| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Department of Pathology, Committee on Immunology, and
Department of Medicine, University of Chicago, Chicago, IL 60637;
Department of Immunology, University of Washington, Seattle, WA 98195; and
§
Genetics Institute, Cambridge, MA 02140
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-producing phenotype in the presence of IL-12.
These results indicate that signals generated by B7.1 and B7.2 are
qualitatively similar, but that B7.1 is quantitatively stronger than
B7.2. Further, our results indicate that the activation state of the
responding T cell may influence the efficiency with which the T cell
can respond to a costimulatory signal provided by either B7.1 or
B7.2. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Whether B7.1 and B7.2 can costimulate T cells equivalently has remained controversial. Differences in the ability of these molecules to augment T cell responses have been observed in several model systems. Some published studies, involving either human or murine models, have suggested that B7.1 can provide a stronger costimulus than B7.2 for T cell activation, as measured by proliferation, lymphokine production, induction of cytolysis, and rejection of certain tumors (7, 8, 9, 10). In addition, costimulation with B7.2 has been reported to preferentially induce differentiation of T cells into an IL-4-producing or T2-like response, suggesting a qualitative difference between the T cell signals generated by costimulation with either B7.1 or B7.2 (11, 12). In contrast, other studies have reported that these molecules appear to be functionally equivalent (13, 14). These disparate findings have been difficult to reconcile using normal lymphocyte populations.
Given the importance of B7.1 and B7.2 in regulating immune responses, a clearer understanding of the capabilities of each in directing T cell responses is critical. It seems likely that differing results have been obtained by various investigators because different model systems have been used. We postulate that at least one factor contributing to the discrepant results of previous studies is the use of mixed populations of responding T cells because naive and primed T cells may differ in their requirements for B7 costimulation (15, 16). To explore these hypotheses, we have developed a model system for examining the requirements for B7 costimulation of naive and previously stimulated CD8+ T cells in vitro. A monoclonal T cell population expressing a naive surface phenotype was obtained by breeding a TCR transgenic mouse with recombinase-activating gene-2-deficient (RAG-2-/-)3 mice. Strict control over the B7 costimulus was achieved by generating transfectants expressing defined levels of B7.1 or B7.2 or by using B7.1-Ig or B7.2-Ig fusion proteins at defined concentrations. This system was used with the intention of 1) allowing strict control over the type and amount of B7 molecule presented to the T cells; 2) ensuring that when initially stimulated, the responding T cell population had not previously encountered a TCR stimulus in the periphery, and thus was truly naive; and 3) minimizing and controlling the potential influence of adhesion or other accessory molecules on the abilities of the B7 costimulatory molecules to drive T cell responses. We attempted to answer three questions. 1) Are there differences between B7.1 and B7.2 in the costimulation of primary CD8+ T cell responses? 2) If so, are the differences quantitative or qualitative in nature? 3) Do naive T cells and T cells previously stimulated with Ag differ in their requirements for B7 costimulation?
Using this model system, we found quantitative differences in the abilities of B7.1 and B7.2 to costimulate CD8+ T cell responses. In naive cells, B7.1 provided a stronger costimulus than B7.2 for proliferation and IL-2 production. The addition of exogenous IL-2 to B7.2-stimulated T cells increased their proliferative response, suggesting that the difference in the abilities of B7.1 and B7.2 to induce proliferation reflects a difference in their respective abilities to induce IL-2 production by primary T cells. In contrast to those of naive cells, we found that the responses of primed T cells to costimulation by B7.1 and B7.2 were quantitatively similar. In addition, primed T cells proliferated and produced IL-2 in response to a TCR stimulus alone in the absence of B7 costimulation. Finally, both B7.1 and B7.2 were capable of supporting the differentiation of naive CD8+ T cells into an IL-4-producing, Tc2 phenotype, arguing against a qualitative difference between the fundamental effects of these two costimulatory molecules.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
All mice were housed in the University of Chicago Barrier
Animal Facility under pathogen-free conditions. The 2C transgenic mice
were obtained from Dr. Dennis Loh (17). These mice are haplotype
H-2b, and 2C T cells are reactive against the
H-2Ld alloantigen complexed with an octapeptide, p2Ca, from
-ketoglutarate dehydrogenase. The RAG-2-/- mice were a
gift from Dr. Celeste Simon (University of Chicago, Chicago, IL) (18).
