl butane, and embedded in TISSUE-TECtm (Miles Laboratories, Elkhart, IN, USA). Freshly cut sections (5 m) were mounted on clean glass slides coated with poly-L-lysine (Sigma Chemical Co., St. Louis, MO, USA), and rapidly air-dried and stored at −80°C until utilized. Sections were washed three times in phosphate-buffered saline (PBS), fixed with 10% neutral-buffered formalin at room temperature for 10 min, washed in PBS, and finally immersed in ethanol-acetic acid mixed at a 2:1 ratio at −20°C for 5 min.
Detection of apoptosis The Apoptag In Situ Apoptosis Detection Kit (Oncor Inc., Gaithersburg, MD, USA) was used to detect apoptosis. Residues of digoxigenin-nucleotides are catalytically added to DNA by terminal deoxynucleotidly transferase (TdT) [17] with the incorporated nucleotides forming a random heteropolymer of digoxigenin-11-dUTP and dATP [18]. This mixed molecular biological–histochemical system allows for sensitive and specific staining of the 3′-OH ends localized in apoptotic bodies. To quench endogenous peroxidase, the fixed sections were soaked in 2.0% hydrogen peroxide in PBS for 5 min at room temperature and rinsed twice with PBS for 5 min. After the addition of an equilibration buffer, the sections were incubated with TdT enzyme and incubated for 1 h at 37°C in a humidified chamber. Sections were subsequently incubated in a stop/wash buffer for 30 min at 37°C, with gentle shaking every 10 min. After washing three times, the slides were incubated with peroxidase labeled anti-digoxigenin antibody
A series of mAb’s that stain distinct lineages of thymic epithelial cells were studied herein. These antibodies included a pan epithelium (MTS1 Ab), cortical epithelium Ab (MTS44), and an anti-keratin Ab. These mAbs were selected because abnormalities of thymic cortical epithelial cells in murine lupus have been previously reported using these mAbs [13–15]. MTS1 and MTS44 were prepared from the fusion of P3-NS1-Ag4–1 (NS-1) cells with spleen cells or popliteal lymph node cells from LOU/M rats immunized with enriched mouse thymic stromal cell suspensions [19], and have been previously designated as belonging to classifications I and IIIB, respectively, of the clusters of thymic epithelial staining (CTES) classification system [20]. The fixed sections described above were incubated for 15 min at RT with normal sheep or goat serum diluted 1:5 in PBS for blocking of non-specific background, and then stained using an indirect immunofluorescence method or a standard avidin–biotin detection system [21]. To stain by indirect immunofluorescence, the sections were incubated with each MTS Ab for 1 h at RT in a moist chamber, then washed three times in PBS for 5 min with gentle shaking. They were then incubated with a 1:100 dilution of FITC-conjugated sheep anti-rat Ig (The Binding Site, Birmingham, UK) for 30 min at RT, washed and mounted with Slow Fade reagent (Molecular Probes, Eugene, OR, USA). In the case of anti-keratin Ab, FITC-conjugated anti-rabbit Ig (Tago Inc., Burlingame, CA, USA) diluted 1:100 was used as a secondary antibody. To stain with avidin–biotin, the appropriate sections were incubated with each MTS Ab for 1 h at room temperature in a moist chamber, then washed three times in Tris-buffered saline (TBS) for 5 min with gentle shaking. They were then incubated with biotinylated goat anti-rat Ig (Tago Inc.) diluted 1:200 in TBS for 30 min at RT and washed with TBS. The alkaline phosphatase conjugated streptavidin–biotin complex (AB complex/AP) (Dako, USA) was freshly prepared and applied to sections for 30 min. After washing in TBS, the substrate (Vector Red or Vector Blue; Vector Inc., Burlingame, CA, USA) was applied to the sections and incubated for 10–15 min. Levamisole was added to the substrate to block endogenous alkaline phosphatase activity. After washing in TBS, the sections were mounted with Slow Fade reagent (Molecular Probes), and a cover slip was applied. In the case of anti-keratin Ab, biotinylated goat
Thymic microenvironment
anti-rabbit Ig (Tago Inc.) diluted 1:200 was used as secondary antibody. The slides were viewed using an Olympus BH-2 with a Bio-Rad MRC 600 laser confocal microscope (Bio-Rad Laboratories, Hercules, CA, USA) with a GHS filter block and a No. 2 neutral density filter for observation. The resulting confocal images were analysed using an SOM program integrated with the Bio-Rad confocal system.
Thymocyte apoptosis induced by dexamethasone (Dex) in vitro Thymocytes from 1-month-old BALB/c, C57BL/6 or NZB mice were obtained by gently pressing the thymus in Hank’s buffer solution. Thymocytes were cultured with 24-well microculture plates with or without 0.01, 0.1 or 1 M of Dex (dexamethasonewater soluble, Sigma Chemical Co.) at a final cell concentration of 5×106 cells/ml in RPMI 1640 supplemented with 10% FCS. The cultures were incubated in an atmosphere of 5% CO2/95% air at 37°C, collected at 0, 3, 8 or 15 h following incubation, and fixed with 4% neutral-buffered formalin (NBF) for 15 h at 4°C. The fixed cells were mounted on clean glass slides coated with poly-L-lysine after washing with PBS, and rapidly air dried. These thymocytes were then assayed for the frequency of apoptotic cells as described above. The apoptotic cells were counted in randomly selected fields at a magnification of ×400. The percentage of apoptotic cells in over 500 counted cells was expressed as mean±SE of nine mice in three repeat experiments.
Purification of thymic nurse cells (TNC) Thymic lymphoid cells and TNC were isolated from thymic tissue as previously described [22]. Briefly, a total of 9–10 thymii were removed and 4–6 incisions made in the lobe of each thymus. The tissues were then placed in RPMI 1640. Free exuding thymocytes were washed from the thymus at 4°C by gentle shaking with a magnetic stirrer. The washed thymii were suspended for 5 min at 30°C in 5 ml of a freshly made digestion medium consisting of RPMI 1640 containing 0.15% (w/v) collagenase (type IA, EC3.4.24.3) (Sigma) and 0.01% (W/V) DNase (type IV, EC3.1.21.1) (Sigma). The resulting cells and debris were allowed to settle and the supernatant fluid removed. The cells and cell debris containing stromal fragments were then digested for 5 min at 30°C in 5 ml of a freshly made digestion medium consisting of RPMI 1640 containing 0.3% (w/v) collagenase/ dispase (EC3.4.24.3/EC3.4.24.4) (Sigma) and 0.01% (W/V) DNase (type IV, EC3.1.21.1). Again, the suspension was allowed to settle and the supernatant removed. The stromal fragments were then further digested by incubation for 30 min at 37°C in 5 ml of a freshly made digestion medium consisting of RPMI 1640 containing 0.3% collagenase/dispase and 0.01% DNase with shaking and pipeting every 5 min. After
327
digestion, the cells were centrifuged at 800 rpm for 5 min. The pellets were suspended in PBS with 0.05 M EDTA to disaggregate the cells and fetal bovine serum (FBS) was added. The cell suspensions were then incubated on ice for 30 min to separate TNC. This disaggregation procedure was repeated to obtain purified TNC (>95%). Lymphocytes (TNC-L) were released from TNC in culture with RPMI-1640 medium containing HEPES buffer pH 7.2, 10% FBS and 50 M 2-mercaptoethanol for 8 h at 37°C and 5% CO2/95% air. Total thymocytes were obtained by gently pressing the thymii through a 35 m mesh strainer and used as control for comparison with TNC-L for FACS analysis. Flow cytometry Monoclonal anti-murine CD3, CD4, CD8, CD25, CD28, CD44, interleukin-2 receptor chain (IL-2R), TCR- and TCR- specific antibodies were purchased from Pharmingen (San Diego, CA, USA). Biotinylated clone F23.1 mAb detects murine TCR-V 8.1, 8.2 and 8.3. Rat monoclonal antibody MTS32 or MTS33 detects medullary thymocytes and epithelial cells [19]. Single cell suspensions of total thymocytes or TNC-L at 107/ml were dispensed in individual tubes in a volume of 0.1 ml in PBS containing 0.1% BSA and 0.1% NaN3 (PBS–BSA–NaN3). Cells were incubated at 4°C for 30 min with a primary reagent, washed with PBS–BSA–NaN3 three times and then resuspended in 0.1 ml of PBS–BSA–NaN3. The cell suspensions were incubated at 4°C for an additional 30 min with an appropriate developing reagent. Optional dilutions of each of the reagents were previously determined. The thymocytes were stained with 0.1 ml of a 1:100 dilution of FITC conjugated anti-mouse CD4, a 1:100 dilution of phycoerythrin (PE)-conjugated anti-mouse CD8 and then one of the biotin-conjugated anti-mouse antibodies, CD3, CD25, CD28, CD44, TCR- or F23.1. The cells were washed and then incubated with 0.1 ml of a 1:50 dilution of Tricolor streptavidin. For TCR- determination, aliquots were stained with 0.1 ml of a 1:100 dilution of FITC conjugated anti-mouse TCR-, PE-conjugated anti-mouse CD8 and biotin-conjugated anti-mouse CD4 followed by a wash and the addition of 0.1 ml of a 1:50 dilution of Tricolor streptavidin. For IL-2R analysis, cells were stained with 0.1 ml of a 1:100 dilution of FITC-conjugated anti-mouse IL-2R, PE-conjugated anti-mouse CD45R and biotinconjugated anti-mouse CD25, followed by 0.1 ml of a 1:50 dilution of Tricolor streptavidin. For the delineation of MTS antibody reactivity, aliquots of the cells were individually stained with 0.1 ml of undiluted MTS32 or MTS33 followed by 0.1 ml of a 1:50 dilution of an FITC conjugated sheep anti-rat Ig (The Binding Site). The cells were then washed three times with PBS–BSA–NaN3 and resuspended in 0.5 ml of PBS–BSA–NaN3 for flow cytometric analysis of 5,000 cells on a FACScan (Becton Dickinson, San Jose, CA, USA) using forward and side scatter to separate out dead cells and debris.
