| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Division of Molecular Immunology, The National Institute for Medical Research, Mill Hill, London, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
6 wk, despite the presence of
selecting MHC molecules. This explains the failure to accumulate high
numbers of peripheral T cells and suggests that the MHC-bound ligand(s)
responsible for initiating survival signals is limiting for the
selection and maintenance of A18 transgenic CD4 T
cells. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We have developed a transgenic TCR (A18 TCRtg) model to study thymic selection of cells expressing a TCR specific for a peptide derived from a natural self Ag in the mouse; this is the serum protein complement C5, recognized in the context of MHC class II H2-Ek (13). Thymocytes in these A18 TCRtg mice are selected into the CD4 lineage in C5neg H-2k mice and are deleted at the late DP stage in C5pos H-2k mice. In the periphery, however, only very few CD4 T cells can be detected. Generation of CD4 SP thymocytes in A18 TCRtg mice is comparable with normal mice, but does not show the substantial skewing into the CD4 compartment reported for some other CD4 TCR transgenic mice (14, 15). While overselection into the CD4 lineage is not necessarily a feature of transgenic mice carrying MHC class II-restricted TCRs, CD4 T cells appear to accumulate in the periphery of these mice, even on a Rag-/- background (Refs. 4 and 16; and P. M. Allen, personal communication). In the A18 mice, however, the number of peripheral CD4 T cells falls short of what is found in other transgenic strains.
In this paper, we address the question of whether the paucity of peripheral T cells is the consequence of abnormal selection processes in the thymus. Two possible scenarios could apply. One is that the A18 TCR avidity for H2-Ek and the positively selecting ligand(s) may be too low and the A18 TCR is therefore not efficiently selected. This could cause defective final maturation at the CD4 SP stage and compromise the export of CD4 T cells into the periphery. Alternatively, the A18 TCR could have too high an avidity for the selecting MHC/ligand complexes and therefore be on the brink of negative selection, from which only few CD4 T cells escape.
Previous data obtained from the analysis of F1 mice generated by breeding A18 Ragneg TCR mice with BM3 Ragneg TCR mice are compatible with both scenarios. In these F1 mice, constitutively expressing both the MHC class II-restricted A18 TCR and an MHC class I-restricted, H-2Kb-specific TCR, we observed a substantial increase in the number of CD4 cells in thymus and periphery (17). Thus, in a low avidity scenario for the A18 TCR, the BM3 TCR could have mediated positive selection. In a high avidity scenario a reduction in expression levels for the A18 TCR due to expression of a second receptor could have avoided negative selection. The results presented in this paper indicate that the A18 TCR is appropriately selected, but not overselected like some MHC class II-restricted transgenic receptors. The CD4 SP thymocytes that are generated are mature and efficiently exported into the periphery. Instead, the lack of CD4 T cell accumulation in lymphoid organs is due to the short life span of these cells. We discuss these findings in conjunction with the recent observations concerning peripheral T cell selection and survival.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Mice of strain A/J and A18 TCRtg Rag1neg C5neg (on an A/J H-2a background) were kept in conventional animal facilities at the National Institute for Medical Research, Mill Hill. A18 TCRtg Rag1neg C5neg mice were bred to Rag1neg C5negSWR (H-2q) mice to generate H-2k/q coexpressing mice. As controls, A18 TCRtg Rag1neg C5neg mice heterozygote for the TCR were generated by crossing them to Rag1neg C5negA/J mice.
Flow cytometry and monoclonal Abs
Analytical flow cytometry was conducted using a FACScan (Becton
Dickinson, Mountain View, CA), and the data were processed using
Cellquest software (Becton Dickinson). Three-color stainings were
performed with FITC-, phycoerythrin, and biotin-conjugated mAbs
followed by streptavidin Red670 (Life Technologies, Paisley, U.K.).
