(Received for publication, July 7, 1996, and in revised form, January 20, 1997)
From the Department of Immunology and Oncology, Centro Nacional de Biotecnología, Consejo Superior de Investigaciones Cientificas, Universidad Autónoma Campus de Cantoblanco, Cantoblanco, 28049-Madrid, Spain
Peripheral blood T lymphocytes require two
signals to enter and progress along the cell cycle from their natural
quiescent state. The first activation signal is provided by the
stimulation through the T cell receptor, which induces the synthesis of
cyclins and the expression of the high affinity interleukin-2 receptor. The second signal, required to enter the S phase, is generated upon
binding of interleukin-2 to the high affinity interleukin-2 receptor. However, resting T cells already express intermediate affinity
interleukin-2 receptors. As shown here, T cell
stimulation through intermediate affinity receptors is capable of
inducing cell rescue from the apoptosis suffered in the absence of
stimulation. Characterization of the signaling pathways utilized by
interleukin-2 receptors in resting T cells, indicated that
pp56lck, but not Jak1 or Jak3, is activated upon receptor
triggering. Compelling evidence is presented indicating that
phosphatidylinositol 3-kinase associates with the intermediate affinity
interleukin-2 receptor and is activated upon interleukin-2 addition.
Bcl-xL gene was also found to be induced upon
interleukin-2 receptor stimulation. Finally, pharmacological inhibition
of phosphatidylinositol 3-kinase blocked both interleukin-2-mediated
bcl-xL induction and cell survival. We conclude that
interleukin-2 receptor mediates T-cell survival via a
phosphatidylinositol 3-kinase-dependent pathway, possibly
involving pp56lck and bcl-xL as upstream and
downstream effectors, respectively.
Interleukin-2 (IL-2)1 plays a major
role in T cell biology, as activated T cells depend on this cytokine
for their proliferation and effector functions. IL-2 effects are
mediated through interaction with a specific transmembrane receptor
(IL-2R) composed of three different molecules, ,
, and
(1).
The
and
chains are constitutively expressed in T lymphocytes
and bind IL-2 with intermediate affinity. The
chain, in contrast,
is only expressed upon T cell activation and, together with the
and
chains, forms the high affinity IL-2 receptor (1). It has been
shown that heterodimerization of the IL-2R
and
chains is
required for IL-2-induced cellular proliferation (2, 3). The IL-2R has
no intrinsic enzymatic activity; therefore, its ability to transmit
intracellular signals relies on its association with signaling
molecules. IL-2R ligation triggers the activation of several tyrosine
kinases, including src-kinases (4, 5), Jak kinases (6-8), and
pp72syk (9). While Jak kinases are essential for IL-2-induced
proliferation, the involvement of src-kinases on T cell proliferation
following IL-2R binding is not so clear (4). IL-2 also regulates
phosphatidylinositol 3-kinase (PI3K) activation (10-12) and induces
p21ras (13), c-raf (14), and pp70S6K (15). After the
initial phase mediated by these set of cytoplasmic molecules, several
transcription factors including Stat3, Stat5, c-myc,
c-fos, and c-jun (16, 17) are also induced that
putatively control gene expression leading to cell division. All these
signaling cascades have been described to be triggered by high affinity
IL-2 receptors and are involved in inducing cell proliferation.
However, it has been recently shown that the intermediate affinity IL-2
receptor mediates protection against radiation-induced cell death (18,
19). We confirm here that although unable to promote cell division, the
intermediate affinity IL-2R induced cell survival. In fact, IL-2
treatment rescued resting T cells from the spontaneous cell death
observed when these cells are cultured without stimulus. We
subsequently analyzed the signaling pathways triggered by intermediate affinity IL-2R in resting T cells and found that pp56lck is
induced upon
IL-2R stimulation. In addition, data is presented indicating that the activation of PI3K is required for IL-2-mediated cell survival and that bcl-xL behaves as a downstream
effector of this enzyme.
