International Immunology, Vol. 8, No. 12, pp. 1927-1936
© 1996 Oxford University Press
CD28-B7 interactions function to
co-stimulate clonal deletion of
double-positive thymocytes
Derk Amsen and Ada M. Kruisbeek
Division of Immunology, The Netherlands Cancer Institute, Antoni van Leeuwenhoek Huis,
Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands
Keywords: central tolerance, co-stimulation, negative selection
Abstract
Negative selection of thymocytes only occurs if next to signals through the TCR, additional
antigen-presenting cell (APC)-derived signals are also provided. It has been unclear which
molecular interactions lead to the generation of these signals. In particular, the involvement of
CD28 and its ligands B7-1 and B7-2 has been controversial. In the present study, we re-address
this issue and first confirm that cross-linking CD28 molecules on thymocytes can indeed
complement TCR-derived signals for induction of deletion upon TCR engagement with antibodies.
Furthermore, we extend these findings by documenting that also peptide agonist-induced deletion
can be co-stimulated by antibody-mediated engagement of CD28. Additionally, blocking B7-1 or
B7-2 reduces negative selection induced by both anti-CD3 and peptide agonist in suspension
cultures and in fetal thymic organ culture. At the same time, prominent co-stimulation of TCRinduced deletion could be provided by a B7-negative cell line. Together these results definitively
demonstrate that CD28-B7 interactions can function to co-stimulate induction of clonal deletion,
while yet to be identified B7-independent co-stimulatory signals can fulfil this function as well.
Introduction
Development of the mature T cell repertoire in the thymus is
characterized by two selection processes resulting in a pool
of mature T cells that is self restricted but not autoreactive.
Thymocytes that express TCR with inherent affinity for complexes of self peptides bound to self MHC are rescued from
programmed cell death in a process called positive selection
(1-6). Additionally, negative selection by induction of
apoptosis occurs for those thymocytes bearing receptors
that react strongly to self (7-9). Both processes depend on
interactions between the TCR on CD4 + CD8 + CD3 + thymocytes and self peptides associated with MHC molecules on
thymic APC (10, 11).
It is clear that a decisive role in positive and negative
selection lies in the affinity of the TCR for a given peptideMHC complex together with the density at which this peptide
is presented (12-14). On the other hand, it has been suggested that different types of antigen-presenting cells (APC)
are associated with these two selection processes, implying
that additional variables are involved. Radiation bone marrow
chimera studies indicated that, with respect to induction of
positive and negative selection, the thymus appears compart-
mentalized. Thus, the radiation-resistant thymic epithelium
was found responsible for positive selection, whereas negative
selection was induced by APC derived from the bone marrow
(i.e. cells of hematopoietic origin) (15-18). The differential
ability of thymic APC to invoke either positive or negative
selection may be the result of several features. First, the MHC
levels on the APC can be expected to influence the density
at which peptide-MHC complexes can occupy TCR on the
thymocytes. Second, expression of adhesion molecules is
likely to be involved, as for instance antibodies against LFA-1
and ICAM-1 have been shown to block negative selection
(19). Finally, expression of co-stimulatory molecules might be
a crucial determinant, making APC specifically suited for one
of the two selection processes.
In line with a proposed role for co-stimulation in thymocyte
selection, it has been demonstrated that signals through the
TCR are by themselves insufficient for clonal deletion (20,21).
It is unclear, however, whether the co-stimulatory molecules
involved in clonal deletion are the same as those operative
in activation of mature T cells. In favor of this notion is the
observation that cells which abundantly express ligands for
Correspondence to. A. M. Kruisbeek
Transmitting editor. A. Singer
Received 24 May 1996, accepted 23 August 1996
1928
CD28 co-stimulates clonal deletion
co-stimulatory molecules for activation such as B7-1 and
B7-2 are also the most efficient APC for deletion (19,22,23).
Moreover, co-stimulatory receptors involved in activation of
mature T cells, such as CD28 and CD27, are expressed at
high levels on double-positive (DP) thymocytes (24,25). While
this suggests that similar co-stimulatory interactions may be
involved in thymic deletion and peripheral activation, the role
of CD28-B7 interactions in this process is controversial.
Although cross-linking CD28 molecules on thymocytes was
shown to augment anti-CD3-mediated deletion (21), several
investigators failed to detect an effect on deletion when
engagement of B7 was prevented by blocking antibodies
(20,26,27). Also, CD28-deficient mice are capable of deleting
superantigen-reactive and peptide-specific thymocytes (2830). These studies may imply that the co-stimulatory interactions involved in activation and deletion are different. However, it has become clear that CD28 is not the only T cell
surface molecule capable of co-stimulating mature T cell
activation. This is illustrated by the observation that in these
same CD28-deficient mice peptide-specific peripheral T cell
responses can still be initiated (29), even though sustained
proliferative responses are defective (28, 29). Several other
interactions other than those between CD28 and B7-1 and
B7-2 can co-stimulate activation of mature T cells, including
heat shock antigen with its unidentified counter-receptor (31),
CD27 with CD70 (32), gp39 with CD40 (33) and perhaps
other receptor ligand pairs of the tumor necrosis factor
receptor (TNFR) family (34-37). The sensitivity of thymocytes
for deleting signals may be so high that such other, potentially
less dominant, co-stimulatory molecules can sufficiently compensate for a lack of CD28-B7 interactions.
The aim of the present study was to re-address the hypothesis that co-stimulatory interactions involved in activation
can also fulfil this function for clonal deletion. We reasoned
that if our hypothesis was correct, it should be possible to
induce clonal deletion in a co-stimulation-deficient setting
through TCR triggering together with ligation of CD28. For
this, an in vitro suspension culture system was used, in which
thymocytes from 2B4 TCR-transgenic mice (6) were stimulated
in various ways. These transgenic thymocytes respond to
presentation of the 88-104 C-terminal peptide from pigeon
cytochrome c (p17) in the context of I-Ek class II molecules
by rapid onset of apoptosis. With this culture system, we
show that CD28 on DP thymocytes is indeed capable of
generating signals sufficient to complement TCR-derived
signals, such that a full stimulus for induction of apoptosis
ensues. We demonstrate that CD28 can perform this function
not only when the TCR is triggered by antibody-mediated
cross-linking, as previously reported (21), but also in an
antigen-specific setting. Moreover, we show that blocking the
ligation of CD28 with its ligands, B7-1 and B7-2, reduces the
percentage of cells that are deleted upon peptide agonist
presentation both in suspension culture and in fetal thymic
organ culture (FTOC). Therefore our data suggest, in contrast
to earlier reports by us and others (20,26-30), that costimulatory interactions involved in activation of mature T cells
also operate in negative selection. The finding that blocking
of B7-1/2 causes a substantial reduction in peptide-induced
deletion even in the intact thymus implies a physiological role
for CD28-mediated signals in negative selection.
Methods
Mice
All mice were bred at the Netherlands Cancer Institute under
specific pathogen-free conditions. Experiments were performed with thymuses from (2B4ax2B4P)F! TCR-transgenic
mice (6) on the C57BL/6 background. For suspension culture
experiments, 3- to 6-week-old mice were used.
Antibodies and reagents
For tissue culture the following Protein A-purified antibodies
were used: GK1.5 (anti-CD4; 38); 11-4-1 (anti-Kk; 39), 14-44s (anti-l-Ek; 40); 1G10 (anti-B7-1; 43), GL1 (anti-B7-2; 44)
and GL3 (anti-y6TCR; 45) were kindly provided by Drs L H.
Glimcher, K. S. Hathcock and L. A. Gravestein respectively.
