| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
*
Immunology and
Biological Structure, University of Washington, Seattle, WA 98195;
Virginia Mason Research Center, Seattle, WA 98101; and
§
Merck Research Laboratories, Rahway, NJ 07065
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
mAbs, suggesting that the
functional TSLP receptor complex uses IL-7R. In contrast, three
different Abs to the common cytokine receptor 
-chain had no effect
on the response of these cells to TSLP, indicating that the functional
TSLP receptor complex does not use the common cytokine receptor
-chain. Both cytokines induced activation of Stat5, but only IL-7
induced activation of the Janus family kinases Jak1 and Jak3. In fact,
TSLP failed to activate any of the four known Janus family kinases,
suggesting that Stat5 phosphorylation is mediated by a novel mechanism.
Taken together, these data support the idea that TSLP can make unique
contributions to B lymphopoiesis and indicate that it does so by
mechanisms distinct from IL-7. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain (IL-7R) also
interfere with B cell development both in vitro and in vivo (5).
Importantly, disruption of the IL-7R gene (6) or treatment of mice
with anti-IL-7R Ab (5) arrests B cell development before the
pro-B cell stage. Hence, interference with IL-7R function has more
severe consequences than interference with IL-7 function. These results
have been interpreted to suggest the existence of another cytokine that
uses the IL-7R as part of its receptor complex (6). Sharing of
cytokine receptor subunits by different cytokines has been well
documented, including coutilization of the IL-2Rß and - subunits
by IL-2 and IL-15 (7) and of the IL-4R by IL-4 and IL-13 (reviewed in
8 , and the sharing of the IL-2R -chain (also known as the
common -chain
(c))3 by IL-4,
IL-9, and IL-7 in addition to IL-2 and IL-15 (reviewed in 9 .
Cytokine signaling mechanisms have been investigated extensively in
recent years, and perhaps the best studied is that of the IL-2/IL-2R
complex. Here, ligand binding induces the tyrosine phosphorylation of
the receptor (reviewed in Refs. 10 and 11) and the activation and
association of the Janus family kinases Jak1 and Jak3 (12, 13, 14). Jak1
has been shown to specifically associate with the IL-2R ß-chain,
while Jak3 interacts with c (12, 13, 14). Activation of
these kinases leads to tyrosine phosphorylation and activation of the
STAT family of transcription factors (specifically Stat5 and Stat3 for
IL-2 signaling), which dimerize and migrate to the nucleus where they
influence transcription of numerous genes (15). The IL-7R complex
includes the IL-7R -chain and c (16, 17, 18). This
sharing of a common receptor chain with the IL-2 system suggests the
possibility that at least part of the signaling machinery used in IL-2
responses may also be mobilized by IL-7 binding to its receptor
complex. Indeed, IL-7 induces the activation of the Janus family
kinases Jak1 (10) and Jak3 (19), which phosphorylate and activate the
Stat5 transcription factors (20).
We previously identified a novel cytokine activity in conditioned medium from a thymic stromal cell line (Z210R.1) and have designated this factor thymic stromal lymphopoietin (TSLP). TSLP activity in conditioned medium from Z210R.1 cells supported the development of IgM+ B cells from fetal liver lymphocyte progenitors and induced proliferation of the factor-dependent cell line NAG8/7 (21). IL-7 and TSLP have been shown to have overlapping functions in the sense that both can costimulate thymocytes and mature T cells, sustain the NAG8/7 cell line, and support B lymphopoiesis in long term cultures of fetal liver cells, but the ability of TSLP to promote the development of B220+/IgM+ immature B cells distinguishes the two, as IL-7 facilitates the development of B220+/IgM- pre-B cells (21). We therefore sought to define the mechanisms by which IL-7 and TSLP exert their distinctive effects by examining the signaling responses that each cytokine evokes.
We have undertaken a series of experiments to characterize the
biochemical signals induced by TSLP and to distinguish them from
responses to IL-7. Our studies demonstrate that both cytokine receptors
employ the IL-7R -chain as part of their signaling complexes.