328
Y. Takeoka et al.
Table 1. Numbers of apoptotic cell complexes in murine thymii
BALB/c C57BL/6 NZB MRL/lpr BXSB/Yaa C3H/gld
Salinea
LPSa
3.5±0.6b 3.1±0.7 1.0±0.4c 8.4±2.3 4.8±1.1 4.4±1.2
129.3±6.9 136.8±9.9 21.9±1.9b,c,d 56.1±7.8c,d 65.1±9.6c,d 17.4±3.3c,d
a The data are expressed as mean ±SEM of apoptotic cells/five fields. b Statistically significant (t-test), P<0.005. c Compared to BALB/c. d Compared to C57BL/6.
Results
Induction of apoptosis among thymocytes cultured in vitro with dexamethasone (Dex) The quantitative analysis of the kinetics of Dexinduced apoptotic DNA strand breaks in cultured thymocytes from control or murine lupus was performed. After 3 h of incubation with Dex, 10–20% of thymocytes from normal or SLE-prone mice were positive for apoptosis, compared to about 4% of saline treated thymocytes. After 15 h of incubation with Dex, the percentages of thymocytes with DNA strand breaks approached 70–80%, compared to 20% of saline-treated thymocytes incubated for the same time. Thymocytes from NZB mice expressed no significant difference of apoptosis induction at each concentration of Dex in vitro as compared to BALB/c or C57BL/6 mice, except at 1 M of Dex after 8 h culture.
Detection of apoptotic cells
Number of thymic nurse cells (TNC)
The apoptotic-positive cells were determined using the apoptosis detection system and consisted usually of groups of two or more cells. The apoptotic cells were primarily found in the thymic cortex, especially the outer cortex; but were not found in the thymic medulla. LPS administration in vivo resulted in a significantly increased level of apoptosis (Table 1, Figure 1). The numbers of apoptotic cells in BALB/c or C57BL/6 mice was greater than 100 per five fields (Figure 1). However, following LPS injection, the thymii of NZB (Figure 1) or C3H/gld mice had fivefold fewer apoptotic cell complexes compared with those seen in BALB/c or C57BL/6 thymic tissue (P<0.01, Student’s t-test). The number of apoptoticpositive thymic cell complexes for MRL/lpr or BXSB/ Yaa mice was also significantly lower than that on BALB/c or C57BL/6 mice (Table 1); these were again most prominent in the outer thymic cortex. The experimental conditions studied herein focus on developing thymocytes.
A total of 38.9×104 cells/tissue weight of thymus (g) and 38.7×104 cells/tissue weight of thymus (g) were obtained from the thymii of BALB/c and C57BL/6 mice respectively. In contrast, thymii from NZB mice yielded 27.3×104 cells/tissue weight of thymus (g), or about 30% less (P<0.05, Student’s t-test).
Thymic stromal architecture The thymic epithelial architecture of murine strains treated with LPS was compared to those of saline treated mice. In the cortical region, MTS44 demonstrated a different staining pattern of cortical epithelium in LPS-injected BALB/c (Figure 2) or C57BL/6 mice. The network of cortical epithelial cells became diffuse and many large atypical epithelial cells were also observed in both normal murine strains. In contrast, NZB (Figure 2) and C3H/gld mice had severe abnormalities of the cortical network. LPStreated NZB and C3H/gld mice expressed similar changes as compared to saline-treated mice, and had a few large atypical epithelial cells. In MRL/lpr (Figure 2) and BXSB/Yaa mice, large atypical epithelial cells were observed within the cortical area. MTS1 (Figure 3) and anti-keratin antibodies (data not shown) which recognized pan epithelial cells had similar results compared to MTS44.
Flow cytometric analysis Frequency analysis demonstrated an increase in CD3 + cells in the TNC-L fraction of thymic tissue from NZB mice as compared to similarly obtained fraction of TNC-L of BALB/c mice (Table 2). This difference in the yield of CD3 + cells was to a large extent due to the increase of CD3low density+ cells. The frequency of CD3 + cells within the TNC-L of C57BL/6 mice was also significantly higher than that seen in BALB/c mice. However, no differences were seen in unfractionated thymocytes. Analysis for CD3, CD4 and CD8 expression revealed that all subpopulations were defined in comparison with total thymocytes (Table 2). Doublepositive CD4 + 8 + cells were the dominant population and significantly decreased in TNC-L of all strains compared to total thymocytes, but not between TNC-L of each strain. CD4 − 8 − cells in TNC-L were significantly increased compared to total thymocytes; CD4 + 8 − cells in each strain were also significantly increased. In the case of CD4 − 8 + cells in TNC-L, NZB mice expressed significant increases compared to total thymocytes, but this was not so for BALB/c or C57BL/6 mice. CD3 + 4 − 8 − cells of TNC-L in all strains were increased and almost all were included in a low frequency population of CD3 + . NZB mice had an increase in CD3 + 4 + 8 − , CD3 + 4 − 8 + or CD3high4 − 8 + cells of TNC-L. In analysis for expression of T cell receptor antigens (Table 2), TNC-L of NZB mice had a significant increase in TCR- positive cells as compared to total thymocytes. Percentages of CD44 + cells were decreased in TNC-L of a high expression (BALB/c) or low expression strain (C57BL/6 or NZB)
Thymic microenvironment
329
Figure 1. Apoptosis staining of thymus from BALB/c (A, B), C57BL/6 (C, D) and NZB (E, F) mice treated with saline (A, C, E) or LPS (B, D, F). Dark cells which reflect apoptosis, were observed in the cortex (confocal microscope, ×100; zoom 1.3).
(Table 3); the data depended on the subpopulation of CD4 + 8 + in CD44 + cells. CD44 + 4 + 8 − cells in TNC-L of NZB mice were increased and were similar to BALB/c mice.
Discussion It is now generally accepted that thymocytes recognizing self-antigens are eliminated by the process of negative selection in the thymus by apoptosis. Apoptotic cells have a characteristic morphology, including marked condensation of chromatin associated with shrinkage of the nucleus and fragmentation of DNA [23]. This fragmentation is inferred from gel electrophoresis of a pooled DNA extract and is the basis for the in situ visualization at the single cell level [24, 25].
Injection of LPS results in increased levels of thymocyte apoptosis. DNA fragmentation of thymocytes is dependent on the dose of injected LPS and reaches a peak 18 h after injection. In the present study, the apoptotic cells were located within the thymus but were not observed in epithelial-free regions (EFR). Generally, epithelial cells were found to surround the apoptotic cells. Electron micrographic techniques have shown that isolated TNC are the sites of thymic apoptosis [26]. In addition, an epithelial cell line, B/c. TEC-1, can form TNC and induce thymocyte apoptosis using in vitro culture systems [12]. The induction of apoptosis by LPS in murine lupus, especially in NZB and C3H/gld mice, was significantly lower than that of normal strains. The existing diffused cortical network was not changed by treatment with LPS, and there were a few TNC-like large epithelial cells in NZB or C3H/gld thymii. In MRL/lpr or BXSB/Yaa mice, the induction of
330
Y. Takeoka et al.
Figure 2. MTS44, which recognizes cortical epithelium, staining of thymus from BALB/c (A, B), NZB (C, D) or MRL/lpr (E, F) mice using AB complex/AP. In BALB/c treated with LPS (B), staining of epithelial cells in cortex was reduced, cortical epithelial free regions and large atypical epithelial cells were formed. Panel A represents staining of the normal cortical epithelial network in saline treated BALB/c mice. Panel C (NZB) or E (MRL/lpr) represents staining of the cortical epithelial network in saline treated murine lupus. Panel D (NZB) or F (MRL/lpr) represents the epithelial staining in LPS treated murine lupus. The cortical epithelial network was markedly reduced in panel C and E with increased areas of non-staining. These abnormalities were also observed in LPS treated NZB mice thymus. Arrowheads point to cortical epithelial cell-free regions. The bar in panel F indicates 250 m (confocal microscope, ×100; zoom 1.3).
apoptosis and epithelial morphology was intermediate between NZB and control mice. In contrast, there was no significant difference of Dex treatment in vitro in the induction of apoptosis in NZB thymocytes. It has been already reported that in vitro LPS treatment of thymocytes from normal mice does not result in apoptosis [24]. The induction of apoptosis in vivo by LPS may require factors other than direct effect of corticoids on thymocytes, because sensitivity of thymocytes from NZB mice against the Dex induced apoptosis is the same in control mice. In addition, there is no evidence for a disturbance of corticosteroid production in New Zealand mice [27]. The role of the thymic nurse cell as the site of
thymic apoptosis and apoptotic cell clearance has attracted recent attention [28]. It appears that the TNC consists of one thymic epithelial cell surrounded by up to 200 thymocytes enclosed by an epithelial cytoplasmic process [22]. Although further research is needed, it appears that the TNC is involved in both positive and negative selection [29]. However, it has been very difficult to correlate the function of thymic nurse cells in vitro with in vivo structure and activity. This has been particularly a problem in humans where in the past some have felt that the TNC may be no more than an in vitro isolation artifact; more recent data suggest that TNCs have a similar function to that in the mouse [28].
