Delineation of Signals Required for Thymocyte Positive
Fabio R. Santori* and Stanislav Vukmanovic ´2*†
Peptide/MHC complexes capable of inducing positive selection in mouse fetal thymic organ cultures fail to do so in suspension
culture. Furthermore, this type of culture does not promote initial stages of differentiation, such as coreceptor down-modulation,
unless peptides used for stimulation have (at least) weak agonist activity. We show in this study that signals provided in suspension
culture by nonagonist peptide/MHC complexes on the surface of macrophages, even though apparently silent, are sufficient to
promote complete phenotypic differentiation when CD4?CD8?thymocytes are subsequently placed in a proper anatomical set-
ting. Furthermore, the synergistic actions of suboptimal concentrations of phorbol esters and nonagonist peptide/MHC complexes
can make the initial stages of positive selection visible, without converting maturation into negative selection. Thus, the correlation
between efficiency of positive selection and the degree of coreceptor down-modulation on CD4?CD8?thymocytes is not linear.
Furthermore, these results suggest that the unique role of thymic stromal cells in positive selection is related not to presentation
of self-peptide/MHC complexes, but most likely to another ligand. The Journal of Immunology, 2004, 173: 5517–5523.
thymocytes receive survival and differentiation signals (1). Ulti-
thymocytes differentiateinto mature
CD4?CD8?or CD4?CD8?T cells depending on whether MHC
class II or class I molecules, respectively, were engaged. In addi-
tion to CD4?CD8?, CD4?CD8?, and CD4?CD8?cells, a num-
ber of intermediate transitional stages of maturation can be found
in the thymus in vivo. These additional stages are CD4lowCD8low,
CD4?CD8low, or CD4lowCD8?and are a result of the complex
transition of CD4?CD8?into the CD4?CD8?or CD4?CD8?
cells after engagement with self-peptide/MHC complexes. Al-
though the precise precursor-product relationship between the
transitional stages is not entirely clear (2, 3), there is a general
agreement that down-modulation of both coreceptors is a transi-
tional phase of thymocyte maturation regardless of whether MHC
class I or class II molecule is the inducing stimulus.
CD4?CD8?cells is not the only possible outcome of TCR en-
gagement by self-peptide/MHC complexes in the thymus. Thy-
mocytes expressing TCRs with potentially harmful reactivity to
self-peptide/MHC complexes are eliminated through a process
called negative selection (4–6). The intermediate stages, if any,
between interaction of CD4?CD8?thymocytes with peptide/
MHC complexes that induce negative selection have been ex-
tensively studied in vitro. Peptide/MHC complexes that pro-
cell differentiation in the thymus requires TCR recogni-
tion of the self-peptide/MHC complexes. As a result of
this interaction, known as positive selection, CD4?CD8?
mote negative selection in vivo induce down-modulation of
CD4 and CD8 in vitro (7), mimicking the effect of positive
selection in vivo. Hence, “coreceptor dulling” has been pro-
moted as a surrogate assay of negative selection (7). However,
although apoptosis can be clearly detected within the popula-
tion of thymocytes stimulated in vitro with ligands that promote
negative selection (8–10), the numbers of thymocytes with
down-modulated CD4/CD8 exceeds the number of apoptotic
cells (10–12). This finding suggests that a proportion of cells
undergoing coreceptor dulling induced by negatively selecting
ligands in vitro may have paradoxically initiated differentiation
into one of the mature thymocyte subpopulations. This conclu-
sion is consistent with the results showing enhanced positive
selection of thymocytes stimulated in vitro with negatively se-
lecting ligands and then assembled in MHC-disparate thymus
reaggregation cultures (12).
Despite the fact that these experiments suggest that coreceptor
dulling is a marker of thymocyte activation, rather than cell death,
attempts to recreate in vitro coreceptor down-modulation after ac-
tivation of CD4?CD8?thymocytes with positively selecting li-
gands were initially unsuccessful. This was most likely because
positively selecting ligands deliver weaker signal than negatively
selecting ones (13). However, culture of CD8 and TCR, double-
transgenic, ?2-microglobulin (?2m)?/?(or TAP1?/?) thymocytes
with positively selecting peptides successfully induced coreceptor
down-modulation in vitro (14). The purpose of using a coreceptor
transgenic strain was to increase the strength of signal induced by
peptide, whereas the purpose of ?2m?/?or TAP1?/?deficiency
was to prevent any selection events in vivo. This assay was used
to discover endogenous self-peptides involved in positive selection
of the OT-I TCR (14, 15). Subsequently, coreceptor down-modu-
lation without the help of the CD8 transgene was used to detect an
endogenous peptide involved in positive selection of another TCR
(16). However, coreceptor dulling caused by this peptide was
barely detectable. Given that this peptide exhibited weak agonist
activity, it could be anticipated that peptides with weaker stimu-
latory potential, but still capable of efficient induction of positive
selection, would be missed using the screening without artificial
signaling enhancement. It is clear that coreceptor dulling is a very
useful assay for easy detection of ligands that could be potentially
*Michael Heidelberger Division of Immunology, Department of Pathology and New
York University Cancer Center, New York University School of Medicine, New
York, NY 10016; and†Center for Cancer and Immunology Research, Children’s
Research Institute, Children’s National Medical Center, Washington, DC 20010
Received for publication January 14, 2004. Accepted for publication August 24, 2004.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported by National Institutes of Health Grants AI48837 and
AI41573 (to S.V.) and National Cancer Institute Core Support Grant 5P30CA16087.