Rapamycin was kindly
donated by Dr. J. Luengo (SmithKline Beecham); LY294002 was purchased
from Biomol (Plymouth Meeting, PA). The following polyclonal antisera
were used: rabbit anti-N-terminal pp56lck, rabbit anti-Jak3
(kindly donated by Dr. J. O'Shea, Frederick Cancer Research and
Development Center, MD), rabbit anti-Jak1 (obtained from UBI, Lake
Placid, NY), and rabbit anti--p85 (kindly donated by Dr. B. Schaffhausen, Tufts University School of Medicine, Boston, MA). The
following monoclonal Abs were used: 4G10 anti-phosphotyrosine (donated
by Dr. B. Drucker, D.F.C.I., Harvard Medical School, Boston, MA) and
SPVT3b anti-human CD3 (kindly donated by Dr. F. Sanchez-Madrid, S. Immunología, H. Princesa, Madrid, Spain). Horseradish
peroxidase-conjugated Abs and chemiluminiscence developing kit were
from Amersham Corp. (UK). Reagents for cell cycle analysis were from
Coulter Corp. (Miami, FL). Peripheral blood lymphocytes (PBL) were
isolated from healthy donor buffy coats as described (20). Cells were
grown in RPMI containing 10% fetal calf serum, 2 mM
L-glutamine, 10 mM Hepes, pH 8, 50 units/ml
penicillin, and 50 units/ml streptomycin (RPMIc) at 37 °C, 5%
CO2, in a humid atmosphere.
For cell cycle analysis, PBL were seeded at 106 cell/ml and cultured for either 24 or 72 h in RPMIc, in the presence or absence of recombinant IL-2 (kindly donated by Hoffman La Roche, Switzerland). Activated T cells were prepared by culturing purified T cells in wells coated with SPVT3b, anti-human CD3 (30 ng/cm2) during 72 h. For experiments using LY294002, cells were preincubated for 1.5 h with this inhibitor prior to incubation with 100 units/ml recombinant IL-2. After induction, cells were collected, washed in PBS, permeabilized, and stained with propidium iodide using a DNA-prep kit (Coulter). The percentage of subdiploid DNA-containing cells was determined by flow cytometry (Coulter).
For DNA synthesis analysis, cells were cultured for 72 h in 96-well plates with or without recombinant IL-2 at the indicated doses or in 96-well plates previously coated with 10 ng/well of the SPVT3b mAb. Then, 1 µCi/well [3H]thymidine (2 Ci/mmol, Amersham) was added for the last 24 h of culture, and [3H]thymidine incorporation was estimated by scintillation counting, as described previously (20).
For biochemical analysis, freshly isolated PBL or T cells activated during three days (see above) were washed, resuspended in RPMI with 1% bovine serum albumin at 107 cells/ml, and incubated for 1 h at 37 °C. Cells were subsequently washed and resuspended in RPMI at 107 cells/ml, and aliquots of 5 × 106 cells were incubated with 500 units/ml IL-2 for the indicated periods of time before lysis.
Immunoprecipitations and Kinase AssaysCells were lysed in Triton X-100 lysis buffer (50 mM HEPES, pH 8.0, 150 mM NaCl, and 1% Triton X-100 with protease inhibitors) as described (21). Cleared lysates were used for Western blot analysis or immunoprecipitated with the appropriate Abs as described (21). For pp56lck, Jak1, and Jak3 assays, cellular extracts were prepared in radioimmune precipitation buffer (22). pp56lck in vitro kinase assays in immunoprecipitates were performed, as described previously (21), using enolase as exogenous substrate. The PI3K in vitro kinase assay was performed as described elsewhere (11), using PI as exogenous substrate. Phospholipids were resolved in thin layer chromatography (Silica Gel 60, Merck, Germany) and developed in chloroform:methanol:ammonia (9:7:2 v/v). Radioactive products were visualized by autoradiography.
Fresh ex vivo isolated resting T cells
undergo apoptotic cell death upon in vitro culture in the
absence of stimulation (Fig. 1A). This
apoptotic process was significant at 24 h and affecting more than
50% of the cells at 72 h (Fig. 1A). Addition of IL-2 to the cultures prevented apoptotic cell death in a
dose-dependent manner (Fig. 1A). In contrast,
only a small increase in the percentage of cells in S+G2/M,
indicative of cell division, was observed upon IL-2 addition (from 2%
in medium alone to 12.6% with the higher IL-2 dose, Fig.
