| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Institute of Virology and Immunobiology, Würzburg, Germany; and
Division of Molecular Immunology, National Institute for Medical Research, London, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-specific Abs or TCR-specific agonist
peptides to induce negative selection. In the presence of the
MEK-specific pharmacological inhibitors PD98059 or UO126, cell recovery
was enhanced and deletion of DP thymocytes was drastically reduced.
Furthermore, development of CD4 SP T cells was blocked, but
differentiation of mature CD8 SP T cells proceeded in the presence of
agonist peptides when MEK activity was blocked. Thus, our data indicate
that the outcome between positively and negatively selecting signals is
critically dependent on MEK activity. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chains are rearranged and expressed, and selection
processes, which minimally involve interactions of the TCR and its
coreceptors with the appropriate MHC/self-peptide ligands, will decide
whether immature thymocytes continue to differentiate into
CD4+ or CD8+ lineage T
cells, a process known as positive selection, or whether they will die
by apoptosis due to negative selection or neglect (2, 3).
Indeed, only a minority of DP thymocytes receives appropriate signals
for further maturation, whereas the majority of thymocytes (90–95%)
will die by apoptosis, ensuring that useless and potentially
self-reactive T cells do not enter the peripheral pool of T cells. The
avidity/affinity of TCR-peptide/MHC and other receptor ligand
interactions will determine survival or death depending on the TCR
signaling events that are initiated and the signaling pathways that are
involved or dominate during individual stages and time points of
selection (4).
In recent years it has become evident that signal transduction in
peripheral T cells as well as in thymocytes is controlled by the
regulation of enzyme activation and the organization of enzymatic
complexes with adaptor, scaffold, and anchor proteins (5).
TCR proximal signaling events involve such important molecules as Lck
(6), Zap-70 (7), Slp76 (8, 9),
Lat (10), Cbl (11), Itk (12, 13), or Csk (14) and the more distal signaling
molecules p21ras (15), Vav
(16, 17, 18), phospholipase C 
(PLC) (19),
protein kinase C (PKC) (20), and calcineurin (21, 22). Genetic elimination or functional inactivation of these and
other individual components has resulted in either altered positive or
negative selection (or both as in the case of Vav, Zap-70, and CD45
(23)). It is still largely unknown how these proximal TCR
signaling events connect to and interconnect with signaling pathways
further downstream and eventually regulate transcription factors such
as members of the NF-AT (24), NF-
B (25),
E2A (26), or IFN regulating factor (IRF) (27)
families. However, several intriguing observations have shown that the
highly conserved mitogen-activated protein (MAP) kinase cascades, the
extracellular signal-regulated kinase (ERK), the c-Jun N-terminal
kinase (JNK), and the p38 MAP kinase cascades, which have been found to
regulate growth, apoptosis, and differentiation of peripheral T cells
(28), also play a major role in thymocyte selection.
Specifically, using dominant-negative ras (29)
and mek1 (30, 31) constructs in transgenic
mice, gain-of-function mutations of erk2 (32)
and mek1 (33), or erk2 knockout mice
(34), the p21ras-Raf-MAPK/ERK
kinase (MEK)-ERK cascade has been shown to be involved in positive
selection but appeared to be dispensable for negative selection.
Moreover, in vitro retroviral gene transfer experiments have
demonstrated a central role for the MAP kinase kinase 6 (MKK6)-p38
pathway in negative selection (33), and most recently,
using dominant-negative constructs in transgenic mice, JNK has also
been found to regulate negative selection in thymocytes
(35). Because TCR signaling might rule over survival or
death and coreceptor signaling over lineage choice, we reevaluated the
involvement of the MEK-ERK pathway in negative selection by culturing
thymic lobes from newborn TCR transgenic mice under conditions that
have been shown to induce negative selection of thymocytes. We used
either nominal agonist peptides specific for class I- or class
II-restricted transgenic TCRs or anti-CD3 Abs and analyzed the
deletion of DP thymocytes when thymic lobes were cultured in the
absence or presence of the MEK-specific inhibitors PD98059 or UO126.
Here, we show that in newborn thymocytes MEK signaling is involved in
negative selection. Deletion of DP thymocytes by either agonist
peptides or anti-CD3 Abs was dramatically reduced by MEK
inhibitors. In addition, differentiation of mature CD8 single positive
(SP) T cells took place in the presence of the apoptosis-inducing
reagents in numbers equal to those in unstimulated control cultures
when MEK activity was blocked. We also show that apoptosis induced by
-irradiation or hydrocortisone is significantly reduced by
inhibition of MEK. Our data suggest that MEK signaling regulates cell
survival and positive selection vs negative selection and cell death.
Thus, it appears that TCR/MHC signals that involve a high level and/or
continuous activation of MEK-ERK lead to or contribute to negative
selection, whereas intermediate MEK-ERK signaling allows positive
selection of CD4 and CD8 T cells to occur and TCR-induced MEK signaling
below a threshold may result in death due to neglect.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The TCR transgenic mice used were MHC class I-restricted F5 TCR (36) backcrossed onto a RAG-1-deficient (RAG-1neg) background (37) specific for influenza nucleoprotein restricted by H-2Db and MHC class II-restricted DO11.10 TCR (38) specific for chicken (OVA) in the context of I-Ad. To yield F5/RAG-1neg neonates, male or female TCR homozygous RAG-1neg mice were bred with RAG-1neg mice. D011.10 mice used were homozygous for TCR. B10 mice were purchased from Charles River Breeding Laboratories (Sulzfeld, Germany). ß2-Microglobulin-deficient (ß2m-/-) mice (39) and I-Aß-deficient (class II-/-) (40) mice were intercrossed to obtain ß2m-/- x class II-/- (MHC-/-) mice.
