| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
or IL-4
*
Lautenberg Center for General and Tumor Immunology, Hebrew University-Hadassah Medical School, Jerusalem, Israel; and
Laboratory of Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and IL-4
by culture in the presence of their cognate Ag, APC, and appropriate
cytokines. In this study, we show that commitment to IFN- production
on the part of rigorously purified naive CD4 T cells can occur without
cell division. Indeed, even entry into S phase is not essential.
Moreover, both CD4 and CD4/CD8 thymocytes from TCR-transgenic mice
(5CC7 mice) on a Rag2-/- background can
acquire IFN--producing capacity when stimulated by peptide, APC, and
IL-12. These cells can do so without dividing and some acquire
IFN--producing activity without entry into S phase. Not only is cell
division not required for acquisition of cytokine-producing potential,
cell populations that have undergone the same numbers of divisions can
have quite different proportions of IFN-- or IL-4-producing cells,
depending on the duration of priming or, in the case of IL-4, on the
concentration of peptide. Thus, cell division is not a clock for the
expression of these cytokines. Factors associated with priming
conditions including strength of stimulation, duration of priming, and
number of divisions each play a role. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. Upon stimulation through their TCR, preferably with ample
costimulation, such cells acquire the capacity to secrete IL-4 or
IFN-. IL-4-dependent activation of Stat6 plays a major role in
differentiation into IL-4-producing cells (1, 2, 3, 4, 5) and is
opposed by IL-12/IFN- (6, 7, 8), whereas differentiation
into IFN--producing cells is strikingly enhanced by IL-12-dependent
Stat4 activation (9, 10) and is opposed by IL-4 (6, 8, 9, 11).
Development of cells that are capable of secreting IL-4 or IFN- has
been reported to be associated with the acquisition of DNase I
hypersensitivity in the IL-4 or IFN- genes (12, 13),
respectively, and with binding of NFAT-1 to key sites in the two genes
(14). Demethylation of CpGs in the second intron of the
IL-4 gene occurs in association with acquisition of competence to
secrete IL-4 (15). However, there is a CpG in the IFN-
promoter that is demethylated in both resting and activated CD4 T cells
as well as in both Th1 and Th2 cells but which is fully methylated in B
cells and other cell types (16), suggesting that some
accessibility of chromatin may be established very early in the
development of CD4 T cells.
It has been argued that DNA synthesis is essential for acquisition of
competence to transcribe the IL-4 and IFN- genes and that the number
of cell divisions can be regarded as a "clock" through which one
can evaluate the likelihood that a given cell will secrete IL-4 and/or
IFN- upon subsequent challenge (15, 17). For IFN-,
it has been suggested that there is a direct relationship between
division number and competence to transcribe, whereas for IL-4, it has
been proposed that a minimum number of three divisions is required
before any cells can transcribe IL-4. However, others have argued that
although DNA synthesis is important in IL-4 and IFN- production,
cell division is not essential and commitment to IL-4 or IFN-
production may occur quite early among cultured cells
(18).
In this study, we show that although cell division plays a role in the
frequency of cells that produce IL-4 or IFN- upon challenge, it is
only one of several important factors. Furthermore, among peripheral
CD4 T cells highly purified for a naive phenotype that have not entered
S phase or undergone cell division in response to Ag stimulation, a
substantial proportion develop into IFN- producers. Indeed, the
capacity of naive CD4 T cells to acquire IFN--producing capacity
without entering S phase is also observed for
CD4+ thymocytes. Both CD4+
and CD4+/CD8+ thymocytes
from TCR-transgenic mice respond to peptide plus APC with priming for
IFN- production even without completion of a cell cycle.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
5CC7 TCR-transgenic mice on Rag2-/- B10.A background were obtained from Taconic Farms (Germantown, NY). Female B10.A mice were purchased from The Jackson Laboratory (Bar Harbor, ME). All mice were maintained under pathogen-free conditions at the National Institute of Allergy and Infectious Diseases Animal Facility and were used at 6–12 wk of age.
Cytokines and Abs
Human IL-2 was a gift from Cetus (Emeryville, CA). Mouse
IL-4 was obtained from a recombinant baculovirus prepared in our
laboratory. Mouse IL-6 was purchased from PeproTech (Rocky Hill, NJ).
