| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Clinical Studies, University of Pennsylvania, School of Veterinary Medicine, Philadelphia, PA 19014
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
on their cell
surface become susceptible to repertoire selection. Of these,
thymocytes that bear TCRs that interact weakly with self-MHC/peptide
complexes are positively selected, whereas those reacting with high
affinity are negatively selected and either die apoptotically in situ
(clonal deletion) or are induced into a state of nonresponsiveness
(anergy) (1). It is not clear how the strength of these
MHC/TCR interactions results in differential death or survival signals.
The number of TCRs that are engaged during the intrathymic selection
process and the longevity of this interaction are thought to be
important factors influencing the outcome of T cell selection
(2, 3, 4). The possibility that quantitative or qualitative
differences in intracellular signaling pathways may influence the fate
of DP thymocytes also has received much attention (5).
Several studies have identified differentially expressed signaling
molecules and pathways in thymocytes undergoing positive vs negative
selection (6, 7, 8, 9, 10), and abnormal thymocyte selection
has been reported in mice with mutations in specific signaling
molecules (11, 12, 13). By contrast, the extrinsic signals
that lead to the initiation of intracellular signaling cascades and
ultimately determine the fate of developing thymocytes still remain to
be fully elucidated.
One manner in which signals delivered by TCR-MHC/peptide interactions
could be modulated is by the action of cytokines produced by the thymic
stroma or by thymocytes themselves. Cross-linking of TCRs on DP
thymocytes induces the production of various cytokines
(14) and numerous studies have shown that cytokines such
as IL-7 and IL-2 can profoundly affect thymocyte growth and survival
(15). However, the precise role of IL-2/IL-2R signaling in
thymocyte development remains controversial due to apparently
contradictory studies providing evidence for both the promotion and
inhibition of thymocyte growth by this signaling pathway
(16). The normal generation of T cells reported in
IL-2-deficient (IL-2-/-) mice (17)
seems to be inconsistent with a direct involvement of IL-2/IL-2R
interactions in intrathymic T cell development. However, the presence
of dysregulated thymically derived pathogenic T cells (18)
and CD4 T cell-mediated autoimmunity (19, 20) in
IL-2-/- mice suggests that IL-2 could be
involved in thymic selection. Indeed, the lymphoid hyperplasia and
autoimmunity present in gnotobiotic (germfree)
IL-2-/- mice is consistent with a defect in
central tolerance leading to uncontrolled T cell responses directed
against self- rather than environmental Ags (21). The
impaired ability of thymocytes in IL-2-/- mice
to undergo anti-CD3 Ab-mediated apoptosis (22) and the
presence of autoimmune disorders and developmental defects in
IL-2R-/- (CD25-/-;
Refs. 23, 24), IL-2R-/-
(CD122-/-; Ref. 25), and
Jak3-/- (26, 27, 28) mice similar to
that seen in IL-2-/- mice agree with this
interpretation. The observation that thymic expression of a CD122
transgene in CD122-deficient mice prevents lethal autoimmunity
(29) further emphasizes the importance of IL-2/IL-2R
interactions in the establishment and maintenance of thymic tolerance.
Yet, the finding that IL-2 deficiency has no effect on the deletion of
self-reactive thymocytes in MHC class I-restricted TCR-transgenic mice
(30) suggests that IL-2/IL-2R interactions may be
restricted to or used specifically for the selection of MHC class
II-restricted thymocytes. Although collectively these observations
provide a compelling argument for the involvement of IL-2/IL-2R
signaling in T cell development, evidence for IL-2/IL-2R interactions
being directly involved in thymocyte selection is lacking.
To address this issue, we have reexamined the involvement of IL-2 in thymocyte development and tested the hypothesis that IL-2/IL-2R interactions play a role in the deletion of self-reactive, Ag-specific thymocytes. By using nontransgenic and transgenic IL-2-sufficient and -deficient animal model systems, our studies show that IL-2 is expressed in situ in response to TCR engagement and is subsequently bound and internalized by apoptotic thymocytes expressing high-affinity IL-2Rs. We also demonstrate that disruption or blockade of IL-2/IL-2R interactions rescues some MHC class II-restricted thymocytes from Ag-induced apoptosis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mice used in these experiments were maintained as specific pathogen-free (SPF) or germfree colonies at the University of Pennsylvania (Philadelphia, PA). Germfree C57BL/6-IL-2-/- mice were housed in isolator cages within a flexible film isolator in the gnotobiotic facility of the Biology Department, University of Pennsylvania. The germfree status of IL-2+/+, IL-2+/-, and IL-2-/- animals was verified by bacteriologic, histologic, and serologic analysis of tissues and fluids at autopsy as described previously (21). IL-2-/- mice on the BALB/C background were purchased from The Jackson Laboratory (Bar Harbor, ME) and were mated with H-2d DO11.10 TCR-transgenic mice (31) to obtain DO11.10 x IL-2-/- mice. Similarly, IL-2-/- mice on the C57BL/6 background were mated with H-2b H-Y TCR-transgenic mice to obtain H-Y x IL-2-/- mice. IL-2-/- mice were identified from samples of tail DNA by PCR (17). Heterozygote and/or wild-type littermates were used as controls. Presence of TCR transgenes were confirmed by flow cytometric detection of transgene-encoded TCR on PBL, lymph node cells, splenocytes, or thymocytes. Use of laboratory animals conformed to institutional and National Institutes of Health (Bethesda, MD) guidelines.
