| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Mucosal Immunity Section, Laboratory of Clinical Investigation, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20852
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. The self-MHC-reactive clones also inhibited proliferation of
primary CD4+ T cells and TT-specific T cell clones, but in
this case the inhibition was mediated by both IL-10 and TGF-ß. In
further studies, the interactions between self-MHC-reactive T cell
clones and non-T cells that led to suppressor cytokine production have
been explored. We found that prestimulation of non-T cells for 8 h
with PWM or for 48 h for rCD40L results in non-T cells capable of
inducing self-MHC-reactive T cell to produce high levels of TGF-ß and
IL-10. In addition, these prestimulation times coincided with peak
induction of HLA-DR and costimulatory B7 molecule (especially CD86)
expression on B cells. Finally, addition of CTLA-4/Fc or blocking
F(ab')2 anti-CTLA-4 mAb, plus optimally stimulated
non-T cells, to cultures of self-MHC-reactive clones inhibited the
induction of TGF-ß but not IL-10 or IFN- production. In summary,
these studies show that activated self-MHC-reactive T cells have the
cytokine phenotype of Th3 or T regulatory cell 1 and thus may be
important regulatory cells that mediate oral and peripheral tolerance
and prevent the development of autoimmunity. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It is now well-established that immune responses driven by Th1 T cells and Th2 T cells (7, 8, 9, 10) are sometimes also influenced by a third T cell type whose main function is counterregulation or suppression of immune responses mediated by Th1 and Th2 (11, 12). It has been shown, for instance, that the induction of oral tolerance by the feeding of relatively low amounts of myelin basic protein leads to the induction of immunoregulatory T cells, which prevent the development of experimental autoimmune encephalitis (11, 12). Such T cells have a unique cytokine production profile in that they produce high levels of TGF-ß without necessarily producing either Th1 or Th2 cytokines; therefore, they have been dubbed Th3 T cells (12). Th3 cells producing TGF-ß have also been shown to occur in experimental models of colitis or diabetes or in HgCl2-induced autoimmune disease, and in these instances it is thought that such T cells play an important role in disease prevention or cure (13, 14, 15, 16). Recently, another type of regulatory T cell has been identified (in both mice and humans) that may be related to the aforementioned Th3 T cell. This cell, termed T regulatory cell 1 (Tr1), is induced in vitro by stimulation of T cells in the presence of IL-10 and is a T cell that produces high levels of IL-10 (17).
The recognition that self-MHC-reactive T cells can act as suppressor T cells as well as the recent interest in Th3/Tr1 T cells prompted us to re-examine self-MHC-reactive T cells to more completely characterize the mechanism of the suppressive capability. To this end, we developed self-MHC-reactive T cell clones as in previous studies and then showed that stimulation of these clones by activated B cells leads to production of high levels of TGF-ß and IL-10 and that such production accounts for their suppressor activity. We also showed that such cytokine production depended on interaction between B7 molecules on the activated B cell and CTLA-4 on the stimulated self-MHC-reactive T cell. These studies thus characterize a unique T cell subset that may play an important immunoregulatory role during inflammation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
FITC-conjugated anti-CD4 (Leu 3a + 3b), anti-CD19 (B43),
anti-HLA-DR (G46–6), PE-conjugated CD86 (B7-2, FUN1), CD152
(CTLA-4, BNI3), anti-HLA-DR, and FITC- or PE-conjugated isotypic
control mAbs were purchased from PharMingen (San Diego, CA).
PE-conjugated anti-CD80 (B7-1, L307.4) was purchased from Becton
Dickinson Immunocytometry Systems (San Jose, CA). Anti-CD3 (OKT3)
hybridoma was obtained from American Type Culture Collection (Manassas,
VA), and mAb was subsequently purified from ascites. Anti-CD28 mAb
(9.3) was kindly provided by Dr. C. June (Bethesda Naval Hospital,
Bethesda, MD). F(ab')2 anti-CTLA-4 (8H5) was
obtained from Ancell (Bayport, MN). Neutralizing anti-TGF-ß1,
-ß2, and -ß3 mAb was purchased from Genzyme (Cambridge, MA); and
anti-IL-2, anti-IFN-, and anti-IL-10 mAb and recombinant
human CTLA-4/Fc were obtained from R&D Systems (Minneapolis, MN). PWM
and recombinant human IL-2 were obtained from Life Science Technologies
(Baltimore, MD). TT was purchased from List Biological Laboratories
(Campbell, CA) and was provided by Wyeth-Lederle Vaccines and
Pediatrics (Pearl River, NY). Recombinant CD40 ligand (rCD40L)-trimer
was kindly provided by Immunex (Seattle, WA).