Mouse IL-12 was purchased from R&D Systems (Minneapolis, MN). Anti-CD3
(145-2C11) and anti-Thy1 (30-H12) Abs were obtained from cultured
hybridoma cells. Anti-CD3 Abs were further purified by chromatography
over protein A. Anti-IL-4 (11B11), anti-IFN- (XMG1.2),
anti-CD28 (37.51), and anti-CD16/CD32 (2.4G2) Abs were prepared
by Harlan Bioproducts for Science (Indianapolis, IN).
Fluorochrome-conjugated anti-CD62L (Mel14), anti-CD8 (53-6.7),
anti-B220 (RA3-6B2), anti-Iak (11-5.2),
anti-IL-4 (11B11), anti-IFN- (XMG1.2),
anti-5-bromo-2'-deoxyuridine
(BrdU)2 (3D4),
anti-V
1 (RR8-1), anti-V
3 (KJ25), and streptavidin were
purchased from PharMingen (San Diego, CA). Biotinylated 3G11 was
purchased from PharMingen. The BrdU flow kit was purchased from
PharMingen.
Naive CD4 cell preparation
CD4 cells were isolated from axillary, inguinal, and mesenteric lymph nodes of 5CC7 TCR-transgenic mice by negative selection, as previously described (8), using fluorescein-conjugated mAbs to B220, CD8, and I-Ak followed by incubation with magnetic beads coated with anti-FITC Abs (PerSeptive Biosystems, Framingham, MA). The depleted cells were centrifuged on a discontinuous 50–70% Percoll density gradient (four layers: 50, 60, 66, and 70%). Dense cells (those at the 66–70% interface and below) were collected and used for generation of effector cells. The cell preparations were >95% CD4+. The cells were stained with FITC-anti-62L and biotinylated 3G11, followed by PE-conjugated streptavidin. Naive CD62Lhigh3G11high cells were further purified by electronic cell sorting using a FACStar Cell Sorter (Becton Dickinson, San Jose, CA).
CFSE labeling and cell division analysis
Cells were washed twice in PBS and resuspended in PBS at a concentration of 2 x 107 cells/ml. An equal volume of freshly diluted 1.25 µM CFSE (Molecular Probes, Eugene, OR) in PBS was added for 4 min at room temperature; staining was stopped by the addition of an equal volume of FCS. The cells were immediately washed three times in culture medium. The percentage of cells in various CFSE peaks was determined by analysis with a FACScalibur and CellQuest software (Becton Dickinson).
Stimulation of naive CD4+ cells in vitro
T-depleted splenocytes as APC were prepared from B10.A mice
using anti-Thy-1 Abs plus rabbit complement (Cedarlane
Laboratories, Hornby, Ontario, Canada). Priming was performed by
culturing 1–4 x 105 CFSE-labeled or
unlabeled sorted CD4+ cells with 1–4 x
106 APC in 24-well tissue culture dishes
containing 1 ml of complete RPMI 1640 supplemented with 10% FCS, 2 mM
glutamine, 5 µM 2-ME, 100 U/ml penicillin, 100 µg/ml streptomycin,
1 mM sodium pyruvate, and 10 U/ml IL-2. When indicated, IL-6 (10
ng/ml), IL-4 (1000 U/ml (
500 pg/ml)), IL-12 (10 ng/ml), neutralizing
anti-IL-4 Abs (11B11; 10 µg/ml), or other cytokines and Abs, as
specified, were added to the culture. The culture medium was replaced
daily with fresh medium containing the appropriate cytokines and
anticytokine Abs. The cytochrome c peptide (88–104;
Structural Biology Section, Peptide Synthesis Laboratory, National
Institute of Allergy and Infectious Diseases, Rockville, MD) was used
at 0–100 µM as specified. At various times, the cells were
transferred to anti-CD3- and anti-CD28-coated 24-well tissue
culture dishes to assess their cytokine-producing capacity.
Intracellular staining of cytokines
Primed cells were transferred at various times after
stimulation, as specified, to anti-CD3- and anti-CD28-coated
24-well plates in 1 ml of culture medium containing IL-2 (10 U ml);
after 4 h of incubation, 2 µM monensin (Calbiochem, La Jolla,
CA) was added and 2 h later the adherent cells were removed by
pipetting in 1 ml of PBS and transferred to a tube containing 1 ml of
8% paraformaldehyde (Electron Microscopy Sciences, Fort Washington,
PA) in PBS. After 10 min at room temperature, the fixed cells were
washed once in PBS and then permeabilized by two washings in saponin
buffer (0.1% saponin (Calbiochem), 0.1% BSA (Sigma, St. Louis, MO),
and 10 mM HEPES in PBS) and stained for intracellular cytokines by
incubation for 30 min at room temperature with either 200 ng of
allophycocyanin-conjugated anti-IFN- or
allophycocyanin-conjugated anti-IL-4 mAbs. The stained cells were
washed three times in saponin buffer and then analyzed by flow
cytometry.