The 2C mice were bred onto a RAG-2-/- background, and the
progeny were screened for expression of the 2C transgene and lack of B
cells by immunofluorescence using FITC-coupled 1B2 (anti-2C
clonotype, D. Loh) and phycoerythrin-coupled DS-1 (anti-IgM,
PharMingen, San Diego, CA) Abs. Splenic T cells from these 2C x
RAG-/- mice (2C/RAG-/-) were purified as
described below and used in experiments. FITC- or phycoerythrin-coupled
Abs against murine ICAM-1, LFA-1, H-2Ld, CD4, CD8, CD80,
CD86, CD44, CD62L, and IgM used for FACS analysis of P815 transfectants
and T cells were purchased from PharMingen. The hybridoma-producing
anti-CD3
mAb, 145-2C11, was a gift from Dr. Jeffrey Bluestone
(University of Chicago). The Ab was purified from culture supernatant
using a protein A-agarose column (Pharmacia Biotech, Piscataway, NJ).
Cell lines producing anti-CD11c mAb, M1/70, and
anti-I-Ab mAb, 25-9-3s, were purchased from American
Type Culture Collection (Manassas, VA). These Abs were used in T cell
purification as culture supernatant. The H-2Ld-expressing
mouse mastocytoma cell line, P815, was transfected with B7.1 or B7.2
and was used as a stimulator cell line (19). Recombinant human IL-2 was
a gift from Cetus Oncology Corp. (Emeryville, CA). Recombinant murine
IL-4 was a gift from Immunex Corp. (Seattle, WA). rIL-12 was a gift
from Genetics Institute (Cambridge, MA).
P815 transfectants
cDNA constructs containing the sequences for murine B7.1 and B7.2, which were originally isolated by Dr. Gordon Freeman, were obtained from Dr. J. Bluestone (20, 21). Murine B7.1 was cloned into the eukaryotic vector pNA', a modified form of pHßAPr-neo that contains the human ß-actin promoter and confers resistance to neomycin (22, 23). The cDNA sequence was originally removed as an EcoRI fragment from pBSKS2-/B71 and inserted into the EcoRI site of pcDNA3 (Invitrogen, San Diego, CA). A subclone containing the B7.1 sequence in the correct orientation was selected; B7.1 was removed with a KpnI, XbaI digest and ligated into pNA' that had been digested with KpnI and XbaI. The resulting subclones were screened for correct orientation in relation to the ß-actin promoter, and one subclone was selected for transfection into P815. The cDNA sequence for murine B7.2 was removed as an XbaI fragment from pcDNA3/B7-2 and inserted into the XbaI site of pHA', a modified form of pNA' that confers resistance to hygromycin (22). A subclone containing B7.2 in correct orientation in relation to the ß-actin promoter was selected for transfection into P815.
pNA'/B7.1 was linearized with ScaI, and pHA'/B7.2 was linearized with XmnI. Both constructs were transfected into P815 cells by electroporation. pNA'/B7.1 transfectants were selected on 1 mg/ml G418 (Gemini Bioproducts, Calabasas, CA). pHA'/B7.2 transfectants were selected on 0.5 mg/ml hygromycin B (Boehringer Mannheim, Indianapolis, IN). Resulting antibiotic-resistant cells were screened for B7.1 and B7.2, and bulk populations of transfected cells were sorted for different levels of expression by FACS using FITC-coupled anti-B7.1 and anti-B7.2 (1G10 and GL-1, respectively; PharMingen). Surface expression of B7.1 and B7.2 on subclones expressing varying levels of the molecule was determined by immunofluorescence using saturating amounts of FITC-coupled murine CTLA-4 Ig (provided by Dr. Jeffrey Bluestone).
Preparation of B7-Ig fusion proteins, B7.1-Ig and B7.2-Ig
Murine B7.1-Ig and murine B7.2-Ig expression plasmids were constructed by joining the DNA encoding the signal and extracellular domains of murine B7.1 or B7.2 to the DNA encoding the hinge CH2-CH3 domains derived from a murine IgG2a Ab. The Ab hinge cysteines remained intact such that the expressed B7.1-Ig or B7.2-Ig was dimeric.