For the induction of thymocyte apoptosis in vivo, mice were injected
i.p. with 50 µg of anti-CD3 Ab (145-2C11, 500A2, or 29B), 0.1
µg/g of recombinant human IL-2 (Proleukin; Chiron, Emeryville, CA),
anti-CD3 plus IL-2, or PBS, and 12 h later, thymi were
removed. For corticosteroid-induced thymocyte apoptosis, mice were
injected i.p. with 150 µg of dexamethasone (Sigma, St. Louis, MO),
and thymi were removed 3 h later for analysis. Adult or neonatal
(<3 days of age) DO11.10-transgenic mice were injected i.p. daily with
1 mg (neonates) or 10 mg (adults) of OVA (Sigma) or PBS for either 1 or
2 days, and thymi were removed 12 h after the final injection. To
block IL-2/IL-2R interactions during thymocyte selection in vivo, mice
(n = 3–4) were treated with a mixture of
anti-IL-2R Abs consisting of 0.5 mg each of rat anti-mouse
anti-IL-2R (clone TM1) and anti-IL-2R (PC.61.5.3) Abs.
Control groups received a mixture of isotype matched control Abs
(anti--galactosidase (GL113) and anti-human TCR
(1B))
or rat IgG. On days 1 and 2, mice were injected with Abs alone followed
by two further injections of Abs together with OVA on days 3 and 4, and
the thymi were removed for analysis on day 5.
Bone marrow radiation chimeras
Two million DO11.10 and BALB/c bone marrow cells were mixed at a 1:1 ratio and injected i.v. into lethally irradiated (2 x 550 rad) BALB/c mice. Ten weeks later, chimeras were injected i.p. with PBS or OVA (as described above), and 12 h later, thymocytes were analyzed for expression of surface Ags and apoptosis by flow cytometry.
Flow cytometry
The following Abs and secondary reagents were used for three-
and four-color flow cytometry and were purchased from Life Technologies
(Grand Island, NY), BD PharMingen (San Diego, CA), or Caltag (San
Francisco, CA), or isolated from hybridoma culture supernatants and
conjugated to FITC (Sigma), biotin (Sigma), or Cy-5
(Biological Systems, Pittsburgh, PA) in our laboratory: anti-CD4
(H129.19)-Red613 or -PE; anti-CD8 (53-6.7)-FITC, -PE, or -Cy-5;
anti-CD3
(500-A2)-FITC; anti-H-Y-transgenic TCR V3.2
(T3.70)-biotin; anti-CD25 (PC61.5.3)-FITC or -Cy5; anti-CD122
(Tm1)-biotin, -PE, or -FITC; and anti-DO11.10 clonotype TCR
(KJ1-26)-biotin.
All of the following procedures were performed at 4°C in the dark with FACS staining buffer (1x PBS, 2% FCS, and 1% penicillin/streptomycin) unless otherwise specified. Pooled or individual thymi were gently dispersed, the cell suspension was passed through fine nytex mesh to exclude debris, and the resulting single-cell suspension was washed twice. Cells (500,000–2,000,000) were stained in 50 µl of buffer in V-bottom 96-well plates for 20 min with the anti-FcR Ab, 2.4.G.2, and then washed twice. The cells then were stained with primary Abs, washed twice, and incubated for 20 min with appropriate secondary reagents and directly conjugated Abs. Finally, the cells were washed twice and fixed with 200 µl of 1% paraformaldehyde/PBS before analysis on a FACScan, FACSCalibur, or FACStar flow cytometer (Becton Dickinson, San Jose, CA) using CellQuest software (Becton Dickinson).
Detection of apoptotic cells
Apoptotic cells were identified in cell suspensions of isolated thymocytes or in situ in frozen sections of thymi by a modified version of a DNA polymerase-mediated dUTP nick translation-labeling (DUNTL) assay (32). Briefly, 5-µm frozen sections were fixed in ice-cold acetone, washed in TBS (10 mM Tris, pH 7.3, 0.15 M NaCl), incubated with labeling buffer (50 mM Tris, pH 7.4, 10 mM MgSO4, 0.1 mM DTT, 1 nmol/ml dATP/CTP/GTP, and 0.7 nmol/ml dTTP) containing 2 U of DNA polymerase (Promega, Madison, WI) and 40 pmol of dUTP-digoxigenin (Boehringer Mannheim, Indianapolis, IN) and incubated at 37°C for 90 min. As a control, duplicate sections were incubated with DNA polymerase that had been denatured by heating at 70°C for 10 min in the presence of 2 mM EDTA. Sections then were washed in TBS buffer, incubated with 5 µg/ml alkaline phosphatase-conjugated sheep anti-digoxigenin Fab (Boehringer Mannheim), washed, and then incubated with the alkaline phosphatase substrate, Vector Red (Vector Laboratories, Burlingame, CA). Sections were counterstained with Methyl Green (Vector Laboratories) before mounting and photomicroscopy.
For flow cytometric detection of apoptotic cells, previously stained and fixed cells were washed in permeabilization buffer (0.1% Triton X-100, 0.1% sodium citrate) and resuspended in labeling buffer containing 2 U of active or inactivated DNA polymerase and 40 pmol of FITC-12-dUTP (Boehringer Mannheim) and incubated at 37°C for 60 min. Cells then were washed in staining buffer and analyzed by flow cytometry. Apoptotic cells were also detected in cell suspension using FITC-annexin V (BD PharMingen).