Cell cycle analysis
Primed cells were transferred to anti-CD3- and anti-CD28-coated 24-well plates 13–14 h or 37 h after priming initiation. After 5 h and 15 min, BrdU was added to wells to achieve a final concentration of 10 µM. Forty-five minutes later, the nonadherent cells were removed by washing with PBS containing 3% FCS. Adherent cells were removed by pipetting in 2 ml of PBS + 3% FCS, spun down, and the pellet was fixed and processed for staining with the BrdU flow kit reagents (PharMingen) according to the manufacturer’s protocol.
The CFSE-labeled cells were stained with PE anti-BrdU and APC
anti-IFN-. Unlabeled cells were stained with PE anti-V3
to identify transgenic cells, FITC anti-BrdU, 7-aminoactinomycin D
(7-AAD) and APC anti-IFN-. The stained cells were washed three
times in saponin buffer and then analyzed for cell cycle and cytokine
production on a FACScalibur (Becton Dickinson). At least 50 x
103 transgenic cells were collected in each
sample.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Naive CD4+ T cells from lymph nodes of
5CC7/Rag2-/- mice were obtained by cell
sorting using high levels of expression of CD62L and 3G11 as markers
for the naive state (8). These cells were labeled with
CFSE and were primed with varying concentrations of cytochrome
c peptide, ranging from 0 to 100 µM, in the presence of
IL-2 and IL-6 with IL-12, and anti-IL-4 (Th1 conditions) or with
IL-4, anti-IL-12, and anti-IFN- (Th2 conditions). After
various times of priming, they were stimulated for 6 h with
immobilized anti-CD3 and anti-CD28, and the number of cells
with cytosolic IFN- or cytosolic IL-4, as well as mean fluorescence
intensity (MFI) of IFN- staining, was measured. In addition, the
fluorescence intensity of CFSE was determined as a measure of cell
division.
To determine the degree of CFSE fluorescence that marked cells that had
not divided, we also placed such cells in culture with IL-2 alone.
Since very few of such cells enter S phase or divide, the predominant
peak of CFSE in this population defines those cells that had not
divided. To more rigorously verify that cells in the peak chosen as the
"0 division" peak had not divided, we used a BrdU-labeling approach
(Fig. 1
). Naive cells were stimulated
with cytochrome c peptide (1.0 µM) in the presence of APC,
IL-2, IL-12, and anti-IL-4. A total of 10 µM BrdU was added at
13 h of culture. The cells were harvested at 37 h and stained
for DNA content using 7-AAD and for BrdU incorporation. Cells that were
in the putative 0 division CFSE peak had few if any BrdU-labeled cells
that had less than a 2x content of DNA, based on staining with 7-AAD;
by contrast, among cells in the "1 division" and "2 division"
CFSE peaks, a large percentage of the BrdU+ cells
had less than 2x content of DNA. This argues that the latter cells had
completed at least one cycle after they had been labeled with BrdU. By
contrast, essentially all cells in the 0 division peak that had been in
S phase during the labeling period had not yet completed a cell cycle.
This strongly supports our identification of the 0 division peak as
correct.
|
(Fig. 2). Sixty and 80%
of cells that had undergone one or two divisions produced IFN-
whereas cells that had not divided had substantially fewer (15%)
IFN- producers. Plotting the amount of IFN- produced per cell, as
determined by MFI of positive cells, reveals a smooth increase with
increased numbers of divisions, from 700 for cells that have not
divided to 2700 for cells that have divided five times.
|
). Cells that had not divided or divided
once or twice were poorer IL-4 producers.
|
- or
IL-4-producing potential (13, 15). Note, however, that
even the cells that had not divided contain a measurable number of
IFN--producing cells, substantially in excess of those cultured with
IL-2, IL-6, IL-12, and anti-IL-4 without Ag for 85 h. The
latter cell population was entirely made up of cells that had not
divided; <5% produced IFN- in response to challenge with
anti-CD3 and anti-CD28. The frequency of IL-4 producers among
the 0 division group in response to priming with 0.1 µM peptide under
Th2 conditions is only slightly greater than the frequency in cells
cultured in the absence of cytochrome c peptide.