2Address correspondence and reprint requests to Dr. Stanislav Vukmanovic ´, Center
for Cancer and Immunology Research, Children’s Research Institute, Children’s Na-
tional Medical Center, 111 Michigan Avenue NW, Washington, DC 20010-2970.
E-mail address: firstname.lastname@example.org
The Journal of Immunology
Copyright © 2004 by The American Association of Immunologists, Inc. 0022-1767/04/$02.00
involved in positive or negative selection. More importantly, a
more thorough characterization of the assay will lead to a better
understanding of the early stages of positive and/or negative se-
lection. We developed a novel coreceptor dulling assay relevant
for positive selection, and we used this assay to characterize early
events during positive selection.
Materials and Methods
Mice and peptides
TAP1?/?mice were obtained from The Jackson Laboratory (Bar Harbor,
ME). C57BL/6 as well as B10.D2 H-Y TCR transgenic, RAG-2?/?mice
were purchased from Taconic Farms (Germantown, NY). Custom-synthe-
sized peptides were purchased from Research Genetics (Huntsville, AL).
Coreceptor down modulation dulling assay
C57BL/6 mice were injected i.p. with 3 ml of thioglycolate solution. After
3–4 days mice were killed, and the activated macrophages were collected
with cold PBS. The cells were washed, resuspended in PM medium (17)
supplemented with 20% FCS and plated in flat-bottom, 96-well plates at a
density of 2 ? 105cells/well. Macrophages were incubated for at least 1 h
at 37°C, then they received 4 ? 105thymocytes/well derived from either
male or female HYB10.D2 RAG?/?mice. The cultures were pulsed with
the desired concentration of peptide and incubated overnight at 37°C. The
next day thymocytes were collected and stained for FACS analysis with
anti-mouse CD4-PE and CD8-CyChrome Abs (BD Pharmingen, San
Diego, CA). Dulling was quantified based on the shift from the gate of
untreated DP thymocytes from those of peptide-treated DP thymocytes.
Peptide-treated thymocytes were also collected and tested for apoptosis
using an FITC-based TUNEL assay kit (Roche, Indianapolis, IN).
Fetal thymus organ cultures (FTOC)3
The FTOCs were performed using gestation day 16 fetuses derived from
time-mated pregnancies of H-Y TCR transgenic mice on the TAP1?/?
background (17). Fetal thymus lobes were cultivated on sponge-supported
filters (Millipore, Bedford, MA) in medium supplemented with peptide at
300 ?M (unless otherwise stated). Cultures were arranged so that one lobe
was treated as experimental, whereas the other lobe from the same fetus
was treated as a control. As a negative control we used peptides that bind
H-2Dbwell, but do not induce positive selection of H-Y thymocytes. After
10 days, lobes were dissociated, and cells and fetal thymuses were screened
with anti-mouse V?8-FITC (BD Pharmingen) to ascertain expression of
H-Y TCR and anti-H-2Kb-PE (BD Pharmingen) to ascertain for TAP1?/?
status. Fetal lobes were further analyzed by triple FACS staining with
anti-mouse CD4-PE, CD8-CyChrome and CD24-FITC mAbs (BD Pharm-
ingen). Remaining cells were used in the proliferation assay as described
Reaggregation thymus organ cultures (RTOC)
Fetal thymus lobes (days 14.5/15.5) from C57BL/6 TAP1?/?mice were
placed on nitrocellulose filters suspended on gelfoam sponges embedded
with PM10 medium containing 1.35 mM deoxyguanosine for 5 days.