Immunohistochemistry
Frozen thymic sections were incubated with TTBS (TBS with 0.2% Triton X-100) containing mouse IgG (10 µg/ml) and 5% normal rabbit serum to block nonspecific reactivity of primary Abs. Sections then were incubated with TTBS containing a mixture of rat anti-mouse IL-2 Abs (5 µg each of JES6-1A12 and JES6-5H4; BD PharMingen), washed with TTBS, and incubated with biotinylated-F(ab')2 of a rabbit anti-mouse IgG Ab (Vector Laboratories). Bound Ab was visualized with a biotin-avidin-alkaline phosphatase complex and with the Vector Red substrate. Sections were counterstained with Methyl Green before photomicroscopy. As controls, anti-mouse IL-2 Abs were preincubated with recombinant murine IL-2 (BD PharMingen), and isotype-matched rat Abs of irrelevant specificity (Caltag Laboratories) were used.
IL-2 binding and internalization
Two million freshly isolated IL-2-/- thymocytes were incubated on ice with 4 µg of hamster anti-mouse CD3 (2C11-145) for 30 min, washed with cold PBS, and placed in microtiter plates at 37°C for 8 h in the presence of 4 µg of anti-hamster Ig. During the last 2 h of culture, 25 ng of biotin-labeled recombinant human IL-2 (R&D Systems, Minneapolis, MN) that had been preincubated with 0.2 µg of streptavidin PE (Caltag) was added to each well. Cells then were stained with anti-CD4 and anti-CD8 Abs and annexin V and immediately analyzed by FACS. In control cultures, PE-labeled IL-2 was replaced with either 1) PE-labeled control protein (soybean trypsin inhibitor) of similar m.w. as IL-2, 2) PE-IL-2 mixed with 100-fold excess of recombinant human IL-2, 3) PE-labeled IL-2 preincubated with 40 µg of polyclonal goat anti-human IL-2 antisera, or 4) PE-IL-2 followed 30 min later (chase) by addition of excess rIL-2.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The hypothesis that IL-2 is actively involved in thymocyte apoptosis was tested directly by determining the consequence of IL-2-deficiency on TCR/CD3-mediated thymocyte death in IL-2-/- mice. Although SPF IL-2-/- mice have been used previously to investigate this hypothesis (22, 33), we wanted to exclude any differences in the susceptibility of IL-2-/- thymocytes to activation-induced cell death (AICD) being attributable to thymic pathology and loss of cortical thymocytes that occurs in SPF IL-2-/- mice (34). Consequently, gnotobiotic (germfree) IL-2-/- mice in which the onset of lymphoid disorders is significantly delayed (21) were used for these experiments. Mice of 4–6 wk of age in which thymus cellularity was comparable to that of wild-type littermates and in which thymic pathology is absent (21) were used.
Gnotobiotic wild-type (n = 11) and
IL-2-/- mice (n = 9) were
injected with anti-CD3 Ab, and 12 h later, thymi were removed.
One thymic lobe was used to detect apoptotic cells (Figs. 1
and 2)
and IL-2 protein production during thymocyte apoptosis (Fig. 2) by
immunohistochemistry. The second lobe was used for flow cytometric
analysis of thymocyte apoptosis and IL-2R expression (Fig. 3). Very few apoptotic cells were evident
in sections of thymi from anti-CD3-treated
IL-2-/- mice (Fig. 1A). By contrast,
large numbers of apoptotic cells were readily detected in the thymic
cortex of anti-CD3-treated IL-2+/+
littermates (Fig. 2, A and B). The similarity in
the cellularity and composition of thymocytes in the germfree
IL-2-/- and IL-2+/+ mice
excludes the possibility that the induction of thymocyte apoptosis is
attributable to abnormalities in the thymocyte subset distributions and
loss of DP thymocytes seen in older SPF IL-2-/-
mice (34).
|
|
|
C). The induction of apoptosis required both IL-2 and
anti-CD3, because IL-2 alone did not result in an increase in the
number of apoptotic cells above that seen in anti-CD3-treated
IL-2-/- mice (Fig. 1B). By counting
apoptotic clusters in several randomly selected microscopic fields, the
level of apoptosis seen in the thymi of IL-2-/-
mice administered anti-CD3 and IL-2 was similar to that seen in the
thymi of wild-type mice treated with anti-CD3 (Figs. 1C
and 2, A and B). Detection of apoptotic cell
clusters by DUNTL was specific because no staining was present when DNA
polymerase was inactivated before use. However, clusters of apoptotic
cells were still visible with the nuclear counterstain Methyl Green
(Fig. 1D). These in situ findings were corroborated by the analysis of apoptotic cells with a flow cytometry-based DUNTL assay. The frequency of apoptotic cells recovered from thymi of anti-CD3-treated IL-2-/- mice (4.1 ± 1.4%) was not significantly different from that seen in thymi of mice treated with PBS alone (3.4 ± 1.2% and 3.1 ± 1.1%, respectively, for IL-2-/- and IL-2+/+ mice). In addition, the frequency of apoptotic cells in the thymi of anti-CD3-treated IL-2+/+ mice (13.3 ± 3.5%) was significantly (p < 0.005) higher than those present in thymi of IL-2-/- mice (4.1 ± 1.4%). Importantly, the combination of IL-2 and anti-CD3 resulted in levels of thymocyte apoptosis (14.0 ± 3.5%) in IL-2-/- mice similar to that seen in anti-CD3-treated IL-2+/+ mice (13.3 ± 3.4%). One implication of these findings is that because IL-2 appears to be necessary for AICD of thymocytes, it might be produced locally in the thymus during this process.