Thymic microenvironment
331
Figure 3. MTSI, which recognizes pan epithelium, staining of thymus from BALB/c (A, B), NZB (C, D) or MRL/lpr (E, F) mice using AB complex/AP. M=medulla, C=cortex. In BALB/c treated with LPS (B), staining of epithelial cells in cortex was reduced, cortical epithelial free regions and large atypical epithelial cells were formed. Panel A represents staining of the normal cortical epithelial network in saline treated BALB/c mice. Panel C (NZB) or E (MRL/lpr) represents staining of the cortical epithelial network in saline-treated murine lupus. Panel D (NZB) or F (MRL/lpr) represents the epithelial staining in LPS treated murine lupus. The cortical epithelial network was markedly reduced in saline treated NZB and MRL/lpr mice with increased areas of non-staining. These abnormalities were also observed in LPS treated NZB mice thymus. Arrows point to cortical epithelial cell-free regions. The bar in panel F indicates 250 m (confocal microscope, ×100; zoom 1.3).
The thymus of NZB mice undergoes premature involution and in vitro studies suggest that cultured thymic epithelia from NZB mice underwent accelerated, preprogrammed degeneration [30]. Recently, NZB mice were found to have distinctive and well-defined abnormalities of thymic architecture involving stromal cells defined by staining with monoclonal antibodies specific for the thymic microenvironment [12]. The thymus of other SLE mouse strains, MRL/lpr, BXSB/Yaa, C3H/gld and (NZB×NZW) F1 mice, were also known to have similar abnormalities of epithelial architecture [14, 15]. C3H/gld mice had the most severe abnormalities of cortical thymocytes among these SLE mice, with sig-
nificant changes in CD4 + 8 + double-positive (DP) and single-positive (SP) T cells [14]. The percentage of DP cells was significantly increased. On the other hand, CD4 + 8 − and CD4 − 8 + SP cells were significantly decreased in C3H/gld mice. Other SLE mice also showed a similar, but not statistically significant change in SP and DP thymocyte populations. Thymocytes in the thymic cortex are primarily CD3 + 4 + 8 + [31] and begin the conversion to SP status at the cortico–medullary interface. The positive selection of CD4 + thymocytes is related to their adherence to cortical epithelia, which express MHC class II antigens [32]. Positive selection of CD8 + cells also presumably occurs in the cortex [33]. These data
332
Y. Takeoka et al.
Table 2. FACS analysis of CD3, 4, or 8 expression on TNC-lymphocytes Positive cells (%)a Subsets
CD3 + CD3high CD3low CD4 + 8 + (DP) CD4 + 8 − (SP) CD4 − 8 + (SP) CD4 − 8 − (DN) CD3 + 4 + 8 + CD3 + 4 + 8 − CD3 + 4 − 8 + CD3 + 4 − 8 − CD3high4 + 8 + CD3high4 + 8 − CD3high4 − 8 + CD3high4 − 8 − CD3low4 + 8 + CD3low4 + 8 − CD3low4 − 8 + CD3low4 − 8 −
Total thymocytes
TNC-L
BALB/c
C57BL/6
NZB
BALB/c
C57BL/6
NZB
55.3±1.9b 15.7±2.5 39.7±1.7 82.0±2.0 12.4±1.2 2.5±0.4 3.0±0.9 40.2±1.6 12.2±1.2 1.9±0.4 0.7±0.1 6.2±1.0 8.1±1.7 1.0±0.4 0.1±0.03 34.0±1.6 4.2±1.2 0.9±0.2 0.6±0.1
53.8±3.5 14.0±2.8 39.9±4.2 84.7±3.1 9.5±1.5 2.4±0.6 2.8±0.9 41.9±3.4 9.3±1.4 1.6±0.3 0.6±0.1 7.3±2.5 5.6±0.9 0.7±0.2 0.1±0.01 34.6±3.6 3.7±1.5 0.9±0.3 0.6±0.1
53.5±2.5 17.5±2.9 35.7±4.1 85.5±1.7 7.9±1.0 2.3±0.3 4.3±1.5 42.4±2.0 7.7±1.0 1.9±0.2 1.3±0.5 10.5±3.2 5.9±0.9 0.9±0.1 0.1±0.05 32.0±3.5 1.9±0.4 1.0±0.1 0.8±0.4
53.5±2.1 11.8±1.5 41.8±2.0 70.0±1.2d 16.7±0.8 3.1±0.4 8.9±0.9d 32.6±1.4d 11.8±1.2 2.2±0.4 3.8±0.4d 2.7±0.4 6.5±1.2 1.1±0.2 0.3±0.1 29.9±1.4 6.6±0.8 1.2±0.3 3.5±0.4d
61.1±1.4c 18.3±1.9 45.2±2.0 64.7±2.3d 18.9±1.4d 4.8±0.9 10.4±1.1d 34.0±1.8 15.5±2.1 3.6±0.8 5.2±0.8d 4.0±0.5 10.9±1.7 1.8±0.5 0.4±0.5d 30.0±1.9 6.0±0.7 1.8±0.3 4.7±0.8d
67.7±1.8c,d 16.5±0.6 51.3±2.2c,d 66.1±2.6d 17.6±1.7d 6.1±1.1d 9.3±0.7d 38.9±3.5 16.4±1.9d 5.1±0.9c,d 6.2±0.8d 2.9±0.2 9.7±1.0 2.2±0.4d 0.7±0.1c,d 36.0±3.4 6.8±1.0 2.9±0.9 5.5±0.7d
a
Data are expressed as mean±SEM of % positive cells. Statistically significant (Student’s t-test, P<0.01) as compared with BALB/c in each group of total thymocytes or TNC-L (c), or with each strain in group of total thymocytes (d). b
Table 3. FACS analysis of TCR, MTS or other CD28, 44, 45, IL2R on TNC-lymphocytes Positive cells (%)a Subsets
TCR- + TCR- + TCR-V8 + MTS32 + MTS33 + MTS33high CD44 + CD44 + 4 + 8 + CD44 + 4 + 8 − CD44 + 4 − 8 + CD44 + 4 − 8 − CD28 + CD45R + CD25 + IL-2R +
Total thymocytes
TNC-L
BALB/c
C57BL/6
NZB
BALB/c
C57BL/6
NZB
46.4±2.2b 1.1±0.1 14.7±0.7 94.7±1.7 87.0±4.4 45.6±4.9 58.2±4.4 46.0±3.4 10.2±2.4 1.2±0.3 1.4±0.4 64.9±6.1 0.9±0.1 7.9±1.6 1.6±0.2
45.7±4.4 1.0±0.2 13.9±1.0 94.8±1.6 89.0±3.6 62.6±1.8 40.2±2.7c 31.8±2.6c 6.3±1.0 1.0±0.1 1.6±0.7 62.0±4.8 0.5±0.1c 12.2±2.3 1.3±0.3
39.8±2.6 0.6±0.1c 15.1±0.7 90.1±3.4 81.2±5.6 30.7±2.7d 40.7±1.6c 30.4±1.7c 6.0±1.0 1.4±0.1 1.5±0.3 63.0±5.5 0.4±0.02c 6.8±2.1 0.8±0.2c
48.1±2.4 7.2±0.6e 18.2±0.3e 90.6±2.1 88.9±2.9 53.6±5.6 38.1±1.5e 25.9±1.2 9.2±0.8 1.2±0.1 1.2±0.1 70.2±4.3 1.8±0.3 9.0±0.9 5.3±0.8e
61.1±1.4 8.1±0.6e 16.7±0.5 91.6±1.5 95.3±1.2 76.5±4.1 25.8±1.4c,e 11.7±1.1c,e 10.7±0.6e 1.0±0.2 1.6±0.4 69.0±6.6 2.3±0.4e 16.6±2.5 8.0±1.3e
57.0±1.6c,d,e 6.4±0.1e 18.9±0.8e 91.2±3.3 90.0±1.4 53.4±7.0e 23.4±1.7c,e 10.1±1.5c,e 9.2±0.3e 1.4±0.3 1.4±0.03 72.3±6.3 1.7±0.5 6.9±0.6c 7.8±2.6
a
Data are expressed as mean±SEM of % positive cells. Statistically significant (Student’s t-test, P<0.01) as compared with BALB/c (c), with C57BL/6 (d) in group of total thymocytes or TNC-L, or with each strain in group of total thymocytes (e). b
imply that abnormalities in the thymic cortex may affect T cell maturation by altering the mechanisms of positive selection, and can be interpreted as indicating that decreased TNC DP formation of thymocytes results in an accumulation following a decrease of apoptosis.
Our findings demonstrate abnormalities among the induction of apoptosis and the formation of TNC-like epithelial cells in SLE mice. As a result, developing thymocytes may not properly bind to the thymic epithelium as TNC and therefore fail to undergo normal positive selection via apoptosis. This may ultimately
Thymic microenvironment
result in abnormal thymocyte education, the loss of self-tolerance induction and autoimmune disease.