Recombinant Chinese hamster ovary (CHO) cell lines, 74-18/0.02 and 73-15/0.02, expressing B7.1-Ig and B7.2-Ig, respectively, were grown in DMEM/Ham’s F-12 medium containing 10% FBS, 0.02 mM methotrexate, and 1.0 mg/ml G418. The cell lines were grown to confluence in roller bottles, and growth medium was replaced with serum-free secretion medium. After 24 h, the secretion medium was removed and clarified through 0.22-µm filters. This medium containing B7.1-Ig or B7.2-Ig was passed over a recombinant protein A-Sepharose Fast Flow column (Pharmacia). The column was eluted with 20 mM citrate, pH 3.0. Concentrations of B7.1-Ig and B7.2-Ig were determined by spectrophotometry (280 nm), and relative concentrations, m.w., and purity of the fusion protein preparations were confirmed with SDS-PAGE.
T cell purification and stimulation
Splenic 2C/RAG-/- T cells used in these studies were purified using a two-step, negative selection protocol. First, spleens from 2C/RAG-/- mice were macerated and washed once in DMEM supplemented with 10% FCS (DMEM-10; Life Technologies, Gaithersburg, MD). Large pieces of debris were removed by filtering the suspension through sterile Nitex mesh (Tetko, Briarcliff Manor, NY). Adherent cells were then removed by adherence to tissue culture dishes (Costar, Cambridge, MA) at 37°C for 2 h. After the plate adherence step, the cells were harvested and washed once with DMEM-10. The harvested cells were incubated with a mixture of anti-CD4 (RL1724.4), anti-MAC-1, and anti-I-Ab at 4°C for 20 min. The cells were washed once in DMEM-10 and resuspended in DMEM without FCS (10 ml/spleen). Rabbit complement (Pel-Freeze, Brown Deer, WI) was added to the suspension at a final dilution of 1/10. This mixture was incubated at 37°C for 45 min. Dead cells were then removed by centrifugation over Ficoll-Hypaque, and the purity of the live T cells was determined by immunofluorescence using anti-TCR (1B2) and anti-CD8 (24). Purified T cells were counted and used in experiments.
In some experiments, previously stimulated T cells were studied. These effector T cell populations were generated by in vitro stimulation of purified primary 2C/RAG-/- T cells with P815/B7.1-H. Cultures were initiated by stimulating 5 x 104 purified splenic CD8+ T cells (2C/RAG-/-) with 3.5 x 105 stimulator cells (P815/B7.1-H) in 24-well tissue culture dishes (Costar) in the presence or the absence of exogenous cytokines. After 5–7 days in culture, the T cells were removed from culture wells and purified by centrifugation over Ficoll-Hypaque for use in experiments.
Proliferation assays and detection of lymphokine production
Purified T cells were stimulated with P815, P815/B7
transfectants, or plate-bound anti-CD3 mAb in the presence or the
absence of immobilized B7.1-Ig or B7.2-Ig. For stimulation with P815
and P815/B7 transfectants, 5 x 103 purified T cells
were stimulated in 96-well round-bottom plates (ICN Biomedicals,
Aurora, OH) with varying numbers of stimulator cells that had been
treated previously with mitomycin C (50 µg/ml/107 cells
for 90 min at 37°C; Sigma, St. Louis, MO) to inactivate the cells. At
various times after the initiation of cultures, 0.5 µCi of
[3H]thymidine was added to the cells in microwells for
8 h, at which time the plates were frozen and thawed, and the
wells were harvested for determination of [3H]thymidine
incorporation using a Packard cell harvester and plate reader (Packard,
Meriden, CT). For lymphokine production, 5 x 104
purified T cells were stimulated with 3.5 x 105 P815
cells or P815/B7 transfectants in 24-well plates (ICN Biomedicals).
Supernatant was removed at 24 and 48 h to determine levels of IL-2
and IFN-, respectively.