Lobes were then collected, and deoxyguanosine and FCS were carefully
washed out with serum-free medium. The lobes where then homogenized
by trypsinization until a cell suspension was formed. All cell clumps, de-
bris, and DNA were removed, and the remaining cells were counted, spun
down, and resuspended in fresh PM10 medium. These cells were used as
a source of thymic stromal cells. To generate RTOCs, 2 ? 105stromal cells
were mixed with 1 ? 105HY?thymocytes derived from HY?mice on the
B10.D2 RAG2?/?background that were previously incubated in suspen-
sion cultures with TAP1?/?microphages, peptide, and/or PMA as de-
scribed above for the dulling assay. The mixed cells were spun, and ag-
gregates were reconstituted over nitrocellulose filters suspended in gelfoam
sponge embedded in fresh PM10 medium. After 72 h, the reaggregates
were homogenized, and cells were stained for FACS analysis with anti-
mouse CD4-PE, CD8-CyChrome and CD24-FITC Abs (BD Pharmingen).
Synergistic action of self-peptide and suboptimal doses of
phorbol ester stimulates coreceptor down-modulation
We asked whether using suboptimal concentrations of pharmaco-
logical stimulators of pathways known to be involved in positive
selection, such as the Ras/MAPK pathway (18, 19), could replace
the requirement for the use of coreceptor transgene. Phorbol esters
activate the Ras/MAPK pathway through protein kinase C activa-
tion (20) and induce coreceptor down-modulation in immature thy-
mocytes (21). We therefore tested whether presence of phorbol
ester (PMA) could help induce coreceptor down-modulation by a
peptide that induces positive selection. As shown in Fig. 1, PMA
at a concentration of 1.0 ng/ml, but not 0.5 ng/ml, induced core-
ceptor down-modulation in H-Y TCR transgenic, RAG-2?/?thy-
mocytes on a nonselecting background (H-2d). Similarly, culture
of the same thymocytes with a self-peptide, Ube1x509–517, capable
of inducing positive selection (17) failed to induce coreceptor
down-modulation (Fig. 1). However, Ube1x509–517and 0.5 ng/ml
PMA synergized to induce significant coreceptor down-modula-
tion. We call this synergistic effect assisted coreceptor dulling to
distinguish it from the dulling induced by peptides that promote
negative selection and require no signaling enhancement (either
pharmacological or genetic). An example of such unassisted dull-
ing is given in Fig. 1C using Smcy738–746peptide. Time-course
experiments have indicated that assisted dulling could be observed
as early as 2 h after initiation of cultures (data not shown). Fur-
thermore, the ability to induce assisted dulling was not restricted to
peptide Ube1x509–517. For example, some variants of Smcy738–746
that were able to induce positive selection in FTOC also induced
assisted dulling (Fig. 2). Thus, stimulation of CD4?CD8?thymo-
cytes with suboptimal PMA concentrations acts synergistically
with signals initiated by peptides that induce positive selection to
induce coreceptor down-modulation.
Assisted dulling correlates better with positive selection than
with antagonist activity
The abilities of peptides to promote positive selection and to in-
duce TCR antagonist activity in peripheral T cells overlap in many
cases, but these two activities are clearly distinguishable. Thus, it
is important to determine whether suboptimal PMA stimulation
enhances weak signals in general or specifically those that induce
positive selection. We have recently characterized ligands capable
of stimulating antagonist activity, positive selection, or both ac-
tivities in the H-Y TCR transgenic model (17). We therefore asked
whether suboptimal PMA concentrations would amplify signals
generated by peptides that have TCR antagonist activity, but do not
promote positive selection. Peptide ARX54–62(VSNLNRQFL) is
identical with Ube1x509–517(KSNLNRQFL) except at position 1,
where lysine in the latter is substituted by valine in the former
peptide. This difference produces distinct biological activities of
the two peptides: Ube1x509–517antagonizes Ag-induced CD8?
cell proliferation and promotes thymocyte positive selection,
whereas ARX54–62induces antagonist activity, but not positive
selection (17). ARX54–62was unable to induce CD4/CD8 core-
ceptor down-modulation, either alone or with the help of 0.5 ng/ml
PMA (Fig. 3). All other peptides that act as TCR antagonists but
do not promote positive selection, including but not limited to
Ube1x509–517K1M, also fail to induce assisted dulling (Fig. 4).
We analyzed the correlation between the capacity of a peptide to
induce positive selection and its antagonist or assisted dulling ac-
tivity. Our data suggest that there is a positive correlation between
3Abbreviations used in this paper: FTOC, fetal thymus organ culture; ?2m, ?2-mi-
croglobulin; RTOC, reaggregation thymus organ culture.
5518TWO SIGNALS FOR THYMOCYTE POSITIVE SELECTION
assisted dulling and positive selection (r ? 0.729). Assisted dull-
ing is also more sensitive than antagonist assays in the detection of
peptides that promote positive selection (r ? 0.65). However, this
difference in sensitivity between assisted dulling and antagonism is
reduced when only strong antagonists are included in the analysis
(r ? 0.69). Therefore, assisted dulling detects peptides that pro-
mote positive selection as well as those with strong antagonist
activity. These results suggest that strong, but not weak, antagonist
activity is associated with thymocyte positive selection in the H-Y
TCR transgenic model.