Nonetheless, whereas TSLP upon binding to its receptor complex
activates both Stat5a and Stat5b, this activation event does not
involve any of the known Janus family kinases. Taken together, these
results suggest that although TSLP-generated signals require a receptor
unit also used by IL-7, the biochemical changes induced in responding
cells are dramatically different. Aside from providing insight into the
physiological differences that attend IL-7 as opposed to TSLP
treatment, our results demonstrates that there exist alternative
mechanisms for STAT activation in hemopoietic cells.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Medium (RPMI 1640; Life Technologies, Gaithersburg, MD) was
supplemented with L-glutamine, nonessential amino acids,
2-ME, and FBS (10% for long term bone marrow cultures (LTBMC) and 15%
for colony assays). Agar was purchased from Difco (Detroit, MI), and
methylcellulose (1500 centipoise) was purchased from Sigma (St. Louis,
MO). Recombinant TSLP was provided by Dr. D. Williams, Immunex
(Seattle, WA), and recombinant murine IL-3, IL-10, and IL-7 were
purchased from Genzyme (Cambridge, MA). Cytokines and media were tested
for the presence of endotoxin by the Limulus lysate assay
and were found to be negative. Anti-IL-7R Ab A7R34 has been
previously described (5) and was provided by Dr. S. Nishikawa.
The anti-c Ab TUGm2 has also been described (22) and
was purchased from PharMingen (San Diego, CA). Other
anti-c Abs (4G3 and 3E12) were characterized by He
and co-workers (23, 24) and were provided by Dr. T. Malek
(University of Miami). Abs to Stat5a, Stat5b, Jak1, Jak2, Jak3, and
Tyk2 were purchased from Santa Cruz Biotechnology (Santa Cruz, CA).
Biotinylated 4G10 anti-phosphotyrosine Ab and rabbit anti-Jak3
antisera were obtained from Upstate Biotechnology (Lake Placid, NY).
Streptavidin and anti-rabbit Ig conjugated to horseradish
peroxidase were purchased from Amersham (Arlington Heights, IL). Abs
for flow cytometry were obtained from Caltag (Burlingame, CA;
anti-B220-phycoerythrin), Southern Biotechnology (Birmingham, AL;
anti-IgM-FITC and anti-Ig
-FITC), or Becton Dickinson (San
Jose, CA; anti-Ig-phycoerythrin).
Colony assays
Cleaned femurs and tibias were ground in a mortar and pestle to recover bone marrow cells. After floating on a cushion of 50% (v/v) FBS/complete medium to remove bone chips and debris, the cells were added to agar (0.3%) or methylcellulose (1.8%) and medium containing cytokines at concentrations previously determined to be maximal (2.5 ng/ml IL-7 or 5 ng/ml TSLP). Replicate 1-ml cultures (105 cells/culture) were harvested after 1 wk in culture (37°C, 5% CO2). Agar disks containing colonies were transferred to glass slides, covered with strips of filter paper, and dried overnight at room temperature. The filter paper was then flooded with 10% neutral buffered formalin, the paper strips were removed, and after washing in PBS, the slides were stained with Hemal stain. Colonies were counted microscopically. Replicate methylcellulose colonies were diluted with medium, transferred to test tubes, and washed repeatedly by centrifugation to remove the methylcellulose. Washed cells were then processed for flow cytometry.
Long term bone marrow cultures
These cultures were prepared as originally described (25). Cytokines were added at the initiation of the cultures. Half the medium in the cultures was replaced at weekly intervals with fresh medium containing the appropriate cytokine. Cultures were harvested 5 wk after initiation.
Fetal liver cultures
Day 14/15 fetal livers were harvested aseptically and mechanically dissociated by pressing them between the frosted ends of sterile glass microscope slides. After repeated washing, cells were plated in methylcellulose at 1.5 x 105 cells/ml (1-ml cultures) as described above for colony assays. Cytokines at the same concentrations as those used in the colony assays were added at the start of the cultures. Cultures were harvested as described for bone marrow colony assays 8 days after initiation.
Proliferation assays
Proliferation assays using NAG8/7, HT-2, and BA/F3 cells were essentially as previously described (21). Specifically, assays were initiated by washing cells four times in medium and plating at 5 x 104 cells/well in 96-well flat-bottom plates with the indicated quantities of cytokine and Ab. Replicate cultures were set up in at least three wells for all experiments. Cells were cultured for 2 days, then pulsed with 1 µCi/well [3H]thymidine (6.7 Ci/mmol; New England Nuclear, Boston, MA), and cultured overnight. Plates were harvested using a Tomtec cell harvester (Tomtec, Orange, CT) and counted using a Wallac beta plate scintillation counter (Wallac, Gaithersburg, MD). Data are expressed as the mean counts per minute incorporated in replicate cultures. Primary cell proliferation assays were set up and run as described above, except that cells were plated at a density of 5 x 105 cells/well.