For stimulation of T cells with anti-CD3 mAb and B7-Ig fusion
protein, 96-well flat-bottom tissue culture plates (Costar) were
precoated for 2 h at 37°C with varying concentrations of
anti-CD3 and/or B7-Ig or control Ig (murine IgG2a; Zymed, South San
Francisco, CA) in Dulbecco’s PBS (Life Technologies) in a final volume
of 50 µl/well. After washing the wells twice with Dulbecco’s PBS,
2.5 x 104 purified T cells were added to each well,
and the plates were centrifuged for 2 min at 500 x g
and placed in the incubator at 37°C. Supernatants were removed at 24
and 48 h for determination of IL-2 and IFN-, respectively. The
amounts of these lymphokines in the supernatants were determined by
ELISA (Endogen, Cambridge, MA).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To generate a model system in which all T cells had a naive
phenotype and expressed the same TCR, 2C transgenic mice were bred to
RAG-2 -/- mice. T cells from the 2C/RAG-/-
mice express a transgenic TCR specific for the -ketoglutarate
dehydrogenase octapeptide, p2Ca; this TCR is detected by the clonotypic
mAb, 1B2. The CD3+ cells from these mice were 100%
1B2+, confirming that the T cells both expressed the
transgene and did not express detectable endogenously rearranged TCR
genes (data not shown). Also, the 2C/RAG-/- splenocytes
lacked IgM+ cells, indicating that the animals were RAG-2
deficient (data not shown). After purification, the T cells used in
experiments were >98% CD8+1B2+ by
immunofluorescence analysis (Fig. 1
A). Because the T cells of
the mice were monoclonal and specific for an alloantigen, it was very
unlikely that they had encountered a TCR stimulus in the periphery.
Indeed, the splenic T cells from these mice were all
CD44low and CD62Lhigh, consistent with a naive
phenotype (Fig. 1, B and C). Stimulation of these
cells with the alloantigen-bearing cell line, P815, for 24 h in
vitro induced an increase in CD44 and a decrease in CD62L expression
(Fig. 1, B and C).
|
D). Subclones of B7.1- and
B7.2-transfected lines binding comparable amounts of CTLA-4 Ig were
selected to use as stimulator lines. Two B7.1 transfectants and two
B7.2 transfectants were selected that expressed either high or low
levels of the molecule on the cell surface (P815/B7.1-H and
P815/B7.1-L, and P815/B7.2-H and P815/B7.2-L, respectively).
P815/B7.1-L and P815/B7.2-L expressed comparable levels of CTLA-4 Ig
binding at the cell surface (Fig. 1, E and F).
Likewise, P815/B7.1-H and P815/B7.2-H expressed comparable, yet higher,
levels of CTLA-4 Ig binding at the cell surface (Fig. 1, E
and F). The parental P815 and all P815 transfectants
expressed comparable levels of the alloantigen, H-2Ld, as
well as of ICAM-1 and LFA-1, as determined by immunofluorescence
analysis (data not shown). Because these cell lines differed only in
the type and amount of B7 molecule expressed at the cell surface, a
quantitative assessment of the relative capacities of B7.1 and B7.2 to
stimulate primary CD8+ T cells could be made. Primary stimulation of naive 2C/RAG-/- T cells
To determine whether B7.1 or B7.2 could equivalently costimulate
primary T cells, purified naive 2C/RAG-/- T cells were
cultured with either P815 or the transfectants expressing high or low
levels of B7.1 and B7.2. Naive 2C/RAG-/- T cells
stimulated with P815 did not proliferate appreciably (Fig. 2A). In contrast, the T cells
did proliferate when B7 was expressed on the stimulator cells. However,
quantitative differences were observed between B7.1 and B7.2 (Fig. 2, A and B). Although P815/B7.1-L and P815/B7.1-H
cells stimulated comparable vigorous levels of proliferation, the B7.2
transfectants were consistently inferior. In fact, stimulation with
P815/B7.2-L cells resulted in levels of proliferation that were little
above background in some experiments. Induction of IL-2 production
paralleled proliferation, with the P815/B7.2-L transfectant stimulating
marginally greater IL-2 secretion than the wild-type P815 cells (Fig. 2C). Induction of IFN- production by the T cells was
slightly different. Both the B7.1 transfectants induced greater IFN-
production than transfectants expressing comparable levels of B7.2;
however, the differences were only around twofold (Fig. 2D).