Whole rat Ig was obtained from Caltag (South San Francisco,
CA). In some instances, ammonium sulfate precipitates from
supernatants of hybridomas were used; they were grown in
media containing nutridoma-NS serum-free media supplement instead of FCS: 53-6.7 (anti-CD8; 41); 37.51 (anti-CD28;
24) and 145-2C11 (anti-CD3e; 42).
The following antibodies and reagents were used for flow
cytometry: 2.43-FITC (anti-CD8; 46), anti-CD4-phycoerythrin
(Caltag), KJ25-FITC (anti-Vp3; 47), anti-B7-1-FITC (PharMingen, San Diego, California), anti-B7-2-FITC (PharMingen)
and streptavidin-tricolor (Caltag).
Pigeon cytochrome c peptide 88-104 (p17) was produced
by the automatic peptide synthesizer Milligen 9050 as
described (48). Synthesis was carried out at the Biochemistry
Department of the Netherlands Cancer Institute.
Cell lines
DAP3.3C1 (49) is an MHC class ll-negative fibroblast cell line
derived from H-2k mice; DCEKhi7 is a daughter cell line of
DAP3.3C1 transfected with l-E k (J. Miller and R. H. Germain,
unpublished results). P815 is a mastocytoma cell line (ATCC);
PIEK was generated from P815 by co-transfection of two
pcEXV-3 plasmids containing mouse MHC class II E a and Epk
(50), respectively, together with the mouse invariant chain
gene (51) under control of a p actin promotor in pFM92
plasmid. Cells were transfected by electroporation. LEC is a
B cell lymphoma derived from CBA/Ht mice (52).
In vitro deletion assay
Thymocytes were brought into single cell suspension in
Iscove's modified Dulbecco's medium (Gibco/BRL, Paisley,
UK) supplemented with 10% FCS (BioWhittaker, Verviers,
Belgium), 2X10~5 M 2-mercaptoethanol (Merck, Darmstadt,
Germany), 100 U/ml penicillin and 100 |ig/ml streptomycin.
Between 2 and 4X10 5 thymocytes were cultured in 96-well
flat-bottom plates for 12-18 h (as indicated in the figure
legends) at 37°C in a humidified incubator containing 5%
CO 2 in air. Spontaneous appearance of EtBr+ cells was
between 25 and 40%. For antigen-specific deletion thymocytes were incubated with 1x10 5 APC (and 1X10 5 costimulators where indicated) per well. Antibodies and peptides
were added soluble in these experiments. In some experiments APC have been fixed with ECDI (Sigma, St Louis, MO).
Fixation was carried out essentially as described (53). Briefly,
1-2x10 7 cells were washed three times in PBS and sub-
CD28 co-stimulates clonal deletion
sequently incubated for 1 h on ice in 1 ml of 0.9% NaCI
containing 75 mM ECDI. After this the cells were washed four
times in medium.
For antibody-mediated deletion, wells were coated for at
least 2 h with 50 nl of a 50 ng/ml antibody solution in PBS.
Subsequently, plates were washed three times with PBS
before thymocytes were added. DP thymocytes were purified
for these experiments by panning on Petri dishes that had
been coated with 25 (ig/ml 53-6.7 antibody in PBS overnight.
Thymocytes were incubated on these plates for 1 h in medium
at 4°C. Subsequently, non-adherent cells were removed by
repeated gentle addition and removal of medium until no
more floating cells were seen under the microscope. After
this the adhering cells were isolated by vigorous pipetting.
Flow cytometry
Maximally 1X10 6 cells per sample were resuspended in PBS
containing 0.5% BSA and 0.02% sodium azide. Antibody was
added in 20 nl per sample, after which the cells were
incubated on ice for 30 min followed by two wash steps.
Incubation with biotin-labeled antibodies was followed by
incubation with streptavidin-TriColor. For measurement of
deletion thymocytes were after surface staining exposed to
1 ng/ml ethidium bromide (Sigma) in 50 \i\ for 30 min as
described (54).
Between 1X10 4 and 2.5X104 cells were analyzed per
sample on a Becton Dickinson FACScan using Lysys II
software. DCEK and PIEK APC were gated out electronically
on the basis of FSC/SSC pattern.
Apoptosis measurement was carried out according to the
protocol published by Nicoletti et al. (55) for staining nuclei
with propidium iodide in hypotonic buffer.
FTOC
Thymic lobes from (C57BL/6 2B4axB10.A 2B4P)F, embryos
at day 16 of gestation were excised and placed on the
surface of 0.8 urn polycarbonate membrane Nucleopore filters
(Poretics, Livermore, CA). Filters were supported by a gelfoam
gelatin sponge (Upjohn, Kalamazoo, Ml). Each sponge was
placed in a well of a six-well tissue culture plate and soaked
in 3 ml of culture medium. The culture medium was freshly
made Iscove's supplemented with 20% FCS, non-essential
amino acids (Gibco/BRL), 2X10~ 5 M 2-mercaptoethanol
(Merck), 100 U/ml penicillin and 100 ng/ml streptomycin, and
4 mM L-glutamine. Starting from 1 day after excision of the
lobes, antibodies against B7-1 and B7-2 (5 ng/ml each),
whole rat Ig (20 |xg/ml) or 11-4-1 (20 ng/ml) were added daily.
On day 3, p17 or anti-CD3 were added to the organ cultures.
After 15 h thymic lobes were harvested for analysis.
Results
Clonal deletion of thymocytes requires co-stimulation
In vitro co-culture of thymocytes, derived from TCR-transgenic
mice, together with APC expressing appropriate MHC molecules is a well established system for studying clonal deletion
(56,57). The usage of TCR-transgenic animals allows visualization of peptide-mediated events, whereas the in vitro culture
system makes it possible to reduce the complexity of the
1929
U
EtBr
Fig. 1. In vitro clonal deletion assay. Thymocytes from 2B4 TCRtransgenic mice were used (on the non-selecting C57BL/6
background), of which -95% express the transgenic receptor specific
for the 88-104 C-terminal peptide from pigeon cytochrome c (p17)
in the context of I-Ek. These thymocytes were cultured for 12-18 h
together with an l-Ek-transfected fibroblast cell line (DCEK hi7) with
different concentrations of p17. Shown are dot plots of CD8 (detecting
all DP thymocytes) versus EtBr (detects loss of viability). The
concentration of p17 used is shown in the upper right corner of each
dot plot. Numbers above the CD8 + EtBr and the CD8 + EtBr + regions
indicate the percentages of cells falling into those regions.
thymic microenvironment. In the present study we made use
of thymocytes derived from mice that are transgenic for the
TCR of the 2B4 clone (6). Over 90% of these thymocytes
recognize the 88-104 C-terminal peptide (p17) from pigeon
cytochrome c in the context of I-Ek. Positive selection of
thymocytes with this receptor occurs in mice expressing I-Ek,
but not in mice with an H-2b haplotype. In the absence of a
positively selecting restriction element accumulation takes
place of CD4 + CD8 + CD3 + cells, generating a large window
for selection studies. Therefore, all our suspension culture
experiments were performed with thymocytes from 2B4-transgenic mice crossed onto the non-selecting C57BL/6 background.
Initiation of deletion of thymocytes is characterized by
down-modulation of CD4 and CD8 molecules (56). However,
this phenomenon has also been observed in thymocytes not
committed to apoptosis (20). Therefore, we chose to screen
our deletion assays by two-color flow cytometry measuring
CD8 expression and EtBr uptake. Because in 2B4 TCRtransgenic mice on the non-selecting C57BL/6 background
very few CD4 + single-positive thymocytes exist, CD8 is a
suitable marker for detection of DP cells. All or most CD8~
are double-negative precursors in this setting. EtBr is a DNA
intercalating dye which can only enter into cells that have
reduced membrane integrity, due to loss of viability (54).