Activation of Stat5 and JAK kinases
NAG8/7 cells were washed four times in medium, cytokine starved for 4–5 h, and then stimulated at 107 cells/ml with either 25 ng/ml IL-7 or 100 ng/ml TSLP. After the indicated times, cells were harvested by centrifugation, washed once in cold PBS, and lysed in 25 mM Tris (pH 8.0)/150 mM NaCl/1% Triton X-100/1 mM Na2VO4 plus protease inhibitors TNT buffer at 108 cells/ml as previously described (26). Insoluble material was pelleted in a microfuge at 4°C, and supernatants were transferred to fresh microfuge tubes. Stimulated cell lysates were prepared in the same way for BA/F3 cells, except that cells were stimulated with 25 ng/ml murine IL-3. Total mouse splenocyte preparations were stimulated with 25 ng/ml murine IL-10 as described above.
Lysates from 1–2 x 107 cells (2 x 107 cells were used for primary mouse splenocyte lysates) were then incubated overnight at 4°C with 0.5 µg of the indicated Ab plus 25 µl of a 30% slurry of protein G-Sepharose (Pharmacia, Piscataway, NJ). Pellets were washed twice in 1 ml of TNT buffer and resuspended in SDS loading buffer. Samples were split, subjected to SDS-PAGE, and electroblotted to nitrocellulose membranes (Bio-Rad, Hercules, CA). Blots were then blocked in 5% BSA in Tris-buffered saline plus 0.1% Tween-20 (TBST) for anti-phosphotyrosine blots or in 5% milk in TBST for other Abs. For phosphotyrosine blots, biotinylated 4G10 was used at a 1/5000 dilution in 1% BSA/TBST and streptavidin conjugated to horseradish peroxidase at 1/5000 in the same buffer. Abs to other proteins were also used at 1/5000 dilutions in 1% milk/TBST with either anti-rabbit Ig (1/3000) or anti-goat Ig (1/2000) conjugated to horseradish peroxidase. After extensive washing in TBST, blots were developed using chemiluminescent detection (New England Nuclear, Boston, MA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
TSLP was initially identified as an activity in conditioned medium (CM) from a thymic stromal cell line (Z210R.1) that supported the development of IgM+ B cells in long term fetal liver cultures (21). This activity was distinct from IL-7 by two criteria: 1) IL-7 did not promote the development of IgM+ B cells under those conditions; and 2) the NAG8/7 pre-B cell line derived from fetal liver lymphoid precursors proliferated in response to Z210R.1 CM even in the presence of neutralizing Abs to IL-7 (21). Since the NAG8/7 cell line that responded to TSLP is a pre-B cell line and because the original characterization of TSLP activity had used fetal liver as a source of lymphocyte precursors in long term cultures that produced B-lineage lymphocytes (21), B cell development was chosen as a model system to compare the biological effects of TSLP and IL-7.
Methylcellulose cultures with fetal liver lymphocyte precursors were
established including either purified IL-7 or TSLP, and the resultant
cells were recovered and analyzed by flow cytometry. While IL-7 induced
the development of B220+/IgM- pre-B
lymphocytes, TSLP addition promoted the development of substantial
numbers of B220+/IgM+ cells that also expressed
light chains on their surfaces (Fig. 1
). This result confirms our previous
observation that CM from Z210R.1 cells promoted the development of
B220+/IgM+ cells in an IL-7-independent manner
(21) and supports the conclusion that TSLP was responsible for the
production of these cells. Moreover, Fig. 1 indicates that while both
IL-7 and TSLP can promote B lymphopoiesis in fetal liver cultures, the
phenotypes of the cells that develop are dramatically different. Our
results differ from those of Ray et al. (27), who reported that in
4-day cultures, cells that developed in IL-7 and TSLP were
indistinguishable. The reason for the discrepancy in these results is
not known, but may relate to the additional period in culture we have
employed (8 days) or the enrichment for "bipotential precursors"
employed by Ray and co-workers (27).
|
Although the liver is the primary site of fetal B lymphopoiesis,
postnatal B cell development takes place primarily in the bone marrow.