Anti-CD4 phycoerythrin (H129.19) was purchased from PharMingen (San
Diego, CA) and anti-CD8 (YTS 169.4), anti-TCR Vß8.3 (7G8.2),
and anti-HSA (YBM5.10) were conjugated with biotin or FITC using
standard procedures. F3 mAb was used as supernatant and detected with
biotinylated anti-rat IgM (Dianova, Hamburg, Germany). Cells were
preincubated with unlabeled mAb to Fc
RII/III (2.4G2) to minimize
unspecific staining. All stainings were performed on ice and washed
with washing buffer (PBS, 2% FCS, 0.1% azide). For DNA analysis,
single-cell suspensions were first stained with anti-CD4 and
anti-CD8 mAb. 7-Amino-actinomycin D (7-AAD; Sigma, St.
Louis, MO) was then added in washing buffer containing 0.3%
saponin (Sigma) for 30 min at room temperature, and the samples were
analyzed immediately after using linear settings for fluorescence 3 to
assess 7-AAD staining.
BrdU labeling and detection
Mice were injected i.p., twice within 4 h, with 1 mg BrdU (Sigma) in PBS for analysis of the selection kinetics in the thymus. For continuous BrdU labeling, mice received one i.p. injection with 1 mg BrdU in PBS and then were given 0.8 mg/ml BrdU in the drinking water, which was changed every three days. Single-cell suspensions were stained with anti-CD4 and anti-CD8 mAb, resuspended in 25 µl of ice cold 0.15 M NaCl, and fixed by the dropwise addition of ice cold 95% ethanol for 30 min on ice. After washing with PBS, the samples were fixed in 100 µl PBS/1% paraformaldehyde/0.01% Tween for 30 min at room temperature followed by 30 min on ice and washed in PBS. Subsequently, the samples were treated with DNaseI (0.15 M NaCl, 4.2 mM MgCl2, 10 mM HCl, 5 µg/ml DNaseI (Boehringer, Mannheim, Germany); 200 µl/sample) for 30 min at 37 C. After washing with PBS, the cells were resuspended in 45 µl of PBS/5% FCS/0.5% Tween, and 5 µl of anti-BrdU mAb (Becton Dickinson) was added for 30 min at room temperature. After a final washing, cells were analyzed on the flow cytometer.
Dexamethasone treatment
Mice were injected twice i.p. with 1 mg dexamethasone (Sigma) in 250 ml PBS within 24 h and analyzed 24 h after the second injection.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The percentage of CD4 SP thymocytes in A18 TCRtg mice on the
positively selecting H-2a, C5neg background is
9%, and thymic cellularity is not drastically different from
nontransgenic AJ mice. In the periphery, however, very few CD4 T cells
can be detected (Table I
). Spleens from
A18 TCRtg mice have a low cellularity, 4 x 106,
and the percentage of CD4 T cells is 3-fold lower than in A/J mice.
In lymph nodes, the percentage of CD4 T cells is only slightly lower
than in A/J mice, but the cellularity is reduced 50-fold.
|
During positive selection into the CD4 lineage, thymocytes
down-regulate the differentiation marker F3Ag. This down-regulation
occurs in the DP stage and is directly correlated to the efficacy of
positive selection; i.e., the more pronounced the skewing into the CD4
compartment, the more pronounced the down-regulation of F3Ag at the DP
stage (18). We stained thymocytes from wild-type A/J and A18 TCRtg mice
with anti-CD4, anti-CD8, and F3 mAb (Fig. 1). Expression of the tgTCR mediates
selection into the CD4 compartment, comprising 10% in A/J and A18
TCRtg mice. Gating on mature F3Agneg thymocytes shows the
enrichment in SP and positively selected
CD4lowCD8low DP cells (19). No significant
differences in wild-type and A18 TCRtg mice were observed, indicating
that there is no overselection into the CD4 lineage for this TCR.