1A). To further demonstrate that prevention of cell death in
resting T cells is independent of DNA synthesis, we tested whether or
not the same IL-2 doses that promote cell survival (see Fig.
1A) were able to trigger significant DNA synthesis. The
highest IL-2 dose used (100 units/ml) induced only an
[3H]thymidine incorporation of approximately 10% of the
value obtained using anti-TCR antibodies (in the presence or absence of
exogenous IL-2, Fig. 1B). In all the samples analyzed, a
small percentage of the population ranging from 5 to 8%, depending on
the donor, are already activated T cells, expressing the IL-2R chain (data not shown). Therefore, proliferation observed in the
presence of IL-2 is likely to correspond to these cells. These results indicate that, although lacking the
chain, intermediate affinity IL-2Rs are capable of delivering a signal that prevents the apoptotic cell death of resting T cells cultured in the absence of TCR
stimulation.
pp56lck Is Activated upon IL-2 Binding to
Previous reports have been dedicated to characterizing the signaling pathways triggered by high affinity IL-2R (for a review, see Ref. 23). These studies have been carried out either in T cells stimulated via the TCR or in cell lines (CTLL2 or transfected BaF/3 cells). However, all these cells differ from resting T cells in the expression of several signaling molecules such as c-raf, c-myc, and others (14, 24) (our data not shown). To identify the signaling pathways triggered by the intermediate affinity IL-2R in resting T cells and implicated in cell survival, we have carried out a systematic analysis of which of the previously described high affinity IL-2R-induced events are observed upon ligation of the intermediate affinity IL-2R.
We first compared the induction of tyrosine kinases in resting T cells
(expressing intermediate affinity IL-2R) and in previously activated T
cells (expressing high affinity IL-2R). To this end, tyrosine
phosphorylation of cellular proteins was analyzed in anti-phosphotyrosine Western blot. As shown in Fig. 2,
activation of tyrosine kinases was essentially different in the two
cell types analyzed. While in activated cells, the activation of
tyrosine kinases was faster and induced the phosphorylation of two
major bands of 90 (25) and 130 kDa; in resting T cells, the activation was slower and induced the phosphorylation of two bands of
approximately 56 and 92 kDa. These results suggest that stimulation of
tyrosine kinases triggered by the intermediate affinity IL-2R differs
from that observed upon high affinity IL-2R stimulation.
Considering that the anti-phosphotyrosine Western blot does not address
the activation of a particular tyrosine kinase, we subsequently
examined individually the activation of pp56lck, Jak1, and
Jak3, previously implicated in high affinity IL-2R signaling (4-9).
pp56lck expression levels in resting or activated T cells was
analyzed by Western blotting comparing the amount of pp56lck
present in similar volumes of both lysates (normalized for protein concentration). This analysis indicated that pp56lck is
three-fold more abundant in lysates of activated T cells (data not
shown). 150 µg of protein from lysates of resting cells and 50 µg
of protein from lysates of activated T cells were used to compare the
specific enzymatic activity of pp56lck in both cell types.
Using this amount of cellular protein, we immunoprecipitated similar
amounts of pp56lck from the two cell types (Fig.
3B). Under these conditions, pp56lck
from resting cells was consistently found to be much more active than
that obtained from activated T cells (Fig. 3A).
Interestingly, pp56lck kinase activity increased upon IL-2
treatment not only in activated T cells (4) (Fig. 3A) but
also very significantly in IL-2-treated resting T cells (Fig.
3A), indicating that, as was the case with high affinity
IL-2R, intermediate affinity IL-2R triggers pp56lck
activation.
We then analyzed the role of the constitutively expressed Jak1 tyrosine
kinase (26). We tested whether the intermediate affinity IL-2R
expressed in resting T cells would induce Jak1 phosphorylation, as
reported for the high affinity IL-2R (6, 7). As shown in Fig.