Antibodies
mAbs were purified and conjugated to FITC or biotin in our own laboratory or were bought from PharMingen (San Diego, CA) unless stated otherwise.
Thymus organ culture
Neonatal (day of birth) thymus lobes were cultured in RPMI
1640/10% FCS medium as described (41, 42) with or without
the indicated agonist peptides: for D011.10 mice, OVA peptide 323–339
(38) and for F5 mice, nucleoprotein 68 (NP68) peptide
366–374 (36) or with anti-CD3 Ab (145.2C11; Ref.
43). MEK inhibitors PD98059 (Calbiochem, San Diego, CA and
New England Biolabs, Beverly, MA) or UO126 (Promega, Madison, WI) were
added to thymus lobe cultures 1–2 h before Ab or peptide addition.
Unless otherwise stated in the figure legends, inhibitors were
replenished daily for the first 2–3 days of culture after which lobes
were transferred to fresh culture medium without Ab or inhibitor for
the final 1–3 days of culture to allow reexpression of down-modulated
molecules. Single-cell suspensions prepared from cultured lobes were
stained with FITC-CD8 (YTS169.4; Ref. 44), PE-CD4
(GK1.5; PharMingen), and biotin-heat-stable Ag (HSA; M1/69;
PharMingen), biotin-Vß11 (KT11.5) for F5 TCR (45) or
clonotypic, biotin-labeled KJ1-26.1 for D011.10 TCR (46)
before staining with Streptavidin Red 670 (Life Technologies,
Gaithersburg, MD). A total of 10,000–20,000 live events (gated on
forward and side scatter profiles) were analyzed on a FACScan (Becton
Dickinson, San Jose, CA).
Proliferation assay
Thymocyte single-cell suspensions were obtained from equivalent numbers of thymus lobes cultured either with 3 µM NP68 peptide or in medium only with and without 50 µM MEK inhibitor PD98059 as described above. Aliquots of thymocytes were stained with anti-Vß11-BIO, CD8-FITC, and CD4-PE to determine the percent input of DP, CD8 SP, and double negative (DN) cells. A total of 1.8 x 105 thymocytes from each lobe culture were stimulated in triplicate in 96-well flat-bottom tissue culture plates with 1 x 106 H-2b splenic B10 cells. B10 splenic cell suspensions were irradiated with 3000 rad and pulsed with 10 µM NP68 peptide for 30 min at 37°C before being added to thymocytes. Cultures were kept at 37°C for 3 days and were pulsed with 1 µCi [3H]thymidine/well for the last 18 h of culture, harvested, and counted in a beta-scintillation counter. Data are presented as mean values from triplicates and as values corrected for the relative input of CD8 SP cells in the starting population.
Detection of apoptosis
To induce apoptosis, thymocytes from different mouse strains
(1 x 106/ml) were cultured in suspension
with NP68 peptide or hydrocortisone (Sigma, St. Louis, MO) or
thymocytes were -irradiated (3000 rad). PD98059, UO126, or DMSO were
added 1 h before apoptosis-inducing reagents were given.
Thymocytes from suspension cultures or from neonatal thymic organ
culture (NTOC) were stained with annexin V-FITC (PharMingen), according
to the manufacturer’s protocol, and 7-amino actinomycin D (7-AAD) (5
µg/ml; Sigma) for 1 h and were analyzed on the FACS. A total of
10,000 ungated events were acquired and the percentages of cells
staining for annexin V and 7-AAD were determined.
Determination of MAP kinase activation
Thymocytes from F5/RAG-1neg mice (1 x 108 cells/ml) were kept at 37°C in RPMI 1640 medium for 5 h. Cells were preincubated with 100 µM MEK inhibitor PD98059 for 60 min at 37°C in medium before stimulation of 3 x 107 cells with 2C11 Abs or NP68 peptide in concentrations as given and for the indicated times. Cells (3 x 106 thymocytes/sample) were pelleted, resuspended in 2x reducing sample buffer, and resolved on a 12.5% SDS-PAGE gel. After transfer to polyvinylidene difluoride membranes (Amersham Buchler, Germany), proteins were detected with ERK or phospho-ERK-specific Abs (New England Biolabs), goat anti-rabbit HRP (PharMingen) and the enhanced chemiluminescence detection system (Amersham). For detection of JNK or p38 activity, thymocytes from B10 mice were cultured as described above and treated with 100 µM PD98059 or 40 µM UO126 for 1 h before stimulation with PMA (Sigma) or PMA plus ionomycin (Sigma). A total of 3 x 106 cells were lysed for 20 min at 4°C in lysis buffer (20 mM HEPES (pH 7.4), 2 mM EDTA, 50 mM ß-glycerophosphate, 1% Triton X-100, 10% glycerol, 50 mM NaF, 0.04% azide, 1 mM DTT, 1 mM orthovanadate, 2 µM leupeptin, and 4 mM PMSF). JNK and p38 kinase activity were determined using phospho-JNK and phospho-p38-specific Abs (Santa Cruz Biotechnology, Santa Cruz, CA) detected by goat anti-mouse HRP (Dianova, Hamburg, Germany). To control for equal protein loading, blots were reprobed with anti-actin Abs (Sigma).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The p21ras-Raf-MEK-ERK cascade is a central
signaling pathway in thymocyte differentiation because it was shown to
be involved in the transition of
CD4-CD8- thymocytes to DP
thymocytes (47, 48) and to be necessary and sufficient for
positive selection (29, 30, 31, 32, 33, 34, 49). In addition, we and
others have shown that MEK activity is an arbiter in T cell lineage
decision (32, 50) and is involved in the final maturation
of T cells, because in the continuous presence of pharmacological MEK
inhibitors, the up-regulation of the TCR in
HSAlow SP T cells is prevented (51).