Collectively, these data suggest that B7.1 provides a quantitatively
stronger costimulus than does B7.2 for IL-2 production and
proliferation. Although B7.1 appeared to be more potent than B7.2 for
inducing IFN- production, the differences were only about twofold.
|
). The inferior capacity of B7.2
transfectants to induce proliferation appeared to correlate with their
decreased ability to induce IL-2 production. To test this hypothesis,
we added exogenous IL-2 to cultures of T cells stimulated with
P815/B7.2-L. Indeed, when exogenous IL-2 was added to
P815/B7.2-L-stimulated T cells, the proliferative response of the T
cells was increased to a level approaching that induced by P815/B7.1-L
(Fig. 3).
|
Because the responses of the T cells to Ag and B7 costimulation might be influenced by other molecules on the surface of the stimulator cell, 2C/RAG-/- T cells were stimulated with the anti-CD3 mAb, 145-2C11, with or without B7.1-Ig or B7.2-Ig fusion proteins in an APC-free system. This system enabled us to compare B7.1 and B7.2 without potential effects of other molecules expressed on the surface of the stimulator cell line. Moreover, it allowed us to manipulate independently the intensity of the TCR/CD3 stimulus and the B7 costimulus.
Preliminary studies revealed that immobilization of the B7 fusion
proteins was necessary to costimulate the T cells, because the addition
of the B7 fusion proteins in soluble form caused no augmentation of
IL-2 production or proliferation (data not shown). Stimulation of naive
2C/RAG-/- T cells with anti-CD3 mAb alone induced a
proliferative response (Fig. 4A) and very little detectable
IL-2 (Fig. 4B), but appreciable amounts of IFN- (Fig. 4C). The addition of either B7.1-Ig or B7.2-Ig increased all
three responses in a dose-dependent fashion. However, B7.1-Ig was more
potent than B7.2-Ig for induction of IL-2 production and proliferation
when added at a concentration of 20 µg/ml (Fig. 4, A and
B). Although IFN- production increased with the addition
of either B7.1-Ig or B7.2-Ig, there was little difference between the
two (Fig. 4C).
|
D). Anti-CD3
alone induced no detectable IL-2 (Fig. 4E) and low levels of
IFN- (Fig. 4F), both of which were increased with the
addition of either B7.1-Ig or B7.2-Ig. As observed in the previous
experiments, B7.1-Ig induced significantly greater proliferation and
IL-2 production than B7.2-Ig, whereas differences in IFN- induction
were negligible (Fig. 4, D–F).
Stimulation with anti-CD3 mAb and B7-Ig allowed determination of
whether a combination of B7.1 and B7.2 would be additive or
synergistic. From this information, inferences about the nature of the
signals generated by these two stimuli could be made. We reasoned that
if the two molecules were synergistic in their ability to costimulate T
cells, then the signals generated by the two molecules would surely be
different. However, if the two were additive in their effects, this
would suggest that B7.1 and B7.2 generate similar signals.
2C/RAG-/- T cells were stimulated with suboptimal
concentrations of anti-CD3 mAb plus either B7.1-Ig or B7.2-Ig at
concentrations that gave similar functional responses. The two
molecules were additive in their ability to costimulate proliferation
and IL-2 production at the anti-CD3 mAb concentrations tested
(Table I). This suggests that the
costimulatory signals for proliferation and IL-2 production generated
by B7.1 and B7.2 are similar. A corollary to this is that the
differences observed between B7.1 and B7.2 are quantitative in nature,
owing to an inferior ability of B7.2 to costimulate proliferation and
IL-2 production by primary CD8+ T cells.
|
Naive and primed T cells may have different requirements for B7
costimulation. To compare the responses of naive T cells with those of
previously stimulated cells, 2C/RAG-/- T cells were first
stimulated with P815/B7.1-H for 6 days. T cells primed in this fashion
(henceforth called secondary cells) were then restimulated with P815 or
P815 transfected with B7.1 or B7.2. In contrast to naive T cells (Fig. 5A), secondary cells (Fig. 5B) had increased proliferative responses to both B7.1 and
B7.2 transfectants. In addition, whereas B7.1 induced a significantly
greater proliferative response than B7.2 by naive T cells, the
secondary cells proliferated equivalently to P815/B7.1-L and
P815/B7.2-L (Fig. 5B). Strikingly, while wild-type P815
induced no proliferation of naive cells, the secondary cells
proliferated significantly in response to P815, apparently in the
absence of B7 costimulation.