To investigate the differentiative capacity of bone marrow cells in the
presence of TSLP, unfractionated bone marrow cells were cultured in
methylcellulose medium without cytokines or in medium supplemented with
either TSLP or, as a positive control, IL-7. Cultures without cytokine
failed to produce lymphoid colonies (data not shown), whereas addition
of IL-7 supported the development of numerous lymphoid colonies (Fig. 2A). Phenotypic analysis of
the cells in these IL-7-induced colonies showed them to be primarily
B220+/IgM- pre-B cells (Fig. 2C)
with forward light scatter profiles typical of cycling cells (Fig. 2B). In contrast, colony growth in response to TSLP was
about half that observed in IL-7 (Fig. 2A) with a generally
smaller colony size (data not shown). Analysis of the composition of
these colonies revealed a strikingly high percentage of
IgM+ cells (Fig. 2C) that were predominantly
IgD- (data not shown) and displayed reduced forward
scatter profiles, suggesting they were primarily immature B cells that
were no longer cycling (Fig. 2B). Consistent with this
interpretation, we found that these cells failed to respond to
restimulation with either IL-7 or TSLP, whereas cells from IL-7
colonies proliferated in response to IL-7 in secondary cultures
(data not shown). Hence, IL-7 and TSLP appear to influence the
development of B lymphoid cells from adult bone marrow in a manner that
parallels the way they affected B lymphopoiesis in fetal liver
cultures.
|
We also examined the effects of exogenous TSLP in LTBMC and
compared these results to those obtained with addition of IL-7. In the
absence of added cytokine these LTBMC produced primarily
B220+/IgM- pre-B cells (Fig. 3). Consistent with previous reports
(1, 2, 3), bone marrow cultures supplemented with exogenous IL-7 produced
large numbers of B220+/IgM- IL-7-dependent
cells (Fig. 3). These cells showed increased forward light scatter
profiles, suggesting that they were a population of rapidly cycling
pre-B lymphoblasts (data not shown). Addition of exogenous TSLP to bone
marrow cultures yielded IgM+ B cells that phenotypically
resembled immature B lymphocytes (Fig. 3). Many of these cells also
expressed light chain, but the cells were produced in fewer numbers
than was observed with added IL-7 (Fig. 3) and showed reduced forward
light scatter profiles (data not shown). While the total number of
cells produced in cultures supplemented with TSLP was considerably
lower than that in cultures supplemented with IL-7, a determination of
the absolute number of cells revealed that this was primarily at the
expense of the B220+/IgM- population (Fig. 3).
Moreover, the increased percentage of IgM+ cells in these
cultures actually reflected a minimal increase in the absolute number
of IgM+ cells produced. These data correlate well with
information obtained from colony forming assays (Fig. 2
) and give the
same phenotypic readout as that of fetal liver cultures. However, the
production of IgM+ cells in fetal liver cultures can all be
attributed to de novo development, since there were no IgM+
cells at their initiation (data not shown). Thus, it appears that
addition of IL-7 and TSLP to three different culture systems designed
to examine B lymphopoiesis produced the same effects: IL-7 favored the
development of large numbers of B220+/IgM-
pre-B cells, while TSLP produced relatively fewer
B220+/IgM+ immature B cells. Taken together
these data support the idea that TSLP uniquely supports the development
of immature B cells.
|
-chain, but not the c
Because of the observed differences in biologic activity of
IL-7 and TSLP, we initiated studies to examine signaling pathways
activated by the two cytokines. As previously noted, NAG8/7 is a
factor-dependent cell line that proliferates in response to both IL-7
and TSLP (21). Fig. 4A
demonstrates that NAG8/7 cells proliferate in response to both
cytokines, but that murine IL-7 is a somewhat more potent mitogen.
Earlier indications that NAG8/7 cells responded weakly to IL-7
reflected the human IL-7 used in those assays (21). To distinguish TSLP
from IL-7, we performed proliferation assays in the presence of
neutralizing Abs to IL-7 and found proliferation induced by TSLP to be
unaffected (data not shown) (21). As a further test of this, we also
performed proliferation assays in the presence of Abs to IL-7R.
Administration of this Ab to mice in vivo or inclusion in LTBMC
inhibits IL-7-dependent responses (5), and hence, we reasoned that this
would be an effective way to specifically affect the IL-7 response by
this cell line. As expected, we found that high concentrations of Ab
inhibited IL-7-mediated proliferation of NAG8/7 cells (Fig. 4B). Lower concentrations of anti-IL-7R caused a
consistent augmentation of proliferation that may reflect enhanced
receptor aggregation. In any case, this effect was dependent on IL-7,
since the Ab alone did not induce proliferation at any concentration
tested (data not shown). Importantly, we also saw dramatic reductions
in TSLP-induced proliferation by treatment with anti-IL-7R Ab
(Fig. 4C). This Ab did not affect the proliferation of the
IL-2-dependent cell line, HT-2, or of the IL-3-dependent cell line,
BA/F3, indicating that this was not a general toxic effect of the Ab
(data not shown). Taken together these results suggest that the TSLP
receptor (TSLP-R) required the IL-7R -chain as part of its
functional signaling complex.