Cell division and DNA synthesis are not essential for acquiring
IFN--producing capacity
A more striking demonstration that cells can acquire
IFN--producing potential without cell division is shown in Fig. 4. Naive cells were cultured for 15 or
44 h with 0 or 1 µM cytochrome c peptide. They were
incubated with IL-2, IL-6, IL-12, and anti-IL-4. Essentially, none
of the cells cultured without peptide for 44 h had undergone any
cell divisions and <3% produced IFN- upon challenge. The cells
cultured for 44 h with 1 µM cytochrome c peptide
under Th1-inducing conditions were largely distributed into populations
that had undergone zero, one, or two divisions. More than 40% of the
cells that had not divided produced IFN- upon challenge. Thus, when
relatively short priming periods are employed, one can observe the
appearance of cells competent to transcribe IFN- even if no cell
division has occurred. The failure of the cells to produce IFN- if
they had not been cultured with cytochrome c peptide and the
fact that they had been highly enriched in cells with the markers of
the naive state strongly argues that these cells acquired
IFN--producing potential during this culture period.
|
-producing
capacity; only 5.7% of these cells produced IFN-. This result
supports the conclusion that the IFN--producing capacity of the 0
division cells at 44 h is acquired as a result of the priming
culture.
Not only was cell division not required to achieve competence to
produce IFN-, it was also possible for some cells to acquire
IFN--producing activity without entering S phase. Naive
TCR-transgenic CD4 T cells from
5CC7/Rag2-/- mice were cultured for 13 or
37 h with cytochrome c peptide (1 µM), T-depleted
spleen cells, IL-2, IL-6, IL-12, and anti-IL-4. At the end of the
culture period, they were placed in wells containing immobilized
anti-CD3 and anti-CD28 for 6 h; BrdU was added for the
final 45 min of the culture. In these experiments, monensin was
omitted. Cells in various stages of the cycle
(G0-G1, early S, late S,
and G2-M) were analyzed for IFN- content. The
majority of cells primed for 14 h that had not divided were in
G0-G1 (89%); only 2.5% of
these cells produced IFN-+ cells upon
challenge (Fig. 5A). Although
the percentage of IFN-+ cells was higher in
the cells that were in early and late S phase, there were extremely
small numbers of cells in both of these populations.
|
+ upon challenge with
anti-CD3 and anti-CD28 for 6 h (Fig. 5
B).
Eighteen percent of the cells that had not divided were in early S
phase (i.e., were BrdU+ and had an 7-AAD
intensity of less than two times the
G0-G1 level). Fifty-nine
percent of these cells produced IFN- upon challenge. Since many of
these cells had less than the maximal amount of BrdU, it is likely that
they had entered S phase after the beginning of the BrdU pulse and thus
were in G0-G1 when they
began transcribing IFN-.
Cells in late S phase and in G2-M of the first
cycle had a frequency of IFN-+ cells of 64 and
69%, respectively, while cells that had completed one or two divisions
displayed a frequency of IFN-+ cells of 83 and
86%, respectively (latter data not shown). Thus, there is an increase
in the frequency of IFN--producing cells with cell division, as well
an increase in MFI, but some cells that had not entered S phase had
acquired IFN--producing capacity.
Thymocytes from 5CC7/Rag2-/- mice can
be induced to produce IFN- without cell division or DNA synthesis
In view of the capacity of naive cells to acquire
IFN--producing capacity without DNA synthesis, we wondered whether
their predecessors might also have such potential. Thymocytes were
prepared from 5CC7/Rag2-/- donors and
cultured for 40 h with 1 µM cytochrome c peptide in
the presence of T-depleted spleen cells and IL-2 with IL-12 and
anti-IL-4 (Th1 conditions) or with IL-4, anti-IL-12, and
anti-IFN- (Th2 conditions). They were then challenged with
immobilized anti-CD3/anti-CD28 for 7 h and the percentage
of IFN-+ cells was determined. In addition,
cells that had not been primed in vitro were tested for "acute"
production of IFN-.
Among CD4+ (single-positive) cells, 27% of those
primed under Th1 conditions produced IFN-, whereas <1% of those
primed under Th2 conditions or unprimed did so (Fig. 6). Among
CD4+/CD8+ cells, 68% of
those primed under Th1 conditions produced IFN- whereas only 13% of
those primed under Th2 conditions did so and <1% of the unprimed
cells made IFN-.
|
Having shown that both single- and double-positive thymocytes could
develop into IFN- producers after a short period of Th1 priming, we
wished to determine whether cell division or DNA synthesis was
required. Thymocytes were prepared from
5CC7/Rag2-/- donors and labeled with
CFSE. They were cultured for 14 or 37 h with or without 1 µM
cytochrome c peptide in the presence of T-depleted spleen
cells, IL-2, IL-6, IL-12, and anti-IL-4. They were then challenged
with immobilized anti-CD3/anti-CD28 for 6 h and pulsed
with BrdU for the last 45 min of the culture. Among cells that had not
divided, those cultured without peptide contained very few
IFN--producing cells (0.5% at 14 h; 0.3% at 37 h).