Generation of SR T cell clones and Ag-specific T cell clones
Self-MHC-reactive human T cell lines and clones were derived
from peripheral blood T cells repetitively stimulated with TT Ag in the
presence of irradiated (4000 rad) autologous non-T cells (non-Tx) cells
and rIL-2 as previously described (4, 5). In brief, using
previously described Ficoll gradient and SRBC rosetting techniques, T
cell and non-T cell populations were isolated from the peripheral blood
of a normal volunteer who had previously undergone TT immunization. The
T cell population contained >95% T cells, and the non-T cell
population contained <0.5% T cells. A CD4+ T
cell subset was then isolated from the T cell population by negative
selection using a human CD4 subset minicolumn (R&D Systems); a
population containing more than 93% CD4+ T cells
was obtained. Equal numbers of CD4+ T cells and
non-Tx cells were then cocultured at a concentration of 2 x
106 cells/ml in complete RPMI 1640 medium
(BioWhittaker, Walkersville, MD) supplemented with 10 mM HEPES,
penicillin (100 U/ml), streptomycin (100 µg/ml), and 10% FCS
(Biofluids, Rockville, MD) in 24-well culture plates (Costar, Corning,
NY) with TT stimulation (2 Lf/ml). In some experiments,
heat-inactivated autologous serum was used instead of FCS in the
induction phase of these cultures (during the first 8 days of
cultures). Eight to 10 days later, viable cells were isolated by
density gradient centrifugation on Lymphocyte Separation Medium (ICN
Biomedicals, Aurora, OH) and restimulated with fresh TT, non-Tx cells,
and IL-2 (2–4 ng/ml). TT and non-Tx cells were replenished every 8–10
days, and IL-2 was replenished every 3–4 days. Eight to 10 days after
the last restimulation, the stimulated CD4+ T
cells were cloned by limiting dilution at one cell per well in the
presence of 1 x 105 non-Tx cells with TT
and IL-2 in round-bottom 96-well microplates (Nalge Nunc International,
Rochester, NY). Wells containing expanding colonies of 
5 x
104 cells on days 14–28 were transferred to
larger wells and expanded by further exposure to TT, non-Tx cells, and
IL-2. Cloned T cells were assessed for their proliferative response to
non-Tx cell alone or to TT and non-T cells to determine their status as
SR or Ag-reactive (TT-reactive) T cells. The three self-MHC-reactive T
cell clones (SR-1, SR-2, and SR-3) and two TT-specific T cell clones
(TT-1 and TT-2) were subsequently maintained for long periods of
time.
Assay of self-reactivity and Ag-reactivity of clonal T cells
Each T cell clone was cocultured at a concentration of 1–2.5 x 104 cells/well with non-Tx at a concentration of 5 x 104 cells/well in the presence or absence of TT (2 Lf/ml) for 72 h in 0.2 ml 10% FCS containing complete RPMI 1640 medium or in 15% autologous serum containing X-VIVO 20 medium (BioWhittaker), which is free from FCS, in 96-well round-bottom microplates (in triplicate). The cells were pulsed with 1 µCi/well of [3H]thymidine (Amersham Pharmacia Biotech, Chicago, IL) for the last 8 h of culture, and then they were harvested and assessed for thymidine deoxyribose incorporation in a liquid scintillation counter.
Cell culture and ELISA for Ig production
SR or Ag-reactive T cell clones were cocultured at a
concentration of 5 x 104 cells/well with an
equal concentration of non-T cells with or without PWM (20 µg/ml),
rCD40L (2 µg/ml), or TT (2 Lf/ml) and with or without one of several
neutralizing anti-cytokine mAbs (anti-TGF-ß1, anti-IL-10,
anti-IFN-, anti-IL-2, or IgG1 control MOPC-21; all added to
achieve a concentration of 10 µg/ml) for 8 days. In some experiments,
CD19+ B cells were positively selected by
magnetic beads (Dynal, Lake Success, NY) and stimulated with
Staphylococcus aureus Cowan I bacteria (SAC) (Pansorbin;
Calbiochem, La Jolla, CA) and IL-2 (18) in the presence or
absence of supernatants of self-MHC-reactive T cell cultures. At the
end of culture, supernatants were collected, and IgG in the supernatant
was assayed by ELISA as previously described (18) using
affinity-purified (mouse Ig-absorbed) goat anti-IgG Ab and alkaline
phosphatase-conjugated goat anti-IgG Ab (Sigma, St. Louis, MO) in
96-well ELISA plates (Immulon 1; Dynatech, Chantilly, VA). After
washing, colorimetric substrate p-nitrophenyl phosphate
(Sigma) was added and A450 was
determined using a microplate ELISA reader (MR5000; Dynatech).
Suppressor activity of T cell proliferative response
To examine the capacity of the SR T cell clones to suppress T cell proliferation, culture supernatants were obtained from cultures containing equal numbers (5 x 105 cells/well each in 24-well plates) of SR T cell and non-Tx cell clones, which had been precultured with or without PWM for 8 h and washed three times to remove PWM. These supernatants were then added (at a 50% final concentration) to cultures containing fresh autologous or allogeneic CD4+ T cells (5 x 104 cells/well) under stimulation with immobilized anti-CD3 and soluble anti-CD28 in 96-well plates. Alternatively, the supernatants were added to cultures of TT-1 cells (2.5 x 104 cells/well) under stimulation with TT in the presence of an equal number of fresh non-Tx cells. Cultures were terminated at 72 h, and cell proliferation was measured by the incorporation of 3H during the last 8 h of cultures.