3C, Jak1 was tyrosine phosphorylated upon IL-2 treatment in
activated T cells but was not modified in IL-2-treated resting T cells,
even though both cells contain similar amounts of Jak1 (Fig. 3D). With
regard to Jak3 kinase, it has been described that resting T cells do
not virtually express this enzyme (26). To confirm these data, we have
compared the amount of Jak3 present in resting and in activated T
cells, which require Jak3 to proliferate in response to IL-2 (27, 28).
As shown in Fig. 4A, while resting T cells
express similar amounts of p85-PI3K than activated cells, as much as
2 × 108 resting T cells are required to slightly
detect Jak3 expression under overexposure conditions. Therefore,
resting T cells only express less than 5% of the amount of Jak3
expressed by activated T cells. This Jak3 most likely corresponds to
that expressed by the 5-8% of preactivated cells present in the pool
of resting T cells (see above). Since in resting T cells IL-2 drives
survival of approximately 50% of the population, the low amounts of
Jak3 expressed in these cells do not prove, but strongly suggest, that Jak3 is not involved in the early signals triggered by IL-2R in
resting T cells. Finally, to make sure that Jak3 does not participate in
IL-2R-mediated survival at a later step of the activation process, we also analyzed whether or not IL-2 could trigger its expression in resting cells. As shown in Fig. 4B, no
induction of Jak3 expression was observed in resting T cells upon IL-2
treatment (analyzed at 24 and 48 h post-induction). In contrast,
Jak3 expression was induced by TCR stimulation in resting T cells at 24 and 48 h (Fig. 4B) (6). As expected, when
TCR-stimulated T cells were induced via high affinity IL-2R, Jak3
underwent the described tyrosine phosphorylation (data not shown) (6,
7). Thus, Jak3 does not seem to be involved in IL-2-mediated cell
survival. The lack of Jak3 involvement in
IL-2R signaling in
resting T cells is consistent with lack of Jak1 phosphorylation (Fig. 3C) since both kinases are thought to be regulated by
transphosphorylation (27). In conclusion, of the tyrosine kinases
analyzed, only pp56lck, but not Jak1 or Jak3, seems to be
induced by
intermediate affinity IL-2R in resting T cells.
IL-2-mediated Cell Survival Requires PI3K Activity
pp56lck has been shown to regulate PI3K
activation (29, 30). The fact that IL-2R triggered
pp56lck induction (Fig. 3A) and the observation that
PI3K mediates cell survival triggered by neural growth factor receptor
(31) and insulin growth factor receptor (32) prompted us to investigate whether or not PI3K activation could mediate the survival of resting T
cells induced through
IL-2R. We first tested whether incubation of resting T cells with IL-2 would induce the early
association/activation of PI3K with the IL-2R. The presence of PI3K on
IL-2R
immunoprecipitates was analyzed by Western blot using an
antibody against p85, the regulatory subunit of PI3K (33). As shown in
Fig. 5A, p85-PI3K was associated with the
IL-2R
chain, even before IL-2 addition to the cells, and became
early and transiently activated upon IL-2 binding to the
IL-2R
(Fig. 5B). Therefore, stimulation of intermediate affinity
IL-2R mediates the activation of its associated PI3K.
To analyze whether or not PI3K activation in response to IL-2R
ligation was related to the IL-2-induced survival effect in resting T
cells, cells were incubated with IL-2 in the presence of PI3K
inhibitors. Given that wortmannin has been shown to inhibit other
enzymes besides PI3K (34, 35), we utilized LY294002 (36). In contrast
with the PI3K pool associated to
IL-2R (Fig. 5B), the
total cellular PI3K present a steady-state kinase activity as evaluated
by an in vitro lipid kinase assay in anti-p85-PI3K immunoprecipitates (Fig. 6A). A slight
increase in PI3K activity was observed upon IL-2 treatment,
consistently with the small percentage of total PI3K associated
with IL-2R. Under these experimental conditions, high LY294002
concentrations blocked virtually all cellular PI3K activity (Fig.