|
). Trypsinized thymocytes
now showed a distinct CD4 SP population, with a mean fluorescence
intensity (MFI) value for CD8 identical to the DN population (MFI,
23.2) compared with an MFI value of 56 in the PBS-treated control CD4
SP population. These results demonstrate that A18 CD4 SP
thymocytes have switched off the CD8 gene and can therefore be regarded
as fully selected and mature. The low level of CD8 expression ovserved
is most probably a consequence of the increase in selection kinetics
(see below) due to the presence of a constitutively rearranged
transgenic TCR.
|
). As expected,
thymi from treated mice contained only immature DN (TCRlow
and predominantly HSAhigh) and mature CD4 SP cells
(TCRhigh and HSAlow), while the bulk of DP
thymocytes was eliminated due to dexamethasone-induced apoptosis.
|
(21). Reduction of the selecting MHC ligand leads to decreased positive selection of A18 TCRtg CD4 SP thymocytes
The following experiment was designed to address the question of
whether a proportion of A18 thymocytes is lost due to high avidity
interactions during positive selection, resulting instead in negative
selection. A18 TCRtg mice were crossed with Rag1-/-
H-2q mice, which are also C5 negative. The H-2q
background is neutral and nonselecting for A18 CD4 T cells, and mature
A18 T cells are not activated by class II-bearing APC from the thymus
or periphery of H-2q mice (data not shown). In the
resulting F1 generation, H-2k expression is
reduced due to codominant expression of the MHC H-2q
allele. In case the A18 thymocytes are on the brink of negative
selection, reduction of one of the ligands participating in positive
selection would be expected to lead to an increase in the number of
thymic CD4 SP cells. In contrast, if the A18 TCR is normally weakly
positively selected, the reduction in selection ligands should lead to
a further reduction in CD4 SP cells. Figure 4 shows that the latter is the case. In
an H-2k/q F1, the amount of CD4 SP cells is
reduced nearly threefold compared with an A18 TCRtg heterozygous
H-2k/k mouse (2.5 to 6.4%, respectively). However, no
significant changes in the peripheral CD4 compartments are observed,
and CD4 spleen cells from both mice respond identically to stimulation
with the antigenic peptide (data not shown). Interestingly, a
significant population of CD8 T cells are generated, in agreement with
previous findings reporting that the lack of CD4 selection can result
in an increased selection into the CD8 compartment (5). These results
suggest that the paucity of peripheral CD4 T cells in A18 TCRtg mice is
not due to negative selection of thymocytes.
|
To investigate the kinetics of CD4 T cell production in the
thymus, mice were injected with the thymidine analogue bromodeoxyuracil
(BrdU), and BrdU incorporation in the different thymic populations was
chased during a 4-day period. BrdU is mainly incorporated during the
proliferative phase at the DN/DP transition so that newly generated and
selected thymocytes can be traced with an anti-BrdU Ab. Table II shows the results of two experiments.
Within 2 days, an average of 21% of CD4 thymocytes in A18 TCRtg mice
were BrdUpos, compared with only 6% in nontransgenic mice.
On day 4, the CD4 SP compartment of nontransgenic mice contained 17%
of recently selected cells, compared with 38% in the A18 thymus. These
data demonstrate that A18 CD4 SP thymocytes are generated very fast,
first appearing within 2 days, whereas A/J thymocytes need at least 4
days to generate an equivalent percentage of newly selected CD4 SP
thymocytes. Constitutive expression of a rearranged TCR chains on most
A18 thymocytes probably accelerates the positive selection process at
the DP stage. We conclude that the DP/SP transition is not a limiting
step in the generation of A18 T cells.