In the course of these earlier experiments, we observed that the
cellularity of neonatal lobe cultures was increased in the presence of
MEK inhibitors, suggesting that MEK might also influence negative
selection and/or thymocyte death by neglect. We routinely used neonatal
(day 0) thymic lobes (NTOC) from TCR transgenic mice starting with a
situation in which thymocytes have expanded normally to the DP subset,
thus minimizing any additional effects of the MEK inhibitors on DP
thymocyte precursors. First, we examined the influence of the MEK
inhibitor PD98059 (52, 53) on the negative selection of
neonatal thymocytes from F5/RAG-1neg mice, which have an MHC class
I-restricted TCR and differentiate to the CD8 lineage only. The F5 TCR
recognizes a nonamer peptide from influenza nucleoprotein (NP68) that
acts as an agonist on peripheral T cells and induces negative selection
of transgenic thymocytes (54). Neonatal lobes were
cultured in medium or with varying concentrations of NP68 peptide in
the presence or absence of the inhibitor PD98059. Inhibition of MEK has
clear effects on thymus differentiation, as shown by the results
represented both as FACS profiles from individual lobe cultures in Fig. 1
and as absolute cell numbers in Table I. Addition of NP68 lead to a 2- to
5-fold reduction in the recovery of viable cells compared to the
recovery from medium controls as determined by trypan blue exclusion.
Most prominent was the drastic decline of the DP population with up to
a 15-fold reduction in the presence of NP68. The number of CD8 SP T
cells was also strongly reduced, most likely because of the depletion
of their DP precursors. In the presence of the MEK inhibitor PD98059,
total cell numbers were generally higher than those in the medium
controls and, strikingly, cultures treated with NP68 plus PD98059
reached values approaching those from lobes cultured in medium only,
indicating that peptide-induced apoptosis of thymocytes was greatly
reduced. This was reflected by a 4- to 18-fold increase in cell numbers
for the DP thymocyte population, allowing the conclusion that in the
presence of the MEK inhibitor fewer DP thymocytes had been induced to
die by peptide stimulation. It should be emphasized that the increase
in the number of DP cells in NP68 plus PD98059 cultures surmounted the
increase in medium plus PD98059 cultures, i.e., peptide-induced
deletion clearly had been reduced by PD98059 treatment, and the
increase was not solely due to rescue from death by neglect. In
addition, substantial numbers of CD8 SP T cells were detected in NP68
plus PD98059-treated cultures, indicating that many DP thymocytes
rescued from apoptosis could differentiate further into CD8 lineage T
cells. As shown in Table I, the inhibition of MEK not only allowed
maturation of CD8 SP T cells in the presence of deleting NP68 peptide,
but the number of HSAlow CD8 SP T cells was
increased compared with medium only cultures and was similar to
cultures treated with PD98059. Thus, inhibition of MEK to a great
extent abrogates the effects induced by the nominal peptide, changing
the negative signal into signals appropriate for unimpaired or enhanced
development of CD8 SP T cells.
|
|
B). Almost all of the CD8 SP T
cells showed low HSA expression, and TCR levels were comparable to
those of CD8 SP T cells from medium controls, indicating that CD8 T
cells differentiating in the presence of NP68 and PD98059 indeed
developed and matured normally. CD8 SP T cells, which differentiated in the presence of NP68 peptide and PD98059, are responsive to Ag
To test whether these phenotypical mature CD8 SP T cells were
functional, cells were harvested from NTOCs from F5/RAG-1neg mice that
were cultured in the presence of 3 µM NP68 with or without PD98059,
and equal numbers of cells were stimulated with APCs plus antigenic
peptide NP68. Proliferation measured on day 3 of culture is shown in
Fig. 2 without and after adjustment of
values to the percentage of CD8 SP T cells contained in the total lobe
suspensions at the onset of secondary culture. Thymocytes from NP68
peptide only-treated lobe cultures gave an unexpectedly high
proliferative response similar to that of thymocytes from untreated
lobe cultures. However, it should be pointed out that the recovery of
viable cells in the peptide-treated lobe cultures was about 3-fold
lower than in medium cultures (0.9 x 106 vs
2.8 x 106), whereas NP68 plus MEK
inhibitor-treated cultures had a cell number close to that of the
medium control (2.1 x 106) so that in spite
of a similar reponse after adjustment to the same cell density,
effectively less CD8 SP cells had matured in the peptide only-treated
lobe cultures. In addition, the DN population in NP68-treated cultures
is overrepresented, and because a high percentage of these DN cells
have a mature phenotype (data not shown), they might contribute to the
proliferative response. Proliferation of CD8 SP cells, which
differentiated in the presence of the MEK inhibitor and peptide, was
comparable to that of cells cultured with inhibitor alone. The
proliferative response of these CD8 cells was lower than that from
lobes cultured in medium or NP68 only. This might be due to an
incomplete recovery from prior PD98059 treatment in the lobe cultures,
because in the presence of PD98059, proliferation was strongly affected
(see Fig. 2). The latter also indicates that CD8 SP T cells from lobe
cultures treated with peptide and inhibitor did not arise by expansion
of a "rare" population but rather differentiated de novo.