6A). Inhibition of PI3K activity by LY294002 was similarly
observed in the presence or absence of IL-2 (not shown). We
subsequently tested the effect of LY294002 for IL-2-mediated T cell
survival. In the absence of IL-2, LY294002 induced the apoptotic cell
death of approximately 10% of the population. This 10% may represent
a toxic effect or reflect, as other authors propose (31), that basal
PI3K activity is required for cellular survival (Fig. 6B).
In the presence of 100 units/ml IL-2, a dose-response inhibition by
LY294002 of IL-2-mediated survival was observed (a representative
experiment is shown in Fig. 6B). In fact, in the absence of
any exogenous additives, the percentage of cell death was 40%. In the
presence of 100 units/ml IL-2, this number decreases to 15%, but
incubation with 100 units/ml IL-2 and 50 µM LY294002
induces an increase in cell death from 15% to 45%. This result
demonstrates the ability of LY294002 to block IL-2-induced survival
beyond its putative toxic effect, affecting only up to 10% of the
population. Representation of residual PI3K activity (at 24 h), in
comparison with IL-2-induced survival upon LY294002, shows a clear
correlation between both parameters (Fig. 6C). While 5 µM LY294002 only partially blocks PI3-kinase activity and
IL-2-induced survival, 50 µM is able to block almost
totally IL-2-induced survival and PI3K activity. At the intermediate
doses, we consistently found in different experiments a partial effect
in both LY294002 ability to block IL-2-induced-cell survival and PI3K
activity. Therefore, abrogation of PI3K activity and inhibition of
IL-2-induced cell survival are observed in parallel, indicating that
intermediate affinity IL-2R mediate cell survival via a
PI3K-dependent pathway.
Bcl-xL Is a Downstream Effector of PI3K
In an
attempt to characterize putative downstream effectors of PI3K in our
system, we first considered pp70S6K. This enzyme has been shown to be
necessary for S phase transition and is known to be blocked by PI3K
inhibitors (37). We analyzed its putative involvement in
IL-2R-mediated survival by using a pp70S6K inhibitor, rapamycin (15).
Under conditions of pharmacological inhibition of pp70S6K that block
60% of IL-2-induced cell division, rapamycin did not abrogate
IL-2-induced survival (Fig. 7), suggesting that pp70S6K
does not participate in cell survival. Another interpretation of this
result is that more pp70S6K activity is required for cell proliferation
than for cell survival. However, data from Yao and Cooper (38), as well
as other reports showing PI3K-mediated cellular responses independent
of pp70S6K (39, 40), support that, most likely, this enzyme is not
involved in
IL-2R-mediated survival but is required for cell
cycle progression.
Bcl-2-related proteins have also been found to be regulated by
IL-2R stimulation (41-43). These genes have been involved as
survival mediators (44) and related to PI3K activation in other
cellular systems (45). The possible involvement of these proteins in
intermediate affinity IL-2R-mediated cell survival, was therefore
analyzed. A marginal increase in bcl-2 protein levels was observed in
IL-2-treated cells (Fig. 8), making unlikely a significant involvement of this protein in IL-2-mediated survival. Another member of this family, bcl-xL, has also been shown
to negatively regulate apoptotic processes (46). Analysis of
bcl-xL expression by Western blot in IL-2-treated resting
cells indicated that this protein is significantly induced upon IL-2
addition (Fig. 8). Moreover, while bcl-2 expression was poorly
affected, bcl-xL up-regulation was abrogated in the
presence of the PI3K inhibitor LY294002 (Fig. 8). Thus,
bcl-xL behaves as a putative downstream effector of PI3K.
The expression of bax was also analyzed and found to be not
significantly modulated in response to IL-2 (not shown). In conclusion,
while bcl-2 may play a partial role in mediating
IL-2R-induced
survival, bcl-xL is most likely an effector of this
cellular response.
The results presented in this report indicate that IL-2 binding to
the intermediate affinity IL-2R promotes rescue from apoptosis of
resting T cells cultured in vitro in the absence of TCR
stimulation. Analysis of the signaling molecules implicated in
IL-2R-mediated survival, addressed here for the first time, indicates
that pp56lck, but not Jak kinases, is induced upon
receptor ligation. We have previously described that pp56lck is
required for activation of TCR-associated PI3K (30) and regulates
IL-2R-induced PI3K activation (29). Thus, we have also
analyzed whether PI3K was stimulated in this system. We found that PI3K
is constitutively associated to
IL-2R and became activated in
response to IL-2 binding. The fact that pharmacological inhibition of
PI3K blocked IL-2-mediated survival strongly suggests that PI3K
activation is required for the ability of IL-2 to abrogate cell death.