Cell culture and ELISA for cytokine production
For assessing the capacity to produce cytokines, each T cell
clone at a concentration of 5 x 105
cells/well was stimulated with immobilized anti-CD3 (10 µg/ml)
and soluble anti-CD28 (1 µg/ml) or was cocultured with an equal
number of resting or prestimulated non-Tx cells for varying periods
with PWM (20 µg/ml) or CD40L (2 µg/ml) in a total volume of 1 ml in
24-well plates. In some experiments, non-T cells were further separated
by magnetic beads into B cells using positive selection with
anti-CD19-coated beads or into monocyte/macrophages (M
) using
negative selection with mAbs to CD2, CD7, CD16, CD19, and CD56
(Monocyte Negative Isolation Kit; Dynal), and then they were
prestimulated with PWM. The purified B cells obtained were >95%
CD19-positive, and the monocyte-M were >90% CD14-positive. In
other experiments, T cell clones were stimulated with coimmobilized
anti-CD3 (5 µg/ml) and anti-CTLA-4 (BNI3; 5 µg/ml) or
MOPC-21 (IgG1; 5 µg/ml; Sigma) with or without soluble anti-CD28
(1 µg/ml). Culture supernatants were collected after 48 h or, in
the case of TGF-ß1, after 72 h. In cultures set up for the assay
of TGF-ß1 production, the FCS concentration was reduced to 3%, and
wells containing culture medium without cells were included to measure
"background" TGF-ß1. Cytokines secreted into culture fluid were
assayed by commercial ELISA kits according to the manufacturer’s
protocol. IL-2, IFN-, IL-4, IL-5, and IL-10 were measured by
Quantikine Immunoassay kits (R&D Systems). TGF-ß1 was measured with
the TGF-ß1 Emax immunoassay kit (Promega, Madison, WI). The total
amounts of TGF-ß1 (latent and bioactive TGF-ß1) produced by cells
were assayed by conversion of latent TGF-ß1 into bioactive TGF-ß1
by acid treatment. Briefly, culture supernatants were diluted with four
volumes of Dulbecco’s PBS and then were acidified with 1 µl of 1N
HCl for each 50 µl of diluted sample. Samples were incubated for 15
min at room temperature, neutralized by adding 1 µl of 1N NaOH per 50
µl of sample, and processed for sandwich ELISA according to the
manufacturer’s instructions. The amounts of total TGF-ß1 produced by
cells were determined by subtracting the amount of TGF-ß1 contained
in the 3% FCS in the medium (90–110 pg/ml) that had been incubated
in wells without cells. In some experiments, non-T cells prestimulated
for an indicated period with or without PWM or rCD40L were exposed to
brefeldin A (5 µg/ml; Sigma) treatment for the last 3 h. The
non-T cells prestimulated in this way were then irradiated and were
cocultured with SR or TT-specific clones in the presence or absence of
rCTLA-4/Fc (5 µg/ml) or F(ab')2 anti-CTLA
(5 µg/ml).
Flow cytometric analysis
After prestimulation with PWM or rCD40L as described above, non-T cells were enriched for CD19+ B cells by magnetic beads (Dynal, Lake Success, NY) (18) and stained with the combinations of FITC anti-CD19 vs PE anti-HLA-DR, FITC anti-CD14 vs PE anti-HLA-DR, FITC anti-HLA-DR vs PE anti-CD80, and FITC anti-HLA-DR vs PE anti-CD86, respectively. In addition, cells in cocultures of a SR T cell clone and non-T cells stimulated with PWM, rCD40L, or immobilized anti-CD3 and anti-CD28 for 8 h were stained with FITC anti-CD4 vs PE anti-CTLA-4. The stained cells were then analyzed with a FACScan flow cytometer (Becton Dickinson, Mountain View, CA) using Cellquest software.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Using previously described methods (4, 5, 6), SR and Ag
(TT)-reactive T cells were generated in cultures of
CD4+ T cells (from a TT preimmunized
HLA-DR9+/+ individual) repetitively stimulated by
TT in the presence of irradiated non-T cells (non-Tx cells).
Subsequently, the expanded T cells were cloned by limiting dilution,
and the specificity of each clone was characterized by proliferation
assay. As shown in the experiment displayed in Table I
(which is representative of four
similar experiments), this procedure gives rise to an array of T cell
clones, as defined by proliferative responses to non-Tx alone or to
non-Tx plus TT. Eight of 18 of the generated clones, exemplified by
clone AT-1, exhibited some proliferation when restimulated with non-Tx
cells alone as well as enhanced proliferation when stimulated with
non-Tx plus TT; these clones thus appeared to be both SR and
Ag-reactive to varying degrees. In contrast to these "mixed"
clones, two clones of the 18 clones generated, exemplified by clones
TT-1 and TT-2, exhibited little or no proliferation when restimulated
with non-Tx cells, but they exhibited robust proliferation when
restimulated with non-Tx cells plus TT; these cells were thus strictly
Ag-reactive T cells. In addition, another two clones of the 18 clones
generated, exemplified by clones SR-1 and SR-2, exhibited equally
strong proliferation when restimulated with non-Tx cells or with non-Tx
cells plus TT; therefore, these cells were self-MHC-reactive cells.
Finally, six of 18 clones, exemplified by BT-2, responded poorly to
either non-Tx cells or non-Tx cells plus TT, and they eventually died
out. Overall, these data, plus those obtained from three other similar
experiments, indicate that 11% of the clones generated under these
culture conditions were self-MHC-reactive clones similar to SR-1
and SR-2.
|
, that
these clones also proliferate well when stimulated by non-T cells in a
FCS-free medium (X-VIVO 20), which contains human serum albumin and
15% autologous human serum rather than FCS. In addition, as also shown
in Table II, a self-MHC-reactive T cell clone, exemplified by SR-3,
could be generated in cultures of cells repetitively stimulated by TT
in which the first two cycles of stimulation were conducted in FCS-free
medium (X-VIVO 20). Finally, the self-MHC-reactive T cell clones SR-1
and SR-2 did not proliferate when cocultured with non-Tx cells obtained
from three individuals with different HLA-DR types (allogeneic T
cells), and proliferation induced by self non-Tx cells was completely
inhibited by the presence of anti-HLA-DR in the culture (data not
shown).
