|
|
||||||||


* Department of Immunology and
Howard Hughes Medical Institute, University of Washington, Seattle, WA 98195
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
5 TCR
-chain (V
5)-only transgenic (Tg) mice due to chronic TCR triggering mediated by a superantigen encoded by the defective endogenous mammary tumor virus (Mtv)-8 (3, 5, 6, 7, 8, 13, 14).
The Mtv-8-encoded superantigen interacts with V
5+ TCRs and, although it fails to eliminate V
5-expressing T cells in the thymus of H-2b mice, it drives the deletion of peripheral V
5+CD4+ T cells in wild-type and V
5 Tg C57BL/6 (B6) mice (3, 7, 8). In V
5 Tg mice, the decline of V
5+CD4+ T cells is paralleled by the appearance of CD4+ T cells that express TCR
-chains other than V
5 (3, 14). The generation of these V
5–TCR
+ cells does not require the thymus (5, 8) and has been linked to peripheral RAG expression and V
to DJ
rearrangements, indicating that these novel surface TCR
-chains result from TCR gene rearrangement in the lymphoid periphery (14). For unknown reasons, postrevision T cells in V
5 Tg mice do not express V
5 on the surface despite expressing V
5 mRNA (our unpublished observation). Revised T cells display a diverse repertoire of TCR
-chains with N regions that are shorter than thymically generated sequences (13, 14). Postrevision T cells accumulate over time in unmanipulated and thymectomized V
5 Tg mice and can comprise up to 40% of peripheral CD4+ T cells, indicating that revision gives rise to functional, long lived T cells that are maintained in the periphery. Nonetheless, it remains largely unknown into which CD4+ T cell subsets revised T cells are capable of differentiating.
The random nature of TCR gene rearrangement inherently has the potential to create T cells carrying autoreactive TCR. TCR revision takes place in germinal centers (6) that could provide an environment for selecting against self-reactivity, but it is not known to what extent self-tolerance is enforced among T cells undergoing TCR revision. The potential for generating self-reactive TCR led us to hypothesize that revised T cells might be prone to convert into Treg (1, 16). However, we now report that Treg do not undergo Mtv-8 mediated peripheral deletion, are excluded from TCR revision, and are underrepresented among revised T cells. In contrast, revised T cells contain a 20-fold higher frequency of IL-17-producing Th17 T cells than is normally found in B6 mice. Thus, our data indicate that TCR revision gives cells a strong bias toward differentiating into Th17 T cells. The link between Th17 differentiation and the generation of autoreactive TCR upon peripheral revision (15, 17) might be vital for understanding the mechanisms underlying the promotion of autoimmunity by Th17 T cells (18).
| Materials and Methods |
|---|
|
|
|---|
V
5 Tg B6, non-Tg littermate control B6, C57BL/6-Tg(TcraTcrb)425Cbn/J (OT-II TCR Tg B6), Mtv-8+ V
5 Tg, and Mtv– V
5 Tg mice were all bred and maintained in specific pathogen-free facilities at the University of Washington (Seattle, WA) and used at 4–45 wk of age. V
5 Tg B6 mice carry Mtv 8, 9, 17, and 30, Mtv-8+ mice carry only Mtv-8, and Mtv– mice are negative for all known Mtv. The latter two strains are both H-2b and were derived by crossing and intercrossing V
5 Tg to wild-derived Mtv– mice (3). All experiments were performed in compliance with University of Washington Institutional Animal Care and Use Committee regulations.
Detecting and phenotyping Foxp3+ and Th17 T cells
RBC-lysed single cell suspensions from spleens were first incubated with an FcR-blocking 2G24 Ab and then surface stained with Abs specific for CD4, CD8, CD25, CD44, CD69, CD62L, CD103, GITR, ICOS, TCR
, or V
5, all from BD Biosciences or eBioscience. Subsequent intracellular staining for Foxp3-expressing cells was performed using a Foxp3 staining kit (eBioscience) in accordance with the provided manual.