IL-2 is produced in situ on TCR engagement
To determine whether IL-2 can be produced in the thymus under
conditions that promote thymocyte apoptosis, sections of thymi from
C57BL/6-IL-2+/+ mice (n = 6)
injected 12 h previously with anti-CD3 Abs (see above) were
evaluated immunohistochemically for the presence of IL-2. As noted
above, clusters of thymocytes with fragmented and aggregated nuclei
were readily identified in the thymic cortex by Methyl Green
counterstain (Fig. 2, A and B). Anti-IL-2 Ab
staining was detected throughout the cortex of the thymi of
anti-CD3-treated mice, with the most intense staining being found
in areas adjacent to clusters of apoptotic thymocytes (Fig. 2, A and inset A1). Staining of
adjacent thymic sections showed that most of the IL-2 protein staining
was in proximity to or within phagocytic F4/80+
and/or Mac-1+ macrophages (data not shown). This
pattern of anti-IL-2 staining was specific because no staining was
observed with anti-IL-2 Abs previously incubated with excess
recombinant murine IL-2 (Fig. 2B). In addition, the
anti-IL-2 staining pattern was shown not to be due to nonspecific
binding of Ab to dying cells because isotype-matched control Abs of
irrelevant specificity gave no staining (data not shown).
Because anti-CD3 engagement of the TCR represents a
nonphysiological interaction that bypasses the requirement for
MHC/peptide interactions with the TCR, we repeated the anti-IL-2
staining on frozen sections of thymi from adult DO11.10 TCR-transgenic
mice that had received a single dose (10 mg) of Ag (OVA). In line with
our analyses of IL-2 expression in thymi of anti-CD3-treated
wild-type mice, IL-2 protein was found throughout the thymi of
OVA-treated DO11.10 mice (Fig. 2D). In contrast to the
cortical localization of apoptotic thymocytes in anti-CD3-treated
wild-type mice (Fig. 2), apoptotic thymocytes and IL-2 protein were
colocalized in the medulla as well as the cortex of OVA-treated
DO11.10-transgenic mice. The staining pattern was shown to be specific
and attributable to the presence of Ag because administration of PBS to
DO11.10 mice (Fig. 3E) or OVA to BALB/c mice (Fig. 2F) failed to reveal any thymocyte death or IL-2 staining.
To determine whether the thymic IL-2 production resulted from thymocyte
apoptosis mediated by other apoptotic stimuli, wild-type mice were
treated with dexamethasone in vivo and their thymi analyzed for IL-2
expression. Although administration of dexamethasone induced rapid
thymocyte apoptosis in situ, we were unable to detect any IL-2 protein
in these thymi (Fig. 2C). These results suggest that IL-2
expression is not a general byproduct of thymocyte death but is a
result of TCR engagement.
Apoptotic thymocytes dying as a result of TCR engagement display characteristics of IL-2/IL-2R interactions
Thymic expression of individual chains of the IL-2R complex is
either restricted to specific subsets of thymocytes (CD25, CD122), or
is expressed at all stages of thymocyte development (CD132). Although
thymocytes undergoing spontaneous apoptosis in vitro have also been
shown to up-regulate CD25 (35) and CD122 can be induced by
self-recognition in the thymus (36), whether or not these
receptors are expressed alone and if they are functional is not known.
To investigate this further, we established the IL-2R phenotype of
thymocytes dying in situ as a result of anti-CD3-induced apoptosis.
The majority of apoptotic (dUTP+) thymocytes from
anti-CD3-treated wild-type mice were DP, with a smaller proportion
being CD4 single-positive (SP; Fig. 3). Significantly, dying thymocytes
uniformly expressed high levels of CD122, with a subset of these cells
also expressing CD25. By contrast, very few viable
(dUTP-) thymocytes expressed detectable levels
of CD122 or CD25. Limited analysis of the kinetics of IL-2R expression
by thymocytes undergoing apoptosis in vivo indicates that up-regulation
of CD25 and CD122 by apoptotic DP thymocytes can be detected within
2–3 h of treatment and that the highest levels of expression are seen
after 
16 h (data not shown).
Dying thymocytes bind and internalize IL-2
To confirm the ability of IL-2R-expressing thymocytes to bind and internalize IL-2, an IL-2 binding assay was developed. Thymocytes from IL-2-deficient mice were exposed to anti-CD3 and cross-linking Abs in vitro for a total of 8 h with PE-labeled recombinant human IL-2 (PE-IL-2) protein being added for the last 2 h of culture. In control cultures, PE-IL-2 was replaced with either a PE-labeled protein (soybean trypsin inhibitor) of similar m.w. as IL-2, PE-IL-2 mixed with 100-fold excess of recombinant human IL-2, PE-labeled IL-2 preincubated with 40 µg of polyclonal goat anti-human IL-2 antisera, or PE-IL-2 followed 30 min later (chase) by addition of excess rIL-2. The purpose of these controls was to distinguish between IL-2 that is surface bound vs IL-2 that is bound and then internalized, which would be indicative of labeled cells expressing high-affinity IL-2Rs that are required for internalizing IL-2 and generating IL-2R-mediated intracellular signals (37). Dying thymocytes were detected by staining with FITC-annexin V.