Altogether, the data show that CD8 T cells that matured in the presence
of agonist peptide plus MEK inhibitor are capable of responding to
Ag.
|
-induced negative selection of DP thymocytes is blocked
by MEK inhibition
CD3 engagement of the surface TCR/CD3 complex by Abs is
known to induce apoptosis (55), a process that is thought
to mimic Ag-induced negative selection in the thymus. To analyze
whether Ab-induced apoptosis can be antagonized by the inhibition of
the MEK pathway, newborn lobes from F5/RAG-1neg mice were cultured with
different anti-CD3 concentrations in the absence or presence of
PD98059 or UO126, a second, recently described noncompetitive inhibitor
of MEK (56). As illustrated in Fig. 3, UO126 inhibition of MEK has the same
consequence as inhibition by PD98059, even though UO126 blocks MEK
activity on downstream targets and PD98059 blocks activation by Raf. In
both cases, deletion of DP thymocytes is drastically reduced and
maturation of CD8 SP T cells is comparable to that of medium cultures.
UO126 treatment results in an even more pronounced inhibition of
deletion of DP thymocytes than PD98059 does, which is consistent with
its ability to block constitutively active MEK and its reported higher
affinity for MEK. It should be noted that in the presence of UO126, the
number of differentiating CD8 SP cells was lower compared with that in
PD98059-treated cultures, although more DP cells were rescued from
death. In this context we observed that daily addition of higher doses
of UO126 resulted in massive death of thymocytes (data not shown). One
possible explanation is that thymocytes are highly sensitive to changes
in the level of MEK activity and that certain thresholds of MEK
activity have to be surpassed for positive selection to occur.
|
To analyze whether deletion of DP thymocytes bearing class
II-restricted TCRs could be influenced by inhibition of MEK, neonatal
lobes from DO11.10 TCR transgenic mice were cultured with chicken OVA
peptide 323–339 (38), as shown in Fig. 4, and as absolute cell numbers in Table II. In Fig. 4, neonatal lobes were
cultured with 1 µg/ml OVA peptide and 50 µM PD98059 for 2 days, and
thymocytes were stained for expression of CD4, CD8, and maturation
markers on day 5. Addition of OVA peptide led to a 5-fold reduction in
the recovery of live cells and an 
10-fold decline in the number of
DP thymocytes. In the presence of PD98059, lobe cellularity was close
to that in the medium control, and the number of DP thymocytes reached
about half of that in the medium control. In addition to this
protective effect on OVA peptide-induced deletion, which led to 5- to
20-fold increases in DP cell numbers after PD98059 or UO126 addition
compared with maximal 4-fold increases of DP cells in medium-only
cultures, we observed an enhancement of CD8 cell maturation in MEK
inhibitor-treated cultures (see Table II). As shown in the histograms
of Fig. 4, the majority of CD8 T cells induced by PD98059 in the
presence or absence of OVA peptide had down-regulated the HSA marker,
and all had up-regulated the TCR, indicating that they were bona fide
selected cells rather than immature CD8 cells arrested at their
transition to the DP stage. Together, the data from Fig. 4 and Table II
show that signals delivered by engagement of class II-restricted TCRs
with MHC/agonist peptide, which normally initiate negative selection of
DP thymocytes, can be altered or dampened by inhibition of MEK to
signals that are appropriate for positive selection and maturation of
CD8 SP T cells bearing class II-restricted TCRs.
|
|
Because deletion of DP thymocytes was greatly affected by
inhibition of MEK with the pharmacological inhibitors, we measured
ongoing apoptosis in day 3 NTOCs treated for 2 days with
anti-CD3 Abs and MEK inhibitor (Fig. 5Aa). Apoptotic cells were
determined by staining thymocyte suspensions with annexin V and 7-AAD.
Cultures in medium without or with inhibitor UO126 displayed similar
levels of apoptosis at day 3 with the levels clearly enhanced by
inclusion of anti-CD3 Abs. Importantly, addition of UO126
stabilized the percentage of dying cells at about the level seen in
cultures left in medium only. In Fig. 5A,
b–d, F5/RAG-1neg thymocytes were treated in
suspension culture with varying concentrations of NP68 without or with
PD98059 or UO126 for 10 or 16 h. In all medium cultures,
significant apoptosis that was enhanced by NP68 treatment in a
dose-dependent fashion was detected. Although neither UO126 nor PD98059
totally blocked cell death or the increase in apoptosis after peptide
treatment, the level of apoptosis was markedly lower for all points.
Thus, under conditions in which no signals or deleting signals only are
provided, apoptosis can be diminished by lowering MEK activation.
|
B, thymocytes from B10, DO11.10,
or MHC-/- mice were gamma-irradiated and
cultured in the presence or absence of different concentrations of
PD98059 or UO126. Sixteen hours after irradiation, 70–90% of
thymocytes from all strains were apoptotic, whereas MEK
inhibitor-treated cultures showed markedly reduced cell death reaching
apoptosis values similar to those from nonirradiated control cultures
in which 30–40% of cells stained positive for 7-AAD/annexin
V.