Preserved pattern of TCR engagement in the presence of
suboptimal phorbol ester stimulation
We have recently demonstrated that the pattern of TCR contacts
used to induce positive selection by Ube1x509–517is different from
that required to induce negative selection by Smcy738–746(22).
Positive selection requires TCR contacts at the N terminus (posi-
tions 1 and 2) and the C terminus (positions 6, 7, and 8) of the
peptide. Negative selection does not require contacts at the N ter-
minus of the peptide, but, instead, uses contacts at central position
down-modulation. Peritoneal exudate
C57BL/6 mice were cocultured with
RAG2?/?thymocytes in the absence or
the presence of 0.5 or 1 ng/ml PMA.
After 16–20 h of coculture, thymocytes
were gently suspended, stained with
CD4- and CD8-specific mAbs, and an-
alyzed by flow cytometry. The percent-
age of cells staining brightly for both
CD4 and CD8 after each treatment is
indicated in A. Shown are the means of
duplicate cultures. The same data are
expressed as the percent dulling, which
is the number of CD4?CD8?thymo-
cytes with reduced levels of CD4/CD8
relative to the number of CD4?CD8?
thymocytes observed in untreated con-
trol samples (B). The same cocultures
described in A and B were performed in
the presence of 10 ?M Ube1x509–517,
Smcy738–746, and/or 0.5 ng/ml PMA, as
indicated (C). The percentage of cells
staining brightly for both CD4 and CD8
5519 The Journal of Immunology
4. It was therefore of interest to determine whether assisted dulling
induced by Ube1x509–517required N-terminal or central TCR con-
tacts. To address this question we used variants of the Ube1x509–517
peptide with single substitutions of original amino acids by alanine
at each position of the peptide except positions 5 and 9, which are
anchor positions for peptide binding to MHC. All variant peptides
bind H-2Db(22). As shown in Fig. 4, positions 1, 2, 6, 7, and 8
were essential for assisted dulling, whereas position 4 was not
required. As was the case for induction of positive selection in
FTOCs (22), only conservative replacement at position 1 (K to R)
preserved the functional activity of the peptide. In contrast, the
presence of amino acid that has a poor capacity to form
noncovalent bonds (M) disabled the activity of the peptide.
Thus, TCR contacts required to induce assisted dulling are same
as those required for positive selection.
Signal enhancement by suboptimal phorbol ester stimulation
does not lead to thymocyte apoptosis
The assisted coreceptor down-modulation may represent an initial
stage of positive selection that cannot be completed in vitro. Al-
ternatively, increased strength of the signal could convert a posi-
tively selecting ligand into a negatively selecting one. To test the
latter possibility, we determined whether apoptosis is induced in
thymocytes stimulated with Ube1x509–517and suboptimal PMA.
Although ?50% of H-Y thymocytes stimulated with Smcy738–746
stained positively in the TUNEL assay (Fig. 5A), cells stimulated
with combination of Ube1x509–517and PMA did not differ from
control thymocytes (Fig. 5B). Thus, the synergistic actions of PMA
and positively selecting peptide do not lead to negative selection.
Suboptimal phorbol ester treatment does not impair positive
We next wanted to determine how the presence of suboptimal
concentrations of PMA affected the ability of Ube1x509–517to in-
duce positive selection. H-Y TCR transgenic thymocytes were
treated with Ube1x509–517and PMA in suspension culture, as de-
scribed above, and were then used for RTOCs with TAP1?/?thy-
mic stromal cells. After 3 days of culture, the RTOCs were ana-
lyzed for the presence of CD4?CD8?thymocytes. Thymocytes
stimulated with B6 macrophages in the absence of added peptides
promoted selection of CD4?CD8?T cells (Fig. 6). This is prob-
ably due to the effect of endogenous self-peptides, as the same
result occurred in a similar study using dendritic cells as APCs
(23). Importantly, a significant increase in the number of
CD4?CD8?thymocytes was observed in RTOCs that used cells
treated with Ube1x509–517alone. Ube1x509–517and PMA, when
added together, acted synergistically to produce an additional in-
crease in the number of CD4?CD8?thymocytes, although the
contribution of PMA was relatively minor over that of the peptide
alone. To eliminate the effect of endogenous self-peptides and ob-
serve the isolated effects of Ube1x509–517and suboptimal PMA,
we used TAP1?/?macrophages during the primary thymocyte
culture. As expected, TAP1?/?macrophages promoted selection
of CD4?CD8?thymocytes only if loaded with Ube1x509–517(Fig.