References 1. Boyd R.L., Tucek C.L., Godfrey D.I., Izon D.J., Wilson T.J., Davidson N.J., Bean A.G.D., Ladyman H.M., Ritter M.A., Hugo P. 1993. The thymic microenvironment. Immunol. Today 14: 445–459 2. Smith C.A., Williams G.T., Kingston R., Jenkinson E.J., Owen J.J.T. 1989. Antibodies to CD3/T-cell receptor complex induce death by apoptosis in immature T cells in thymic cultures. Nature 337: 181–184 3. Murakami M., Tsubata T., Okamoto M., Shimizu A., Kumagai S., Imura H., Honjo T. 1992. Antigen-induced apoptosis death of Ly-1 B cells responsible for autoimmune disease in transgenic mice. Nature 357: 77–79 4. Telford W.G., King L.E., Franker P.J. 1991. Evaluation of glucocorticoid-induced DNA fragmentation in mouse thymocytes by flow cytometry. Cell Prolif. 24: 1447–1459 5. Migliorati G., Nicoetti I., Pagliacci M.C., D’Adamo L., Riccardi C. 1993. Interleukin-2 induces apoptosis in mouse thymocytes. Cell. Immunol. 146: 52–61 6. Kizaki H., Ohnishi Y., Azuma Y., Mizuno Y., Ohsaka F. 1992. 1--D arabinosylcytosine and 5-azacytidine induce internucleosomal DNA fragmentation and cell death in thymocytes. Immunophamacology 24: 219–227 7. Suzuki K., Tadakuma T., Kizaki H. 1991. Modulation of thymocytes apoptosis by isoproterenol and prostaglandin E2. Cell. Immunol. 134: 235–240 8. Zhang Y-H., Takahashi K., Jiang G.-Z., Kawai M., Fukada M., Yokochi T., 1993. In vivo induction of apoptosis (programed cell death) in mouse thymus by administration of lypopolysaccharide. Infect. Immun. 61: 5044–5048 9. Wekerle H., Ketelsen U.-P. 1980. Thymic nurse cells—la-bearing epithelium involved in T-lymphocyte differentiation? Nature 283: 402–404 10. Wekerle H., Ketelsen U.-P., Ernst M. 1980. Thymic nurse cells—Lymphoepithelial cell complexes in murine thymuses: morphological and serological characterization. J. Exp. Med. 151: 925–944 11. Kyewski B.A. 1986. Thymic nurse cells: Possible sites of T-cell selection. Immunol. Today 7: 374–378 12. Hiramine C., Hojo K., Koseto K., Nakagawa T., Mukase A. 1990. Establishment of a murine thymic epithelial cell line capable of inducing both thymic nurse cell formation and thymocyte apoptosis. Lab. Invest. 62: 41–54 13. Watanabe Y., Naiki M., Wilson T.J., Godfrey D.I., Chiang B.-L., Boyd R.L., Ansari A., Gershwin M.E. 1993. Thymic microenvironmental abnormalities and thymic selection in NZB.H-2bm12 mice. J. Immunol. 150: 4702–4712 14. Takeoka Y., Yoshida S.H., Van de Water J., Boyd R.L., Suehiro S., Ansari A.A., Gershwin M.E. 1995. Thymic microenvironmental abnormalities in MRL/MP-lpr/lpr, BXSB/MpJ Yaa and C3H/Hej-gld/gld mice. J. Autoimmun. 8: 145–161 15. Takeoka Y., Whitmer K.J., Chen S.-Y., Ansari A.A., Boyd R.L., Shultz L., Suehiro S., Gershwin M.E. 1995. Thymic epithelial cell abnormalities in (NZB×H-2u) F1 mice. Clin. Immunol. Immunopathol. 76: 297–307
333
16. Takeoka Y., Taguchi N., Kotzin B.L., Bennett S., Vyse T.J., Boyd R.L., Naiki M., Konishi J.-E., Ansari A.A., Shultz L.D., Gershwin M.E. 1999. Thymic microenvironment and NZB mice: The abnormal thymic microenvironment of New Zealand mice correlates with immunopathology. Clin. Immunol. Immunopathol. 90: 388–398 17. Zychlinsky A., Zheng L.M., Liu C-C, Young J.D.-E. 1991. Cytolytic lymphocytes induce both apoptosis and necrosis in target cells. J. Immunol. 146: 393–400 18. Schmitz G.G., Walter T., Seibl R., Kessler C. 1991. Nonradioactive labeling of oligonucleotides in vitro with the hapten digoxigenin by tailing with terminal transferase. Anal. Biochem. 192: 222–231 19. Godfrey D.I., Izon D.J., Tucek C., Wilson T.J., Boyd R.L. 1990. The phenotypic heterogeneity of mouse thymic stromal cells. Immunology 70: 66–74 20. Kampringa J., Berges S., Boyd R.L., Brekelmans P., Colic M., Van Ewijk W., Kendall M., Ladyman H., Nieuwenhuis P., Ritter M., Schuurmar H.-J., Tournefier A. 1989. Thymic epithelial antibodies: immunohistological analysis and introduction of CTES nomenclature. Summary of the Rolduc Epithelium Workshop. Thymus 13: 65–173 21. Wood G.S., Warnke R. 1981. Suppression of endogeneous biotin-avidin detection systems. J. Histochem. Cytochem. 29: 1196–1204 22. Andrews P., Boyd R.L. 1985. The murine thymic nurse cell: an isolated thymic microenvironment. Eur. J. Immunol. 15: 36–42 23. Arends M.J., Wyline H. 1991. Apoptosis: mechanisms and roles in pathology. Int. Rev. Exp. Pathol. 32: 223–254 24. Gavrieli Y., Sherman Y., Ben-Sasson S.A. 1992. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J. Cell Biol. 119: 493–501 25. Fehsel K., Kro¨ ncke K.D., Kolb H., Kolb-Bachofen V. 1994. In situ nick-translation detects focal apoptosis in thymuses of glucocorticoid- and lipopolysaccharidetreated mice. J. Histochem. Cytochem. 42: 613–619 26. Aguilar L.K., Aguilar-Cordova E., Carwright, Jr. J., Belmont J.W. 1994. Thymic nurse cells are sites of thymocyte apoptosis. J. Immunol. 152: 2645–2651 27. Whitmer K.J., Ansari A., Gershwin M.E. 1994. The pathogenesis of autoimmunity in New Zealand Mice. Semin. Clin. Immunol. 8: 35–53 28. Hiramine C., Nakagawa T., Miyauchi A., Hojo K. 1996. Thymic nurse cells as the site of thymocyte apoptosis and apoptotic cell clearance in the thymus of cyclophosphamide-treated mice. Lab. Invest. 75: 185–201 29. Faas S.J., Rothstein J.L., Kreider B.L., Rovera G., Knowles B.B. 1993. Phenotypically diverse mouse thymic stroma cell lines which induce proliferation and differentiation of hematopoietic cells. Eur. J. Immunol. 23: 1201–1214 30. Mitamura K., Kang H., Tomita Y., Hashimoto H., Sawada S., Horie T. 1991. Impaired tumour necrosis factor- (TNF-) production and abnormal B cell response to TNF- in patients with systemic lupus erythematosus (SLE). Clin. Exp. Immunol. 85: 386–391 31. Gershwin M.E., Ikeda R.M., Kruse W.L., Wilson F., Shifrine M., Spangler W. 1978. Age-dependent loss in New Zealand mice of morphological and functional
334
characteristics of thymic epithelial cells. J. Immunol. 120: 971–979 32. Boyd R.L., Wilson T.J., Ward H.A., Gershwin M.E. 1992. Phenotypic characterization of chicken thymic stromal elements. Dev. Immunol. 2: 51–66
Y. Takeoka et al.
33. Berg L.J., Pullen A.M., de St Growth F.F., Mathis D., Benoist C., Davis M.M. 1989. Antigen/MHC-specific T cells are preferentially exposed from the thymus in the presence of their MHC ligand. Cell 58: 1035–1046
Journal of Autoimmunity (1999) 13, 325–334
Apoptosis and the Thymic Microenvironment in Murine Lupus Yuichi Takeoka1,2, Nobuhisa Taguchi1,2, Leonard Shultz3, Richard L. Boyd4, Mitsuru Naiki2, Aftab A. Ansari5 and M. Eric Gershwin1 1
Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, School of Medicine, Davis, California 95616, USA 2 Institute of Bio-Active Science, Nippon Zoki Pharmaceutical Co. Ltd., Yashiro, Hyogo 673–1461, Japan 3 Jackson Laboratories, Bar Harbor, ME 04609, USA 4 Department of Pathology and Immunology, Monash University, Prahran, Victoria 3181, Australia 5 Department of Pathology, Emory University, School of Medicine, Atlanta, GA 30322, USA Received 18 January 1999 Accepted 14 June 1999 Key words: apoptosis, autoimmunity, murine lupus, thymic microenvironment
The thymus of New Zealand black (NZB) mice undergoes premature involution. In addition, cultured thymic epithelial cells from NZB mice undergo accelerated preprogrammed degeneration. NZB mice also have distinctive and well-defined abnormalities of thymic architecture involving stromal cells, defined by staining with monoclonal antibodies specific for the thymic microenvironment. We took advantage of these findings, as well as our large panel of monoclonal antibodies which recognize thymic stroma, to study the induction of apoptosis in the thymus of murine lupus and including changes of epithelial architecture. We studied NZB, MRL/lpr, BXSB/Yaa, C3H/gld mice and BALB/c and C57BL/6 as control mice. Apoptosis was studied both at basal levels and following induction with either dexamethasone or lipopolysaccharide (LPS). The apoptotic cells were primarily found in the thymic cortex, and the frequency of apoptosis in murine lupus was less than 20% of controls. Moreover, all strains of murine lupus had severe abnormalities of the cortical network. These changes were not accentuated by dexamethasone treatment in cultured thymocytes. However, the thymus in murine lupus was less susceptible to LPS-induced apoptosis than control mice. Finally we note that the number of thymic nurse cells (TNC) was lowest in NZB mice. Our findings demonstrate significant abnormalities in the induction of apoptosis and the formation of TNC-like epithelial cells in SLE mice, and suggest that the abnormalities of the thymic microenvironment have an important role in the pathogenesis of murine lupus. © 1999 Academic Press
Introduction
positive and negative selection; the latter being more pronounced in thymic medulla. The process involved in negative selection, known as apoptosis, has generated considerable interest. Apoptosis has been shown to occur not only in immature thymocytes secondary to ligation of the TCR with appropriate peptide bearing MHC molecules, but also to be initiated by glucocorticoid hormones. The role of glucocorticoids is highlighted by the finding that in vivo administration of LPS increases glucocorticoid secretion by the adrenal gland, leading to a marked increase in numbers of apoptotic cells within the murine thymus within 24 h [5–8]. Investigation of LPS-mediated apoptosis has permitted evaluation of the relative frequency of cells that are susceptible to apoptosis. Our laboratory has previously documented severe abnormalities in the thymic tissue of murine lupus compared with several control strains [9–16]. In this study, we have extended our observations by focusing on changes of thymic architecture following the induction of LPS-mediated apoptosis in murine lupus.