Finally, we also show here that bcl-xL is induced upon
IL-2R stimulation (Fig. 8) and that this induction is blocked
upon PI3K inhibition (Fig. 8). Together these results allow the
proposal of the model depicted in Fig. 9: ligation of intermediate affinity
IL-2R induces both pp56lck and
PI3K activation. PI3K activity is then required for the induction of
bcl-xL, an important mediator of cell survival.
To the best of our knowledge, no previous publication has characterized
the effect of IL-2R ligation on the death of normal T cells in
the absence of TCR stimulation. However, at least three previous
reports have analyzed the IL-2 involvement in T cell rescue from
radiation-induced cell death (18, 19, 47). Boise et al. (18)
describe that preincubation of normal human peripheral T cells with
IL-2, IL-4, or IL-7, but not with IL-1, IL-3, or IL-6, decreases the
apoptosis observed in response to
radiation. Mor and Cohen (47),
have reported that IL-2 mediates protection to
radiation in
antigen-specific T cells; however, they do not detect protection in
resting T cells. Finally, Seki et al. (19) show that
CD8+ and NK cells are the most radiosensitive of resting T
cells but could be rescued by IL-2 treatment. These observations,
together with the description here that IL-2 protects from the cell
death suffered by normal T cells when cultured in the absence of TCR stimulation (Fig. 1), support the conclusion that intermediate affinity
IL-2R is capable of inducing T cell survival. In addition to this
mechanism, others are likely to mediate survival of resting T cells,
given that a significant proportion of the resting cells survive in the
absence of IL-2 (Fig. 1A).
We have performed a systematic study of IL-2R-induced signaling
pathways. As proposed in the model in Fig. 9, these data suggest that
pp56lck is one of the signals involved in IL-2R-mediated
survival in resting T cells. This is in agreement with the observation
that constitutive active pp56lck mutants decrease the apoptosis
induced by growth factor removal (42). In contrast, the involvement of
pp56lck on IL-2-induced cell division is controversial, since a
mutant of the
chain lacking the acidic region (which does not
associate pp56lck) is, however, capable of mediating a
proliferative signal (4). The other tyrosine kinases involved in high
affinity IL-2R signaling are the Jak kinases (6-8), which have been
shown to be required for IL-2-mediated proliferation (27, 28). The
mechanism by which these enzymes are activated entails their
phosphorylation on tyrosine residues. The current model proposes that
Jak1 and Jak3 transphosphorylate each other (27). The low levels of
Jak3 expression, and the concomitant lack of Jak1 and Jak3 activation in our system suggest that these proteins do not participate in IL-2-mediated survival. In addition to IL-2-mediated survival, bcl-2
induction by IL-2 in BAF/3 also seems to be independent of Jak3 as
bcl-2 up-regulation is not altered by overexpression of a dominant
negative Jak3 in these cells (28).
Our conclusion on PI3K activation involvement in T cell rescue from
apoptosis (Fig. 6C) concurs with the implication of PI3K in
survival discussed by Ward et al. (45) for CD28-mediated signals. Moreover, studies in rodent fibroblast (38), in neural growth
factor receptor signaling in PC12 cells (29), and insulin growth factor
receptor signaling in myeloid progenitors cells (32), in which
activation of PI3K prevents apoptosis, also support this view. The
mechanism through which PI3K is activated is not evident. In this
regard, several cytokine receptors, such as IL-4R and IL-7R that share
the chain with the IL-2R (48, 49) have been shown to enhance the
resistance of resting T cells to radiation-induced cell death as
effectively as IL-2 (18, 19). Studies with the IL-7 receptor show that
PI3K activation is mediated by the
chain and its associated Jak3
tyrosine kinase (50). However, the low levels of Jak3 expression in our
system and the fact that this protein is not up-regulated upon IL-2
treatment makes unlikely the implication of this mechanism in PI3K
activation. Another possible mechanism for PI3K activation involves the
IL-2R
chain and src-kinases (29, 51). We previously described that
pp56lck activation regulates PI3K activation upon IL-2
triggering of high affinity IL-2R (29). In addition, the serine-rich
region required for pp56lck activation is also required for
activation of PI3K (52). Finally, the fact that pp56lck
tyrosine kinase and correlatively PI3K are activated upon IL-2 binding
to
IL-2R (Fig. 3A) suggests that pp56lck may
contribute to induce PI3K activation.