6). PMA acted synergistically to produce an additional increase in
the number of CD4?CD8?thymocytes, albeit the contribution of
PMA was again relatively minor. Thus, suboptimal concentrations
of PMA during primary culture do not convert peptide-induced
positive selection into negative selection.
Stronger signals during positive selection are known to raise the
threshold of activation by Ag (24, 25). To determine whether co-
treatment with PMA may affect the responsiveness of selected
cells to Ag in a similar manner we tested proliferative responses of
thymocytes derived from RTOC to Smcy738–746. However, to
date, we have not observed functional responses to Ag by
CD4?CD8?thymocytes isolated from RTOCs treated either with
Ube1x509–517and PMA or with Ube1x509–517alone (data not
shown). These findings suggest that positive selection in suspen-
sion culture, although effective for phenotypic maturation, may not
be optimal for generating functionally competent cells in this ex-
Coreceptor down-modulation characterizes the initial stages of
both positive and negative selection. Hence, induction of corecep-
tor down-modulation in in vitro suspension cultures is used as a
surrogate assay for both of these processes. Under these culture
conditions, ligands that induce positive selection induce none or
very weak coreceptor down-modulation unless the signal is am-
plified in a nonspecific manner. Amplification of the signal can be
achieved by coreceptor overexpression (14). However, the require-
ment for TCR and coreceptor transgenes as well as for H-2 back-
ground that cannot induce positive selection results in a complex
FTOC and assisted dulling. Peptide variants of Smcy738–746(R4A, R6A,
and Q7A) that do not induce strong proliferation of mature T cells or
thymocyte negative selection, but bind H-2Dbwere tested for positive se-
lection in FTOC (A) and induction of assisted dulling (B). Shown are the
mean and SD of percentage of CD4?CD8?CD24lowthymocytes found
after 10 days of treatment of FTOC with 300 ?M peptide (A) or percent
dulling of thymocytes treated with 0.5 ng/ml PMA and 10 ?M of the
indicated peptides (B).
Altered antigenic epitopes induce positive selection in
itive selection and not with TCR antagonist activity. Peritoneal exudate
cells from thioglycolate-treated C57BL/6 mice were cocultured with H-Y
TCR transgenic, B10.D2, RAG2?/?thymocytes in the absence or the pres-
ence of 1 ?M Ube1x509–517(induces positive selection and TCR antagonist
activity), ARX54–62(induces TCR antagonist activity only), or Uty246–254
(induces neither positive selection nor TCR antagonist activity) and/or 0.5
ng/ml PMA, as indicated. After overnight culture, thymocytes were stained
with CD4- and CD8-specific mAbs. Shown are the mean percent dulling
and SD calculated relative to the numbers of CD4?CD8?thymocytes in
cultures treated with medium only.
Assisted coreceptor down-modulation correlates with pos-
5520TWO SIGNALS FOR THYMOCYTE POSITIVE SELECTION
and time-consuming breeding. It would be advantageous, there-
fore, if at least one breeding component could be replaced with an
externally applicable component. Obvious candidates that could
replace the need for CD8 transgene are pharmacological agents
that could strengthen TCR signaling pathways known to be in-
volved in positive selection. We showed that amplification can be
achieved by the use of suboptimal concentrations of phorbol esters.
We also found that signal amplification does not convert positive
selection into negative selection, but only makes the productive
interaction of thymocytes with positively selecting ligands visible.
We demonstrated that induction of assisted coreceptor down-mod-
ulation correlated better with positive selection than with antago-
nist activity. However, signal amplification did not significantly
improve the completion of thymocyte maturation in MHC-defi-
cient RTOC, suggesting that the efficiency of positive selection
does not depend on the degree of coreceptor down-modulation.
Our results using RTOC demonstrate that positive selection can
be split into at least two events: one early event that is MHC-
dependent but apparently cell type-independent, and a late event
that is apparently MHC-independent but cell type-dependent. Sim-
ilar conclusions have been reached in previous studies using
RTOCs, where positive selection was completed in two phases
(23). However, some minor differences can also be noted between
the two sets of results. Although, in our studies, the second stage
was apparently MHC-independent, MHC was required in some,
but not all, experimental conditions for phenotypic maturation in
studies by Yasutomo et al. (23). Some differences in experimental
protocols exist that could potentially explain different findings.