|
To compare the export of CD4 cells from thymus to periphery in A18
TCRtg and nontransgenic A/J mice, we gave them BrdU in the drinking
water for 8.5 days to ensure continuous labeling of cells. After the
labeling period, 95% of the A18 and 70% of the A/J CD4 SP thymocytes
were BrdUpos (data not shown). This suggests that the
mature nondividing and therefore BrdUneg CD4 SP thymocytes,
present before the addition of BrdU, must either have left the thymus
or have died during the 8.5-day labeling period. To discriminate
between these two possibilities, BrdU-labeled CD4 T cells were traced
in the lymph nodes. Figure 5 shows that
18% of the CD4 T cells in A18 TCRtg mice have incorporated BrdU,
compared with 4% in A/J lymph nodes. BrdU incorporation in CD4 T cells
is not due to proliferation in this population, since analysis with the
DNA-binding dye 7-AAD (Fig. 6) shows that
in A18 TCRtg mice neither CD4 SP thymocytes nor CD4 T cells in lymph
nodes contain a significant number of cells in cycle. Nontransgenic A/J
mice, on the other hand, have some cycling CD4 cells in their lymph
nodes, as would be expected in mice kept in a conventional animal house
facility. Taken together, these results imply that the paucity of
peripheral A18 CD4 T cells is not due to excessive cell death at the
CD4 SP stage nor to inefficient export from the thymus. Instead, the
high proportion of recent thymic emigrants found 8 days after BrdU
labeling suggests that there is a high turnover of A18 CD4 T cells in
the periphery.
|
|
Since the low numbers of peripheral A18 CD4 T cells are not due to
inefficient thymic positive selection and export, other mechanisms must
be responsible for this phenotype. A18 CD4 T cells are phenotypically
naive; i.e., they are CD44low, CD25low,
CD69neg, CD62Lhi, and CD45RBhi
(data not shown), which makes it unlikely that are undergoing deletion
after interaction with an unknown ligand in the periphery. To address
the question of their life span, A/J and A18 TCRtg mice were
thymectomized, and the number of CD4 T cells in the blood was
determined over time (Fig. 7). During the
7 wk of the experiment, no reduction of A/J CD4 T cells could be
observed, although there was some degree of experimental variation on
the different time points. On the other hand, more than half of the
thymectomized A18 mice showed a progressive loss of peripheral CD4
cells, suggesting a t1/2 for these T cells
between 4 and 6 wk. Although there is variation between mice, the trend
in A18 TCRtg mice is for a decrease in CD4 T cells over time. The
number of A18 CD4 T cells observed in the blood correlated with the
numbers found in spleen and lymph nodes analyzed at the end of the 7-wk
period (data not shown). We conclude that the most likely reason for
the low numbers of CD4 T cells found in the periphery of A18 TCRtg mice
is their reduced life span following export from the thymus.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The A18 TCRtg mice are an exception to the so far-described MHC class
II-restricted TCRtg mice. They are selected reasonably well into the
CD4 lineage, comprising 10% of the thymus, and the thymus
cellularity is comparable to nontransgenic controls, but the amount of
peripheral tg CD4 T cells is very low. On average, spleen and lymph
nodes of the A18 mice contain 5 x 105 CD4 T cells.
In contrast, DO-11-10 mice on a scid background have
1.5 x 107 CD4 T cells in the spleen and lymph
nodes (M. Jenkins, personal communication). Our data rule out
gross abnormalities in the thymic selection processes as a reason for
the low numbers of peripheral T cells. One concern was the possibility
that the relatively low numbers of CD4 SP thymocytes generated in
comparison with other CD4 TCR transgenic strains reflected a degree of
negative selection due to a too high avidity of the transgenic TCR for
positively selecting ligands. While there is no direct way to
distinguish between cell loss due to failed positive selection or
negative selection, the data presented in this paper, as well as
previously published observations, make it very unlikely that this is
the case. 1) TUNEL staining of thymus from C5-negative A18 TCRtg mice
revealed very few apoptotic cells in contrast to thymus from
C5-positive A18 TCRtg mice (28). 2) Negative selection in the presence
of the self Ag C5 does not take place until the very late DP stage,
although all DP thymoctes and a proportion of DN thymocytes express
TCR, suggesting that the A18 TCR interaction with its ligand is of
relatively low avidity (13), (iii) in this paper, we show that lowering
the avidity of interaction further by reducing expression of the MHC
ligand results in a further decrease in CD4 SP thymocytes. The opposite
effect would have been expected, had the relatively low number of CD4
SP thymocytes been the consequence of negative selection due to avidity
interactions that were too high.