|
c, which is essential for IL-7 responses. We tested
three different mAbs against c (TUGm2, 4G3, and 3E12)
(22, 23, 24). In cultures including TUGm2 and 4G3, we consistently observed
inhibition of IL-7-mediated proliferation at high Ab concentrations and
in most cases observed an augmentation of the IL-7 response at lower Ab
levels (Fig. 4B and data not shown). Consistent with
previous reports, the 3E12 Ab did not affect IL-7-mediated
proliferation. In contrast, the response of NAG8/7 cells to TSLP was
unaffected by any of the three mAbs to c (Fig. 4C and data not shown). From these results, we concluded
that the TSLP-R complex includes the IL-7R -chain, but that it does
not require c, in contrast to the IL-7R complex, which
requires the IL-7R -chain and c. These results
suggest a relationship between IL-7 and TSLP similar to that observed
for IL-4 and IL-13, where IL-4 responses are mediated by the IL-4R and
c, whereas IL-13 responses depend on a unique receptor
component (the IL-13R) and the IL-4R (reviewed in 8 . TSLP triggers tyrosine phosphorylation of Stat5
Because our results showed that both IL-7 and TSLP use the
IL-7R -chain, we undertook a series of experiments designed to
compare the biochemical changes induced by treatment of cells with IL-7
and TSLP. We reasoned that some features of the signal transduction
mechanism would be shared by virtue of a common receptor component, but
that others may be different because the functional receptor complexes
are distinct. Consistent with previous reports in other cell systems
(20), we readily observed IL-7-induced tyrosine phosphorylation of
Stat5 (Stat5a and Stat5b) in NAG8/7 cells (Fig. 5A). We also observed Stat5
activation after treatment of these cells with TSLP (Fig. 5A), supporting the hypothesis that at least certain aspects
of the signaling pathways used by these two cytokines overlap. The
level of tyrosine-phosphorylated Stat5 from TSLP-treated cells was
slightly lower than that in IL-7-treated cells, and the kinetics of
accumulation were somewhat slower (Fig. 5A and data not
shown). This suggested that tyrosine phosphorylation of Stat5 was
induced less efficiently by TSLP than by IL-7, perhaps reflecting
different levels of receptor expression or participation of alternative
signaling pathways (see below). The antiserum used for these
experiments has been reported to recognize both Stat5 isoforms (Stat5a
and Stat5b, which are the protein products of closely related, but
distinct, genes), although it was raised against a Stat5b peptide (see
Materials and Methods). While these two isoforms have
slightly different mobilities on SDS-PAGE (Stat5a migrates at about 94
kDa and Stat5b at about 92 kDa), we could not clearly distinguish this
difference on the 10% gels employed for this analysis (Fig. 5A). Hence, it was formally possible that only one isoform
was being activated, and the other was not, by IL-7 and/or TSLP. To
examine this possibility we used an antiserum that specifically
recognized Stat5a only and analyzed the immune complexes with 8%
acrylamide gels (Fig. 5B). Cell extracts that had been
immunodepleted of Stat5a were then subsequently immunoprecipitated with
the antiserum that recognized both isoforms. In this way we were able
to assess the phosphorylation state of each Stat5 isoform. Moreover,
the use of 8% polyacrylamide gels allowed a distinction to be made
between the two based on their migrations. We found that TSLP and IL-7
activated both the Stat5 isoforms to a similar extent (Fig. 5B) and hence differential activation of these transcription
factors is unlikely to account for the different biological responses
observed in IL-7-treated vs TSLP-treated bone marrow cultures. We also
noted a mobility shift in Stat5b in extracts from cells treated with
IL-7 that was less prevalent in extracts from TSLP-treated cells (Fig. 5B and data not shown). The biological importance of this
observation is not yet known.