|
C) or anti-CD3 mAb and B7-Ig fusion proteins (Fig. 6, A and B). In
addition, the difference in IL-2 production induced by B7.1 vs B7.2
costimulation in naive cells was much less evident in secondary cells
(Figs. 5C and 6, A and B). Also, in
sharp contrast to naive cells, the secondary cells produced significant
quantities of IL-2 in response to P815 or to anti-CD3 mAb alone,
even without B7 costimulation (Figs. 5C and 6, A
and B). Similar results upon secondary stimulation were
observed when the primary stimulation was performed with P815/B7.2-H
(data not shown). Neither proliferation nor IL-2 production by
secondary cells induced by parental P815 could be blocked with CTLA-4
Ig, suggesting that activation of secondary T cells by P815 is truly B7
independent (data not shown). These results suggest that primed
(secondary) T cells have a relaxed requirement for B7 costimulation,
such that TCR ligation alone is sufficient to induce a substantial
response, which is augmented comparably by either B7.1 or the weaker
ligand, B7.2.
|
production (Fig. 7, A and B). T
cells stimulated in the primary culture with B7.1, with or without the
addition of exogenous IL-4 or IL-12, were restimulated in secondary
cultures with B7.1 transfectants without the addition of any exogenous
cytokines. Likewise, T cells initially stimulated with B7.2
transfectants were restimulated with B7.2 transfectants without the
addition of exogenous cytokines. We found that CD8+ T cells
primed with either B7.1 or B7.2 in the absence of exogenous cytokines
or in the presence of exogenous IL-12 produced only background levels
of IL-4 upon restimulation (Fig. 7A). In contrast, T cells
stimulated with either B7.1 or B7.2 transfectants in the presence of
exogenous IL-4 produced increased amounts of IL-4 at similar levels
upon restimulation (Fig. 7A). These data indicate that B7.1
and B7.2 were equivalent in their ability to prime naive T cells for
IL-4 production when the T cells were stimulated in the presence of
exogenous IL-4.
|
upon restimulation of T cells stimulated in the
primary culture with IL-12 was greater than that of T cells
stimulated in the primary culture in the presence of IL-4 or in
the absence of exogenous cytokines (Fig. 7B). This pattern
of IFN- production was observed regardless of whether primary
stimulation was conducted with B7.1 or B7.2 transfectants. From these
observations, we conclude that both B7.1 and B7.2 can support the
differentiation of naive CD8+ T cells toward an
IL-4-producing or IFN--producing phenotype. Collectively, our
results indicate that B7.1 and B7.2 are quantitatively, rather than
qualitatively, different in their ability to costimulate T cells, and
that this quantitative difference is particularly evident when a naive
T cell population is studied. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In the human system, B7.1 binds to CD28 and CTLA-4 more avidly and with a slower off-rate compared with B7.2 (28). B7.2, which is a lower avidity ligand, would be expected to have a shorter occupancy time, resulting in the reduced probability of making a productive interaction with the receptor, thus resulting in a reduced capacity to stimulate the cell through that receptor. Several lines of evidence suggest that this is the case with B7/CD28 interactions. One recent study showed that B7.1 was significantly more effective than B7.2 at inducing the down-regulation of CD28 expression upon engagement of CD28 with comparable amounts of either ligand (29). While the functional importance of this phenomenon is unknown, the observation demonstrates a difference in a very early event that might depend on the strength of the interaction between the ligand and its receptor. In other studies using B7 fusion proteins that were mutated to affect their binding affinity to CD28, a direct correlation between strength of adhesion/binding and the ability to promote IL-2 production by T cells was demonstrated (30).