Cells primed with peptide showed 3.5% IFN--producing cells in the
undivided population at 14 h. At 37 h, the 0 division
population contained 30% IFN-+ cells (Fig. 7) whereas the 1 division population
contained 76% IFN-+ cells (data not shown).
Indeed, in the 0 division population, 29.9% of the
G0-G1 and 61.7% of the
early S phase cells were IFN-+ (Table I). Thus, even thymocytes from
TCR-transgenic/Rag 2-/- mice have the
capacity to develop into IFN--producing cells without dividing.
|
|
Not only is cell division or DNA synthesis not essential for
IFN- production, both duration of priming and concentration of Ag
also play roles in determining the capacity to produce IFN- and
IL-4. Among cells primed for 85 h with peptide concentrations
ranging from 0.1 to 100 µM and with IL-2, IL-6, IL-12, and
anti-IL4 that had divided between one and five times, a similar
proportion produced IFN- upon challenge. However, the amount of
IFN--produced per cell, as measured by the MFI of IFN- staining
in these cells, varied strikingly with the concentration of peptide
used for priming. Among cells that had divided five times that had been
primed for 85 h, those primed with 0.1 µM peptide had an MFI of
4200, whereas those primed with 100 µM peptide had an MFI of 1500
(Fig. 8A). Among those primed
to be IL-4 producers for 85 h, 40% of cells that divided five
times that were primed with 0.1 µM peptide produced IL-4, whereas
only 5% of comparable cells primed with 100 µM peptide produced IL-4
(Fig. 8B). Thus, the concentration of priming peptide is an
independent variable determining the likelihood that individual cells
will produce cytokine even if the cytokine environment, the duration of
priming, and the number of cell divisions are kept constant.
|
producers than
those primed for 64 or 85 h (Fig. 9A). Among cells that had
divided once or twice, those primed for 44 or 64 h had a higher
percentage of IFN- producers than did cells primed for 85 h.
This presumably reflects the further division of highly stimulated
cells so that the cells that had divided zero, one, or two times at
85 h may largely represent cells that responded less vigorously
than cells that divided zero, one, or two times over the shorter
priming periods. Nonetheless, it argues that frequency of division is
not the sole determinant of cytokine-producing potential. A similar
analysis of cells primed to be IL-4 producers shows that cells that
divided one through five times during an 85-h priming period with 1
µM peptide had a higher frequency of IL-4 producers than cells primed
under identical conditions for 64 h that had divided a similar
number of times (Fig. 9B). Here, one would argue that
priming for IL-4 production requires time, perhaps to integrate the
total extent of signaling, so that very few cells can be primed to
produce IL-4 during the shorter interval even though they have
responded quite well as measured by cell division. Thus, cell division
is not a rigid clock of commitment to cytokine production, although it
does appear to be one of the critical variables.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and
IL-4. It is well known that acquisition of competence to produce these
cytokines requires TCR engagement and the presence of inducing
cytokines.
Acquisition of such competence is associated with changes in
chromatin structure as revealed by analysis of DNase I
hypersensitivity, methylation state, and transcription factor binding.
In cells differentiating to the Th2 phenotype, sites in the IL-4 gene
and the IL-4/IL-13 intergenic region display DNase I hypersensitivity
(12, 13); hypomethylation of several CpGs in the second
intron of the IL-4 gene has been reported (13); and NFAT
has been shown to be associated with a segment of the IL-4 gene
(14). Similarly, increased accessibility of the IFN-
gene occurs during Th1 differentiation (12).
Two groups reported that the process through which competence to
transcribe the IL-4 and IFN- genes was acquired was tightly linked
to cell division. Both Bird et al. (15) and Gett and
Hodgkin (17) argued that cell division was required for
acquiring such competence and that there was an increase in the
capacity to produce IL-4 or IFN- with each cell division, leading
them to propose that there was a finite probability of opening the IL-4
or the IFN- locus with each cell division. Effectively, cell
division was regarded as a clock that could allow measurement of the
acquisition by individual cells of the capacity to transcribe these
cytokines. This explanation would also explain the observation that
IL-4 was often monoallelically expressed in individual Th2 cells
(19, 20, 21).