After TCR/CD3 cross-linking, a small proportion of thymocytes were
labeled with PE-IL-2 (Fig. 4A). Of note, the majority of
annexin+ thymocytes did not stain with PE-IL-2.
These cells may be dying via an IL-2-independent mechanism or may have
bound, internalized, and degraded the PE-IL-2 before flow cytometric
analysis. The observation that spontaneous death of wild-type or
IL-2-deficient thymocytes in culture is not augmented by the addition
of rIL-2 or prevented by anti-IL-2R Abs (our unpublished
observations) is consistent with the interpretation that once removed
from the thymic microenvironment, thymocytes undergo apoptosis via
IL-2-independent mechanisms.
|
90%) of viable (annexin-)
thymocytes that bound IL-2 were double negative (DN). Among apoptotic
thymocytes, the binding of IL-2 was more widespread, comprising DN,
CD4SP, and DP thymocytes (Fig. 4B). Interestingly, virtually
no IL-2 labeling was seen among CD8SP thymocytes (Fig. 4B).
The detection of IL-2 binding was specific because a control protein
(labeled with PE to the same extent as PE-IL-2) did not bind (Fig. 4G), and binding of PE-IL-2 was abrogated by preincubation
with anti-IL-2 antisera that blocks IL-2/IL-2R interactions (Fig. 4H).
The proportion of IL-2+
annexin+ thymocytes cultured in the presence of
100-fold excess native IL-2 (1.7%; Fig. 4C) was similar to
that in cultures containing PE-IL-2 alone (1.1%; Fig. 4A),
demonstrating that PE-IL-2 binding is efficient. This suggests but does
not prove that this binding could involve receptors of high affinity.
The slight increase in the proportion of thymocyte labeled with PE-IL-2
in the presence (1.7%) vs the absence (1.1%) of native IL-2 could be
attributable to nonlabeled IL-2 promoting internalization of
surface-bound PE-IL-2. Consistent with this interpretation is the
finding that excess native IL-2 when added after incubating with
PE-IL-2, which serves to promote the internalization of surface-bound
PE-IL-2, resulted in a similar increase in the proportion of
IL-2+ annexin+ cells
(1.9%; Fig. 4E). Interestingly, the inclusion of excess
native IL-2 also produced a change in the profile of
IL-2+ annexin+ thymocytes.
Compared with cells incubated with PE-IL-2 alone, there were fewer
IL-2+ annexin+ DN cells and
a higher proportion of IL-2+
annexin+ CD4SP and DP cells in cultures
containing excess native IL-2. Thus, in contrast to the binding of
PE-IL-2 to CD4SP and DP thymocytes binding of PE-IL-2 to apoptotic DN
cells can be blocked or reversed. Although we have not directly
measured the affinity of IL-2R expressed by dying thymocytes, the
efficient binding and internalization of PE-IL-2 by apoptotic CD4SP and
DP cells together with their known expression of CD25 and CD122 (Fig. 3) suggests that the binding of PE-IL-2 to at least some of these cells
may involve high-affinity IL-2Rs.
Collectively, the results presented so far demonstrate that during thymocyte apoptosis, IL-2 is expressed in situ in the thymus and is bound and internalized by dying DP and CD4SP thymocytes. The ability of exogenous IL-2 to overcome the resistance of cortical IL-2-/- thymocytes to TCR/CD3-mediated AICD suggests that IL-2/IL-2R signaling contributes to the inductive phase of thymocyte apoptosis. This was investigated further by determining the consequences of disrupting IL-2/IL-2R interactions on the deletion of Ag-specific thymocytes.
Blockade of IL-2/IL-2R interactions disrupts apoptosis of Ag-specific CD4+ thymocytes
Two experimental approaches involving the use of TCR-transgenic mice were used to address the role of IL-2/IL-2R interactions in the Ag-specific deletion of thymocytes. In the first, anti-IL-2R Abs were administered in conjunction with Ag to DO11.10 TCR-transgenic mice. In the second approach, MHC class I- and class II-restricted TCR-transgenic mice were crossed with IL-2-/- mice of appropriate MHC haplotypes, and deletion of transgenic IL-2-deficient thymocytes was assessed in the presence of the selecting Ag.
Newborn (<3 days old) DO11.10 mice were administered a mixture (0.5 mg
each) of anti-CD25 and anti-CD122 or isotype-matched control
Abs on days 1 and 2 followed by two further injections of Abs plus 1 mg
of OVA on days 3 and 4, with the thymi and spleen being removed for
analysis on day 5. The results in Fig. 5
and Table I show that Ab-mediated
disruption of IL-2/IL-2R interactions significantly reduces the
deletion of Ag-specific thymocytes as indicated by cellularity and
distribution of CD4 and CD8 thymocyte subsets in DO11.10 mice treated
with OVA and anti-IL-2R Abs. Administration of Ag alone resulted in
the loss of DP thymocytes bearing low to intermediate levels of
transgenic TCR (Fig. 5A). By contrast, anti-IL-2R Abs
protected a large proportion of TCRlow DP
thymocytes from AICD. However, isotype-matched control Abs of
irrelevant specificity did not effect the deletion of DP
thymocytes.
|
|
A) allowed us to exclude the possibility that
thymocyte death caused by Ag administration was attributable to
peripheral T cell activation and the production of glucocorticoids
(GCs). Although the finding that intrathymic T cell development
proceeds normally in GC receptor (GCR)-/- mice
(38) convincingly demonstrates that GCR signaling does not
play a central role in thymocyte selection, we attempted to quantitate
the level of thymocyte death resulting from Ag- vs non-Ag-specific
mechanisms with bone marrow chimeras.