Staining with this dye strongly separates viable from non
viable cells (Fig. 1), thereby allowing sensitive measurement
of deletion.
Presentation of p17 on DCEK, a fibroblast cell line transfected with I-Ek, leads to a dose-dependent decrease in live
DP cells, identified as CD8 + EtBr. This decrease is mirrored
by an increase in CD8'°EtBr+ cells (Fig. 1). In some experi-
1930
CD28 co-stimulates clonal deletion
ments the loss of CD8 + EtBr cells was not associated with a
proportional increase in CD8 + EtBr + cells. Possibly, this was
the result of a specific disappearance of these latter cells
either because of adherence to the APC, phagocytosis or
desintegration. An alternative explanation could, however, be
that the loss of CD8 + EtBr cells was partially due to downregulation of CD8 molecules by thymocytes not committed to
die. On the basis of the cell yields we obtained we could not
definitively discriminate between these possibilities. For this
reason we discarded experiments in which the loss of
CD8 + EtBr cells was not matched by an increase in
CD8 + EtBr + cells.
In order to demonstrate the requirement for co-stimulatory
signals in induction of deletion, we analyzed the capacity
of 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (ECDI)treated DCEK to function as APC for induction of deletion.
Fixation of APC destroys their ability to co-stimulate activation
of mature T cells, without severe disruptive effects on peptide
presentation (53). Moreover, it has been shown that it is
possible to generate responses of mature T cells with fixed
APC in a three cell system. In this setting, peptide presentation
occurs by the fixed APC, whereas co-stimulation is delivered
in trans by APC that lack permissive MHC molecules (58).
Clearly, fixation of DCEK abrogates their ability to delete
thymocytes (Fig. 2). That this is due to destruction of costimulatory capacity and not the result of disruption of peptide
presentation is demonstrated by the fact that the deletional
response is restored by addition of DAP3.3C1 (Fig. 2). These
cells do not express MHC class II (59) and therefore cannot
present p17 to the thymocytes (data not shown). Thus, in
confirmation of earlier results (20), induction of negative
selection requires co-stimulation, which can be provided in
trans. Not all cells are capable of delivering such co-stimulatory signals in this setting, as illustrated for instance by the
inability of P815 mastocytoma cells to co-stimulate in trans
(Fig. 2).
0
20
40
60
80
100
120
pl7 cone (ng/ml)
unfixed DCEK
fixed DCEK
fixed DCEK+P815
fixed DCEK+DAP3
Fig. 2. MHC class ll-deficient fibroblast can co-stimulate deletion
induced by fixed DCEK. Either 1X105 fresh (closed squares) or ECDI
fixed (open symbols) DCEK were incubated with 4 X 1 0 5 thymocytes
from C57BL/6 2B4 TCR-transgenic mice and p 17 ± 1 x 105 DAP3.3C 1
(open diamonds) or P815 (open circles) in 96-well flat-bottom plates.
After 14-18 h samples were stained for CD8 and EtBr, and measured
by flow cytometry. Percentage of specific deletion was calculated as
follows: 100x(%CD8 + EtBr cells without p17 - %CD8 + EtBr cells
at experimental p17 concentration/%CD8+EtBr cells without p17).
Background percentages of CD8 + EtBr cells in the absence of p17
were: thymocytes with DCEK, 44.52%; thymocytes with fixed DCEK
and p815, 38.30%; thymocytes with fixed DCEK and DAP3.3C1,
43.11%; thymocytes with fixed DCEK alone, 41.36%.
Co-stimulatory signals for deletion can be provided by CD28
The above data document that activation and deletion share
a requirement for co-stimulation. An important co-stimulatory
signal involved in activation of mature T cells is generated by
ligation of CD28 on the T cell with its B7-1 and B7-2 counterreceptors on the APC (60-64). Blocking this interaction
reduces T cell responses, whereas antibody-mediated crosslinking of CD28 can complement signals through the TCR
and prevent induction of anergy in T cell clones (65). It was
reported earlier (21) that also on DP thymocytes, engagement
of CD28 molecules leads to the generation of signals that
can complement those generated by anti-CD3 for full induction
of deletion. We first confirmed (data not shown) that antiCD28 co-stimulates anti-CD3-induced deletion of DP thymocytes, and also found through propidium iodide staining,
again in agreement with Punt et al. (21), that the specific
deletion of DP thymocytes by CD3 and CD28 cross-linking
occurs via an apoptotic mechanism.
It is unclear whether clonal deletion resulting from crosslinking the TCR on thymocytes with antibodies is representative of deletion that is induced by recognition of antigen on
APC. In Bcl-2-transgenic mice, for example, the former is
blocked whilst the latter appears to be unaffected (66). Thus,
different signal transduction pathways may be activated by
the two different ways of triggering deletion. We therefore
investigated whether cross-linking CD28 molecules on thymocytes could also function to co-stimulate deletion if the signal
through the TCR was provided by engagement of peptide
plus MHC. In order to do this, we chose to use PIEK cells
(P815 transfected with I-Ek) as APC for two reasons. First,
these cells express high levels of Fc receptors (CD16) and
therefore are capable of presenting antibodies to thymocytes.
In addition, the parent cell line of PIEK, P815, was found to
be ineffective in providing co-stimulation in trans (Fig. 2).
Therefore, PIEK could be expected to be an inefficient inducer
of deletion, allowing us to optimize deletion by provision of
co-stimulation. As shown in Fig. 3, presentation of p17 on
PIEK cells to purified thymocytes, without exogenously added
co-stimulation, leads to maximally 40% deletion. CD3 crosslinking on these same thymocytes by plate bound antibody
did not lead to deletion (data not shown), demonstrating
that these thymocytes had been sufficiently depleted of
endogenous APC. While these results indicate that PIEK cells
are not completely co-stimulation-deficient, a large portion of
the thymocytes seems to interact with the p17 presented on
CD28 co-stimulates clonal deletion
o
.1
1
10
pl7 cone (jig/ml)
costimulus:
—~°"— none
— » — anti-CD28
— ° — control hamster IgG
Fig. 3. Anti-CD28 enhances induction of deletion by PIEK. DP
thymocytes from 2B4 TCR-transgenic mice were purified by panning
on anti-CD8-coated plates. Next, these cells (>95% DP) were cultured
together with l-Ek-transfected P815 cells (PIEK) with different
concentrations of p17, alone or in the presence of anti-CD28 or
control hamster antibody (GL3). Shown is the percentage deletion,
calculated as in Fig. 2. PIEK cells had been fixed with ECDI.
Similar results were obtained with unfixed PIEK cells. Background
percentages of CD8 + EtBr cells in the absence of p17 were: in
the absence of antibody, 57.55%; with anti-CD28, 61.44%; with
GL3, 58.80%.
the PIEK cells without receiving sufficient co-stimulation. This
is documented by the fact that they have down-regulated
CD8, yet do not become EtBr+ (results not shown). Addition
of anti-CD28 significantly enhances deletion induced by presentation of p17 on PIEK (Fig. 3). Not only is the peptide
dose-response curve shifted to 100-fold lower concentrations
with anti-CD28, but also the amount of deletion under saturating p17 concentrations is increased when extra co-stimulation
is provided by anti-CD28 antibody. This enhancing effect
does not occur when control hamster antibody is added (Fig.