|
In previous studies it was shown that SR T cells have dual
regulatory function with respect to their capacity to regulate (help or
suppress non-T cell (B cell)) Ig production. This property of SR T
cells was confirmed in the present study. Thus, as shown in Fig. 1A–C, in the absence of a
cell stimulant such as PWM, the self-MHC-reactive T cell clones induced
autologous B cells present in the same culture to produce IgG, whereas
in the presence of PWM, such induction of B cells was greatly
decreased. That this decrease of Ig production was in fact a result of
suppression (rather than lack of help) was shown in previous studies
demonstrating that appropriately stimulated self-MHC-reactive T cell
clones suppress Ig production in cultures of allogeneic T and B cells
stimulated by PWM (5, 6) and that, in the present study,
help is restored in cultures of SR T cell clones and autologous B cells
cultured with PWM by addition of Abs to suppressor cytokines (see
below). Another polyclonal B cell stimulant with a more defined
mechanism of action, CD40L-trimer, also induced non-T cells that
elicited self-MHC-reactive T cell clone suppressor activity. This was
shown best in the studies depicted in Fig. 2, in which non-T cells alone were
prestimulated with CD40L-trimer for 48 h, irradiated, and
cocultured with self-MHC-reactive T cells (SR-2 and SR-3) and fresh
autologous B cells for 8 days. The need for preculture in this instance
probably relates to the fact that CD40L require more time to induce
expression of stimulatory surface molecules on non-T cells than does
PWM (see below). Finally, as shown in Fig. 1, D
and E, in contrast to the above findings with the SR T cell
clones, the Ag-reactive T cell clones TT-1 and TT-2 did not induce
autologous B cells to produce Ig unless TT was added to the culture
and, perhaps more importantly, these clones did not significantly
suppress Ig production when PWM (or CD40 ligand-trimer) was added to
the culture.
|
|
, A–C,
we found that addition of anti-TGF-ß completely restored the
helper function of the self-MHC-reactive T cell clones SR-1, SR-2, and
SR-3 in cultures with non-T cells containing PWM. In contrast,
anti-IL-10 and anti-IFN- did not reverse the suppression,
and in fact, addition of anti-IL-10 (in cultures stimulated with TT
alone) led to decreased Ig production even in the absence of
suppression, possibly because IL-10 is necessary for B cell
differentiation. The addition of anti-IL-2 to cultures of the same
cells in the absence of PWM abolished the helper effect, indicating
that the helper function was IL-2 dependent. Similarly, as shown in
Fig. 2, addition of anti-TGF-ß to cultures of self-MHC-reactive T
cell clones stimulated by non-T cells pre-exposed to rCD40L-trimer also
abolished the suppressor activity of the clones, in this case with
respect to fresh non-T cells. Taken together, these studies show that
self-MHC-reactive, but not Ag-reactive, T cells act as suppressor cells
(with respect to B cell Ig production) in the presence of activated B
cells and that such suppression is mediated by the secretion of
TGF-ß.
In a final series of studies, we also determined whether B cells
are the target of the suppressive activity of the self-MHC-reactive T
cell clones. In these studies, culture supernatants were collected from
cocultures of self-MHC-reactive T cells and non-T cells prestimulated
with or without PWM and added to cultures of highly purified PBL B
cells or a B cell line (CL-1) (19) in which the B cells
were stimulated by SAC and IL-2. As shown in Fig. 3, the supernatant obtained from SR-1 or
SR-3 cells suppressed Ig synthesis by activated purified B cells,
indicating that the self-MHC-reactive SR T cell clones have a direct
suppressive effect on B cells. This conclusion was supported by the
fact that a similar suppressive effect was observed in cultures of CL-1
B cells (data not shown).
|
In further studies of the regulatory function of
self-MHC-reactive T cell clones, we determined whether these clones
could down-regulate T cell proliferation as well as B cell Ig
production. Because effects on the proliferation of indicator T cells
cannot be reliably measured in cocultures of regulatory T cells and
indicator T cells if the regulatory T cells must themselves be
activated (and thus induced to proliferate) to manifest regulatory
function, we determined whether supernatants obtained from cultures of
activated SR T cell clones could inhibit T cell proliferation in
separate cultures of various indicator T cells. This experimental
design was based on the observation that the stimulation of the SR T
cell clones with PWM-prestimulated non-Tx cells for optimal time
periods results in the secretion of cytokines with suppression activity
(see further discussion below). As shown in Fig. 4, A and B,
supernatants harvested from cultures of the self-MHC-reactive T cell
clone SR-l stimulated by PWM or CD40L-prestimulated autologous non-Tx
cells led to moderate (but highly reproducible) inhibition of
proliferation of anti-CD3/anti-CD28-stimulated allogeneic and
autologous CD4+ T cells, whereas the same cell
clone activated by PWM-prestimulated allogeneic non-Tx cells did not so
inhibit. Furthermore, as shown in Fig. 4, C and
D, the supernatants generated from cultures of the
self-MHC-reactive T cell clone SR-1 with the PWM or CD40L-prestimulated
autologous non-Tx cells also inhibited TT-induced proliferation of the
Ag-reactive T cell clone TT-1, and such inhibition was reversed by the
addition of anti-IL-10, anti-TGF-ß, or the combination of
anti-IL-10 and anti-TGF-ß to the culture. These data show
that self-MHC-reactive T cell clones also down-regulate T cell
responses, albeit not as completely as they down-regulate B cell
responses. In addition, they show that such down-regulation is mediated
via the secretion of both IL-10 and TGF-ß. In separate studies, we
found that the proliferation of the self-MHC-reactive T cell clone SR-1
was also decreased when cocultured with prestimulated autologous non-Tx
cells compared with proliferation when cocultured with unstimulated
non-T cells, suggesting that self-MHC-reactive T cells are also
sensitive to the autocrine TGF-ß and IL-10 production (data not
shown).