The analyis of apoptosis of thymocytes from F5/RAG-1neg, class
II-/-, or MHC-/- mice
induced by the glucocorticoid hydrocortisone (Fig. 5C)
showed similar results. UO126 and especially PD98059 substantially
reduced or totally blocked hydrocortisone-induced cell death within the
time period studied, i.e., within 16 h of culture. Thus
glucocorticoids, which have been shown to play an important role in
thymocyte apoptosis (57), might either directly signal via
MEK or use proteins activated or induced by MEK signaling.
Interestingly, survival of MHC-/- thymocytes
was also improved by MEK blockade, suggesting that death by neglect may
involve MEK signaling.
ERK but not JNK and p38 MAP kinase activation is blocked by PD98059 or UO126
Engagement of TCR by agonist peptides or anti-CD3 Abs has
been shown to activate the p21ras-Raf-MEK-ERK
pathway (58). Given that the MEK inhibitors greatly
reduced apoptosis of thymocytes in suspension culture and blocked
peptide- or anti-CD3-induced deletion of DP thymocytes in NTOC
whereas CD8 T cell maturation in NTOC was unimpaired, we analyzed to
what extent ERK activity was diminished by the pharmacological
inhibitors. Thymocytes from F5/RAG-1neg mice were cultured for 5 h
in medium, and PD98059 was added for 1 h before stimulation with
the TCR agonists NP68 peptide or anti-CD3 for 2–5 min. A total
of 3 x 106 cells were analyzed for ERK
activation using phospho-ERK-specific Abs. Both NP68 and
anti-CD3 Abs induced ERK activation in a concentration-dependent
manner, and PD98059 was able to reduce ERK activation to levels
equivalent to those of medium-only controls, i.e., to levels of ERK
activation induced by endogenous selecting ligands that induce CD8
maturation. From these data we can assume that ERK activation was
indeed diminished when PD98059 was applied in NTOCs. Although we
replenished NTOCs daily with high concentrations of inhibitor, we do
not know for how long ERK activity was blocked in each case. Yet, it
seems that inhibition of ERK was high enough to dampen the signal
induced by agonist peptide or anti-CD3 Abs and thus to rescue a
high percentage of DP thymocytes from apoptosis. To exclude that the
high concentrations of PD98059 or UO126 applied in NTOC or suspension
cultures influenced JNK or p38 activity leading to indirect effects on
apoptosis, B10 thymocytes were stimulated with PMA or PMA plus
ionomycin, reagents that have been shown to induce JNK and p38
activation in thymocytes (59). Both MEK inhibitors totally
blocked ERK activation, whereas JNK or p38 activation as measured by
phospho-specific Abs was not affected (Fig. 6, B and C).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
complex, were altered by inhibition of MEK in such a way that a high
percentage of DP thymocytes was rescued from death and could mature
into the CD8 lineage. Inhibition of MEK per se also resulted in a
higher survival rate of thymocytes in thymic organ and thymocyte
sus-pension culture. Previous studies on ERK signaling in thymocytes
concluded that the MEK-ERK cascade is not involved in negative
selection (29, 30, 31, 33). In contrast, we detected a clear
role of MEK in negative selection and we like to explain these
discrepancies by two major differences in culture conditions: first, we
used newborn lobes (rather than fetal thymus lobes) from TCR transgenic
mice, in which 95% of thymocytes had matured to the DP stage in a
normal environment, excluding effects of MEK inhibition on transition
to DP thymocytes. In newborn lobes, many of the DP thymocytes probably
had numerous TCR/MHC/coreceptor and additional receptor-ligand
interactions (60) and might have received some signals for
lineage commitment, positive or negative selection, or death due to
neglect. Negative selection of thymocytes is not a one-hit process, but
thymocytes rather remain susceptible to negative selection over a
period of time (61, 62, 63) in the CD4 lineage even until they
have reached the HSAlowCD4+
stage (64). Thus, the sensitivity of DP thymocytes to
MEK-mediated signals and inhibition of MEK activity might be strongly
dependent on the time point and the amount and type of signals acquired
before (65). Indeed, so far we do not know which subset of
the DP thymocytes is actually rescued from deletion by inhibition of
MEK, "early" or "late" DP thymocytes or both.
Second and probably more important, we added MEK inhibitors daily, for
2–3 days in high concentrations (50 µM for PD98059 and 20 µM for
UO126), whereas in other studies lower concentrations of inhibitor were
used and were only applied once at the beginning of culture. We assume
that by our protocol more extensive and/or longer lasting inhibition of
MEK was achieved. UO126, which has a higher affinity for MEK than
PD98059, generally antagonized the deletion of DP thymocytes from
F5/RAG-1neg mice better than PD98059 (see Fig. 3), resulting in a
higher recovery of viable cells and DP thymocytes. Although the
different mechanisms of MEK inhibition by the two compounds might
include additional effects, these results indicate that DP thymocytes
are highly sensitive to the level of MEK-ERK activation. Higher
concentrations of UO126 blocked positive selection completely and
induced massive cell death instead (data not shown), effects which are
reminiscent of experiments from Alberola-Ila et al. (30, 31), which showed that dominant-negative forms of ras
or mek (each individually but particularly when present in
combination) could block positive selection, and of the strongly
reduced positive selection of both lineages in erk2 knockout
mice (34).