Bcl-2 and bcl-xL have been shown to mediate survival (44)
and to be induced upon high affinity IL-2R ligation (41-43, 47). We
therefore tested the putative involvement of bcl-2 and
bcl-xL in IL-2-mediated survival. Induction of
bcl-xL/bcl-2 in irradiated cells upon IL-2R ligation
has not been previously detected (19). We found a slight increase in
bcl-2 expression that could suggest a partial role of this protein for
IL-2-mediated survival. More clearly, the induction of
bcl-xL upon
IL-2R stimulation and the fact that
blocking IL-2-induced survival with LY294002 also results in abrogation
of bcl-xL up-regulation suggest that PI3K activation
precedes bcl-xL expression and that bcl-xL is probably involved in
IL-2R-induced survival.
IL-2 mediates three different cellular responses. First, it has been
involved in mediating the proliferation of cells preactivated via T
cell receptor and expressing IL-2R (53). Second, it has been
suggested to mediate survival (Fig. 1) (18, 19), and finally, it has
been recently proposed to program T lymphocytes for apoptosis (54).
Regarding the dual role of IL-2 at promoting either survival or cell
proliferation, we describe here that only a set of signals
(pp56lck, PI3K, and bcl-xL) of those triggered for
IL-2-induced cell division are involved in mediating cell survival. The
reason for high affinity IL-2R to trigger different early signals than
that derived from intermediate affinity IL-2R in resting T cells could
be the presence in the high affinity IL-2R of the
chain. However,
chimeric
and
IL-2R chains transfected in cell lines lacking the
chain behave as high affinity IL-2R, inducing proliferation (2, 3). Therefore, it seems that the lack of IL-2R
chain expression is not
responsible for the partial signaling of
IL-2R in resting T
cells. Instead, other proteins not expressed in resting T cells but
expressed in cell lines or induced upon TCR cross-linking in normal T
cells could be responsible for the different biological responses
following TCR ligation. Candidates for these molecules could be
c-myc, Jak3, or c-raf, relatively abundant in
activated T cell and cell lines but not detectable in resting T cells
(our data not shown) (14, 26).
Regarding the ability of IL-2 to program T lymphocytes for apoptosis,
it has been described that incubation of T cells in IL-2 or IL-4 is
required for ulterior TCR-mediated cell death (54). Moreover,
IL-2-/- mice developed splenomegaly and lymphoadenophaty
(55), similar to IL-2R-/- mice (56). We sought to
investigate the signaling pathways responsible for the ability of IL-2
to induce susceptibility to TCR-triggered apoptosis. However, in our
hands, the conditions that induce susceptibility to apoptosis also
trigger cell division. Thus, we favor the hypothesis that the same
intracellular signaling program induces both cellular responses.
The fact that intermediate affinity IL-2 receptors transmit
intracellular signals in resting cells, opens the possibility of a
biological role for the IL-2R. The circulating levels of IL-2
are normally low. However, encounter with antigen primes T cells for
IL-2 production, locally increasing the concentration of this cytokine.
Cells that have not been stimulated by the antigen, and therefore do
not express the IL-2R
chain, could use IL-2 to increase their
survival potential.
We thank Drs. M. A. Rodriguez and P. Torres from the Centro de Transfusiones de la Comunidad de Madrid for kindly providing the buffy coats, Drs. A. Bernard and V. Calvo for critical reading of the manuscript, and C. Mark for editorial assistance.