Bone marrow-derived precursor T cells traffic through the thymus and participate in a phenomenon termed negative and positive selection. This process of positive/negative selection involves the interaction of precursor T cells with various cell lineages, including thymic stromal cells, which form the framework of the thymic tissue [1–4]. The precise precursor T cell and thymic stromal cell lineages, the receptor–ligand interaction and the intracellular molecules that are involved in processes that distinguish negative from positive selection are not well understood. It is generally accepted that precursor T cells upon interaction with cell lineages, including thymic epithelial cells and thymic nurse cells (TNC), undergo both Correspondence to: Dr M. Eric Gershwin, Division of Rheumatology, Allergy and Clinical Immunology, University of California at Davis, School of Medicine, One Shields Avenue, TB192, Davis, CA 95616–8660, USA. Fax: 530–752–4669. E-mail: [email protected] 325 0896–8411/99/070325+10 $30.00/0
© 1999 Academic Press
326
Materials and Methods Mice Female NZB-Bln, MRL/MP-Faslpr (MRL/lpr), C3H/ HeJ-Fasgld (C3H/gld), BALB/c, C57BL/6 and male BXSB/MpJ Yaa (BXSB/Yaa) mice were obtained from The Jackson Laboratory (Bar Harbor, ME, USA). All mice were studied beginning at 4–5 weeks of age.
Y. Takeoka et al.
in a humidified chamber for 30 min at room temperature (RT). For detection, a 0.05% 3,3′-diaminobenzidine tetrahydrochloride substrate solution was applied to the sections. Finally, after washing in PBS, the sections were stained by immunohistochemical techniques as described below. Apoptotic cells were counted using a BH-2 Olympus Microscope at ×400. The data are expressed as mean±SE of apoptotic cells visualized in a total of five distinct fields. Immunohistochemistry
Administration of LPS An LPS preparation, extracted by Westphal method from Salmonella typhosa 0901, was obtained from Difco Laboratories (Detroit, MI, USA). All mice, except controls, were injected intraperitoneally (ip) with LPS suspended in saline at 2.5 mg/kg body weight. Control mice received saline. Groups of eight mice in each experimental cohort were killed 18 h following injection.
Tissue sections Thymii were snap-frozen in dry ice-cold 2-methyl butane, and embedded in TISSUE-TECtm (Miles Laboratories, Elkhart, IN, USA). Freshly cut sections (5 m) were mounted on clean glass slides coated with poly-L-lysine (Sigma Chemical Co., St. Louis, MO, USA), and rapidly air-dried and stored at −80°C until utilized. Sections were washed three times in phosphate-buffered saline (PBS), fixed with 10% neutral-buffered formalin at room temperature for 10 min, washed in PBS, and finally immersed in ethanol-acetic acid mixed at a 2:1 ratio at −20°C for 5 min.
Detection of apoptosis The Apoptag In Situ Apoptosis Detection Kit (Oncor Inc., Gaithersburg, MD, USA) was used to detect apoptosis. Residues of digoxigenin-nucleotides are catalytically added to DNA by terminal deoxynucleotidly transferase (TdT) [17] with the incorporated nucleotides forming a random heteropolymer of digoxigenin-11-dUTP and dATP [18]. This mixed molecular biological–histochemical system allows for sensitive and specific staining of the 3′-OH ends localized in apoptotic bodies. To quench endogenous peroxidase, the fixed sections were soaked in 2.0% hydrogen peroxide in PBS for 5 min at room temperature and rinsed twice with PBS for 5 min. After the addition of an equilibration buffer, the sections were incubated with TdT enzyme and incubated for 1 h at 37°C in a humidified chamber. Sections were subsequently incubated in a stop/wash buffer for 30 min at 37°C, with gentle shaking every 10 min. After washing three times, the slides were incubated with peroxidase labeled anti-digoxigenin antibody
A series of mAb’s that stain distinct lineages of thymic epithelial cells were studied herein. These antibodies included a pan epithelium (MTS1 Ab), cortical epithelium Ab (MTS44), and an anti-keratin Ab. These mAbs were selected because abnormalities of thymic cortical epithelial cells in murine lupus have been previously reported using these mAbs [13–15]. MTS1 and MTS44 were prepared from the fusion of P3-NS1-Ag4–1 (NS-1) cells with spleen cells or popliteal lymph node cells from LOU/M rats immunized with enriched mouse thymic stromal cell suspensions [19], and have been previously designated as belonging to classifications I and IIIB, respectively, of the clusters of thymic epithelial staining (CTES) classification system [20]. The fixed sections described above were incubated for 15 min at RT with normal sheep or goat serum diluted 1:5 in PBS for blocking of non-specific background, and then stained using an indirect immunofluorescence method or a standard avidin–biotin detection system [21]. To stain by indirect immunofluorescence, the sections were incubated with each MTS Ab for 1 h at RT in a moist chamber, then washed three times in PBS for 5 min with gentle shaking. They were then incubated with a 1:100 dilution of FITC-conjugated sheep anti-rat Ig (The Binding Site, Birmingham, UK) for 30 min at RT, washed and mounted with Slow Fade reagent (Molecular Probes, Eugene, OR, USA). In the case of anti-keratin Ab, FITC-conjugated anti-rabbit Ig (Tago Inc., Burlingame, CA, USA) diluted 1:100 was used as a secondary antibody. To stain with avidin–biotin, the appropriate sections were incubated with each MTS Ab for 1 h at room temperature in a moist chamber, then washed three times in Tris-buffered saline (TBS) for 5 min with gentle shaking. They were then incubated with biotinylated goat anti-rat Ig (Tago Inc.) diluted 1:200 in TBS for 30 min at RT and washed with TBS. The alkaline phosphatase conjugated streptavidin–biotin complex (AB complex/AP) (Dako, USA) was freshly prepared and applied to sections for 30 min. After washing in TBS, the substrate (Vector Red or Vector Blue; Vector Inc., Burlingame, CA, USA) was applied to the sections and incubated for 10–15 min. Levamisole was added to the substrate to block endogenous alkaline phosphatase activity. After washing in TBS, the sections were mounted with Slow Fade reagent (Molecular Probes), and a cover slip was applied. In the case of anti-keratin Ab, biotinylated goat
Thymic microenvironment
anti-rabbit Ig (Tago Inc.) diluted 1:200 was used as secondary antibody. The slides were viewed using an Olympus BH-2 with a Bio-Rad MRC 600 laser confocal microscope (Bio-Rad Laboratories, Hercules, CA, USA) with a GHS filter block and a No. 2 neutral density filter for observation. The resulting confocal images were analysed using an SOM program integrated with the Bio-Rad confocal system.
Thymocyte apoptosis induced by dexamethasone (Dex) in vitro Thymocytes from 1-month-old BALB/c, C57BL/6 or NZB mice were obtained by gently pressing the thymus in Hank’s buffer solution. Thymocytes were cultured with 24-well microculture plates with or without 0.01, 0.1 or 1 M of Dex (dexamethasonewater soluble, Sigma Chemical Co.) at a final cell concentration of 5×106 cells/ml in RPMI 1640 supplemented with 10% FCS. The cultures were incubated in an atmosphere of 5% CO2/95% air at 37°C, collected at 0, 3, 8 or 15 h following incubation, and fixed with 4% neutral-buffered formalin (NBF) for 15 h at 4°C. The fixed cells were mounted on clean glass slides coated with poly-L-lysine after washing with PBS, and rapidly air dried. These thymocytes were then assayed for the frequency of apoptotic cells as described above. The apoptotic cells were counted in randomly selected fields at a magnification of ×400. The percentage of apoptotic cells in over 500 counted cells was expressed as mean±SE of nine mice in three repeat experiments.