First, in contrast to Yasutomo et al. (23), we did not purify
CD69highthymocytes after the first culture. Triggering CD69 ini-
tiates signaling (26), and it is possible that the anti-CD69 treatment
used for sorting interfered with or modulated signaling in thymo-
cytes. Also, we used macrophages instead of dendritic cells. These
two cell types may differ in the levels of MHC molecules, adhe-
sion/costimulatory molecules, and adherence to the plastic. Fi-
nally, different types of MHC class I-deficient cells were used for
secondary cultures. The residual levels of MHC class I in TAP1-
deficient cells (this study) are higher than those in ?2m-deficient
cells (27). In this case it would have to be postulated that the MHC
signals in the second phase need not be strong and most likely do
not depend on the presence of specific peptides that initiated
The two-signal selection described above is reminiscent of re-
quirements for stimulation of naive cells. Two signals are required
for full T cell activation: an Ag presented by MHC molecules, and
costimulatory molecules (28). Thymic epithelial cells can readily
promote positive selection (29–34), and although there are reports
of positive selection induced by other cell types, these are always
less efficient than thymic epithelial cells (35–39). The separation of
two events required for thymocyte maturation can explain these
apparent discrepancies in the literature. MHC-dependent signal
can be delivered by any cell type, whereas thymic epithelial cells
offer an equivalent of costimulatory signal. In this scenario, the
signal provided by self-peptide/MHC complexes need not be pre-
sented by the same cell that provides the equivalent of costimula-
tory signal. Thus, the equivalent of costimulation can be provided
udate cells from thioglycolate-treated C57BL/6 mice were cocultured with
H-Y TCR transgenic, B10.D2, RAG2?/?thymocytes in the presence of 0.5
ng/ml PMA alone or in combination with 10 ?M of the original (KSNL-
NRQFL) or one of the indicated variants of Ube1x509–517. After overnight
culture, thymocytes were stained with CD4- and CD8-specific mAbs.
Shown are the mean percent dulling and SD calculated relative to the
number of CD4?CD8?thymocytes in cultures treated with medium only.
TCR contacts essential for assisted dulling. Peritoneal ex-
genic, B10.D2, RAG2?/?thymocytes cultured overnight in the absence
(MC) or the presence of Smcy738–746were analyzed for DNA breaks using
a TUNEL assay (A). The same cells were cultured in the absence or the
presence of 0.5 ng/ml PMA (line overlaps for the most part with the MC
line) or in the presence of Ube1x509–517and PMA (B).
Induction of apoptosis of H-Y thymocytes. H-Y TCR trans-
5521 The Journal of Immunology
in trans, as is the case with the classical costimulation (40). And
because costimulation of mature T cells in trans is less efficient
(41), we expect the costimulation equivalent provided by thymic
epithelial cells also to be less efficient when provided in trans.
Hence, we postulate that in all experimental settings where cell
types other than thymic epithelial cells were found to promote
positive selection, these cells provided the MHC-dependent signal,
whereas thymic epithelial cells provided costimulation equivalent
in trans. This scenario requires relative anatomical vicinity of
MHC donor cells and thymic epithelial cells. Because of the lower
efficiency of trans-delivered costimulatory signals (41), all other
cell types showed reduced efficiency in selection compared with
thymic epithelium. The nature of the epithelial cell ligand is at
present unknown, but classical costimulatory interactions (pro-
vided by CD86/CD80 molecules) have been excluded (42, 43).
We thank John Hirst for FACS analysis.
1. Jameson, S. C., K. A. Hogquist, and M. J. Bevan. 1995. Positive selection of
thymocytes. Annu. Rev. Immunol. 13:93.
2. Lucas, B., and R. N. Germain. 1996. Unexpectedly complex regulation of CD4/
CD8 coreceptor expression supports a revised model for CD4?CD8?thymocyte
differentiation. Immunity 5:461.
3. Brugnera, E., A. Bhandoola, R. Cibotti, Q. Yu, T. I. Guinter, Y. Yamashita,
S. O. Sharrow, and A. Singer. 2000. Coreceptor reversal in the thymus: signaled
CD4?8?thymocytes initially terminate CD8 transcription even when differen-
tiating into CD8?T cells. Immunity 13:59.
4. Kappler, J., N. Roehm, and P. Marrack. 1987. T cell tolerance by clonal elimi-
nation in the thymus. Cell 49:273.
5. Kisielow, P., H. Bluthmann, U. D. Stearz, M. Steimetz, and H. von Boehmer.
1988. Tolerance in T-cell-receptor transgenic mice involves deletion of nonma-
ture CD4?8?thymocytes. Nature 333:742.
tion signals delivered by Ube1x509–517. H-Y
TCR transgenic, B10.D2, RAG2?/?thymo-
cytes were cultured for 3–16 h in the presence
of wild-type or TAP1?/?macrophages loaded
with Ube1x509–517and/or 0.5 ng/ml PMA as
indicated. Thymic stromal cells were isolated
by mild trypsin digestion from neonatal
TAP1?/?thymi treated for 5 days with deox-
yguanosine to eliminate bone marrow-derived
cells. Thymocytes from suspension cultures
and neonatal stromal cells were mixed and
used to form RTOC. After 3 days of secondary
culture, thymocytes were stained with CD4-
and CD8-specific mAbs.