The data are compatible with the assumption that the avidity of TCR/MHC ligand interactions for this transgenic TCR are on the low side and therefore might not favor extensive production of CD4 SP. However, this conclusion seems somewhat at odds with the high sensitivity of functional activation seen in mature T cells (or CD4 SP thymocytes) with nanogram amounts of Ag (13, 21). An alternative explanation could lie in the kinetics of CD4 SP production. A18 SP cells are generated very rapidly, so that within 2 to 3 days, 30 to 45% of the CD4 SP population is derived from cycling DP precursors as defined by BrdU labeling, whereas in nontransgenic controls only a small number of CD4 SP is detected 2 days after BrdU labeling. While in nontransgenic mice, 20 to 30% of the DP pool are still labeled on day 2 after BrdU injection, A18 DP thymocytes have lost the BrdU label nearly completely at that time point. This suggests that the DP stage is very short lived, so that the low cell output from positive selection might be due to a restricted "time window" at the DP stage. In theory, the presence of a single TCR specificity on a positively selecting background should allow 100% positive selection. This is not the case, however, even in TCRtg mice, which generate SP thymocytes with higher efficiency than the A18 (29, 30). One reason for this could be the availability of selecting stromal microenvironments, which were shown to be rate limiting for positive selection, because during thymocyte selection each thymocyte appears to engage with only one rather than multiple stromal cells (31). Whereas in transgenic mice on a Rag-/- background there should not be any competition with other T cells, one commodity that might be limiting is the selecting ligand(s), so that the kinetic feasibility of engaging stromal cells displaying the relevant ligand may be low for A18 thymocytes.
A related problem may face peripheral CD4 T cells. We have excluded the possibility that the low number of peripheral T cells is due to a defect in export from the thymus. Instead, the relatively high percentage (18%) of BrdU-labeled CD4 T cells in A18 TCRtg mice, compared with 4% in A/J controls, suggests that a high turnover of CD4 T cells must take place in A18 lymph nodes. Analysis of peripheral A18 T cells in thymectomized mice supported this assumption, because in half of the thymectomized mice a clear reduction of peripheral CD4 T cells could be observed during a 7-wk time course. A number of recent reports provided evidence that an indefinite life span for naive T cells is not guaranteed once the T cells have left the thymus. Although the life spans vary from one report to another, the general consensus is that for long term survival of naive T cells, continuous engagement of the restricting MHC ligand is required (8, 9, 10, 11, 12). In the A18 TCRtg model, however, peripheral T cell survival is compromised even in the presence of the restricting MHC molecule.
One could envisage that the rapid positive selection of A18 thymocytes
fails to provide some crucial signals in the thymus necessary to make
them responsive to survival signals in the periphery. The requirement
for a thymus in generation of long-lived CD4 T cells even after
completion of positive selection has been shown previously (32).
However, the fact that A18 SP CD4 T cells are corticoid resistant and
functionally mature argues against this hypothesis. Akkaraju et al.
observed that spleen and lymph nodes of transgenic mice with hen egg
lysozyme-specific TCRs also had markedly reduced cellularity on a
Rag-/- background, compared with a Rag+/+
background, and suggest that these mice may not have a normal splenic
architecture, including B cells, which might compromise their survival
(33). Similarly, The A18 mice have more T cells on a Rag+/+
background, but it is difficult to interpret these higher numbers of
CD4 T cells without a clonotypic Ab for their identification because of
the presence of cells with endogenously rearranged TCR or two TCR. We
are currently back-crossing our mice to TCR 
-chain knock-out mice to
investigate the involvement of B cells in T cell survival. However,
abnormally low T cell numbers do not seem to be a general problem in
TCRtg mice on Rag-/- or scid backgrounds, as
was mentioned earlier.