3). Also, upon presentation of anti-CD3 by P815 no effect
was found by addition of anti-CD4 (data not shown), whereas
anti-CD28 significantly enhanced deletion.
Involvement of B7-1 and B7-2 in induction of deletion by APC
Having established that mAb-mediated engagement of CD28
can provide co-stimulatory signals for negative selection, the
question remains whether physiological engagement of this
molecule by its ligands can allow this process to occur. In
several studies, blocking B7 molecules with CTLA4-lg does
not abrogate deletion, implying that engagement of these
molecules is not necessary for deletion (20,26,27). Also,
CD28-deficient mutant mice exhibit normal negative selection
of certain self-reactive T cell specificities (28-30), arguing
against a prerequisite role for CD28 in thymic deletion. The
models used so far, however, may have been insufficiently
sensitive to allow visualization of a contribution of CD28 to
deletion. Status quo analysis of deletion in CD28 mutant mice
would provide ample opportunity for alternative co-stimulatory
pathways to exert their effect (28-30), as would the longer
1931
incubation periods of thymocytes with the deleting antigen
we and others previously used (20,27). Also, visualization of
a role for CD28 in deletion will be critically dependent on the
type of APC expressing a given self-antigen, and the Mis and
class l-restricted peptides used previously (28,30) may have
been presented by APC not expressing ligands for CD28.
Together, these arguments presented an incentive to reexamine whether B7 molecules are involved in negative
selection in the intact thymus in a FTOC model. In contrast
to our previous approach (27), we now first saturated fetal
thymic lobes with antibodies against B7-1 and B7-2, and
subsequently incubated the lobes with peptide or anti-CD3
for only 15 h. When analyzing these lobes with ethidium
bromide, we observed too much variation between groups
treated similarly to allow clear judgement of effects of antiB7 treatment (data not shown). A likely explanation for this is
that small parts of fetal lobes become necrotic, due to
imperfect penetration of nutrients and/or oxygen. As ethidium
bromide staining cannot distinguish between necrotic and
apoptotic cells, we therefore analyzed the results from the
organ cultures by staining with propidium iodide in hypotonic
buffer. Indeed, after staining the samples with propidium
iodide in hypotonic buffer, we observed much less variation
as well as much lower background cell death. Using this
detection method, we observed that anti-CD3 induced
apoptosis in B10.A fetal thymic lobes was substantially
blocked by anti-B7 treatment, but not by treatment with control
antibodies (Fig. 4A). Similarly, we found that anti-B7 treatment
leads to a clear reduction in the amount of deletion induced
by p17 in lobes derived from 2B4 TCR-transgenic mice (Fig.
4B). These results document that, in contrast to earlier reports,
B7 molecules are indeed used by APC in the intact thymus
for induction of negative selection.
Since blocking of B7-1/2 did not abrogate deletion in
the FTOC completely, we argued that skewing of the costimulatory requirements towards a role for B7-1 or B7-2 by
the use of clonal populations of APC expressing B7-1 (DCEK)
or B7-2 (LEC) would more prominently allow visualization of
B7-1/2 involvement. To test this prediction, we performed
experiments in which thymocytes were exposed to p17 on
DCEK or LEC cells in the presence of blocking antibodies
against B7-1 and B7-2. As shown in Fig. 5(A), blocking B7-1
and B7-2 together diminishes the amount of deletion induced
by DCEK. Because we failed to detect B7-2 expression
on DCEK by flow cytometry, the effect of the blockade is
presumably due to blocking the accessibility of B7-1. This
notion is corroborated by the fact that antibodies to B7-1
alone have a comparable reducing effect on deletion to the
effect with both antibodies. In addition, antibodies against
B7-2 have no effect on deletion mediated by DCEK (data not
shown). That B7-2 molecules can perform a similar role as
B7-1 in clonal deletion can be concluded from the results
obtained with LEC as APC. LEC is an H-2k B cell lymphoma
that highly expresses B7-2 but not B7-1 (as determined by
flow cytometry). Addition of anti-B7-2 antibody in the assay
leads to a marked reduction in the amount of deletion (Fig.
5B). Anti-B7-1 does not appear to have any effect on induction
of deletion by this B cell lymphoma. Together, these findings
document a dominant role for B7-1 and B7-2 in deletion when
antigen is presented only by B7-expressing APC.
1932
CD28 co-stimulates clonal deletion
A
PBS
4
Afi CQ
anti CD3
III
5.24
anti B7
ft
Rat IgG
4
IJ
AN
3.47
5.21
PBS
1
1
B
PBS
anti B7
Rat IgG
P 17
38.89
27.63
40.37
PBS
7.59
6.42
8.13
Fig. 4. B7 molecules are involved in induction of deletion in the intact
thymus. Day 16 fetal thymi from C57BL/6 mice (A) or 2B4 TCRtransgenic mice (B10.AXC57BL/6) (B) were cultured in the presence
of a mixture of anti-B7-1 and anti-B7-2 (final concentration 5 ng/ml
each added daily), whole rat IgG (final concentration 20 |ig/ml added
daily) or PBS (added daily) for 3 days. On the third day 5 ng/ml antiCD3 (A) or p17 (B) was added. After 15 h the thymi were harvested
and nuclei were stained with propidium iodide in hypotonic buffer.
Numbers in each histogram indicate the percentage of nuclei
containing subdiploid DNA contents. Results shown are
representative of seven experiments.
In contrast to the results obtained for B7-1 with DCEK and
for B7-2 with LEC as APC, we found no effect of either
antibody on deletion induced by PIEK (Fig. 5C). PIEK cells
do not express either B7-1 or B7-2. These results illustrate
that different types of APC employ different molecules for costimulation of deletion. Thus, while LEC appears to heavily
depend on B7-2 for its co-stimulatory ability, DCEK and
especially PIEK must clearly express other, as yet unidentified
but prominent co-stimulatory molecules. This finding provides
a rational explanation for the lack of B7 participation in deletion
reported previously, as well as for our observation that not all
deletion in the thymus can be blocked by anti-B7-1/2 (Fig. 4).
Discussion
For efficient establishment of tolerance it is necessary that
interactions leading to T cell activation by self antigens in
the periphery are prevented through clonal deletion of the
pertinent thymocytes. It could, therefore, be expected that
qualitatively similar interactions are involved in activation of T
cells in the periphery and negative selection in the thymus.
Indeed, it has been demonstrated that the range of peptide
doses and affinities required for activation and deletion overlap, with a safe margin on the side of tolerance (22,57,67,68).
Also, the potency of different APC for supporting activation
appears to correspond with ability to induce clonal deletion.
From both bone marrow transplantation studies (15) and
in vitro experiments (22,23,69) it can be concluded that bone
marrow-derived cells are the most potent APC for this process.