PMA enhances positive selec-
5522 TWO SIGNALS FOR THYMOCYTE POSITIVE SELECTION
6. Sha, W. C., C. A. Nelson, R. D. Newberry, D. M. Kranz, J. H. Russel, and Download full-text
D. Y. Loh. 1988. Positive and negative selection of an antigen receptor on T cells
in transgenic mice. Nature 336:73.
7. Swat, W., L. Ignatowicz, H. von Boehmer, and P. Kisielow. 1991. Clonal deletion
of immature CD4?8?thymocytes in suspension culture by extrathymic antigen-
presenting cells. Nature 351:150.
8. Iwabuchi, K., K. I. Nakayama, R. L. McCoy, F. Wang, T. Nishimura, S. Habu,
K. M. Murphy, and D. Y. Loh. 1992. Cellular and peptide requirements for in
vitro clonal deletion of immature thymocytes. Proc. Natl. Acad. Sci. USA
9. Tanaka, Y., C. Mamalaki, B. Stockinger, and D. Kioussis. 1993. In vitro negative
selection of ab T cell receptor transgenic thymocytes by conditionally immortal-
ized thymic cortical epithelial cell lines and dendritic cells. Eur. J. Immunol.
10. Pircher,H.,K. Brduscha,U.
R. M. Zinkernagel, H. Hengartner, B. Kyewski, and K.-P. Muller. 1993. Toler-
ance induction by clonal deletion of CD4?8?thymocytes in vitro does not re-
quire dedicated antigen-presenting cells. Eur. J. Immunol. 23:669.
11. Vukmanovic, S., S. C. Jameson, and M. J. Bevan. 1994. A thymic epithelial cell
line induces both positive and negative selection in the thymus. Int. Immunol.
12. Yasutomo, K., B. Lucas, and R. N. Germain. 2000. TCR signaling for initiation
and completion of thymocyte positive selection has distinct requirements for
ligand quality and presenting cell type. J. Immunol. 165:3015.
13. Hogquist, K. A. 2001. Signal strength in thymic selection and lineage commit-
ment. Curr. Opin. Immunol. 13:225.
14. Hogquist, K. A., A. J. Tomlinson, W. C. Kieper, M. A. McGargill, M. C. Hart,
S. Naylor, and S. C. Jameson. 1997. Identification of a naturally occurring ligand
for thymic positive selection. Immunity 6:389.
15. Santori, F. R., W. C. Kieper, S. M. Brown, Y. Lu, T. A. Neubert, K. L. Johnson,
S. Naylor, S. Vukmanovic, K. A. Hogquist, and S. C. Jameson. 2002. Rare,
structurally homologous self-peptides promote thymocyte positive selection. Im-
16. Sasada, T., Y. Ghendler, J. M. Neveu, W. S. Lane, and E. L. Reinherz. 2001. A
naturally processed mitochondrial self-peptide in complex with thymic MHC
molecules functions as a selecting ligand for a viral-specific T cell receptor.
J. Exp. Med. 194:883.
17. Santori, F. R., S. M. Brown, Y. Lu, T. A. Neubert, and S. Vukmanovic. 2001.
Cutting edge: positive selection induced by a self peptide with TCR antagonist
activity. J. Immunol. 167:6092.
18. Alberola-Ila, J., K. A. Forbush, R. Seger, E. G. Krebs, and R. M. Perlmutter.
1995. Selective requirement for MAP kinase activation in thymocyte differenti-
ation. Nature 373:620.
19. Alberola-Ila, J., K. A. Hogquist, K. A. Swan, M. J. Bevan, and R. M. Perlmutter.
1996. Positive and negative selection invoke distinct signaling pathways. J. Exp.
20. Downward, J., J. D. Graves, P. H. Warne, S. Rayter, and D. A. Cantrell. 1990.
Stimulation of p21rasupon T cell activation. Nature 346:719.
21. Nakayama, K.-i., and H. Nakauchi. 1993. Cyclosporin A inhibits the decrease of
CD4/CD8 expression induced by protein kinase C activation. Int. Immunol.
22. Santori, F. R., K. Holmberg, D. Ostrov, N. R. J. Gascoigne, and S. Vukmanovic.
2004. Distinct footprints of T cell receptor engagement with highly homologous
ligands. J. Immunol. 172:7466.