The strength or quality of the TCR signal might also influence the response of T cells to B7.1 and B7.2. Quantitative differences in B7.1 and B7.2 in inducing both proliferation and lymphokine production by T cells have been reported when the T cells were stimulated with a low-affinity weak agonist peptide, but not with a higher-affinity agonist peptide (31). These differences were attributed to the response of the T cells to the different affinity TCR ligands, a higher-strength TCR stimulus masking differences in the abilities of the two costimulatory molecules to induce T cell responses. In our present study, when the strength of TCR ligation was increased in the anti-CD3 mAb plus B7-Ig stimulation of T cells, we did not observe any diminishment of differences between B7.1 and B7.2. This may reflect a difference between stimulation with Ag/APC vs immobilized purified ligands.
Another determinant of the apparent strength with which B7 molecules influence T cell activation is the state of the responding T cell itself. As reported above, naive T cells respond differently to B7.1 and B7.2 costimulation, whereas activated T cells respond to the two molecules equivalently. It is notable that the primed T cells produced IL-2 and proliferated in response to a TCR stimulus alone, in the absence of B7. The transition of naive T cells to a primed phenotype apparently enables them to respond to subsequent TCR stimulation in a B7-independent manner. This may reflect an increased sensitivity of primed T cells to TCR ligation. In a recent study it was shown that CD44high, Ly-6C+ (memory) CD8+ T cells were more sensitive than naive T cells (CD44low, Ly-6C-) to stimulation via the TCR, corroborating our findings (32). While our studies suggest that TCR ligation alone is sufficient for activation of primed T cells, it is formally possible that other CD28-independent molecules up-regulated on the T cell during priming could play a role in T cell activation via T-T contact.
We found that the effects of B7.1 and B7.2 on proliferation and IL-2
production were additive, suggesting that the signals generated by
stimulation with the two molecules necessary for these responses are
similar, consistent with quantitative, rather than qualitative,
differences in B7.1- vs B7.2-induced cellular events. A clearer
understanding of the early biochemical events induced by B7
costimulation will be necessary to definitively resolve this issue. In
contrast with the findings of proliferation and IL-2 production,
differences between B7.1 and B7.2 were less evident when IFN-
production by the T cells was examined. B7.1- and B7.2-generated
signals necessary for IL-2 production may be different from those that
influence IFN- production, but the signaling threshold required for
IFN- production by the T cell also may be lower than that required
for IL-2 production. We have observed a similar phenomenon previously
in anergic T cells (33). These and other data suggest that a hierarchy
of signaling thresholds needed to be crossed to allow the transcription
of different lymphokine genes (34). Because the quantity of signal
required for IFN- may be less than that required for IL-2
production, costimulation provided by low levels of either B7.1 or B7.2
may be sufficient to enable maximum or near-maximum IFN- gene
transcription.
Previous studies reported differences in the abilities of B7.1 and B7.2 to drive differentiation of T cells into lymphokine-producing subsets. Naive T cells serially stimulated with B7.2-transfected stimulator cell lines were found to produce much more IL-4 after several rounds of stimulation (11). In our present study we found no difference after one restimulation. In fact, in our system neither B7.1 nor B7.2 costimulation induced differentiation of naive CD8+ T cells into an IL-4-producing phenotype unless exogenous IL-4 was included in the primary stimulation. An alternative interpretation of the previous reports may relate to the differentiation state of the starting T cell population. Because naive T cells respond less well to B7.2 than to B7.1 costimulation, whereas activated T cells respond to the two equivalently, B7.2 costimulation might preferentially result in the expansion of previously activated T cells. If such previously activated T cells produce some IL-4, then preferential expansion of this IL-4-producing population might result in an apparent skewing of the T cell response toward a Th1/Th2 phenotype by selection. Because in our system very little IL-4 was produced in the absence of exogenous IL-4, no such apparent skewing was observed.
Our findings indicate that the differences in the abilities of B7.1 and B7.2 to costimulate naive CD8+ T cells probably result from quantitative, rather than qualitative, differences in stimulation with these molecules. However, the activation state of the responding T cell has a profound effect on the ability of T cells to respond, such that primed T cells respond to B7.1 and B7.2 similarly. It appears that an activation-induced "resetting" of TCR stimulation and costimulation thresholds occurs in the T cells. An understanding of these events at a molecular level could have implications for understanding the generation of memory T cell responses.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Frank W. Fitch, University of Chicago, 5841 S. Maryland Ave., MC1089, Chicago, IL 60637. E-mail address:
3 Abbreviations used in this paper: RAG-2-/-, recombinase-activating gene-2-deficient; DMEM-10, DMEM supplemented with 10% FCS.