Purification of thymic nurse cells (TNC) Thymic lymphoid cells and TNC were isolated from thymic tissue as previously described [22]. Briefly, a total of 9–10 thymii were removed and 4–6 incisions made in the lobe of each thymus. The tissues were then placed in RPMI 1640. Free exuding thymocytes were washed from the thymus at 4°C by gentle shaking with a magnetic stirrer. The washed thymii were suspended for 5 min at 30°C in 5 ml of a freshly made digestion medium consisting of RPMI 1640 containing 0.15% (w/v) collagenase (type IA, EC3.4.24.3) (Sigma) and 0.01% (W/V) DNase (type IV, EC3.1.21.1) (Sigma). The resulting cells and debris were allowed to settle and the supernatant fluid removed. The cells and cell debris containing stromal fragments were then digested for 5 min at 30°C in 5 ml of a freshly made digestion medium consisting of RPMI 1640 containing 0.3% (w/v) collagenase/ dispase (EC3.4.24.3/EC3.4.24.4) (Sigma) and 0.01% (W/V) DNase (type IV, EC3.1.21.1). Again, the suspension was allowed to settle and the supernatant removed. The stromal fragments were then further digested by incubation for 30 min at 37°C in 5 ml of a freshly made digestion medium consisting of RPMI 1640 containing 0.3% collagenase/dispase and 0.01% DNase with shaking and pipeting every 5 min. After
327
digestion, the cells were centrifuged at 800 rpm for 5 min. The pellets were suspended in PBS with 0.05 M EDTA to disaggregate the cells and fetal bovine serum (FBS) was added. The cell suspensions were then incubated on ice for 30 min to separate TNC. This disaggregation procedure was repeated to obtain purified TNC (>95%). Lymphocytes (TNC-L) were released from TNC in culture with RPMI-1640 medium containing HEPES buffer pH 7.2, 10% FBS and 50 M 2-mercaptoethanol for 8 h at 37°C and 5% CO2/95% air. Total thymocytes were obtained by gently pressing the thymii through a 35 m mesh strainer and used as control for comparison with TNC-L for FACS analysis. Flow cytometry Monoclonal anti-murine CD3, CD4, CD8, CD25, CD28, CD44, interleukin-2 receptor chain (IL-2R), TCR- and TCR- specific antibodies were purchased from Pharmingen (San Diego, CA, USA). Biotinylated clone F23.1 mAb detects murine TCR-V 8.1, 8.2 and 8.3. Rat monoclonal antibody MTS32 or MTS33 detects medullary thymocytes and epithelial cells [19]. Single cell suspensions of total thymocytes or TNC-L at 107/ml were dispensed in individual tubes in a volume of 0.1 ml in PBS containing 0.1% BSA and 0.1% NaN3 (PBS–BSA–NaN3). Cells were incubated at 4°C for 30 min with a primary reagent, washed with PBS–BSA–NaN3 three times and then resuspended in 0.1 ml of PBS–BSA–NaN3. The cell suspensions were incubated at 4°C for an additional 30 min with an appropriate developing reagent. Optional dilutions of each of the reagents were previously determined. The thymocytes were stained with 0.1 ml of a 1:100 dilution of FITC conjugated anti-mouse CD4, a 1:100 dilution of phycoerythrin (PE)-conjugated anti-mouse CD8 and then one of the biotin-conjugated anti-mouse antibodies, CD3, CD25, CD28, CD44, TCR- or F23.1. The cells were washed and then incubated with 0.1 ml of a 1:50 dilution of Tricolor streptavidin. For TCR- determination, aliquots were stained with 0.1 ml of a 1:100 dilution of FITC conjugated anti-mouse TCR-, PE-conjugated anti-mouse CD8 and biotin-conjugated anti-mouse CD4 followed by a wash and the addition of 0.1 ml of a 1:50 dilution of Tricolor streptavidin. For IL-2R analysis, cells were stained with 0.1 ml of a 1:100 dilution of FITC-conjugated anti-mouse IL-2R, PE-conjugated anti-mouse CD45R and biotinconjugated anti-mouse CD25, followed by 0.1 ml of a 1:50 dilution of Tricolor streptavidin. For the delineation of MTS antibody reactivity, aliquots of the cells were individually stained with 0.1 ml of undiluted MTS32 or MTS33 followed by 0.1 ml of a 1:50 dilution of an FITC conjugated sheep anti-rat Ig (The Binding Site). The cells were then washed three times with PBS–BSA–NaN3 and resuspended in 0.5 ml of PBS–BSA–NaN3 for flow cytometric analysis of 5,000 cells on a FACScan (Becton Dickinson, San Jose, CA, USA) using forward and side scatter to separate out dead cells and debris.
328
Y. Takeoka et al.
Table 1. Numbers of apoptotic cell complexes in murine thymii
BALB/c C57BL/6 NZB MRL/lpr BXSB/Yaa C3H/gld
Salinea
LPSa
3.5±0.6b 3.1±0.7 1.0±0.4c 8.4±2.3 4.8±1.1 4.4±1.2
129.3±6.9 136.8±9.9 21.9±1.9b,c,d 56.1±7.8c,d 65.1±9.6c,d 17.4±3.3c,d
a The data are expressed as mean ±SEM of apoptotic cells/five fields. b Statistically significant (t-test), P<0.005. c Compared to BALB/c. d Compared to C57BL/6.
Results
Induction of apoptosis among thymocytes cultured in vitro with dexamethasone (Dex) The quantitative analysis of the kinetics of Dexinduced apoptotic DNA strand breaks in cultured thymocytes from control or murine lupus was performed. After 3 h of incubation with Dex, 10–20% of thymocytes from normal or SLE-prone mice were positive for apoptosis, compared to about 4% of saline treated thymocytes. After 15 h of incubation with Dex, the percentages of thymocytes with DNA strand breaks approached 70–80%, compared to 20% of saline-treated thymocytes incubated for the same time. Thymocytes from NZB mice expressed no significant difference of apoptosis induction at each concentration of Dex in vitro as compared to BALB/c or C57BL/6 mice, except at 1 M of Dex after 8 h culture.
Detection of apoptotic cells
Number of thymic nurse cells (TNC)
The apoptotic-positive cells were determined using the apoptosis detection system and consisted usually of groups of two or more cells. The apoptotic cells were primarily found in the thymic cortex, especially the outer cortex; but were not found in the thymic medulla. LPS administration in vivo resulted in a significantly increased level of apoptosis (Table 1, Figure 1). The numbers of apoptotic cells in BALB/c or C57BL/6 mice was greater than 100 per five fields (Figure 1). However, following LPS injection, the thymii of NZB (Figure 1) or C3H/gld mice had fivefold fewer apoptotic cell complexes compared with those seen in BALB/c or C57BL/6 thymic tissue (P<0.01, Student’s t-test). The number of apoptoticpositive thymic cell complexes for MRL/lpr or BXSB/ Yaa mice was also significantly lower than that on BALB/c or C57BL/6 mice (Table 1); these were again most prominent in the outer thymic cortex. The experimental conditions studied herein focus on developing thymocytes.
A total of 38.9×104 cells/tissue weight of thymus (g) and 38.7×104 cells/tissue weight of thymus (g) were obtained from the thymii of BALB/c and C57BL/6 mice respectively. In contrast, thymii from NZB mice yielded 27.3×104 cells/tissue weight of thymus (g), or about 30% less (P<0.05, Student’s t-test).
Thymic stromal architecture The thymic epithelial architecture of murine strains treated with LPS was compared to those of saline treated mice. In the cortical region, MTS44 demonstrated a different staining pattern of cortical epithelium in LPS-injected BALB/c (Figure 2) or C57BL/6 mice. The network of cortical epithelial cells became diffuse and many large atypical epithelial cells were also observed in both normal murine strains. In contrast, NZB (Figure 2) and C3H/gld mice had severe abnormalities of the cortical network. LPStreated NZB and C3H/gld mice expressed similar changes as compared to saline-treated mice, and had a few large atypical epithelial cells. In MRL/lpr (Figure 2) and BXSB/Yaa mice, large atypical epithelial cells were observed within the cortical area. MTS1 (Figure 3) and anti-keratin antibodies (data not shown) which recognized pan epithelial cells had similar results compared to MTS44.
Flow cytometric analysis Frequency analysis demonstrated an increase in CD3 + cells in the TNC-L fraction of thymic tissue from NZB mice as compared to similarly obtained fraction of TNC-L of BALB/c mice (Table 2). This difference in the yield of CD3 + cells was to a large extent due to the increase of CD3low density+ cells. The frequency of CD3 + cells within the TNC-L of C57BL/6 mice was also significantly higher than that seen in BALB/c mice. However, no differences were seen in unfractionated thymocytes. Analysis for CD3, CD4 and CD8 expression revealed that all subpopulations were defined in comparison with total thymocytes (Table 2). Doublepositive CD4 + 8 + cells were the dominant population and significantly decreased in TNC-L of all strains compared to total thymocytes, but not between TNC-L of each strain. CD4 − 8 − cells in TNC-L were significantly increased compared to total thymocytes; CD4 + 8 − cells in each strain were also significantly increased. In the case of CD4 − 8 + cells in TNC-L, NZB mice expressed significant increases compared to total thymocytes, but this was not so for BALB/c or C57BL/6 mice. CD3 + 4 − 8 − cells of TNC-L in all strains were increased and almost all were included in a low frequency population of CD3 + . NZB mice had an increase in CD3 + 4 + 8 − , CD3 + 4 − 8 + or CD3high4 − 8 + cells of TNC-L. In analysis for expression of T cell receptor antigens (Table 2), TNC-L of NZB mice had a significant increase in TCR- positive cells as compared to total thymocytes. Percentages of CD44 + cells were decreased in TNC-L of a high expression (BALB/c) or low expression strain (C57BL/6 or NZB)
Thymic microenvironment
329
Figure 1. Apoptosis staining of thymus from BALB/c (A, B), C57BL/6 (C, D) and NZB (E, F) mice treated with saline (A, C, E) or LPS (B, D, F). Dark cells which reflect apoptosis, were observed in the cortex (confocal microscope, ×100; zoom 1.3).
(Table 3); the data depended on the subpopulation of CD4 + 8 + in CD44 + cells. CD44 + 4 + 8 − cells in TNC-L of NZB mice were increased and were similar to BALB/c mice.
Discussion It is now generally accepted that thymocytes recognizing self-antigens are eliminated by the process of negative selection in the thymus by apoptosis. Apoptotic cells have a characteristic morphology, including marked condensation of chromatin associated with shrinkage of the nucleus and fragmentation of DNA [23]. This fragmentation is inferred from gel electrophoresis of a pooled DNA extract and is the basis for the in situ visualization at the single cell level [24, 25].