The efficacy of these cells in negative selection could in part
be explained by the high MHC levels they express as well as
by the localization of these cells in the thymic medulla and
cortico-medullary junction. In these regions the DP thymocytes
have started to up-regulate their TCR levels, thereby perhaps
allowing higher avidity interactions. A contribution to the high
efficiency might also be made by co-stimulatory molecules
expressed on bone marrow-derived cells. Moreover, the fact
that co-stimulatory molecules contribute importantly to the
activating capacity of professional APC, together with the
above noted similarity of the interactions leading to deletion
or activation, would suggest that the co-stimulatory molecules
involved in the two processes may be the same or at least
overlap. Despite many recent studies addressing this issue
(20-28) it remains difficult to assign a role for specific costimulatory molecules in thymic deletion. We here re-examine
the possible contribution of CD28-B7 interactions in negative
selection. Engagement of the T cell surface molecule CD28
with its ligands B7-1 and B7-2 leads to generation of a
dominant co-stimulatory signal for activation of mature T cells
(60-65). A role for these molecules in negative selection is
suggested by their expression pattern in the thymus. CD28
is expressed at high levels on double-positive thymocytes
(24), whereas B7 is highly expressed (as measured with
CTLA-4-lg) on thymic DC, macrophages and B cells (27)
as well as on certain medullary epithelial cells (70,71). In
accordance with this expression pattern, our data demonstrate that ligation of CD28 can complement TCR-derived
signals in thymocytes for induction of apoptosis, both if TCR
cross-linking occurs by anti-CD3 (as reported previously,
21) or by peptide presentation. Moreover, we show that
engagement of B7-1 and B7-2 is involved in deletion induced
by certain APC in suspension culture experiments, as well as
in the intact thymus. Thus, our data strongly suggest that, at
least with respect to involvement of CD28-B7 interactions,
co-stimulatory events associated with activation of mature T
cells and clonal deletion of thymocytes indeed overlap.
The notion that CD28-B7 interactions are involved in negative selection seems to be inconsistent with the phenotype of
CD28-deficient mice. In these mice clonal deletion induced
by Mis antigens or by agonist peptide in a TCR-transgenic
setting appeared unaffected (28-30). It is conceivable, therefore, that the results we obtained from blocking B7 molecules
are the consequence of prevented ligation thereof with other
thymocyte surface molecules than CD28, such as CTLA-4.
However, although CTLA-4-deficient mice have a dramatic
phenotype in the thymus (72,73), perhaps suggesting a role
for this molecule in negative selection, we were unable to
detect CTLA-4 expression on DP thymocytes. More importantly, we did not observe any effect from anti-CTLA-4 antibodies, alone or cross-linked, on deletion in suspension
culture (data not shown). Possibly, therefore, the thymic
abnormality of CTLA-4-deficient mice is the consequence of
stress associated with the uncontrolled activation of mature
T cells in these mice. Other, yet undiscovered, receptors for
B7 may be expressed by thymocytes, explaining the effects
of the anti-B7 blockade. We contend, however, that these
CD28 co-stimulates clonal deletion
1933
noab
antiKk
anti I-Ek
anti B7-l+anti B7-2
—
•d
100
.1
1
10
pl7 cone (u.g/ml)
3U "
no antibody
anti Kk
anti B7-1
anti B7-2
anti B7-l+anti B7-2
40-
let
e
o
30-
T3
20-
1
o
S*
100<
.
i^-^
.01
.1
1
pl7 cone (u.g/ml)
10
no antibody
RIgG
anti B7-1
anti B7-2
anti B7-l+anti B7-2
—
20-
0
.001
.01
.1
1
pl7 cone (jig/ml)
Fig. 5. (A). Blocking B7 molecules on APC in suspension culture reduces the capacity to induce clonal deletion. Thymocytes from 2B4 TCRtransgenic mice were cultured for 13 h with DCEK cells (as in Fig. 1) and different concentrations of p17, alone or in the presence of 10 ng/ml
anti-K\ 3 ng/ml anti-l-Ek or 3 ng/ml anti-B7-1 and 3 ng/ml anti-B7-2. After 14-18 h, cells were harvested and stained for CD8 versus EtBr.
Data are expressed as percentage deletion and were calculated as in Fig. 2. Background percentages of CD8 + EtBr cells in the absence of
p17 were: without antibody, 60.25%; with anti-Kk, 60.55%; with anti-l-Ek, 57.83%; with anti-B7-1 and anti-B7-2, 60.78%. (B) Similarly, thymocytes
from 2B4 TCR-transgenic mice were cultured and analyzed as in (A) with B cell Iymphoma LEC and different concentrations of p17, alone or
in the presence of 25 ng/ml anti-Kk, 10 ng/ml anti-B7-1, 10 ng/ml anti-B7-2 (GL1) or 10 (ig/ml of both anti-B7-1 and anti-B7-2. Background
percentages of CD8 + EtBr cells were: without antibody, 43.90%; with anti-Kk, 41.90%; with anti-B7-1, 44.06%; with anti-B7-2, 43.61%; with
anti-67-1 and anti-B7-2, 45.84%. (C) Thymocytes from 2B4 TCR-transgenic mice were cultured and analyzed as in (A) with PIEK cells and
different concentrations of p17.
1934
CD28 co-stimulates clonal deletion
effects are the result from preventing CD28 cross-linking on
the thymocytes, since direct antibody mediated engagement
of CD28 results in co-stimulation of deletion. The explanation
for the lack of impairment of negative selection in CD28
knock-out mice could then be that the CD28-B7 interaction
is one of a redundant set of interactions with the capacity to
co-stimulate this process. That other interactions can provide
co-stimulation for deletion is also strongly suggested by our
observation that certain types of APC, such as PIEK cells, do
not express B7 and yet can co-stimulate deletion.
Although we were able to inhibit deletion of DP thymocytes
in suspension culture, it is possible that in vivo B7 does not
participate in negative selection of cells at this stage of
development. Deletion of high-affinity TCR-bearing thymocytes has been reported to occur in the thymic cortex, where
very little B7 expression can be detected in the adult thymus.
In contrast, B7 expression is abundant throughout the thymus
in day 18 embryos. This could be the explanation for the
reduction of deletion we obtained when treating FTOC with
anti-B7 antibodies. In vivo involvement of B7 in negative
selection has been suggested by two previous studies. Thus,
deletion of Vp5+ l-E-reactive thymocytes by a subset of thymic
medullary epithelial cells was found to be correlated with
expression of B7 by these cells (71,74). A role for B7 in
deletion was also suggested when mice were chronically
treated with anti-gp39 (75). In these mice deletion induced
by endogenously expressed antigens was blocked. Since the
same was true in gp39-deficient mice, the authors reasoned
that this blockade was probably due to a lack of CD40mediated signal transduction on the thymic APC, resulting in
insufficient up-regulation of co-stimulatory molecules on these
cells. Indeed, expression of B7-2 was shown to be dramatically
reduced in the thymi of these mice (75), suggesting its
involvement in deletion. Alternatively, CD40 directly co-stimulates deletion through a B7-independent mechanism. Separate non-overlapping functions for gp39- and CD28-mediated
signal transduction pathways are also suggested to operate
in clonal expansion of peripheral T cells (76).
Besides CD40, another member of the TNFR gene family
was recently implicated in negative selection: male CD30deficient mice expressing a transgenic anti-H-Y TCR exhibit
severely reduced clonal deletion in the thymus (77). Although
it was not addressed in this study (77), CD30 might act as a
specific death inducing molecule in thymocytes, similar to
Fas on mature T cells. Alternatively, CD30 may function, like
CD28, as a co-stimulatory molecule. In that event, the CD30deficient mice serve to make the point that more than one set
of interactions can co-stimulate clonal deletion. Indeed, while
deletion of anti-H-Y-specific T cells is diminished in CD30deficient mice, deletion of Mis-reactive thymocytes is not (77).
Chronic anti-gp39 treatment, on the other hand, does inhibit
deletion of Mis-reactive thymocytes (75). Also, thymi from
male anti-H-Y TCR-transgenic CD30-deficient mice are still
much smaller than those from females, implicating negative
selection driven by the participation of other co-stimulatory
molecules (77).