Lethally irradiated BALB/c mice were reconstituted with a 1:1 mixture
of DO11.10 TCR-transgenic and -nontransgenic BALB/c bone marrow cells.
Ten weeks later, groups (n = 5) of chimeric or DO11.10
and BALB/c control animals were injected with OVA, and the level of
apoptosis among TCR-transgenic (KJ1-26+) and
wild-type (KJ1-26-) thymocytes was assessed. The
similarity in the proportion of CD4-expressing thymocytes bearing the
transgenic (40%) vs endogenously rearranged (44%) TCR (Fig. 5B) demonstrates true chimerism of the reconstituted hosts.
Almost two-thirds of apoptotic thymocytes in OVA-treated mice were
KJ1-26+, and the remainder (30%) were
KJ1-26-, demonstrating that the majority of
thymocyte apoptosis after Ag administration to TCR-transgenic mice is a
consequence of TCR engagement on thymocytes. However, some
non-Ag-specific thymocytes also die as a consequence of peripheral or
non-Ag specific effects such as GC and may indirectly involve IL-2. The
results from these experiments also corroborate the results of similar
experiments investigating mechanisms of thymocyte death with influenza
nucleoprotein-reactive F5 TCR-transgenic mice
(39).
In our second approach, DO11.10 mice homozygous for the mutated IL-2
gene (DO11.10 x IL-2-/-) were obtained
from matings of DO11.10 and BALB/c IL-2+/- mice.
At 6 wk of age, CD4 and CD8 thymocyte subset distribution and TCR
expression levels were comparable to that of DO11.10 x
IL-2+/+ mice (Fig. 6, and data not shown). Interestingly,
although BALB/c IL-2-/- mice display severe
anemia and autoimmune disease soon after birth and the majority
die within 5 wk of age (20), DO11.10 x
IL-2-/- mice (n = 8) remained
disease free (our unpublished observations). Presumably, the
increased lifespan and absence of lymphoid hyperplasia and autoimmunity
in these animals is attributable to the skewing of the CD4 T cell
repertoire to a single exogenous non-self Ag.
|
). In contrast, the level of apoptosis in the thymi of
DO11.10 x IL-2-/- mice after OVA
treatment was similar to that seen after administration of PBS to
DO11.10 x IL-2-/- mice. The resistance of
DP thymocytes to AICD in IL-2-deficient DO11.10 mice was confirmed from
comparisons of the cellularity and absolute number of thymocyte subsets
in these and mock-treated animals (Table I). This result is consistent
with the ability of anti-IL-2R Abs to protect Ag-specific DP
thymocytes from apoptosis (Fig. 5) and provides further evidence
for the direct involvement of the IL-2/IL-2 signaling pathway in
thymocyte death and selection.
A different outcome was seen for the deletion of Ag-specific MHC class
I-restricted thymocytes in H-Y x IL-2-/-
mice. There was no significant difference in the cellularity or
thymocyte subset distribution in male H-Y x
IL-2+/+ and H-Y x
IL-2-/- mice (Fig. 7), and the onset and severity of
lymphoid hyperplasia and autoimmune disease was similar to that of
C57BL/6 IL-2-/- mice (data not shown). The
absence of IL-2 had no obvious effect on the deletion of H-Y-specific
CD8+ thymocytes. The cellularity and profile of
H-Y-specific thymocytes in the positive selecting environment of the
female thymi was similar irrespective of the presence or absence
of IL-2.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It is possible that the thymocyte apoptosis we have observed is mediated by GCs released by the adrenal medulla in response to activation of peripheral T cells (reviewed in Ref. 40). However, the fact that Ag-specific deletion is inhibited by anti-IL-2Rs in neonatal DO11.10-transgenic mice that have few circulating responsive peripheral T cells and that the death of the majority of Ag-specific thymocytes in bone marrow chimeras is a consequence of TCR engagement on thymocytes argues against this being the sole explanation for our observations. Indeed, the recent finding that thymocyte selection proceeds normally in GCR-/- mice (38) convincingly demonstrates that GCR signaling does not play a central role in intrathymic T cell development.
The apparent lack of a requirement for IL-2/IL-2R interactions in deletion of MHC class I-restricted HY-specific thymocytes is consistent with the results from other transgenic mouse model systems in which class I-restricted TCRs specific for endogenous or exogenous Ags have been crossed to IL-2-/- or CD25-/- mouse strains. Ag-mediated selection has been shown to occur normally in F5 nucleoprotein-specific MHC class I-restricted TCR-transgenic mice crossed onto the IL-2-/- background (30) and in H-Y TCR-transgenic mice bred onto the CD122-/- (41) and CD132-/- (42, 43) backgrounds. One possible conclusion from these findings is that IL-2/IL-2R-signaling might distinguish between CD4 and CD8 T cell selection. However, caution is warranted in view of unique properties and different mechanisms involved in each experimental model system making it difficult to directly compare the results obtained with different model systems. For example, differences in the properties of the selecting Ag in each model system including its origin (exogenous vs endogenous), nature (free or cell-associated), distribution (cortex vs medulla), concentration, mode, and duration of presentation in the thymus, and the possibility that the involvement of the IL-2/IL-2R signaling pathway is determined by some or all of these factors could account for the different outcomes for CD4 vs CD8 T cell selection in the absence of intrathymic IL-2/IL-2R signaling. Therefore, our findings may be unique to the experimental system and TCR-transgenic mouse strain we have used.