|
The tyrosine phosphorylation of STAT transcription factors in
general has been shown to be conducted by tyrosine kinases of the Janus
family (JAK kinases). In particular, IL-7-induced Stat5 activation is
correlated with activation of Jak1 and Jak3 (10, 19, 20). Treatment of
NAG8/7 cells with IL-7 also produced activation and tyrosine
phosphorylation of Jak1 (Fig. 6A) and Jak3 (Fig. 6B), but failed to activate Jak2 or Tyk2 (Fig. 6C). This was contrasted by NAG8/7 cells stimulated with
TSLP, which did not show detectable activation of either Jak1 or Jak3
(Fig. 6, A and B). Moreover, activation of Jak2
or Tyk2 was not observed in NAG8/7 cells following TSLP stimulation,
indicating that other members of this kinase family did not respond to
TSLP binding to its receptor (Fig. 6C). Since we readily
detected activation of these latter two kinases in other cells using
cytokines known to activate them (specifically, Jak2 activation after
IL-3 treatment of BA/F3 cells and Tyk2 activation after IL-10 treatment
of murine splenocytes; Fig. 6C), our inability to detect
activation did not reflect a failure of the Abs to immunoprecipitate
the kinases. These results taken together suggest that Stat5 tyrosine
phosphorylation in response to TSLP is mediated by either an
unidentified JAK kinase or another mechanism altogether. Our results
also demonstrate a biochemical distinction between IL-7 and TSLP
signaling that may in part underlie the different responses induced by
these cytokines.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain, but not c. This partial receptor
sharing was suggested by Peschon and co-workers (6) as the basis for
the more severe lymphopoietic abnormalities observed in mice rendered
genetically deficient in the IL-7R -chain relative to IL-7-deficient
mice. Hence, we predict that the functional TSLP-R complex will include
at a minimum the IL-7R -chain and a unique TSLP-binding receptor
component. In fact, such a receptor component has recently been cloned
and found to be a member of the hemopoietin receptor family (D.
Williams, unpublished observation). This suggests a relationship
between IL-7 and TSLP that is very similar to that seen for IL-4 and
IL-13, where the functional IL-4R includes the IL-4R -chain and
c, while the IL-13R includes the IL-4R -chain and a
unique IL-13R chain (8, 28). IL-4 and IL-13 show a degree of overlap in
their biochemical and biological effects, but also exhibit distinct
activities. We found this to be the case for IL-7 and TSLP as well.
Perhaps the most remarkable observation we made was that although
tyrosine phosphorylation of Stat5a and Stat5b could be induced by
treatment of NAG8/7 cells with both IL-7 and TSLP (Fig. 5), we could
only detect Jak1 and Jak3 activation in IL-7-treated cells (Fig. 6).
Moreover, we failed to detect activation of any of the four known Janus
family kinases in response to TSLP even though we could detect their
activation in other cell types under the appropriate stimulation
conditions (Fig. 6). This suggests that an as yet unidentified member
of the Janus family may be involved in the response to TSLP or that
TSLP stimulation leads to Stat5 phosphorylation in a JAK
kinase-independent manner. In fact, recent reports suggest that other
PTKs can activate STAT transcription factors independent of Janus
family kinases (29, 30). While we cannot formally exclude the
possibility that Jak1 or any of the other JAK kinases is activated by
TSLP, but with a diminished efficiency, we believe that this is an
unlikely explanation. The efficiency of Stat5 phosphorylation is
reduced by only two- to threefold at most, and yet we could not detect
Jak1 activation in immunoprecipitates from 2 x 107
TSLP-stimulated cells, although activation was readily detected in
lysates from 2.5 x 106 IL-7-stimulated cells (data
not shown). Moreover, even prolonged exposures of the chemiluminescent
immunoblots failed to give any indication of activation of any of the
JAK kinases (data not shown). Taken together, these data strongly argue
against a diminished level of JAK kinase activation by TSLP. It is
somewhat surprising that Jak1 is not activated by TSLP treatment, since
it probably interacts with and phosphorylates the IL-7R -chain.
Jak3, on the other hand, associates with c (12, 13), and
since TSLP activity apparently does not depend on this chain, it is
less surprising that Jak3 appears not to be involved in TSLP signaling.
It is worth noting that while IL-13 stimulation activates Jak1 (which
interacts with the IL-4R -chain in that system), it does not
activate Jak3 (which is activated by IL-4 treatment) (31). This
suggests an important distinction between these two related cytokine
systems. Moreover, the demonstration here that TSLP stimulation can
lead to activation of Stat5 without apparent activation of Jak3 may
provide an explanation for the differential effect of Jak3 or IL-7
deficiency on B and T lymphopoiesis; in both cases B cell development
is more profoundly affected than is thymocyte development (4, 32, 33, 34).
These differences may reflect the bioavailability of TSLP in these
lymphopoietic tissues and/or the relative contribution of a
Jak3-independent action of TSLP to lymphopoiesis in the bone marrow or
thymus.