Richter at. al. (18) challenged the notion that
acquisition of competence to produce IL-4 was strictly linked to cell
division. They reported that some cells that had not divided acquired
the capacity to produce IL-4 when cultured under Th2 conditions, if
cell cycle inhibitors were used to block cell division. They added
inhibitors from the outset of the stimulatory culture and observed that
some cells acquired the capacity to produce IL-4 without dividing.
However, those inhibitors that acted before or during S phase prevented
acquisition of IL-4-producing capacity, leading to the conclusion that
completion of S phase was required. Richter et al. (18)
also showed that some cells could produce IFN- without completing a
cell division. They further argued that the concept that cell division
was a clock that measured the frequency of cells capable of
secreting IL-4 was incorrect. Thus, cell populations that had
undergone the same number of divisions but had been primed for
different periods of time contained different percentages of
IL-4-producing cells, a result that we have also described in this
report.
We (21)3 have argued that expression of IL-4 is not a measure of the competence of a cell to transcribe IL-4, since we have observed that early (67 h) in a Th2 culture, all cells have an equal probability of expressing IL-4, although only a small proportion do so. Thus, basing conclusions regarding acquisition of competence to transcribe IL-4 on the frequency of cells that actually transcribe this cytokine could be quite misleading.
In this study, we have investigated the requirements for cell division
and DNA synthesis to acquire IFN--producing capacity in more detail,
taking advantage of the ability to directly analyze the number of cell
divisions, entry into S phase, and cytokine-producing capacity on the
individual cell level. Thus, using CFSE to count cell divisions, BrdU
uptake to measure entry into, and progress through, S phase and
intracellular staining for cytokines, we can directly determine the
cycle status of cells and their ability to produce cytokines.
We observed that a substantial fraction of naive
CD4+ T cells primed to be Th1 cells acquired the
capacity to produce IFN- by 44 h of culture. Indeed, 43% of
cells that had not divided at 44 h could be stimulated to produce
IFN- by immobilized anti-CD3 and anti-CD28. Even more
striking was the finding that of the 0 division cells that were in
G0-G1 at 44 h, 16%
were capable of secreting IFN-. Furthermore, 59% of the 0 division
cells in early S phase were capable of secreting IFN-. Since BrdU
was present only for the last 45 min of the culture, it is likely that
many of these cells had acquired competence to produce IFN- while
they were still in G1. Thus, in contrast to
others, we provide direct evidence that acquisition of competence to
produce IFN- can be achieved without entering S phase.
Obviously the contention that cells required neither cell division nor
DNA synthesis to acquire competence to produce IFN- requires
substantial confidence that the cells were actually naive and had not
acquired their capacity to transcribe IFN- previously. The cells we
used were derived from TCR receptor transgenic mice on a
Rag2-/- background. This ensures that all
T cells only expressed receptors for the 88–104 peptide of pigeon
cytochrome c and thus should markedly diminish the frequency
of cells that could have been activated by environmental Ags. We
further limited our analysis to lymph node cells and purified these
cells both by density gradient centrifugation to obtain only the
densest cells and by sorting for CD62Lbright,
3G11bright cells. We have previously reported
that 3G11, an Ab recognizing a disialoceramide (22), is an
excellent marker for the naive state and that
CD62Lbright, 3G11bright
lymph node cells produce no detectable IL-4 or IFN- on primary
stimulation (8). We emphasize these points because failure
to be rigorous about purification of naive cells may have led to
unreliable results due to contamination of the population with
previously activated cells that had already undergone commitment to a
given cytokine-producing phenotype in vivo.
We further observed that not only do naive peripheral
CD4+ T cells have the capacity to differentiate
into IFN- producers under appropriate stimulatory conditions but so
do both CD4 and CD4/CD8 thymocytes from the same
5CC7/Rag2-/--transgenic mice. Indeed,
these cells could acquire IFN--producing capacity without cell
division and some of them did so without entering S phase, just as was
true of peripheral CD4 T cells.
There are previous reports that mouse (23) and human
(24, 25) single positive CD4 thymocytes can be stimulated
to become IFN- producers. It has generally been interpreted that the
cells with this capacity are NK T cells or their progenitors
(26). However, our observation of the capacity of CD4 and
CD4/CD8 thymocytes to acquire IFN--producing capacity cannot be
ascribed to NK T cells. Since we used thymocytes from 5CC7-transgenic
mice on a Rag2-/- background, there
should be few if any NK T cells or their progenitors, since virtually
all NK T cells express V14/J281 (27).