23. Yasutomo, K., C. Doyle, L. Miele, and R. N. Germain. 2000. The duration of
antigen receptor signalling determines CD4?versus CD8?T-cell lineage fate.
24. Kawai, K., and P. Ohashi. 1995. Immunological function of a defined T-cell
population tolerized to low-affinity self antigens. Nature 374:68.
25. Sebzda, E., T. M. Kundig, C. T. Thomson, K. Aoki, S.-Y. Mak, J. P. Mayer,
T. Zamborelli, S. G. Nathenson, and P. S. Ohashi. 1996. Mature T cell reactivity
altered by peptide agonist that induces positive selection. J. Exp. Med. 183:1093.
26. Marzio, R., J. Mauel, and S. Betz-Corradin. 1999. CD69 and regulation of the
immune function. Immunopharmacol. Immunotoxicol. 21:565.
27. Ljunggren, H.-G., L. Van Kaer, M. S. Sabatine, H. Auchincloss, Jr.,
S. Tonegawa, and H. L. Ploegh. 1995. MHC class I expression and CD8?T cell
development in TAP1/?2-microglobulin double mutant mice. Int. Immunol.
28. Jenkins, M. K., and R. H. Schwartz. 1987. Antigen presentation by chemically
modified splenocytes induces antigen-specific T cell unresponsiveness in vitro
and in vivo. J. Exp. Med. 165:302.
29. Fink, P. J., and M. J. Bevan. 1978. H-2 antigens of the thymus determine lym-
phocyte specificity. J. Exp. Med. 148:766.
30. Zinkernagel, R. M., G. N. Callahan, A. Althage, S. Cooper, P. A. Klein, and
J. Klein. 1978. On the thymus in the differentiation of “H-2 self recognition” by
T cells: evidence for dual recognition. J. Exp. Med. 147:882.
31. Vukmanovic, S., A. G. Grandea III, S. J. Faas, B. B. Knowles, and M. J. Bevan.
1992. Positive selection of T-lymphocytes induced by intrathymic injection of a
thymic epithelial cell line. Nature 359:729.
32. Hugo, P., J. W. Kappler, D. I. Godfrey, and P. Marrack. 1992. A cell line that can
induce thymocyte positive selection. Nature 360:679.
33. Cosgrove, D., S. H. Chan, C. Waltzinger, C. Benoist, and D. Mathis. 1992. The
thymic compartment responsible for positive selection of CD4?T cells. Int.
34. Anderson, G., J. J. T. Owen, N. C. Moore, and E. J. Jenkinson. 1994. Thymic
epithelial cells provide unique signals for positive selection of CD4?CD8?thy-
mocytes in vitro. J. Exp. Med. 179:2027.
35. Bix, M., and D. Raulet. 1992. Inefficient positive selection of T cells directed by
haematopoietic cells. Nature 359:330.
36. Pawlowski, T., J. D. Elliot, D. Y. Loh, and U. D. Stearz. 1993. Positive selection
of T lymphocytes on fibroblasts. Nature 364:642.
37. Hugo, P., J. W. Kappler, J. E. McCormack, and P. Marrack. 1993. Fibroblasts can
induce thymocyte positive selection in vivo. Proc. Natl. Acad. Sci. USA
38. Zinkernagel, R. M., and A. Althage. 1999. On the role of thymic epithelium vs.
bone marrow-derived cells in repertoire selection of T cells. Proc. Natl. Acad.
Sci. USA 96:8092.
39. Lilic, M., F. R. Santori, E. G. Nielson, A. B. Frey, and S. Vukmanovic. 2002. The
role of fibroblasts in positive selection of T cells. J. Immunol. 169:4945.
40. Jenkins, M. K., J. D. Ashwell, and R. H. Schwartz. 1988. Allogeneic non-T
spleen cells restore the responsiveness of normal T cell clones stimulated with
antigen and chemically modified antigen-presenting cells. J. Immunol. 140:3324.
41. Liu, Y., and C. A. J. Janeway. 1992. Cells that present both specific ligand and
costimulatory activity are the most efficient inducers of clonal expansion of nor-
mal CD4?T cells. Proc. Nat. Acad. Sci. USA 89:3845.
42. Vukmanovic, S., G. Stella, P. T. King, R. Dyall, K. A. Hogquist, J. T. Harty,
J. Nikolic-Zugic, and M. J. Bevan. 1994. A positively selecting thymic epithelial
line lacks costimulatory activity. J. Immunol. 152:3814.
43. Hare, K. J., J. Pongracz, E. J. Jenkinson, and G. Anderson. 2003. Modeling TCR
signaling complex formation in positive selection. J. Immunol. 171:2825.
5523 The Journal of Immunology