Injection of LPS results in increased levels of thymocyte apoptosis. DNA fragmentation of thymocytes is dependent on the dose of injected LPS and reaches a peak 18 h after injection. In the present study, the apoptotic cells were located within the thymus but were not observed in epithelial-free regions (EFR). Generally, epithelial cells were found to surround the apoptotic cells. Electron micrographic techniques have shown that isolated TNC are the sites of thymic apoptosis [26]. In addition, an epithelial cell line, B/c. TEC-1, can form TNC and induce thymocyte apoptosis using in vitro culture systems [12]. The induction of apoptosis by LPS in murine lupus, especially in NZB and C3H/gld mice, was significantly lower than that of normal strains. The existing diffused cortical network was not changed by treatment with LPS, and there were a few TNC-like large epithelial cells in NZB or C3H/gld thymii. In MRL/lpr or BXSB/Yaa mice, the induction of
330
Y. Takeoka et al.
Figure 2. MTS44, which recognizes cortical epithelium, staining of thymus from BALB/c (A, B), NZB (C, D) or MRL/lpr (E, F) mice using AB complex/AP. In BALB/c treated with LPS (B), staining of epithelial cells in cortex was reduced, cortical epithelial free regions and large atypical epithelial cells were formed. Panel A represents staining of the normal cortical epithelial network in saline treated BALB/c mice. Panel C (NZB) or E (MRL/lpr) represents staining of the cortical epithelial network in saline treated murine lupus. Panel D (NZB) or F (MRL/lpr) represents the epithelial staining in LPS treated murine lupus. The cortical epithelial network was markedly reduced in panel C and E with increased areas of non-staining. These abnormalities were also observed in LPS treated NZB mice thymus. Arrowheads point to cortical epithelial cell-free regions. The bar in panel F indicates 250 m (confocal microscope, ×100; zoom 1.3).
apoptosis and epithelial morphology was intermediate between NZB and control mice. In contrast, there was no significant difference of Dex treatment in vitro in the induction of apoptosis in NZB thymocytes. It has been already reported that in vitro LPS treatment of thymocytes from normal mice does not result in apoptosis [24]. The induction of apoptosis in vivo by LPS may require factors other than direct effect of corticoids on thymocytes, because sensitivity of thymocytes from NZB mice against the Dex induced apoptosis is the same in control mice. In addition, there is no evidence for a disturbance of corticosteroid production in New Zealand mice [27]. The role of the thymic nurse cell as the site of
thymic apoptosis and apoptotic cell clearance has attracted recent attention [28]. It appears that the TNC consists of one thymic epithelial cell surrounded by up to 200 thymocytes enclosed by an epithelial cytoplasmic process [22]. Although further research is needed, it appears that the TNC is involved in both positive and negative selection [29]. However, it has been very difficult to correlate the function of thymic nurse cells in vitro with in vivo structure and activity. This has been particularly a problem in humans where in the past some have felt that the TNC may be no more than an in vitro isolation artifact; more recent data suggest that TNCs have a similar function to that in the mouse [28].
Thymic microenvironment
331
Figure 3. MTSI, which recognizes pan epithelium, staining of thymus from BALB/c (A, B), NZB (C, D) or MRL/lpr (E, F) mice using AB complex/AP. M=medulla, C=cortex. In BALB/c treated with LPS (B), staining of epithelial cells in cortex was reduced, cortical epithelial free regions and large atypical epithelial cells were formed. Panel A represents staining of the normal cortical epithelial network in saline treated BALB/c mice. Panel C (NZB) or E (MRL/lpr) represents staining of the cortical epithelial network in saline-treated murine lupus. Panel D (NZB) or F (MRL/lpr) represents the epithelial staining in LPS treated murine lupus. The cortical epithelial network was markedly reduced in saline treated NZB and MRL/lpr mice with increased areas of non-staining. These abnormalities were also observed in LPS treated NZB mice thymus. Arrows point to cortical epithelial cell-free regions. The bar in panel F indicates 250 m (confocal microscope, ×100; zoom 1.3).
The thymus of NZB mice undergoes premature involution and in vitro studies suggest that cultured thymic epithelia from NZB mice underwent accelerated, preprogrammed degeneration [30]. Recently, NZB mice were found to have distinctive and well-defined abnormalities of thymic architecture involving stromal cells defined by staining with monoclonal antibodies specific for the thymic microenvironment [12]. The thymus of other SLE mouse strains, MRL/lpr, BXSB/Yaa, C3H/gld and (NZB×NZW) F1 mice, were also known to have similar abnormalities of epithelial architecture [14, 15]. C3H/gld mice had the most severe abnormalities of cortical thymocytes among these SLE mice, with sig-
nificant changes in CD4 + 8 + double-positive (DP) and single-positive (SP) T cells [14]. The percentage of DP cells was significantly increased. On the other hand, CD4 + 8 − and CD4 − 8 + SP cells were significantly decreased in C3H/gld mice. Other SLE mice also showed a similar, but not statistically significant change in SP and DP thymocyte populations. Thymocytes in the thymic cortex are primarily CD3 + 4 + 8 + [31] and begin the conversion to SP status at the cortico–medullary interface. The positive selection of CD4 + thymocytes is related to their adherence to cortical epithelia, which express MHC class II antigens [32]. Positive selection of CD8 + cells also presumably occurs in the cortex [33]. These data
332
Y. Takeoka et al.
Table 2. FACS analysis of CD3, 4, or 8 expression on TNC-lymphocytes Positive cells (%)a Subsets
CD3 + CD3high CD3low CD4 + 8 + (DP) CD4 + 8 − (SP) CD4 − 8 + (SP) CD4 − 8 − (DN) CD3 + 4 + 8 + CD3 + 4 + 8 − CD3 + 4 − 8 + CD3 + 4 − 8 − CD3high4 + 8 + CD3high4 + 8 − CD3high4 − 8 + CD3high4 − 8 − CD3low4 + 8 + CD3low4 + 8 − CD3low4 − 8 + CD3low4 − 8 −
Total thymocytes
TNC-L
BALB/c
C57BL/6
NZB
BALB/c
C57BL/6
NZB
55.3±1.9b 15.7±2.5 39.7±1.7 82.0±2.0 12.4±1.2 2.5±0.4 3.0±0.9 40.2±1.6 12.2±1.2 1.9±0.4 0.7±0.1 6.2±1.0 8.1±1.7 1.0±0.4 0.1±0.03 34.0±1.6 4.2±1.2 0.9±0.2 0.6±0.1
53.8±3.5 14.0±2.8 39.9±4.2 84.7±3.1 9.5±1.5 2.4±0.6 2.8±0.9 41.9±3.4 9.3±1.4 1.6±0.3 0.6±0.1 7.3±2.5 5.6±0.9 0.7±0.2 0.1±0.01 34.6±3.6 3.7±1.5 0.9±0.3 0.6±0.1
53.5±2.5 17.5±2.9 35.7±4.1 85.5±1.7 7.9±1.0 2.3±0.3 4.3±1.5 42.4±2.0 7.7±1.0 1.9±0.2 1.3±0.5 10.5±3.2 5.9±0.9 0.9±0.1 0.1±0.05 32.0±3.5 1.9±0.4 1.0±0.1 0.8±0.4
53.5±2.1 11.8±1.5 41.8±2.0 70.0±1.2d 16.7±0.8 3.1±0.4 8.9±0.9d 32.6±1.4d 11.8±1.2 2.2±0.4 3.8±0.4d 2.7±0.4 6.5±1.2 1.1±0.2 0.3±0.1 29.9±1.4 6.6±0.8 1.2±0.3 3.5±0.4d
61.1±1.4c 18.3±1.9 45.2±2.0 64.7±2.3d 18.9±1.4d 4.8±0.9 10.4±1.1d 34.0±1.8 15.5±2.1 3.6±0.8 5.2±0.8d 4.0±0.5 10.9±1.7 1.8±0.5 0.4±0.5d 30.0±1.9 6.0±0.7 1.8±0.3 4.7±0.8d
67.7±1.8c,d 16.5±0.6 51.3±2.2c,d 66.1±2.6d 17.6±1.7d 6.1±1.1d 9.3±0.7d 38.9±3.5 16.4±1.9d 5.1±0.9c,d 6.2±0.8d 2.9±0.2 9.7±1.0 2.2±0.4d 0.7±0.1c,d 36.0±3.4 6.8±1.0 2.9±0.9 5.5±0.7d
a
Data are expressed as mean±SEM of % positive cells. Statistically significant (Student’s t-test, P<0.01) as compared with BALB/c in each group of total thymocytes or TNC-L (c), or with each strain in group of total thymocytes (d). b
Table 3. FACS analysis of TCR, MTS or other CD28, 44, 45, IL2R on TNC-lymphocytes Positive cells (%)a Subsets
TCR- + TCR- + TCR-V8 + MTS32 + MTS33 + MTS33high CD44 + CD44 + 4 + 8 + CD44 + 4 + 8 − CD44 + 4 − 8 + CD44 + 4 − 8 − CD28 + CD45R + CD25 + IL-2R +
Total thymocytes
TNC-L
BALB/c
C57BL/6
NZB
BALB/c
C57BL/6
NZB
46.4±2.2b 1.1±0.1 14.7±0.7 94.7±1.7 87.0±4.4 45.6±4.9 58.2±4.4 46.0±3.4 10.2±2.4 1.2±0.3 1.4±0.4 64.9±6.1 0.9±0.1 7.9±1.6 1.6±0.2
45.7±4.4 1.0±0.2 13.9±1.0 94.8±1.6 89.0±3.6 62.6±1.8 40.2±2.7c 31.8±2.6c 6.3±1.0 1.0±0.1 1.6±0.7 62.0±4.8 0.5±0.1c 12.2±2.3 1.3±0.3
39.8±2.6 0.6±0.1c 15.1±0.7 90.1±3.4 81.2±5.6 30.7±2.7d 40.7±1.6c 30.4±1.7c 6.0±1.0 1.4±0.1 1.5±0.3 63.0±5.5 0.4±0.02c 6.8±2.1 0.8±0.2c
48.1±2.4 7.2±0.6e 18.2±0.3e 90.6±2.1 88.9±2.9 53.6±5.6 38.1±1.5e 25.9±1.2 9.2±0.8 1.2±0.1 1.2±0.1 70.2±4.3 1.8±0.3 9.0±0.9 5.3±0.8e
61.1±1.4 8.1±0.6e 16.7±0.5 91.6±1.5 95.3±1.2 76.5±4.1 25.8±1.4c,e 11.7±1.1c,e 10.7±0.6e 1.0±0.2 1.6±0.4 69.0±6.6 2.3±0.4e 16.6±2.5 8.0±1.3e
57.0±1.6c,d,e 6.4±0.1e 18.9±0.8e 91.2±3.3 90.0±1.4 53.4±7.0e 23.4±1.7c,e 10.1±1.5c,e 9.2±0.3e 1.4±0.3 1.4±0.03 72.3±6.3 1.7±0.5 6.9±0.6c 7.8±2.6
a
Data are expressed as mean±SEM of % positive cells. Statistically significant (Student’s t-test, P<0.01) as compared with BALB/c (c), with C57BL/6 (d) in group of total thymocytes or TNC-L, or with each strain in group of total thymocytes (e). b
imply that abnormalities in the thymic cortex may affect T cell maturation by altering the mechanisms of positive selection, and can be interpreted as indicating that decreased TNC DP formation of thymocytes results in an accumulation following a decrease of apoptosis.