Interestingly, the development of mature,
TCRhigh, and HSAlow CD8 SP
T cells in the presence of deleting peptide or anti-CD3 Abs was
normal when MEK activity was blocked. The total number of CD8 SP T
cells that differentiated under these conditions was comparable to that
of medium controls and at lower peptide concentrations even reached the
enhanced maturation observed in cultures treated with PD98059 only.
This effect was most clearly visible in DO11.10 mice, in which MEK
inhibition during agonist peptide treatment blocked CD4 T cell
maturation, inhibited deletion of DP thymocytes, and induced CD8 cell
development. In an earlier study, we could show that signals that favor
CD4 SP cell differentiation, namely limited coligation of CD3 with
the coreceptors, induced CD8 development when MEK activity was blocked
by PD98059 (51). These signals, which behave as weak
agonists on peripheral T cells in the presence of PD98059, led to the
same outcome (CD8 differentiation) as artificial signals delivered by
CD3/CD3 F(ab')2 Abs that behave like antagonists
(42, 66). In this paper we used agonist peptides, which
were specific for class I- or class II-restricted TCR transgenic
thymocytes and agonist anti-CD3 Abs that efficiently induce ERK
activation (Fig. 6), in accordance with results from a number of groups
that showed that these stimuli lead to the activation of the
p21ras/ERK kinases, unlike activation by
antagonist peptides (67, 68, 69). In another case it was found
that antagonists can induce the same levels of ERK activation but that
the duration of activation was shorter for antagonists
(70). In this report we show that addition of MEK
inhibitor reduced ERK activation in thymocytes to medium control levels
and reduced apoptosis in thymocyte suspension cultures, indicating that
the reduction in negative selection and unimpaired/enhanced development
of CD8 T cells in thymus lobe culture was due to decreased MEK activity
or altered kinetics of MEK activation. We show that the high
concentrations of MEK inhibitors used do not influence the activation
of the JNK or p38 kinases, excluding that the reduction of apoptosis or
negative selection is the indirect result of altered activity of these
MAP kinases. In view of our results, we would like to propose the
following scenario: interaction of TCR with agonist
peptide/MHC/coreceptors or CD3 engagement delivers signals that
along the p21ras-MEK-ERK pathway lead to strong
and sustained activation of MEK, overriding the upper threshold of ERK
signaling below which positive selection can occur, resulting in
apoptosis of the majority of DP thymocytes. Because precursors are
deleted, differentiation of CD4 and CD8 T cells is critically
diminished. The only DP thymocytes that might be able to differentiate
further are those that might have committed to CD8 and CD4 T cells and
can differentiate despite strong MEK signaling or even may depend on it
for further immediate differentiation steps to occur. Inhibition of MEK
signaling by the highly specific inhibitors dampens the induced MEK-ERK
signaling to levels that are below the level for negative selection but
that are appropriate for positive selection to occur. Very strong or
prolonged inhibition of MEK would lead to signaling below a threshold
for positive selection and would finally result in apoptosis due to
neglect. On the other hand, thymocytes that might die due to neglect
could receive sufficient signals in the presence of the agonists plus
MEK inhibitor to survive and to be selected. Essentially, in this model
the degree of MEK-ERK signaling would decide over positive or negative
selection, implicating that agonists induce strong ERK activation
whereas antagonists induce moderate ERK activation. Indeed, inhibition
of MEK reduced ERK activation induced by agonists (NP68 or
anti-CD3 treatment) to the levels found in untreated cultures,
i.e., to the levels of ERK activation induced by endogenously selecting
ligands which in the F5/RAG-1neg mice induce CD8 maturation only. When
we treated thymocytes from F5/RAG-1neg mice with the antagonist peptide
NP34, ERK activation did not rise above the levels of medium controls
(data not shown), probably due to background ERK activation by
endogenous ligands. Smyth et al. (71) used F5/RAG-1neg
ß2m-/- mice, a system
in which no positive selection of CD8 cells occurs, to determine ERK
activation after treatment with agonist NP68 or antagonist NP34
peptides presented by Y01 epithelial cells. They showed that NP34
antagonist peptide induces ERK activation above medium levels but that
it was lower than ERK activation by agonist NP68 peptide. Thus, there
seems to be a gradient of MEK activity or ERK activation along which no
selection, positive selection, or negative selection is initiated.
The observed increase in the number of viable cells in PD98059-treated
medium cultures could be due to several causes: decrease in spontaneous
cell death inherent to the culturing of newborn lobes, which generally
show a lower survival rate than fetal lobe organ cultures do; prolonged
survival of DP thymocytes, which are rescued from negative selection or
hormone-induced death; or protection from death by neglect. That the
latter may contribute to increased cell survival is shown by the
reduction of apoptosis in suspension cultures of thymocytes from
MHC-/- or F5/RAG-1neg mice treated with the MEK
inhibitors only. Our novel finding that hydrocortisone-induced
apoptosis of thymocytes signals via MEK feeds into previous
observations from several groups that glucocorticoids play an important
role during thymocyte differentiation and that selection and apoptosis
of thymocytes is critically dependent on the interplay between TCR and
glucocorticoid-mediated stimuli (72, 73, 74). Because both
signal via MEK, the balance might be shifted toward survival and
positive selection or toward negative selection, depending on the
individual contributions. Notably, death of thymocytes after
gamma-irradiation was also greatly reduced by inhibition of MEK.