It is interesting that two CpGs in the IFN- promoter that are
normally methylated in non-T cells have been shown to be hypomethylated
in thymocytes (16). These CpGs are at a site that had been
identified by Young et al. (28) as one to which proteins
found in nuclear extracts of Th1 clones can bind. The recognition that
the IFN- gene has hypomethylated sites in its promoter in thymocytes
and naive T cells, but not in liver cells, taken with our observation
that a substantial proportion of thymocytes and naive CD4 T cells can
produce IFN- without entering S phase strongly suggests that
transcription of the IFN- gene in such cells may be principally
limited not by the state of chromatin but rather by availability of key
transcription factors, such as T-bet, which are induced by
Th1 priming conditions (29). The observed differences in
DNase I hypersensitivity between Th1 and Th2 lines (12)
may reflect later events associated with the acquisition of
"high-rate" IFN- transcription in well-differentiated Th1 cells.
This may be associated with the very large amounts of IFN- produced
per cell in such highly differentiated populations.
Regarding the issue of cell division as a clock, we argue against that
both on the basis of our observations regarding the lack of concordance
between competence to transcribe IL-4 and the actual expression of IL-4
(21)3 as well as our observations
and those of Richter et al. (18) that cell populations
primed to be Th2 cells for various times will have different
frequencies of IL-4-producing cells, despite having undergone the same
number of divisions. In the experiments we presented, cells that had
undergone one to five divisions had a higher frequency of
IL-4-producers if they had been cultured for 85 h than for 64
h. Interestingly, for IFN-, the results were reversed. Cells that
had undergone zero, one, or two divisions had a greater frequency of
IFN- producers if they were cultured for 44 h than if they were
cultured for 85 h.
We would suggest that the IL-4 data can be explained by a constant proportion of cells primed under Th2 conditions (probably all) having accessible chromatin but that an increase in expression of key transcription factors, such as c-maf, occurs as time of priming proceeds. Indeed, it is known that c-maf levels tend to accumulate with duration of priming, consistent with this view (30).
A similar explanation can proposed for the discordance in expression of
IFN-. We postulate that essentially all Th1 cells have or rapidly
acquire a state of accessible IFN- chromatin and that expression
depends upon the levels of key transcription factors, such as
T-bet (30). Cells that respond vigorously
(i.e., those that divide rapidly) may have induced higher levels of the
key factor(s) than those that respond less intensely. This might
account for more IFN- expression by cells that had divided once or
twice at 44 h in contrast to cells that had only completed one or
two cell divisions after 85 h.
A similar argument may be adduced for the finding that different
concentrations of Ag lead to differences in the frequency of
IL-4-producing cells among populations that had undergone the same
number of cell divisions. Thus, one would predict that differences in
the levels of key transcription factors will be found in cells that had
divided three, four, or five times and had been primed with 100 µM
peptide vs those primed with 0.1 or 1 µM peptide. For IFN-,
similar differences are observed, although they are at the level of the
amount of cytokine per cell rather than at the level of the frequency
of cytokine producers within the population. Cells that had undergone
three, four, or five divisions and were stimulated with 100 µM
peptide had a substantially lower MFI of intracellular IFN- than
cells that had undergone the same number of divisions that had been
stimulated with 0.1 µM peptide. Again, one would predict that the
level of expression of key transcription factors will be higher in
cells primed with 0.1 µM peptide than with 100 µM peptide, despite
the fact that they had completed the same number of divisions.
Fundamentally, we argue that cell division will not be a clock for
expression of cytokines in newly primed cells because it is not a clock
for the opening of chromatin. Factors associated with the priming
conditions, strength of stimulation, duration of priming, and number of
divisions, each play a role in determining the probability that a given
set of cells will actually transcribe one or both of their IFN- or
IL-4 genes; in turn, that probability will be governed not only by the
degree of chromatin accessibility but also by the levels of key
transcription factors.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Abbreviations used in this paper: BrdU, 5-bromo-2'-deoxyuridine; 7-AAD, 7-aminoactinomycin D; MFI, mean fluorescence intensity.
3 J. Hu-Li, C. Pannetier, L. Guo, M. Lohning, H. Gu, C. Watson, M. Assenmacher, A. Radbruch, and W. Pace. Rgulation of expression of IL-4 alleles: analysis using a chimeric GFP/IL-4 gene. Submitted for publication.