The question remains why in thymic deletion a requirement
for co-stimulation exists (78). On a theoretical basis one could
envisage a system in which deletion would be induced
independent of co-stimulation. Such a system would presum-
ably allow removal of more TCR specificities and therefore
might offer a safer mechanism for tolerance establishment. A
possible reason for the requirement for co-stimulation could
lie in the contribution it may make to the discrimination by
the thymocyte between positively and negatively selecting
interactions. It seems clear by now that an important determining factor between these two selection processes is the
peptide dose together with the affinity of the TCR for this
peptide + MHC (12-14). On the other hand, whether positive
or negative selection is induced also depends on the nature
of the APC involved, implying a role in selection for accessory
molecules that are differentially expressed. The role of costimulation might, in this respect, either be quantitative or a
qualitative, but this issue cannot be concluded yet from the
currently available information. Although the role of CD28 and
B7 in thymic negative selection seems less prominent than
in activation of mature T cells, presumably due to redundancy,
the results presented in this study clearly suggest that the
co-stimulatory signals involved in negative selection and
activation overlap. Identification of the signal transduction
pathways and transcriptional events linking CD28 engagement on thymocytes to apoptosis will be a major focus for
the future.
Acknowledgements
We thank Drs Karen Hathcock and Laurie Glimcher for providing
1G10 and GL1 antibodies as well as Dr Jim Allison for the 37.51
hybridoma. Furthermore, we thank Dr Frank Momburg for pFM92
containing the mouse li gene; Dr Andrea Sant for pcEXV-3 containing
the mouse MHC class II Ea and Ep genes, and Dr Ron Germain for
providing DCEKhi7 and DAP3.3C1 cell lines. Finally, we thank Drs
Frank Staal and Jannie Borst for critical review of the manuscript,
Brigit Verbeek for technical assistance, and Marie Anne van Halem
for preparation of the manuscript.
Abbreviations
APC
DP
FTOC
PIEK
TNFR
antigen presenting cells
double positive
fetal thymic organ culture
l-Ek-transfected P815 cells
tumor necrosis factor receptor
References
1 Bevan, M. J. 1977. In a radiation chimaera, host H-2 antigens
determine immune responsiveness of donor cytotoxic T cells.
Nature 269:417.
2 Zinkernagel, R. M., Callahan, G. N., Althage, A., Cooper, S., Klein,
P. A. and Klein, J. 1978. On the thymus in the differentiation of
'H-2 self-recognition' by T cells: evidence for dual recognition?
J. Exp. Med. 147:882.
3 Marusic-Galesic, S., Stephany D. A., Longo, D. L. and Kruisbeek,
A. M. 1988. Development of CD4"CD8+ cytotoxic T cells requires
interactions with class I MHC determinants. Nature 333:180.
4 Marusic-Galesic, S., Longo, D. L. and Kruisbeek, A. M. 1989.
Preferential differentiation of T cell receptor specificities based
on the MHC glycoproteins encountered during development.
Evidence for positive selection. J. Exp. Med. 169:1619.
5 Kisielow, P., Teh, H. S., Bluthmann, H. and von Boehmer, H.
1988. Positive selection of antigen-specific T cells in thymus by
restricting MHC molecules. Nature 335:730.
6 Berg, L. J., Pullen, A. M., Fazekas de St Groth, B., Mathis, D.,
Benoist, C. and Davis, M. M. 1989. Antigen/MHC-specific T cells
CD28 co-stimulates clonal deletion
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
are preferentially exported from the thymus in the presence of
their MHC ligand. Cell 58:1035.
Kappler, J. W., Roehm, N. and Marrack, P. 1987. T cell tolerance
by clonal elimination in the thymus. Cell 49:273.
Kisielow, P., Bluthmann, H., Staerz, U. D., Steinmetz, M. and Von
Boehmer, H. 1988. Tolerance in T-cell-receptor transgenic mice
involves deletion of nonmature CD4 + 8 + thymocytes. Nature
333:742.
Murphy, K. M., Heimberger, A. B. and Loh, D. Y. 1990. Induction by
antigen of intrathymic apoptosis of CD4 + CD8 + TCR'° thymocytes
in vivo. Science 250:1720.
Nikolic-Zugic, J. and Bevan, M. J. 1990. Role of self-peptides in
positively selecting the T-cell repertoire. Nature 344:65.
Jacobs, H., von Boehmer, H., Melief, C. J. M. and Berns, A. 1990.
Mutations in the major histocompatibility complex class I antigenpresenting groove affect both negative and positive selection of
T cells. Eur. J. Immunol. 20:2333.
Ashton-Rickard, P. G., Van Kaer, L, Schumacher, T. N. M., Ploegh,
H. L. and Tonegawa, S. 1993. Peptide contributes to the specificity
of positive selection of CD8 + T cells in the thymus. Cell 73:1041.
Hogquist, K. A., Jameson, S. C , Heath, W. R., Howard, J. L,
Bevan, M. J. and Carbone, F. R. 1994. T cell receptor antagonist
peptides induce positive selection. Cell 76:17.
Sebzda, E., Wallace, V. A., Mayer, J., Yeung, R. S. M., Mak, T. W.
and Ohashi, P. S. 1994. Positive and negative thymocyte selection
induced by different concentrations of a single peptide.
Science 263:1615.
Sprent, J., Kosaka, H., Gao, E.-K., Surh, C. D. and Webb, S. R.
1993. Intrathymic and extrathymic tolerance in bone marrow
chimeras. Immunol. Rev. 11:151.
Lo, D. and Sprent, J. 1986. Identity of cells that imprint H-2
restricted T-cell specificity in the thymus. Nature 329:672.
Benoist, C. and Mathis, D. 1989. Positive selection of the T cell
repertoire: where and when does it occur? Cell 58:1027.
Bill, J. and Palmer, E. 1989. Positive selection of CD4 + T cells
mediated by MHC class ll-bearing stromal cell in the thymic
cortex. Nature 341:649.
Carlow, D. A., Van Oers, N. S. C , Teh, S.-J. and Teh, H.-S. 1992.
Deletion of antigen-specific immature thymocytes by dendritic
cells requires LFA-1/ICAM interactions. J. Immunol. 148:1595.
Page, D. M., Kane, L. P., Allison, J. P. and Hedrick, S. M. 1993.
Two signals are required for negative selection of CD4 + CD8 +
thymocytes. J. Immunol. 151:1868.
Punt, J. A., Osborne, B. A, Takaham, Y, Sharrow, S. O. and
Singer, A. 1994. Negative selection of CD4 + CD8 + thymocytes
by T cell receptor-induced apoptosis requires a co-stimulatory
signal that can be provided by CD28. J. Exp. Med. 179:709.
Iwabuchi, K., Nakayma, I.-I., McCoy, R. L, Wang, F., Nishimura,
T., Habu, S., Murphy, K. M. and Loh, D. Y. 1992. Cellular and
peptide requirements for in vitro clonal deletion of immature
thymocytes. Proc. Natl Acad. Sci. USA 89:9000.
Matzinger, P. and Guerder, S. 1989. Does T-cell tolerance require
a dedicated antigen-presenting cell? Nature 338:74.
Gross, J. A., Callas, E. and Allison, J. P. 1992. Identification and
distribution of the co-stimulatory receptor CD28 in the mouse.
J. Immunol. 149:380.
Gravestein, L. A., Nieland, J. D., Kruisbeek, A. M. and Borst, J.
1995. Novel mAbs reveal potent co-stimulatory activity of murine
CD27. Int. Immunol. 7:551.
Tan, R., Teh, S.-J., Ledbetter, J. A., Linsley, P. S. and Teh, H.-S.