Our data also demonstrate that it is unlikely that IL-2/IL-2R interactions mediate negative selection of all MHC class II-restricted thymocytes. Only a subset of apoptotic Ag-specific or anti-CD3-stimulated thymocytes express functional IL-2Rs, and blocking or disrupting IL-2/IL-2R interactions results in incomplete protection from Ag-specific apoptosis. This may reflect the relative inefficiency of deletion in the experimental systems we have used or that IL-2/IL-2R involvement in thymocyte selection is restricted to specific cell populations and/or specific microenvironmental conditions under which developing T cells are exposed to Ag. The observation that superantigen (Mtv)-mediated deletion of thymocytes is incomplete in CD132- (44) and Jak3-mutant mice (28) implies the existence of multiple mechanisms to ensure the escape of as few autoreactive T cell clone as possible from the thymus. In our analyses of H-2b IL-2-/- mice, we too have been unable to detect any abnormalities in deletion of superantigen-reactive thymocyte populations (our unpublished observations).
Collectively, our findings may help explain the underlying basis of the abnormalities described in IL-2- (19, 20, 21, 34), IL-2R chain- (24, 25, 45), and IL-2R signaling molecule-deficient mice (26, 27). Defective negative selection would allow for the escape of autoreactive MHC class II-restricted clones from the thymus that proliferate on encountering self-Ag in the periphery but fail to undergo efficient AICD (23, 24, 46). The occurrence of lymphoid hyperplasia and autoimmune disease in gnotobiotic (germfree) IL-2-/- mice similar to that in SPF IL-2-/- mice (21) demonstrates that these disorders are attributable to a defect in central and/or peripheral tolerance that leads to uncontrolled responses to self and not environmental Ags. The observation that directed expression of CD122 to the thymus in CD122-deficient mice prevents the occurrence of CD4 T cell-mediated autoimmunity (29) is consistent with our hypothesis that IL-2/IL-2R interactions play a critical role in the maintenance of intrathymic central tolerance. Observations regarding defective IL-2/IL-2R-signaling in other unrelated autoimmune diseases such as diabetes (47, 48, 49) corroborate this hypothesis.
The source and use of IL-2 within the thymus during negative selection may be autocrine or paracrine in nature. Considering the short half-life and limited range of IL-2 in vivo (50) and the expression of IL-2 protein in close association with clusters of apoptotic cells in situ in the intact thymus, it is likely that IL-2 is produced locally within the thymus. Preliminary studies to identify the cellular source of IL-2 in the thymus during negative selection indicate that nonapoptotic Ag-specific thymocytes produce IL-2 in culture in response to Ag presented by a population of thymic macrophages (our unpublished observations). Moreover, these studies also indicate that this IL-2 production is potentiated by interactions between CD4SP and DP thymocytes and thymic macrophages. These observations underscore the importance of cross-talk between various thymocyte populations and the thymic stroma and suggest the existence of a feedback mechanism by which CD4 T cells could control the level or rate at which unwanted thymocytes are culled from the repertoire.
Our results may help clarify some of the seemingly contradictory findings regarding the role of IL-2 in thymocyte development (reviewed in Ref. 16). IL-2/IL-2R interactions in the thymus, similar to those in the periphery, are capable of both promoting and inhibiting thymocyte survival. These seemingly paradoxical effects may be explained by a developmental-stage-specific responsiveness of thymocytes to the activity of IL-2. Signals generated via the IL-2R may promote thymocyte differentiation by stimulating the proliferation and expansion of TCR- immature thymocyte populations while they promote the death of TCR+ thymocytes on TCR engagement at the DP and CD4SP selection-susceptible maturational stages. In addition, there may be inherent differences in the production of IL-2 and the mediation of these effects in the presence or absence of thymic APC populations. Indeed, our finding that thymic APC populations are compromised in IL-2-/- mice (34), combined with the requirement for these cells for production of IL-2 might exacerbate the defect in TCR-mediated thymocyte apoptosis in these mice. Finally, some of the controversy regarding the role of IL-2 in thymocyte development also may stem from intrinsic differences between fetal and adult thymic stromal cells and T cell precursor populations. Therefore, we cannot exclude the possibility that there may not be a strict requirement for IL-2/IL-2R interactions in negative selection of fetal thymocytes.
In summary, the results of our analyses demonstrate a role for IL-2/IL-2R interaction in the mediation of thymic negative selection of certain CD4 T cells. Therefore, IL-2R-generated signals may represent a unique and important pathway by which negative selection is modulated, and defects in this signaling pathway may result in loss of lymphoid homeostasis and autoimmunity.
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Simon R. Carding at his current address, School of Biochemistry and Molecular Biology, University of Leeds, Irene Manton Research Building Room 8.91 h, Leeds LS2 9JT, U.K.