Received for publication May 1, 1998. Accepted for publication July 13, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
chains mediate the signal for T cell proliferation. Nature 369:333.[Medline]
This article has been cited by other articles:
|
|
 |
|
  I. J. Thomas, L. G. Petrich de Marquesini, R. Ravanan, R. M. Smith, S. Guerder, R. A. Flavell, D. C. Wraith, L. Wen, and F. S. Wong CD86 Has Sustained Costimulatory Effects on CD8 T Cells J. Immunol., November 1, 2007; 179(9): 5936 - 5946. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. A. Casey and M. F. Mescher IL-21 Promotes Differentiation of Naive CD8 T Cells to a Unique Effector Phenotype J. Immunol., June 15, 2007; 178(12): 7640 - 7648. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Pejawar-Gaddy and M. A. Alexander-Miller Ligation of CD80 Is Critical for High-Level CD25 Expression on CD8+ T Lymphocytes J. Immunol., October 1, 2006; 177(7): 4495 - 4502. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. E. Brown, C. Blank, J. Kline, A. K. Kacha, and T. F. Gajewski Homeostatic Proliferation as an Isolated Variable Reverses CD8+ T Cell Anergy and Promotes Tumor Rejection J. Immunol., October 1, 2006; 177(7): 4521 - 4529. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Arimilli, M. A. Alexander-Miller, and G. D. Parks A Simian Virus 5 (SV5) P/V Mutant Is Less Cytopathic than Wild-Type SV5 in Human Dendritic Cells and Is a More Effective Activator of Dendritic Cell Maturation and Function. J. Virol., April 1, 2006; 80(7): 3416 - 3427. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Kronfeld, E. Hochleitner, S. Mendler, J. Goldschmidt, R. Lichtenfels, F. Lottspeich, H. Abken, and B. Seliger B7/CD28 Costimulation of T Cells Induces a Distinct Proteome Pattern Mol. Cell. Proteomics, December 1, 2005; 4(12): 1876 - 1887. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Liu, P. Hu, L. A. Khawli, and A. L. Epstein Combination B7-Fc Fusion Protein Treatment and Treg Cell Depletion Therapy Clin. Cancer Res., December 1, 2005; 11(23): 8492 - 8502. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. P. O'Keefe and T. F. Gajewski Cutting Edge: Cytotoxic Granule Polarization and Cytolysis Can Occur without Central Supramolecular Activation Cluster Formation in CD8+ Effector T Cells J. Immunol., November 1, 2005; 175(9): 5581 - 5585. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Favre, F. Spertini, and B. Corthesy Secretory IgA Possesses Intrinsic Modulatory Properties Stimulating Mucosal and Systemic Immune Responses J. Immunol., September 1, 2005; 175(5): 2793 - 2800. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Markiewicz, L. N. Carayannopoulos, O. V. Naidenko, K. Matsui, W. R. Burack, E. L. Wise, D. H. Fremont, P. M. Allen, W. M. Yokoyama, M. Colonna, et al. Costimulation through NKG2D Enhances Murine CD8+ CTL Function: Similarities and Differences between NKG2D and CD28 Costimulation J. Immunol., September 1, 2005; 175(5): 2825 - 2833. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. S. Pejawar, G. D. Parks, and M. A. Alexander-Miller Abortive versus Productive Viral Infection of Dendritic Cells with a Paramyxovirus Results in Differential Upregulation of Select Costimulatory Molecules J. Virol., June 15, 2005; 79(12): 7544 - 7557. |
) or
P815/B7.2-L with (
) or without (
) the addition of 20 U/ml
recombinant human IL-2. [3H]Thymidine incorporation was
measured at various times after initiation of cultures as described in
Materials and Methods. Depicted are the mean ± SE
of triplicate cultures from one representative of three independent
experiments.