Finally, we have noted the unique capacity of TSLP to promote the
development of IgM+ immature B lymphocytes in three
different B cell culture systems. It is worth noting that in bone
marrow cultures, the number of IgM+ cells that develop in
TSLP-supplemented cultures is relatively close to the number that
develop in IL-7-supplemented cultures (Fig. 4). The elevated percentage
of IgM+ cells that are produced in TSLP cultures reflect
diminished numbers of B220+/IgM- cells. Hence,
we cannot at this time exclude the possibility that the reason why the
number of IgM+ cells produced in TSLP cultures is not
higher is because of the reduced number of precursor cells available.
Preliminary experiments supplementing long term cultures with both
cytokines suggest this may be true, but since high levels of IL-7 may
also antagonize the development of IgM+ cells (35, 36),
careful titration of cytokine levels may be critical to establish a
ratio that facilitates maximal IgM+ cell development. This
facilitation could be produced at least three ways. It is possible that
TSLP simply gives the appearance of IgM+ B cell development
by causing input IgM+ B cells to proliferate in these
cultures. We believe that this is an unlikely explanation for two
reasons. First of all, in fetal liver cultures there are virtually no
IgM+ B cells in the starting population of cells, and yet
these cultures respond more rapidly than either of the adult bone
marrow culture systems we employed. Second, we note the meager capacity
of TSLP to promote proliferation of any primary cells tested to date,
including unfractionated bone marrow, bone marrow enriched for
B-lineage cells, and peripheral B cells (S. D. Levin and A. Farr,
unpublished observations). TSLP might also appear to promote
differentiation by serving as a "survival factor" for
IgM+ B lymphocytes. While this possibility has not been
excluded as an explanation for enhanced IgM+ cell
development in bone marrow cultures, it is an unlikely explanation for
the development of IgM+ B cells in fetal liver cultures
because there are none at the start (data not shown), none develop in
cultures without added cytokine (data not shown), and very few develop
in the presence of IL-7 (Fig. 1). A third possibility is that TSLP
directly promotes the differentiation of pre-B cells into
IgM+ immature B cells. While both pro-B cells and pre-B
cells express the IL-7R (5) (our unpublished observations), only pro-B
cells respond to IL-7 (37). This raises the possibility that the
function of the IL-7R in pre-B cells could be to facilitate
differentiative responses to TSLP rather than IL-7. The effects of TSLP
could be brought about through several mechanisms. The critical event
in this developmental transition is the productive rearrangement of a
light chain locus, a process dependent on the recombinase-activating
genes, RAG-1 and RAG-2. It should be noted that in some in vitro
culture systems, IL-7 has been shown to keep expression of RAG genes
low and hence prevents light chain rearrangement and developmental
progression (35, 36). We suggest the possibility that TSLP may serve to
antagonize this function of IL-7, perhaps directly by increasing the
level of RAG expression or by competing with IL-7 for binding to its
receptor complex. It is also possible that TSLP promotes accessibility
of the germline light chain genes, which has been shown to be important
in the initiation of rearrangement (38). Recently, it was noted that
the transcription of some germline VH genes and hence their
rearrangement are diminished in pre-B cells from IL-7R-deficient mice
(39). Such a result raises the possibility that this effect actually
reflects a deficiency in TSLP signaling and that a similar mechanism
for regulating VL germline transcription may influence
light chain gene rearrangement. One possible way that TSLP could be
affecting transcriptional events is, of course, through the activation
of Stat5, as we have reported here. However, disruptions of the genes
for both Stat5 isoforms, singly and when superimposed on each other,
had no overt effect on B lymphopoiesis (at least as measured by the
appearance of normal numbers of peripheral B cells), although the
ability of bone marrow cells to form colonies in methylcellulose in
response to IL-7 was compromised (40). Hence, either TSLP also
influences transcription through other mechanisms as well or another
STAT protein is "masking" the Stat5 defects in the knockout animals
by redundant function.
We have demonstrated that TSLP has novel and intriguing properties when tested in in vitro assays of B lymphocyte development. Aside from the interesting aspects of the TSLP-invoked signaling cascade, definition of the in vivo importance of TSLP vs IL-7 in normal lymphopoiesis remains a critical challenge for future investigation.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Andrew G. Farr, Departments of Immunology and Biological Structure, Box 357420, University of Washington, Seattle, WA 98195. E-mail address:
3 Abbreviations used in this paper: c, common -chain; TSLP, thymic stromal lymphopoietin; LTBMC, long term bone marrow cultures; CM, conditioned medium; TSLP-R, thymic stromal lymphopoietin receptor; RAG, recombinase-activating gene.