Although the signaling machinery induced by gamma-irradiation is not
well defined, Kasid et al. (75) showed for human carcinoma
cells that ionizing radiation leads to the activation of Raf-1 and
increased ERK activity. For thymocytes, it is known that in RAG-1 neg
mice treatment with anti-CD3 Abs (76) or
gamma-irradiation can induce the transition of DN thymocytes to the DP
stage (77, 78). Because we have shown that
gamma-irradiation in thymocytes also involves MEK, this could be a
simple explanation for the same outcome after administering such
different stimuli, i.e., mimicking pre-TCR-mediated MEK signaling.
Finally, the reduced cell death in MEK inhibitor-treated cultures could
be due to rescue from death of thymocytes whose TCRs cross-react with
multiple ligands (79, 80) including Ag variants (81, 82), superantigens (83), or related self peptides
(84, 85, 86). Whatever the mechanism of increased cell
survival in MEK inhibitor-treated newborn lobe cultures may be, we like
to emphasize that the increased recovery of DP thymocytes in
agonist plus MEK-treated cultures results from an additional specific
blockade of agonist-induced negative selection.
Elegant studies previously have shown that antagonist peptides can induce positive selection of CD8 T cells, and low concentrations of peptide were found to induce positive selection whereas higher concentrations lead to the deletion of thymocytes (87, 88, 89, 90). Furthermore, antagonist peptides (91) or antagonist-like Abs (42, 92) could block positive selection of CD4 T cells, select class II-restricted T cells into the CD8 lineage (93), inhibit negative selection by agonist peptides (94, 95), or block selection of CD8 SP T cells (96). Because the effects of inhibition of MEK by PD98059 or UO126 (low or lack of ERK activation, block of negative selection, inhibition of CD4, and enhancement of CD8 T cell maturation) are reminiscent of the effects induced by antagonists, one could assume that by competing interactions of TCR with agonist- or antagonist-like self-peptides the overall MEK-ERK signaling or "net result" of ERK activation at a given time and within certain niches (97, 98) is shifted toward the thresholds for either positive or negative selection.
The involvement of ERK signaling in negative selection of course has to be seen in combinatorial action with the p38 or JNK pathways (33, 35, 99, 100, 101) and with costimulatory molecules such as CD28 that preferentially signal through these cascades (23, 102). However, it seems plausible that molecules that modify or induce additional MEK signaling influence the balance toward survival or death. For instance, it was shown that activation of ERK could cancel drug-induced apoptosis that was mediated by the JNK and p38 pathways (103). Another candidate likely to influence the level of ERK activation is CTLA4, which when inhibited was shown to inhibit anti-CD3-induced apoptosis of thymocytes by recruitment of the phosphatase src homology 2-containing protein tyrosine phosphatase-2 (SHP2) (104).
So far we do not know how inactivation of MEK signaling translates into
the survival of DP thymocytes. Strong ERK signaling may override
survival signals provided by members of the Bcl-2 family
(105) or may induce high levels of activated NF-B
(106, 107). The extent of ERK signaling might also be
critical for the activation of caspase systems (108) or
for the activity of the Cdk2 kinase, which recently was shown to be a
master regulator for various different stimuli inducing apoptosis
(109). Formally we cannot exclude that the MEK inhibitors
influence the expression of genes for receptor-ligand pairs expressed
on thymocytes and/or thymic stromal cells such as Notch and its ligand
(110, 111), the Fas-Fas ligand system (112, 113), or the CD40-CD40 ligand system (114) and
thereby indirectly influence negative selection.
Previously we proposed a model in which the discriminatory signals that direct differentiation to either the CD4 or CD8 subsets are a function of the extent to which src family kinases, particularly lck, are activated at the same time as TCR is engaging the ligand in the thymus. TCR engagements that lead to significant activation of lck are most appropriate for directing cells to the CD4 lineage, whereas limited activation of lck results in commitment to the CD8 lineage (42, 92). In addition, we could show that strong ERK signaling is required for CD4 SP maturation, whereas CD8 maturation requires a low level of ERK activation or none at all, whereby signals appropriate for CD4 maturation can be converted to signals inducing CD8 maturation when MEK activity is blocked by PD98059 (51). The data described in this report would support and extend this model in that the degree of ERK activation through the TCR sets thresholds for positive and negative selection. This would be in accordance with our previous finding that engagement of the TCR together with CD4 in very high concentrations did not lead to enhanced CD4 maturation but rather caused apoptosis of some DP thymocytes (41). The amount of lck activation and ERK signaling needed at a given time point to surpass the thresholds for positive or negative selection would depend on the integration of signals from other pathways such as the JNK or p38 signaling cascades. In the absence of ERK signaling, i.e., with ERK signaling below the threshold for positive selection to occur, these cascades could simply dominate or be incomplete, resulting in death. Future studies will resolve how the different MAP kinase pathways crosstalk or feed back on each other during thymic selection.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Ursula Bommhardt, Institute of Virology and Immunobiology, Versbacher Strasse 7, D-97078 Würzburg, Germany. E-mail address:
3 Abbreviations used in this paper: DP, double positive (CD4+CD8+); MAP, mitogen-activated protein; ERK, extracellular signal-regulated kinase; JNK, c-Jun N-terminal kinase; SP, single positive; RAG-1neg, RAG-1-deficient; ß2m-/-, ß2-microglobulin-deficient; class II-/-, I-Aß-deficient; MHC-/-, ß2m-/- x class II-/-; DN, double negative; NP68, nucleoprotein 68; NTOC, neonatal thymic organ culture; MEK, MAPK/ERK kinase; HSA, heat-stable Ag; 7-AAD, 7-amino actinomycin D.