1992. B7 co-stimulates proliferation of CD4~8+ T lymphocytes but
is not required for the deletion of immature CD4 + 8 + thymocytes.
J. Immunol. 149:3217.
Jones, L. A., Izon, D. J., Nieland, J. D., Linsley, P. S. and
Kruisbeek, A. M. 1993. CD28-B7 interactions are not required for
intrathymic clonal deletion. Int. Immunol. 5:503.
Shahinian, A., Pfeffer, K., Lee, K. P., Kundig, T. M., Kishihara, K.,
Wakeham, A., Kawai, K., Ohashi, P. S., Thompson, C. B. and
Mak, T. W. 1993. Differential T cell co-stimulatory requirements in
CD28-deficient mice. Science 261:609.
Lucas, P. J., Negishi, I., Nakayama K., Fields, L. E. and Loh, D. Y.
1995. Naive CD28-deficient mice can initiate but not sustain an
in vitro antigen specific immune response. J. Immunol. 154:5757.
1935
30 Walunas, T. L., Sperling, A. I., Khattri, R., Thompson, C. B. and
Bluestone, J. A. 1995. CD28 expression is not essential for
positive and negative selection of thymocytes or peripheral T cell
tolerance. J. Immunol. 156:1006.
31 Liu, Y, Jones, B., Aruffo, A., Sullivan, K. M., Linsley, P. S. and
Janeway, C. A., Jr. 1992. Heat-stable antigen is a co-stimulatory
molecule for CD4 T cell growth. J. Exp. Med. 175:437.
32 Goodwin, R. G., Alderson, M. R., Smith, C. A., Armitage, R. J.,
VandenBos, T., Jerzy, R., Tough, T. W., Schoenborn, M. A., DavisSmith, T., Hennen, K., Falk, B., Cosman, D., Baker, E., Sutherland,
G. R., Grabstein, K. H., Farrah, T., Giri, J. G. and Peckmann, M. P.
1993. Molecular and biological characterization of a ligand for
CD27 defines a new family of cytokines with homology to tumor
necrosis factor. Cell 73:446.
33 Cayabyab, M., Phillips, J. H. and Lanier, L. L. 1994. CD40
preferentially co-stimulates activation of CD4 + lymphocytes.
J. Immunol. 152:1523.
34 Fanslow, W. C , Clifford, K. N., Seaman, M., Alderson, M. R.,
Spriggs, M. K., Armitage, R. J. and Ramsdell, F. 1994.
Recombinant CD40 ligand exerts potent biologic effects on T
cells. J. Immunol. 152:4262.
35 Alderson, M. R., Armitage, R. J., Maraskovsky, E., Tough, T. W.,
Roux, E., Schooley, K., Ramsdell, F. and Lynch, D. H. 1993. Fas
transduces activation signals in normal human T lymphocytes.
J. Exp. Med. 178:2231.
36 Pollok, K. E., Kim, Y- J., Zhou, Z., Hurtado, J., Kim, K. K., Pickard,
R. T. and Kwon, B. S. 1993. Inducible T cell antigen 4-1BB.
Analysis of expression and function. J. Immunol. 150:664.
37 Smith, C. A., Gruss, H.-J., Davis, T., Anderson, D., Farrah, T.,
Baker, E., Sutherland, G. R., Brannan, C. I., Copeland, N. G.,
Jenkins, N. A., Grabstein, K. H., Gliniak, B., McAlister, I. B.,
Fanslow, W., Alderson, M., Falk, B., Gimpel, S., Gillis, S., Din,
W. S., Goodwin, R. G. and Armitage, R. J. 1993. CD30 antigen,
a marker for Hodgkin's lymphoma, is a receptor whose ligand
defines an emerging family of cytokines with homology to TNF.
Cell 73:1349.
38 Dialynas, D. P., Qyan, Z. S., Wall, K. A., Pierres, A., Quintans, J.,
Loken, M. R., Pierres, M. and Fitch, F. W. 1983. Characterization
of the murine T cell surface molecule, designated L3T4, identified
by monoclonal antibody GK1.5: similarity of L3T4 to the human
Leu-3/T4 molecule. J. Immunol. 131:2445.
39 Oi, V., Jones, P., Gooding, J. and Herzenberg, L. 1978. Properties
of monoclonal antibodies to mouse Ig allotypes. H-2 and la
antigens. Curr. Top. Microbiol. Immunol. 81:115.
40 Ozato, K., Mayer, N. and Sachs, D. H. 1980. Hybridoma cell lines
secreting monoclonal antibodies to mouse H-2 and la antigens.
J. Immunol. 124:533.
41 Nabavi, N., Freeman, G. J., Gault, A., Godfrey, D., Nadler, L. M.
and Glimcher, L. H. 1992. Signalling through the MHC class II
cytoplasmic domain is required for antigen presentation and
induces B7 expression. Nature 360:266.
42 Hathcock, K. S., Laszlo, G., Dickler, H. B., Bradhsaw, J., Linsley,
P. and Hodes, R. J. 1993. Identification of an alternative CTLA-4
ligand co-stimulatory for T cell activation. Science 262:905.
43 Goodman, T. and Lefrancois, L. 1989. Intraepithelial lymphocytes.
Anatomical site, not T cell receptor form, dictates phenotype and
function. J. Exp. Med. 170:1569.
44 Ledbetter, J. A. and Herzenberg, L. A. 1979. Xenogeneic
monoclonal antibodies to mouse lymphoid differentiation antigens.
Immunol. Rev. 47:63.
45 Leo, O., Foo, M., Sachs, D. H., Samelson, L. E. and Bluestone,
J. A. 1987. Identification of a monoclonal antibody specific for a
murine T3 polypeptide. Proc. Natl Acad. Sci. USA 84:1374.
46 Sarmiento, M., Glasebrook, A. L. and Fitch, F. W. 1980. IgG or
IgM monoclonal antibodies reactive with different determinants
on the molecular complex bearing lyt 2 antigen block T cellmediated cytolysis in the absence of complement. J. Immunol.
125:2665.
47 Pullen, A. M., Marrack, P. and Kappler, J. W. 1988. The T-cell
repertoire is heavily influenced by tolerance to polymorphic selfantigens. Nature 335:796.
48 Merrifield, R. B. 1963. Solid phase peptide synthesis. I. The
synthesis of a tetrapeptide. J. Am. Chem. Soc. 85:2149.
1936
CD28 co-stimulates clonal deletion
49 Germain, R. N. and Quill, H. 1986. Unexpected expression of a
unique mixed-isotype class II MHC molecule by transfected
L-cells. Nature 320:72.
50 Miller, J. and Germain, R. N. 1986. Efficient cell surface expression
of class II MHC molecules in the absence of associated invariant
chain. J. Exp. Med. 164:1478.
51 Stockinger, B., Pessara, U., Lin, R. H., Habicht, J., Grez, M. and
Koch, N. 1989. A role of la-associated invariant chains in antigen
processing and presentation. Cell 56:683.
52 Van Pel, A., Vessiere, F. and Boon, T. 1983. Protection against
two spontaneous mouse leukemias conferred by immunogenic
variants obtained by mutagenesis. J. Exp. Med. 157:1992.
53 Jenkins, M. K. and Schwartz, R. H. 1987. Antigen presentation
by chemically modified splenocytes induces antigen-specific T
cell unresponsiveness in vitro and in vivo. J. Exp. Med. 165:302.
54 Lyons, A. B., Samuel, K., Sanderson, A. and Maddy, A. H. 1992.
Simultaneous analysis of immunophenotype and apoptosis of
murine thymocytes by single laser flow cytometry. Cytometry
13:809.