|
In further studies related to the mechanism of the suppressor
effect of self-MHC-reactive T cells, we determined the cytokine
production of such cells as well as the cellular interactions necessary
for such cytokine production. Initially, we determined the cytokines
produced by the various T cell clones after stimulation with the
polyclonal stimulants anti-CD3/anti-CD28 (see Materials
and Methods). As shown in Table III,
both of the Ag-reactive T cell clones TT-1 and TT-2 produced
substantial amounts of IL-2 and IFN-, but not IL-4. On the other
hand, although both of the self-MHC-reactive T cell clones, SR-1 and
SR-2, exposed to the same stimulants produced comparable amounts of
IFN-, they produced considerable amounts of IL-2 when stimulated
with non-Tx/PWM but not when stimulated with
anti-CD3/anti-CD28. Next, we determined cytokine production by
SR T cell clones and Ag-reactive T cell clones after stimulation with
non-Tx cells prestimulated by PWM or rCD40L, i.e., stimulants similar
to those encountered by self-MHC-reactive and Ag-reactive T cell clones
in cultures in which the former manifested suppressor activity. In
these studies, the length of time the non-Tx cells were prestimulated
before coculture with the T cell clones was varied from 0 to 24 h
(or in some cases from 0 to 48 h) to determine the time of peak
expression of stimulatory Ags on the surface of cells in the non-Tx
cell population. As shown in Fig. 5,
stimulation of Ag-reactive TT-1 cells with prestimulated non-Tx cells
led to only a slight increase in IL-10 or TGF-ß production, whereas
stimulation of self-reactive SR-1 and SR-2 cells with prestimulated
non-Tx cells led to substantial increases in both IL-10 and TGF-ß1
production, with the former peaking with non-Tx cells preactivated for
4 h and the latter peaking with non-Tx cells prestimulated for
8 h. Similar studies of cytokine production were conducted with
rCD40L-prestimulated non-Tx cells, but in this case, neither the Ag-
nor the self-MHC-reactive T cell clones produced IL-10 or TGF-ß1.
However, when we prestimulated non-Tx cells for 48 h with rCD40L,
the non-Tx cells now stimulated self-MHC-reactive T cell clones to
produce high levels of TGF-ß1 and IL-10 (data not shown).
|
|
, TT-1 and TT-2 secreted low levels of
TGF-ß1 (compared with SR-1, SR-2, and SR-3) when stimulated over a
wide range of TT concentrations that induce a level of proliferation
equal to or greater than exhibition by SR-1, SR-2, and SR-3 (compare
with Tables I and II). Thus, we conclude that the difference in
TGF-ß1/IL-10 between self-MHC-reactive and Ag-reactive clones is not
due to inadequate activation of the latter.
|
, cocultures of both the
self-MHC-reactive T cell clones SR-1 and SR-2 with PWM-stimulated non-T
cells as well as with purified B cells led to substantial TGF-ß1
secretion, whereas coculture with purified monocytes/M did not.
Therefore, these studies show that B cells themselves are capable of
inducing self-MHC-reactive T cells (20) to produce
TGF-ß1. Whether monocytes/M also have this property awaits studies
in which the M are stimulated with a range of known M
stimulants.
|
Expression of costimulatory molecules on the surface of non-T cells preactivated by PWM or rCD40L and on SR T cell clones activated by PWM or rCD40L
In the next set of studies, we focused on the cellular interactions between activated non-Tx cells and SR T cell clones that resulted in suppressor cytokine production. In this regard, it is known that PWM activates purified non-T cells (as well as purified B cells) (21, 22) so that the question arose as to whether such activation results in increased expression of B7-1 (CD80) and B7-2 (CD86) (23), which then facilitates interaction with T cells via CTLA-4, a recently described signaling pathway of TGF-ß production (24). In an initial exploration of this question, we performed flow cytometric analyses of non-Tx cells (particularly B cells) after preactivation with PWM or rCD40L (25) as well as parallel analyses of T cell clones stimulated with PWM or CD40L to determine the surface expression of B7 and CTLA-4 on these cells at various times after stimulation.
As shown in Fig. 8A, the
expression of B7-1 (CD80) or B7-2 (CD86) on CD19-positive B cells after
PWM stimulation was highest at 8 h after initiation of stimulation
and then declined to nearly baseline levels at 24 h after
initiation of stimulation. Thus, as shown in Fig. 8B, the
time of maximum coexpression of HLA-DR and B7-1/B7-2 corresponds to
time that non-Tx cells prestimulated with PWM are most capable of
inducing TGF-ß production. As shown in Fig. 8C, similar
studies conducted on rCD40L-stimulated non-T cells showed that in this
case expression of CD80 and, more prominently, expression of CD86
occurred more slowly and did not peak until 48 h after initiation
of stimulation. This finding explains why non-Tx cells preactivated
with rCD40L-trimer for only 8 h did not induce suppressor cytokine
production in SR T cell clones, whereas the addition of rCD40L-trimer
into longer-term cultures of SR T cells and non-Tx cells did lead to
suppression of PWM-stimulated Ig secretion.