Given that it is unlikely that the MHC interaction necessary for survival of peripheral T cells depends on recognition of "empty" MHC molecules, this could indicate that, in the periphery as well, the availability of an MHC-bound ligand might be limiting for the A18 TCR. It is possible that the A18 TCR is very restricted in its Ag recognition pattern, i.e., might not be as promiscuous as other tg TCR described. We have never observed any cross-reactivity with potential Ags present in a variety of other MHC haplotypes such as H-2b, H-2d, H-2q, whereas other TCR expressed in transgenic mice have known cross-reactivities (34, 35). The nature of MHC ligands mediating survival signals are not known, but it is logical to assume that self peptides must be of importance. Whether the self peptides that are responsible for peripheral selection are the same ones mediating thymic selection is another unknown variable at present.
It is worth mentioning that C5 in a C5-deficient mouse has no autologous counterpart, due to the complete absence of this protein from the repertoire. In contrast, other transgenic TCR with specificity for molecules like hen egg lysozyme, OVA, or cytochrome c, may in some way be influenced by the presence of the autologous mouse proteins. While presentation of these is not likely to lead to activation or negative selection, it is conceivable that peptides derived from such autologous proteins could yield alternative ligands, which are known to give partial activation signals that can have both beneficial and harmful effects on peripheral T cells (36) In addition, certain cytokines have been reported to promote Ag-independent proliferation of naive T cells; this apparently does not result in up-regulation of standard activation markers (37, 38). Although mature naive CD4 T cells in the A18 TCRtg mice do not appear to cycle, there is considerable variability in the numbers of CD4 T cells in individual mice. Given our present lack of knowledge concerning the identity of ligands responsible for providing survival signals via the TCR, it will be important to determine whether cytokines, perhaps secreted by cells of the innate immune system in response to environmental stimuli, could play a role in promoting T cell survival.
|
Footnotes |
|---|
2 Abbreviations used in this paper: DP, double positive; SP, single positive; 7-ADD, 7-amino acid actinomycin D; HSA, heat stable Ag.
Received for publication April 8, 1998. Accepted for publication June 17, 1998.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
/ß T cell receptor-expressing thymocytes in transgenic mice. J Exp Med 175:1013.This article has been cited by other articles:
|
|
 |
|
  L. Kuffova, M. Netukova, L. Duncan, A. Porter, B. Stockinger, and J. V. Forrester Cross Presentation of Antigen on MHC Class II via the Draining Lymph Node after Corneal Transplantation in Mice J. Immunol., February 1, 2008; 180(3): 1353 - 1361. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Kassiotis, R. Zamoyska, and B. Stockinger Involvement of Avidity for Major Histocompatibility Complex in Homeostasis of Naive and Memory T Cells J. Exp. Med., April 21, 2003; 197(8): 1007 - 1016. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Viret and C. A. Janeway Jr. Self-Specific MHC Class II-Restricted CD4-CD8- T Cells That Escape Deletion and Lack Regulatory Activity J. Immunol., January 1, 2003; 170(1): 201 - 209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Viret, X. He, and C. A. Janeway Jr. On the Self-Referential Nature of Naive MHC Class II-Restricted T Cells J. Immunol., December 1, 2000; 165(11): 6183 - 6192. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. B. Reome, D. S. Johnston, B. K. Helmich, T. M. Morgan, N. Dutton-Swain, and R. W. Dutton The Effects of Prolonged Administration of 5-Bromodeoxyuridine on Cells of the Immune System J. Immunol., October 15, 2000; 165(8): 4226 - 4230. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Ferreira, T. Barthlott, S. Garcia, R. Zamoyska, and B. Stockinger Differential Survival of Naive CD4 and CD8 T Cells J. Immunol., October 1, 2000; 165(7): 3689 - 3694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Murali-Krishna, L. L. Lau, S. Sambhara, F. Lemonnier, J. Altman, and R. Ahmed Persistence of Memory CD8 T Cells in MHC Class I-Deficient Mice Science, November 12, 1999; 286(5443): 1377 - 1381. [Abstract] [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