Received for publication July 23, 1999. Accepted for publication December 15, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ß versus 
lineage commitment. Immunol. Rev. 165:111.[Medline]
1 in an SLP-76-deficient T cell. Science 281:413.1. J. Biol. Chem. 266:12135.
during T-cell activation. Nature 385:83.[Medline]
B: ten years after. Cell 87:13.[Medline]
ß T-cell development and to rapid development of T-cell lymphomas. Mol. Cell. Biol. 17:4782.[Abstract]
. Proc. Natl. Acad. Sci. USA 90:11401. engagement in RAG-1-/--immature thymocytes. Int. Immunol. 8:1159.B. J. Exp. Med. 185:1897.B in promoting double positive thymocyte apoptosis. J. Exp. Med. 189:145.t, a novel isoform of an orphan receptor, negatively regulates Fas ligand expression and IL-2 production in T cells. Immunity 9:797.[Medline]
This article has been cited by other articles:
|
|
 |
|
  S. E. Ziemba, S. L. Menard, M. J. McCabe Jr, and A. J. Rosenspire T-cell receptor signaling is mediated by transient Lck activity, which is inhibited by inorganic mercury FASEB J, June 1, 2009; 23(6): 1663 - 1671. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. K. McNeil, T. K. Starr, and K. A. Hogquist A requirement for sustained ERK signaling during thymocyte positive selection in vivo PNAS, September 20, 2005; 102(38): 13574 - 13579. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Simeoni, V. Posevitz, U. Kolsch, I. Meinert, E. Bruyns, K. Pfeffer, D. Reinhold, and B. Schraven The Transmembrane Adapter Protein SIT Regulates Thymic Development and Peripheral T-Cell Functions Mol. Cell. Biol., September 1, 2005; 25(17): 7557 - 7568. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. L. Kortum, D. L. Costanzo, J. Haferbier, S. J. Schreiner, G. L. Razidlo, M.-H. Wu, D. J. Volle, T. Mori, H. Sakaue, N. V. Chaika, et al. The Molecular Scaffold Kinase Suppressor of Ras 1 (KSR1) Regulates Adipogenesis Mol. Cell. Biol., September 1, 2005; 25(17): 7592 - 7604. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Kwon, H.-S. Jun, L.-Y. Khil, and J.-W. Yoon Role of CTLA-4 in the Activation of Single- and Double-Positive Thymocytes J. Immunol., December 1, 2004; 173(11): 6645 - 6653. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Tsukamoto, A. Irie, and Y. Nishimura B-Raf Contributes to Sustained Extracellular Signal-regulated Kinase Activation Associated with Interleukin-2 Production Stimulated through the T Cell Receptor J. Biol. Chem., November 12, 2004; 279(46): 48457 - 48465. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. F. Reynolds, C. de Bettignies, T. Norton, A. Beeser, J. Chernoff, and V. L. J. Tybulewicz Vav1 Transduces T Cell Receptor Signals to the Activation of the Ras/ERK Pathway via LAT, Sos, and RasGRP1 J. Biol. Chem., April 30, 2004; 279(18): 18239 - 18246. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L.-F. Belanger, S. Roy, M. Tremblay, B. Brott, A.-M. Steff, W. Mourad, P. Hugo, R. Erikson, and J. Charron Mek2 Is Dispensable for Mouse Growth and Development Mol. Cell. Biol., July 15, 2003; 23(14): 4778 - 4787. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. DeRyckere, D. L. Mann, and J. DeGregori Characterization of Transcriptional Regulation During Negative Selection In Vivo J. Immunol., July 15, 2003; 171(2): 802 - 811. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Norment, L. Y. Bogatzki, M. Klinger, E. W. Ojala, M. J. Bevan, and R. J. Kay Transgenic Expression of RasGRP1 Induces the Maturation of Double-Negative Thymocytes and Enhances the Production of CD8 Single-Positive Thymocytes J. Immunol., February 1, 2003; 170(3): 1141 - 1149. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-W. He Orphan nuclear receptors in T lymphocyte development J. Leukoc. Biol., September 1, 2002; 72(3): 440 - 446. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Bettini, H. Xi, J. Milbrandt, and G. J. Kersh Thymocyte Development in Early Growth Response Gene 1-Deficient Mice J. Immunol., August 15, 2002; 169(4): 1713 - 1720. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Shen, Y.-B. Ouyang, C.-K. Qu, A. Alonso, L. Sperzel, T. Mustelin, M. H. Kaplan, and G.-S. Feng Grap Negatively Regulates T-Cell Receptor-Elicited Lymphocyte Proliferation and Interleukin-2 Induction Mol. Cell. Biol., May 15, 2002; 22(10): 3230 - 3236. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Mariathasan, A. Zakarian, D. Bouchard, A. M. Michie, J. C. Zuniga-Pflucker, and P. S. Ohashi Duration and Strength of Extracellular Signal-Regulated Kinase Signals Are Altered During Positive Versus Negative Thymocyte Selection J. Immunol., November 1, 2001; 167(9): 4966 - 4973. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hayden-Martinez, L. P. Kane, and S. M. Hedrick Effects of a Constitutively Active Form of Calcineurin on T Cell Activation and Thymic Selection J. Immunol., October 1, 2000; 165(7): 3713 - 3721. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
). Gray bars (
) give data after
relating proliferation counts to the percentage of CD8 SP T cells in
the various lobe cultures, which were determined at the start of the
assay.