|
and in data not shown, whereas before
stimulation none of the T cell clones expressed CTLA-4 with or without
non-T cell coculture (during 0–72 h), after 8 h of stimulation
with PWM, 24 h of stimulation with anti-CD3/anti-CD28, and
48 h of stimulation with CD40L, the self-MHC-reactive T cell
clones SR-1 and SR-2 induced CTLA-4 expression. On the other hand, the
Ag-reactive T cell clone TT-1 expressed CTLA-4 after stimulation with
anti-CD3/anti-CD28 after 24 h but did not express CTLA-4
after stimulation with PWM or rCD40L (in the absence of TT). In
contrast, CD28 was present on the cell surfaces of the
self-MHC-reactive T cell clones and Ag-specific clones before
stimulation and was not increased by stimulation (data not shown). We
also observed that the CD40L was not expressed on SR-1 and SR-2 cells
unless it was stimulated with PWM or with anti-CD3/anti-CD28
(data not shown) as previously reported (26).
|
Having established that PWM-preactivated B cells express B7-2 and
SR T cell clones express CTLA-4 at a time course consistent with
suppressor cytokine production, it remained for us to explore whether
the interaction between these costimulatory molecules was essential for
such cytokine production. Accordingly, we determined the ability of
PWM-prestimulated non-Tx cells to induce IL-10 and TGF-ß1 when these
cells were cocultured in the presence of soluble CTLA-4/Fc, a molecule
with a high binding affinity for both forms of B7, which therefore
blocks B7 interaction with cell-bound CTLA-4 or CD28. As shown in Fig. 10, whereas coculture of
PWM-prestimulated non-Tx cells with either of the SR T cell clones,
SR-l or SR-2, in the absence of CTLA-4/Fc again led to the production
of substantial amounts of TGF-ß1, coculture in the presence of
CTLA-4/Fc led to a marked inhibition of such production. However,
somewhat surprisingly, addition of CTLA-4/Fc to the coculture did not
inhibit production of IL-10. Finally, coculture of the Ag-reactive T
cell clone TT-1 with PWM-preactivated non-Tx cells or TT and non-T
cells led to only marginal TGF-ß1 production that was only slightly
affected by the addition of CTLA-4/Fc.
|
, we found that this blocking mAb
prevented TGF-ß1 production but had no effect on IL-10 production.
Furthermore, down-regulation of TGF-ß1 production by this mAb was
accompanied by a reciprocal up-regulation of IFN- production. In
further studies, we cultured self-MHC-reactive T cells (SR-1 and SR-2)
or Ag-reactive T cells (TT-1) in wells coated with anti-CTLA-4 mAb
(BNI3) and anti-CD3 to determine whether direct stimulation of
these clones via CTLA-4 leads to TGF-ß1 production. As shown in Fig. 12, CD28 induces marginal TGF-ß1
production in both types of T cell clones.
|
|
In summary, these CTLA-4 blocking and stimulation studies, taken in conjunction with the above-described data on B7 and CTLA-4 expression, establish that the induction of TGF-ß1 production by self-MHC-reactive T cells is at least in part dependent on costimulation via CTLA-4.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In our initial studies of the function of the self-MHC-reactive T cell
clones, we found that during or after culture with activated non-T
cells, these clones suppress both B cell Ig production and T cell
proliferation. With respect to B cell suppression, such suppression is
directly attributable to the secretion to TGF-ß, because inclusion of
anti-TGF-ß into the culture converts the SR T clones into helper
cells that support (help) Ig production. On the other hand, the
suppression of T cell proliferation is attributable not only to
TGF-ß1 but also to IL-10, a cytokine also secreted in large amounts
by the self-MHC-reactive T cells. These results with self-MHC-reactive
T cells contrast with those obtained with Ag-reactive T cells. Thus,
although both self-MHC-reactive and Ag-reactive T cells produce large
amounts of IFN- when activated by anti-CD3/CD28, Ag-reactive T
cells produce only modest (nonsuppressive) amounts of TGF-ß1 or IL-10
when stimulated by specific Ag in a manner analogous to
self-MHC-reactive T cell stimulation by non-T cells. This lack of
TGF-ß1 production cannot be attributed to lack of stimulation because
it was not seen at Ag concentrations inducing optimal proliferation,
i.e., proliferation that was equivalent to or exceeded that seen in the
stimulation of self-MHC-reactive T cells.
In further studies, we established that the suppressor function of self-MHC-reactive T cells, at least that because of TGF-ß secretion, was mediated by stimulation of the T cells via CTLA-4. This was shown by the fact that maximal TGF-ß secretion correlated with expression of B7 (particularly B7-2) and, more importantly, that TGF-ß secretion was blocked by the presence of a nonstimulatory F(ab')2 anti-CTLA-4 mAb. Contrariwise, cross-linking (stimulatory) anti-CTLA-4 Ab (but not anti-CD28 Ab) induced anti-CD3-activated self-MHC-reactive T cells to produce TGF-ß1 in the absence of activated B cells. It is of interest that IL-10 secretion by the SR T cell clones was not similarly related to CTLA-4 signaling, because IL-10 secretion was not blocked by CTLA-4/Fc or anti-CTLA-4 mAb. Recently, an inducible costimulator molecule termed ICOS has been identified that is involved in the superinduction of IL-10 (31). This costimulator molecule does not contain a structural motif that is involved in the CD28/CTLA-4 interaction with B7-1/B7-2 and therefore may be a discrete costimulator involved in the induction of IL-10 secretion by self-MHC-reactive T cells.