4 Present address: Immunex Corporation, Seattle, WA 98101.
Received for publication July 31, 1998. Accepted for publication September 25, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Blood 74:1936. chain of the IL-2 receptor. EMBO J. 14:3654.[Medline]
chain: its role in the multiple cytokine receptor complexes and T cell development in XSCID. Annu. Rev. Immunol. 14:179.[Medline]
c chains with JAK1 and JAK3: implications for XSCID and XCID. Science 266:1042. chains. Proc. Natl. Acad. Sci. USA 91:7271. chain: a functional component of the interleukin-7 receptor. Science 262:1877. chain is required for IL-7 signal transduction. Eur. J. Immunol. 25:399.[Medline]
chain between receptors for IL-2 and IL-4. Science 262:1874.c subunit of the IL-2, IL-4, and IL-7 receptors. J. Immunol. 154:1596.[Abstract]
c subunit of the receptors for interleukins 2, 4, and 7. Proc. Natl. Acad. Sci. USA 92:5689.
. Blood 90:4341.This article has been cited by other articles:
|
|
 |
|
  M. C. Banas, B. Banas, K. L. Hudkins, T. A. Wietecha, M. Iyoda, E. Bock, P. Hauser, J. W. Pippin, S. J. Shankland, K. D. Smith, et al. TLR4 Links Podocytes with the Innate Immune System to Mediate Glomerular Injury J. Am. Soc. Nephrol., April 1, 2008; 19(4): 704 - 713. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Chen, B. L. Esplin, K. P. Garrett, R. S. Welner, C. F. Webb, and P. W. Kincade Retinoids Accelerate B Lineage Lymphoid Differentiation J. Immunol., January 1, 2008; 180(1): 138 - 145. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chappaz, L. Flueck, A. G. Farr, A. G. Rolink, and D. Finke Increased TSLP availability restores T- and B-cell compartments in adult IL-7 deficient mice Blood, December 1, 2007; 110(12): 3862 - 3870. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. I. Brown, J. Hulitt, J. Fish, C. Sheen, M. Bruno, Q. Xu, M. Carroll, J. Fang, D. Teachey, and S. A. Grupp Thymic Stromal-Derived Lymphopoietin Induces Proliferation of Pre-B Leukemia and Antagonizes mTOR Inhibitors, Suggesting a Role for Interleukin-7R{alpha} Signaling Cancer Res., October 15, 2007; 67(20): 9963 - 9970. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Kato, S. Favoreto Jr., P. C. Avila, and R. P. Schleimer TLR3- and Th2 Cytokine-Dependent Production of Thymic Stromal Lymphopoietin in Human Airway Epithelial Cells J. Immunol., July 15, 2007; 179(2): 1080 - 1087. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Rochman, N. Watanabe, K. Arima, Y.-J. Liu, and W. J. Leonard Cutting Edge: Direct Action of Thymic Stromal Lymphopoietin on Activated Human CD4+ T Cells J. Immunol., June 1, 2007; 178(11): 6720 - 6724. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. C. Osborne, S. Dhanji, J. W. Snow, J. J. Priatel, M. C. Ma, M. J. Miners, H.-S. Teh, M. A. Goldsmith, and N. Abraham Impaired CD8 T cell memory and CD4 T cell primary responses in IL-7R{alpha} mutant mice J. Exp. Med., March 19, 2007; 204(3): 619 - 631. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Omori and S. Ziegler Induction of IL-4 Expression in CD4+ T Cells by Thymic Stromal Lymphopoietin J. Immunol., February 1, 2007; 178(3): 1396 - 1404. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-C. Lee and S. F. Ziegler Inducible expression of the proallergic cytokine thymic stromal lymphopoietin in airway epithelial cells is controlled by NF{kappa}B PNAS, January 16, 2007; 104(3): 914 - 919. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. D. Klonowski, K. J. Williams, A. L. Marzo, and L. Lefrancois Cutting Edge: IL-7-Independent Regulation of IL-7 Receptor {alpha} Expression and Memory CD8 T Cell Development J. Immunol., October 1, 2006; 177(7): 4247 - 4251. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. R. Hardy B-1 B cell development. J. Immunol., September 1, 2006; 177(5): 2749 - 2754. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
|