Our findings demonstrate abnormalities among the induction of apoptosis and the formation of TNC-like epithelial cells in SLE mice. As a result, developing thymocytes may not properly bind to the thymic epithelium as TNC and therefore fail to undergo normal positive selection via apoptosis. This may ultimately
Thymic microenvironment
result in abnormal thymocyte education, the loss of self-tolerance induction and autoimmune disease.
References 1. Boyd R.L., Tucek C.L., Godfrey D.I., Izon D.J., Wilson T.J., Davidson N.J., Bean A.G.D., Ladyman H.M., Ritter M.A., Hugo P. 1993. The thymic microenvironment. Immunol. Today 14: 445–459 2. Smith C.A., Williams G.T., Kingston R., Jenkinson E.J., Owen J.J.T. 1989. Antibodies to CD3/T-cell receptor complex induce death by apoptosis in immature T cells in thymic cultures. Nature 337: 181–184 3. Murakami M., Tsubata T., Okamoto M., Shimizu A., Kumagai S., Imura H., Honjo T. 1992. Antigen-induced apoptosis death of Ly-1 B cells responsible for autoimmune disease in transgenic mice. Nature 357: 77–79 4. Telford W.G., King L.E., Franker P.J. 1991. Evaluation of glucocorticoid-induced DNA fragmentation in mouse thymocytes by flow cytometry. Cell Prolif. 24: 1447–1459 5. Migliorati G., Nicoetti I., Pagliacci M.C., D’Adamo L., Riccardi C. 1993. Interleukin-2 induces apoptosis in mouse thymocytes. Cell. Immunol. 146: 52–61 6. Kizaki H., Ohnishi Y., Azuma Y., Mizuno Y., Ohsaka F. 1992. 1--D arabinosylcytosine and 5-azacytidine induce internucleosomal DNA fragmentation and cell death in thymocytes. Immunophamacology 24: 219–227 7. Suzuki K., Tadakuma T., Kizaki H. 1991. Modulation of thymocytes apoptosis by isoproterenol and prostaglandin E2. Cell. Immunol. 134: 235–240 8. Zhang Y-H., Takahashi K., Jiang G.-Z., Kawai M., Fukada M., Yokochi T., 1993. In vivo induction of apoptosis (programed cell death) in mouse thymus by administration of lypopolysaccharide. Infect. Immun. 61: 5044–5048 9. Wekerle H., Ketelsen U.-P. 1980. Thymic nurse cells—la-bearing epithelium involved in T-lymphocyte differentiation? Nature 283: 402–404 10. Wekerle H., Ketelsen U.-P., Ernst M. 1980. Thymic nurse cells—Lymphoepithelial cell complexes in murine thymuses: morphological and serological characterization. J. Exp. Med. 151: 925–944 11. Kyewski B.A. 1986. Thymic nurse cells: Possible sites of T-cell selection. Immunol. Today 7: 374–378 12. Hiramine C., Hojo K., Koseto K., Nakagawa T., Mukase A. 1990. Establishment of a murine thymic epithelial cell line capable of inducing both thymic nurse cell formation and thymocyte apoptosis. Lab. Invest. 62: 41–54 13. Watanabe Y., Naiki M., Wilson T.J., Godfrey D.I., Chiang B.-L., Boyd R.L., Ansari A., Gershwin M.E. 1993. Thymic microenvironmental abnormalities and thymic selection in NZB.H-2bm12 mice. J. Immunol. 150: 4702–4712 14. Takeoka Y., Yoshida S.H., Van de Water J., Boyd R.L., Suehiro S., Ansari A.A., Gershwin M.E. 1995. Thymic microenvironmental abnormalities in MRL/MP-lpr/lpr, BXSB/MpJ Yaa and C3H/Hej-gld/gld mice. J. Autoimmun. 8: 145–161 15. Takeoka Y., Whitmer K.J., Chen S.-Y., Ansari A.A., Boyd R.L., Shultz L., Suehiro S., Gershwin M.E. 1995. Thymic epithelial cell abnormalities in (NZB×H-2u) F1 mice. Clin. Immunol. Immunopathol. 76: 297–307
333
16. Takeoka Y., Taguchi N., Kotzin B.L., Bennett S., Vyse T.J., Boyd R.L., Naiki M., Konishi J.-E., Ansari A.A., Shultz L.D., Gershwin M.E. 1999. Thymic microenvironment and NZB mice: The abnormal thymic microenvironment of New Zealand mice correlates with immunopathology. Clin. Immunol. Immunopathol. 90: 388–398 17. Zychlinsky A., Zheng L.M., Liu C-C, Young J.D.-E. 1991. Cytolytic lymphocytes induce both apoptosis and necrosis in target cells. J. Immunol. 146: 393–400 18. Schmitz G.G., Walter T., Seibl R., Kessler C. 1991. Nonradioactive labeling of oligonucleotides in vitro with the hapten digoxigenin by tailing with terminal transferase. Anal. Biochem. 192: 222–231 19. Godfrey D.I., Izon D.J., Tucek C., Wilson T.J., Boyd R.L. 1990. The phenotypic heterogeneity of mouse thymic stromal cells. Immunology 70: 66–74 20. Kampringa J., Berges S., Boyd R.L., Brekelmans P., Colic M., Van Ewijk W., Kendall M., Ladyman H., Nieuwenhuis P., Ritter M., Schuurmar H.-J., Tournefier A. 1989. Thymic epithelial antibodies: immunohistological analysis and introduction of CTES nomenclature. Summary of the Rolduc Epithelium Workshop. Thymus 13: 65–173 21. Wood G.S., Warnke R. 1981. Suppression of endogeneous biotin-avidin detection systems. J. Histochem. Cytochem. 29: 1196–1204 22. Andrews P., Boyd R.L. 1985. The murine thymic nurse cell: an isolated thymic microenvironment. Eur. J. Immunol. 15: 36–42 23. Arends M.J., Wyline H. 1991. Apoptosis: mechanisms and roles in pathology. Int. Rev. Exp. Pathol. 32: 223–254 24. Gavrieli Y., Sherman Y., Ben-Sasson S.A. 1992. Identification of programmed cell death in situ via specific labeling of nuclear DNA fragmentation. J. Cell Biol. 119: 493–501 25. Fehsel K., Kro¨ ncke K.D., Kolb H., Kolb-Bachofen V. 1994. In situ nick-translation detects focal apoptosis in thymuses of glucocorticoid- and lipopolysaccharidetreated mice. J. Histochem. Cytochem. 42: 613–619 26. Aguilar L.K., Aguilar-Cordova E., Carwright, Jr. J., Belmont J.W. 1994. Thymic nurse cells are sites of thymocyte apoptosis. J. Immunol. 152: 2645–2651 27. Whitmer K.J., Ansari A., Gershwin M.E. 1994. The pathogenesis of autoimmunity in New Zealand Mice. Semin. Clin. Immunol. 8: 35–53 28. Hiramine C., Nakagawa T., Miyauchi A., Hojo K. 1996. Thymic nurse cells as the site of thymocyte apoptosis and apoptotic cell clearance in the thymus of cyclophosphamide-treated mice. Lab. Invest. 75: 185–201 29. Faas S.J., Rothstein J.L., Kreider B.L., Rovera G., Knowles B.B. 1993. Phenotypically diverse mouse thymic stroma cell lines which induce proliferation and differentiation of hematopoietic cells. Eur. J. Immunol. 23: 1201–1214 30. Mitamura K., Kang H., Tomita Y., Hashimoto H., Sawada S., Horie T. 1991. Impaired tumour necrosis factor- (TNF-) production and abnormal B cell response to TNF- in patients with systemic lupus erythematosus (SLE). Clin. Exp. Immunol. 85: 386–391 31. Gershwin M.E., Ikeda R.M., Kruse W.L., Wilson F., Shifrine M., Spangler W. 1978. Age-dependent loss in New Zealand mice of morphological and functional
334
characteristics of thymic epithelial cells. J. Immunol. 120: 971–979 32. Boyd R.L., Wilson T.J., Ward H.A., Gershwin M.E. 1992. Phenotypic characterization of chicken thymic stromal elements. Dev. Immunol. 2: 51–66
Y. Takeoka et al.
33. Berg L.J., Pullen A.M., de St Growth F.F., Mathis D., Benoist C., Davis M.M. 1989. Antigen/MHC-specific T cells are preferentially exposed from the thymus in the presence of their MHC ligand. Cell 58: 1035–1046