The role of the CTLA-4 costimulatory molecule in the negative
regulation of immune responses is now well-established
(32). It has been shown that cross-linking of this
molecule leads to suppression of IL-2 production and inhibition of cell
cycle progression, probably through the activation of a
CTLA-4-associated phosphatase, Src homology 2-containing tyrosine
phosphatase 2, which dephosphorylates TCR-
(33). On
this basis, costimulation of cells via this molecule is now thought to
be a prerequisite for the induction of T cell anergy and/or T cell
deletion. A second function of CTLA-4 in negative regulation is
inherent in recent studies and in the present study showing that
cross-linking of this molecule on T cells leads to TGF-ß production
(24, 34). Thus, in the present study we showed that
Ag-reactive T cells, when stimulated by anti-CD3 and cross-linking
anti-CTLA-4 Ab, produce increased TGF-ß1 even though they do not
produce TGF-ß1 under normal stimulation by APC and Ag. A somewhat
reversed situation occurs with self-MHC-reactive T cells, where
anti-CD3/anti-CTLA-4 results in increased TGF-ß1 production,
but such production is considerably lower than that obtained by
stimulation by self-Ag. Thus, the picture that emerges is that CTLA-4
provides signaling pathways for TGF-ß1 production but that this
pathway has a higher capacity to induce TGF-ß1 in self-MHC-reactive T
cells compared with Ag-reactive T cells (in our study, TT-specific T
cell clones).
In addition to its role as the initiator of a cellular program leading to suppressor cytokine production through interaction with CTLA-4, B7 costimulatory ligands may have a more fundamental role as activators of normally anergic SR T cells. As shown in our previous studies of the autoreactive T cell clone known as MTC-1 as well as in the present study of a similar clone, SR-1, the ability of the cell both to provide help for B cells and to suppress B cells is MHC-restricted, at least in the inductive phase, because neither of these functions is observed with respect to allogeneic B cells, even in the presence of PWM. Thus, there can be no question that the self-MHC-reactive T cell clones under study are in fact reacting with self-MHC, most likely a self-MHC containing a self-peptide in the MHC class II groove (30, 35). Recent studies of the origin of SR (autoreactive) T cells show that these cells recognize Ag with sufficiently low affinity to escape negative selection in the thymus. They are functionally anergic when released into the periphery and thus do not cause disease unless stimulated under particular conditions (36, 37). One such condition relative to the present findings may be the presence of cells expressing high levels of B7 that are thus capable of providing a potent costimulatory signal to T cells via CD28 (38). The latter possibility is suggested by studies showing that transgenic mice carrying a B7-2 transgene under the control of an insulin promotor (and thus expressing B7-2 in islet cells) ultimately develop a lymphoid infiltrate in the vicinity of the islets, whereas mice transgenic for B7-1 under the control of the same promotor do not. Furthermore, if the intensity of stimulation of the autoreactive cells, i.e., islet cell Ag-reactive cells, is increased by increasing their level of TCR signaling by the presence of a second transgene giving rise to MHC-class II expression in the islet cell, this distinction between B7-2 and B7-1 disappears (39). Thus, in this instance as well as in other instances (20), the level of B7 expression may determine whether an autoreactive clone will respond to a self-Ag.
The possible relation of self-MHC-reactive T cells producing suppressor
cytokines to autoimmune disease has recently been explored in several
animal models of autoimmunity (15, 16, 40). In one such
model in which Th2 T cell-mediated autoimmunity is elicited in Brown
Norway rats by administration of HgCl2, the T
cells inducing disease have been shown to be reactive to self-MHC class
II molecules and/or a self-peptide presented by the latter. In
contrast, treatment of Lewis rats with HgCl2 also
elicited SR T cells, but in this case the T cells prevented rather than
caused autoimmune disease when administered to rats undergoing
induction of experimental autoimmune encephalitis or
HgCl2-induced autoimmunity (40).
Although the autoreactive T cells mediating disease in the
HgCl2 autoimmunty model produced Th2 cytokines,
the autoreactive T cells preventing disease produce IFN- and high
levels of TGF-ß, and it is the latter cytokine that was responsible
for disease prevention because the administration of anti-TGF-ß
along with the autoreactive cells abrogated prevention
(16). These and similar studies in animal models of
diabetes establish that self-MHC-reactive T cells producing TGF-ß can
act as counterregulatory cells in autoimmune states
(15).
A second context in which SR T cells producing suppressor cytokines may relate to autoimmunity (and in this case inflammation as well) relates to recent findings concerning the regulatory activity of T cells arising in the mucosal immune system after oral Ag administration (11, 12, 41, 42). In these studies it has been shown that cells termed Th3 cells and producing high amounts of TGF-ß are generated by oral administration of Ag and that such cells are active in the prevention of experimental allergic encephalitis and other autoimmune states. In addition, it has been shown in the SCID adoptive transfer model of chronic colonic inflammation as well as in the chronic colonic inflammations developing in trinitrobenzene sulfonic acid-colitis or the colitis associated with IL-2 deficiency that cells producing TGF-ß can also prevent or even reverse the colitis (14, 43). Finally, it has been shown that T cell clones developed in the presence of IL-10, now termed Tr1 cells, produce IL-10 and/or TGF-ß and prevent colonic inflammation in SCID mice reconstituted with naive CD4+ T cells that otherwise induce colitis (17). Taken together, these studies draw attention to the possibility that counterregulatory cells controlling autoimmunity (15, 16, 39, 44) or inflammation of the gastrointestinal tract are actually self-MHC-reactive T cells. However, further work will be necessary to substantiate this possibility, including the isolation of regulatory cells occurring during (after) the induction of autoimmune inflammation of the specificity of such cells.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Abbreviations used in this paper: TT, tetanus toxoid; Tr1, T regulatory cell 1; CD40L, CD40 ligand; non-Tx cell, irradiated non-T cell; SR, self-reactive; M, macrophage; SAC, Staphylococcus aureus Cowan I bacteria.