Received for publication August 18, 2000. Accepted for publication September 29, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and IL-4 on the in vitro development of human Th1 and Th2 clones. J. Immunol. 148:2142.[Abstract]
T-cell-receptor transgenic system. Proc. Natl. Acad. Sci. USA 89:6065. and IL-4 expression. Cytokine 12:198.[Medline]
-producing CD4+ and CD8+ T cells. Int. Immunol. 11:593. gene. J. Immunol. 153:3603.[Abstract]
This article has been cited by other articles:
|
|
 |
|
  Q. Liu, Z. Liu, C. T. Rozo, H. A. Hamed, F. Alem, J. F. Urban Jr., and W. C. Gause The Role of B Cells in the Development of CD4 Effector T Cells during a Polarized Th2 Immune Response J. Immunol., September 15, 2007; 179(6): 3821 - 3830. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. V. Ganusov, D. Milutinovic, and R. J. De Boer IL-2 Regulates Expansion of CD4+ T Cell Populations by Affecting Cell Death: Insights from Modeling CFSE Data J. Immunol., July 15, 2007; 179(2): 950 - 957. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Jelley-Gibbs, J. P. Dibble, S. Filipson, L. Haynes, R. A. Kemp, and S. L. Swain Repeated stimulation of CD4 effector T cells can limit their protective function J. Exp. Med., April 4, 2005; 201(7): 1101 - 1112. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Sehra, D. Patel, S. Kusam, Z.-Y. Wang, C.-H. Chang, and A. L. Dent A Role for Caspases in Controlling IL-4 Expression in T Cells J. Immunol., March 15, 2005; 174(6): 3440 - 3446. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Liu, Q. Liu, H. Hamed, R. M. Anthony, A. Foster, F. D. Finkelman, J. F. Urban Jr, and W. C. Gause IL-2 and Autocrine IL-4 Drive the In Vivo Development of Antigen-Specific Th2 T Cells Elicited by Nematode Parasites J. Immunol., February 15, 2005; 174(4): 2242 - 2249. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Quemeneur, L. Beloeil, M.-C. Michallet, G. Angelov, M. Tomkowiak, J.-P. Revillard, and J. Marvel Restriction of De Novo Nucleotide Biosynthesis Interferes with Clonal Expansion and Differentiation into Effector and Memory CD8 T Cells J. Immunol., October 15, 2004; 173(8): 4945 - 4952. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. W. Lawrence and T. J. Braciale Activation, Differentiation, and Migration of Naive Virus-Specific CD8+ T Cells during Pulmonary Influenza Virus Infection J. Immunol., July 15, 2004; 173(2): 1209 - 1218. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Seki, M. Miyazaki, W. Suzuki, K. Hayashi, K. Arima, E. Myburgh, K. Izuhara, F. Brombacher, and M. Kubo IL-4-Induced GATA-3 Expression Is a Time-Restricted Instruction Switch for Th2 Cell Differentiation J. Immunol., May 15, 2004; 172(10): 6158 - 6166. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. J. Manfras, S. Reuter, T. Wendland, B. O. Boehm, and P. Kern Impeded Th1 CD4 memory T cell generation in chronic-persisting liver infection with Echinococcus multilocularis Int. Immunol., January 1, 2004; 16(1): 43 - 50. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. D. Roback, M. S. Hossain, L. Lezhava, J. W. Gorechlad, S. A. Alexander, D. L. Jaye, S. Mittelstaedt, S. Talib, J. E. Hearst, C. D. Hillyer, et al. Allogeneic T Cells Treated with Amotosalen Prevent Lethal Cytomegalovirus Disease without Producing Graft-versus-Host Disease Following Bone Marrow Transplantation J. Immunol., December 1, 2003; 171(11): 6023 - 6031. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Lohning, A. Richter, T. Stamm, J. Hu-Li, M. Assenmacher, W. E. Paul, and A. Radbruch Establishment of memory for IL-10 expression in developing T helper 2 cells requires repetitive IL-4 costimulation and does not impair proliferation PNAS, October 14, 2003; 100(21): 12307 - 12312. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Adkins, T. Williamson, P. Guevara, and Y. Bu Murine Neonatal Lymphocytes Show Rapid Early Cell Cycle Entry and Cell Division J. Immunol., May 1, 2003; 170(9): 4548 - 4556. [Abstract] [Full Text] [PDF]
|
|