3 Abbreviations used in this paper: DP, double positive; SP, single positive; DN, double negative; SPF, specific pathogen-free; DUNTL, dUTP-mediated nick translation labeling; AICD, activation-induced cell death; GC, glucocorticoid; GCR, GC receptor.
Received for publication October 16, 2000. Accepted for publication March 9, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
)-deficient T cells expressing a transgenic antigen receptor. J. Immunol. 158:3738.[Abstract]
chain regulates the size and content of the peripheral lymphoid compartment. Immunity 3:521.[Medline]
. Science 268:1472. T cells: a mechanism for the loss of T cells in patients with pulmonary tuberculosis. J. Immunol. 161:1558. chain expression by self-recognition in the thymus. J. Exp. Med. 180:1629.- and -chain cytoplasmic domains is required for signalling. Nature 369:330.[Medline]
-chain-deficient mice. Int. Immunol. 9:1367.-chain in intrathymic and peripheral T cell selection. J. Exp. Med. 183:1111.-chain (c) in thymocyte selection. Immunol. Today 21:88.[Medline]
-chain plays an essential role in regulating lymphoid homeostasis. J. Exp. Med. 185:189.-chain. Proc. Natl. Acad. Sci. USA 92:377.This article has been cited by other articles:
|
|
 |
|
  L. Zheng, R. Sharma, F. Gaskin, S. M. Fu, and S.-T. Ju A Novel Role of IL-2 in Organ-Specific Autoimmune Inflammation beyond Regulatory T Cell Checkpoint: Both IL-2 Knockout and Fas Mutation Prolong Lifespan of Scurfy Mice but by Different Mechanisms J. Immunol., December 15, 2007; 179(12): 8035 - 8041. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. R. Gothert, R. L. Brake, M. Smeets, U. Duhrsen, C. G. Begley, and D. J. Izon NOTCH1 pathway activation is an early hallmark of SCL T leukemogenesis Blood, November 15, 2007; 110(10): 3753 - 3762. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Yu and T. R. Malek Selective Availability of IL-2 Is a Major Determinant Controlling the Production of CD4+CD25+Foxp3+ T Regulatory Cells J. Immunol., October 15, 2006; 177(8): 5115 - 5121. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Selvaraj, D. Bunick, C. Finnigan-Bunick, R. W. Johnson, H. Wang, L. Liu, and P. S. Cooke Gene Expression Profiling of 17{beta}-Estradiol and Genistein Effects on Mouse Thymus Toxicol. Sci., September 1, 2005; 87(1): 97 - 112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Nacsa, Y. Edghill-Smith, W.-P. Tsai, D. Venzon, E. Tryniszewska, A. Hryniewicz, M. Moniuszko, A. Kinter, K. A. Smith, and G. Franchini Contrasting Effects of Low-Dose IL-2 on Vaccine-Boosted Simian Immunodeficiency Virus (SIV)-Specific CD4+ and CD8+ T Cells in Macaques Chronically Infected with SIVmac251 J. Immunol., February 15, 2005; 174(4): 1913 - 1921. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. H. Nelson IL-2, Regulatory T Cells, and Tolerance J. Immunol., April 1, 2004; 172(7): 3983 - 3988. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Xiao, S.-s. J. Sung, S. M. Fu, and S.-T. Ju Combining Fas Mutation with Interleukin-2 Deficiency Prevents Colitis and Lupus: IMPLICATING INTERLEUKIN-2 FOR AUTO-REACTIVE T CELL EXPANSION AND Fas LIGAND FOR COLON EPITHELIAL CELL DEATH J. Biol. Chem., December 26, 2003; 278(52): 52730 - 52738. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Freywald, N. Sharfe, C. Rashotte, T. Grunberger, and C. M. Roifman The EphB6 Receptor Inhibits JNK Activation in T Lymphocytes and Modulates T Cell Receptor-mediated Responses J. Biol. Chem., March 14, 2003; 278(12): 10150 - 10156. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Lindemann, M. Benczik, and S. L. Gaffen Anti-apoptotic Signaling by the Interleukin-2 Receptor Reveals a Function for Cytoplasmic Tyrosine Residues within the Common gamma (gamma c) Receptor Subunit J. Biol. Chem., March 14, 2003; 278(12): 10239 - 10249. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. E. Cheng and P. D. Greenberg Selective Delivery of Augmented IL-2 Receptor Signals to Responding CD8+ T Cells Increases the Size of the Acute Antiviral Response and of the Resulting Memory T Cell Pool J. Immunol., November 1, 2002; 169(9): 4990 - 4997. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. N. D'Souza, K. S. Schluns, D. Masopust, and L. Lefrancois Essential Role for IL-2 in the Regulation of Antiviral Extralymphoid CD8 T Cell Responses J. Immunol., June 1, 2002; 168(11): 5566 - 5572. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-H. Yeh, S. Spicuglia, S. Kumar, A. Sanchez-Sevilla, P. Ferrier, and J. Imbert Control of IL-2R{alpha} gene expression: structural changes within the proximal enhancer/core promoter during T-cell development Nucleic Acids Res., May 1, 2002; 30(9): 1944 - 1951. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. E. Cheng, C. Ohlen, B. H. Nelson, and P. D. Greenberg Enhanced signaling through the IL-2 receptor in CD8+ T cells regulated by antigen recognition results in preferential proliferation and expansion of responding CD8+ T cells rather than promotion of cell death PNAS, March 5, 2002; 99(5): 3001 - 3006. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