55 Nicoletti, I., Migliorati, G., Paglicacci, M. C , Grignani, F. and
Riccardi, C. 1991. A rapid and simple method for measuring
thymocyte apoptosis by propidium iodide staining and flow
cytometry. J. Immunol. Methods 139:271.
56 Swat, W., Ignatowicz, L, Von Boehmer, H. and Kisielow, P. 1991.
Clonal deletion of immature CD4 + 8 + thymocytes in suspension
culture by extrathymic antigen-presenting cells. Nature 351:150.
57 Vasquez, N. J., Kaye, J. and Hedrick, S. M. 1992. In vivo and
in vitro clonal deletion of double-positive thymocytes. J. Exp.
Med. 175:1307.
58 Jenkins, M. K., Ashwell, J. D. and Schwartz, R. H. 1988. Allogeneic
non-T spleen cells restore the responsiveness of normal T cell
clones stimulated with antigen and chemically modified antigenpresenting cells. J. Immunol. 140:3324.
59 Loss, G. E., Elias, C. G., Fields, P. E., Ribaudo, R. K., McKisic,
M. and Sant, A. J. 1993. Major histocompatibility complex class
II restricted presentation of internally synthesized antigen displays
cell-type variability and segregates from the exogenous class II
and endogenous class I presentation pathways. J. Exp. Med.
178:73.
60 Linsley, P. S., Brady, W., Grosmaire, L, Aruffo, A., Damle, N. K.
and Ledbetter, J. A. 1991. Binding of the B cell activation antigen
B7 to CD28 co-stimulates T cell proliferation and interleukin 2
mRNA accumulation. J. Exp. Med. 173:721.
61 Gimmi, C. D., Freeman, G. J., Gribben, J. G., Sugita, K., Freedman,
A. S., Morimoto, C. and Nadler, L. M. 1991. B-cell surface antigen
B7 provides a co-stimulatory signal that induces T cells to
proliferate and secrete interleukin 2. Proc. Natl Acad. Sci. USA
88:6575.
62 Freeman, G. J., Bordello, F, Hodes, R. J., Reiser, H., Hathcock,
K. S., Laszlo, G., McKnight, A. J., Kim, J., Du, L, Lombard, D. B.,
Gray, G. S., Nadler, L. M. and Sharpe, A. H. 1993. Uncovering of
functional alternative CTLA-4 counter-receptor in B7-deficient
mice. Science 262:907.
63 Freeman, G. J., Gribben, J. G., Boussiotis, V. A., Ng, J. W.,
Restivo, V. A., Jr, Lombard, L. A., Gray, G. S. and Nadler,
L. M. 1993. Cloning of B7-2: a CTLA-4 counter-receptor that costimulates human T cell proliferation. Science 262:909.
64 Hathcock, K. S., Laszlo, G., Pucillo, C , Linsley, P. and Hodes,
R. J. 1994. Comparative analysis of B7-1 and B7-2 co-stimulatory
ligands: expression and function. J. Exp. Med. 180:631.
65 Harding, F. A., McArthur, J. G., Gross, J. A., Raulet, D. H. and
Allison, J. P. 1992. CD28-mediated signalling co-stimulates murine
T cells and prevents induction of anergy in T-cell clones. Nature
356:607.
66 Sentman, C. L, Shutter, J. R., Hockenbery, D., Kanagawa, O.
and Korsmeyer, S. J. 1991. Bcl-2 inhibits multiple forms of
apoptosis but not negative selection in thymocytes. Cell 67:879.
67 Pircher, H., Rohrer, U. H., Moskophidis, D., Zinkernagel, R. M.
and Hengartner, H. 1991. Lower receptor avidity required for
thymic clonal deletion than for effector T-cell function. Nature
351:482.
68 Page, D. M., Alexander, J., Snoke, K., Appella, E., Sette, A.,
Hedrick, S. M. and Grey, H. M. 1994. Negative selection of CD4 +
CD8+ thymocytes by T-cell receptor peptide antagonists. Proc.
Natl Acad. Sci. USA 91:4057.
69 Mazda, 0., Watanabe, Y, Gyotoku, J.-l. and Katsura, Y. 1991.
Requirement of dendritic cells and B cells in the clonal deletion
of Mis-reactive T cells in the thymus. J. Exp. Med. 173:539.
70 Nelson, A. J., Hosier, S., Brady, W., Linsley, P. S. and Farr, A. G.
1993. Medullary thymic epithelium expresses a ligand for CTLA4
in situ and in vitro. J. Immunol. 151:2453.
71 Degermann, S., Surh, C. D., Glimcher, L. H., Sprent, J. and Lo,
D. 1994. B7 expression on thymic medullary epithelium correlates
with epithelium-mediated deletion of Vp5+ thymocytes. J.
Immunol. 152:3254.
72 Waterhouse, P., Penninger, J. M., Timms, E., Wakeham, A.,
Shahinian, A., Lee, K. P., Thompson, C. B., Griesser, H. and Mak,
T. W. 1995. Lymphoproliferative disorders with early lethality in
mice deficient in CTLA4. Science 270:985.
73 Tivol, E. A., Borriello F., Schweitzer A. N., Lynch W. P., Bluestone,
J. A. and Sharpe, A. H. 1995. Loss of CTLA4 leads to massive
lymphoproliferation and fatal multiorgan tissue destruction,
revealing a critical negative regulatory role of CTLA4. Immunity
3:541.
74 Burkly, L. C , Degermann, S., Longley, J., Hagman, J., Brinster,
R. L, Lo, D. and Flavell, R. A. 1993. Clonal deletion of Vp5 + T
cells by transgenic I-E restricted to thymic medullary epithelium.
J. Immunol. 151:3954.
75 Foy, T. M., Page, D. M., Waldschmidt, T. J., Schoneveld, A.,
Laman, J. D., Masters, S. R., Tygrett, L, Ledbetter, J. A., Aruffo,
A., Claassen, E., Xu, J. C , Flavell, R. A., Oehen, S., Hedrick,
S. M. and Noelle R. J. 1995. An essential role for gp39, the ligand
forCD40, in thymic selection. J. Exp. Med. 182:1377.
76 Griggs, N. D., Agersborg, S. S., Noelle, R. J., Ledbetter, J. A.,
Linsley P. S. and Tung, K. S. K. 1996. The relative contribution of
the CD28 and gp39 costimulatory pathways in the clonal
expansion and pathogenic acquisition of self-reactive T cells.
J. Exp. Med. 183:801.
77 Amakawa, R., Hakem, A., Kundig, T. M., Matsyama, T., Simard,
J. J. L., Timms, E., Wakeham, A., Mittruecker H.-W., Griesser, H.,
Takimoto, H., Schmits, R., Shahinian A., Ohashi, P. S., Penninger,
J. M. and Mak, T. W. 1996. Impaired negative selection of T cells
in Hodgkin's disease antigen CD30-deficient mice. Cell 84:551.
78 Amsen, D. and Kruisbeek, A. M. 1996. Mechanisms underlying
T-cell tolerance. Curr. Opin. Immunol. 8:233.
79 Kishimoto, H., Cai, Z., Brunmark, A., Jackson, M. R., Peterson,
P. A. and Sprent, J. 1996. Differing roles for B7 and intercellular
adhesion molecule-1 in negative selection of thymocytes. J. Exp.
Med. 184:531.
Note added in proof
While this paper was under review, Kishimoto et al. (79) have
similarly reported a role for B7 molecules and clonal deletion
of double-positive thymocytes induced by MHC class Itransfected Drosophila cells.