Received for publication May 28, 1999. Accepted for publication April 24, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  G. Lizee, L. G. Radvanyi, W. W. Overwijk, and P. Hwu Improving Antitumor Immune Responses by Circumventing Immunoregulatory Cells and Mechanisms. Clin. Cancer Res., August 15, 2006; 12(16): 4794 - 4803. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Shao, A. R. Jacobs, V. V. Johnson, and L. Mayer Activation of CD8+ Regulatory T Cells by Human Placental Trophoblasts J. Immunol., June 15, 2005; 174(12): 7539 - 7547. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Chen, S. S. Yan, J. Colgan, H.-P. Zhang, J. Luban, A. M. Schmidt, D. Stern, and K. C. Herold Blockade of Late Stages of Autoimmune Diabetes by Inhibition of the Receptor for Advanced Glycation End Products J. Immunol., July 15, 2004; 173(2): 1399 - 1405. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Veldman, A. Hohne, D. Dieckmann, G. Schuler, and M. Hertl Type I Regulatory T Cells Specific for Desmoglein 3 Are More Frequently Detected in Healthy Individuals than in Patients with Pemphigus Vulgaris J. Immunol., May 15, 2004; 172(10): 6468 - 6475. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. M. Jovasevic, L. Gorelik, J. A. Bluestone, and M. B. Mokyr Importance of IL-10 for CTLA-4-Mediated Inhibition of Tumor-Eradicating Immunity J. Immunol., February 1, 2004; 172(3): 1449 - 1454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Monsonego and H. L. Weiner Immunotherapeutic Approaches to Alzheimer's Disease Science, October 31, 2003; 302(5646): 834 - 838. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Kitani, I. Fuss, K. Nakamura, F. Kumaki, T. Usui, and W. Strober Transforming Growth Factor (TGF)-{beta}1-producing Regulatory T Cells Induce Smad-mediated Interleukin 10 Secretion That Facilitates Coordinated Immunoregulatory Activity and Amelioration of TGF-{beta}1-mediated Fibrosis J. Exp. Med., October 20, 2003; 198(8): 1179 - 1188. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Bommireddy, V. Saxena, I. Ormsby, M. Yin, G. P. Boivin, G. F. Babcock, R. R. Singh, and T. Doetschman TGF-{beta}1 Regulates Lymphocyte Homeostasis by Preventing Activation and Subsequent Apoptosis of Peripheral Lymphocytes J. Immunol., May 1, 2003; 170(9): 4612 - 4622. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Bommireddy, I. Ormsby, M. Yin, G. P. Boivin, G. F. Babcock, and T. Doetschman TGF{beta}1 Inhibits Ca2+-Calcineurin-Mediated Activation in Thymocytes J. Immunol., April 1, 2003; 170(7): 3645 - 3652. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. R. Singh, F. M. Ebling, D. A. Albuquerque, V. Saxena, V. Kumar, E. H. Giannini, T. N. Marion, F. D. Finkelman, and B. H. Hahn Induction of Autoantibody Production Is Limited in Nonautoimmune Mice J. Immunol., July 1, 2002; 169(1): 587 - 594. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. K. U. Koehler, C. P. Genain, B. Giesser, and S. L. Hauser The Human T Cell Response to Myelin Oligodendrocyte Glycoprotein: A Multiple Sclerosis Family-Based Study J. Immunol., June 1, 2002; 168(11): 5920 - 5927. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. P. Gray, P. Arosio, and P. Hersey Heavy chain ferritin activates regulatory T cells by induction of changes in dendritic cells Blood, May 1, 2002; 99(9): 3326 - 3334. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Saverino, A. Merlo, S. Bruno, V. Pistoia, C. E. Grossi, and E. Ciccone Dual Effect of CD85/Leukocyte Ig-Like Receptor-1/Ig-Like Transcript 2 and CD152 (CTLA-4) on Cytokine Production by Antigen-Stimulated Human T Cells J. Immunol., January 1, 2002; 168(1): 207 - 215. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Nakamura, A. Kitani, and W. Strober Cell Contact-dependent Immunosuppression by CD4+CD25+ Regulatory T Cells Is Mediated by Cell Surface-bound Transforming Growth Factor {beta} J. Exp. Med., September 4, 2001; 194(5): 629 - 644. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. Levings, R. Sangregorio, and M.-G. Roncarolo Human CD25+CD4+ T Regulatory Cells Suppress Naive and Memory T Cell Proliferation and Can Be Expanded In Vitro without Loss of Function J. Exp. Med., June 4, 2001; 193(11): 1295 - 1302. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. K. Levings, R. Sangregorio, F. Galbiati, S. Squadrone, R. de Waal Malefyt, and M.-G. Roncarolo IFN-{{alpha}} and IL-10 Induce the Differentiation of Human Type 1 T Regulatory Cells J. Immunol., May 1, 2001; 166(9): 5530 - 5539. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Sullivan, J. J. Letterio, A. van Elsas, M. Mamura, J. van Amelsfort, S. Sharpe, B. Metzler, C. A. Chambers, and J. P. Allison Lack of a role for transforming growth factor-beta in cytotoxic T lymphocyte antigen-4-mediated inhibition of T cell activation PNAS, February 15, 2001; (2001) 51632398. [Abstract] [Full Text]
|
|
|
|
|
|
T. J. Sullivan, J. J. Letterio, A. van Elsas, M. Mamura, J. van Amelsfort, S. Sharpe, B. Metzler, C. A. Chambers, and J. P. Allison Lack of a role for transforming growth factor-beta in cytotoxic T lymphocyte antigen-4-mediated inhibition of T cell activation PNAS, February 27, 2001; 98(5): 2587 - 2592. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