To detect IL-17 producing T cells, 4 x 106 cells were transferred into 96-well plates in DMEM containing 10% FCS, antibiotics, and 50 µM 2-ME. PMA (0.3 µg/ml; Sigma-Aldrich) and Ionomycin (0.3 µg/ml; Calbiochem) were added and the samples were incubated for 40 min at 37°C. Cultures were supplemented with 7 µg/ml brefeldin A (Sigma-Aldrich) and incubated for an additional 5.5 h. Cells were washed, surface stained, fixed, permeabilized with the BD Cytofix/Cytoperm kit (BD Biosciences), and stained intracellularly with anti-IL-17A (BD Biosciences or eBioscience).
Suppression assays
Splenocytes from 18- to 23-wk-old B6 or V
5 Tg mice and 8- to 10-wk-old B6 mice were stained for CD4, CD8, and CD25 and sorted as CD4+CD25+. Cells from young B6 mice were also sorted into CD4+CD25– populations. CD25– cells (2 x 104) were plated and 2-fold dilutions of CD25+ cells (starting with 2 x 104 per well) were added along with 8 x 104 T-depleted irradiated splenocytes from B6 mice as APC. The cultures were stimulated with 2 µg/ml Con A for 72 h. [3H]Thymidine (1 µCi) was added 12 h before cell harvest.
| Results and Discussion |
|---|
|
|
|---|
5+CD4+ T cells are enriched for Foxp3-expressing T cells
V
5 Tg B6 mice express a rearranged TCR
-chain derived from a Kb/OVA-specific T cell clone. Because these mice exhibit diverse TCR
-chain gene rearrangements (3), their T cell repertoire is polyclonal and contains both CD4+ and CD8+ T cells. However, splenocytes isolated from all but very young V
5 Tg mice show an inverted CD4 to CD8 ratio due to a gradual decline in the number of peripheral CD4+ T cells (Fig. 1A). Peripheral CD8+ T cell numbers remain stable at
10 x 106 cells per spleen.
|
5 Tg mice (Fig. 1B). Interestingly, CD4+ T cells expressing endogenously rearranged V
5 TCR
-chains in B6 mice also include a large fraction of Treg such that
40% of V
5+CD4+ T cells in the 32-wk-old B6 mouse express Foxp3, whereas only
21% of V
5–CD4+ T cells express Foxp3.
To determine whether the increased frequency of Foxp3+ T cells depends on Mtv-8, we analyzed V
5 Tg mice that are negative for all known Mtv or positive solely for Mtv-8. Interestingly, although 30% of CD4+ T cells in the Mtv-8+ mouse express Foxp3, only 8% did so in the age-matched Mtv– mouse (Fig. 1C). Thus, the increased frequency of Foxp3+ cells among V
5+CD4+ T cells depends on Mtv-8.
Revised T cells do not convert into Treg and Foxp3 expression protects cells from deletion and TCR revision
As reported previously (3), Mtv-8 mediates not only the decline of V
5+CD4+ T cells but also the appearance of V
5– revised T cells that are pan-TCR
+. Interestingly, for V
5 Tg mice of all ages, Foxp3 expression is almost exclusively restricted to V
5+ cells, with only a few V
5– cells that are Foxp3+ (Fig. 1B).
To further investigate this skewed representation of Foxp3+ T cells, we determined the frequency of Foxp3-expressing T cells among CD4+V
5+ and CD4+V
5– T cells in V
5 Tg and B6 mice (Fig. 2A). In both, V
5+ T cells are skewed toward expressing Foxp3, such that 40–50% are Foxp3+ whereas only
8% of revised V
5– cells in V
5 Tg mice are Foxp3+. The latter frequency is 2.5 times lower than in V
5– CD4+ T cells from age-matched B6 mice (Fig. 2A). Thus, revised cells are not prone to differentiate into Treg. Furthermore, because <10% of Foxp3+ T cells in 18- to 45-wk-old V
5 Tg mice are V
5– while up to 50% of Foxp3– cells are V
5–, the data also indicate that cells expressing Foxp3 are precluded from revising their TCR (Fig. 2B). Whether the few V
5–Foxp3+ T cells originate from Foxp3+ cells that have undergone TCR revision or from Foxp3– cells that have revised their TCR and subsequently differentiated into Foxp3-expressing cells is not known.
|
5+ T cells is due to the expansion of Treg or to differential sensitivity to Mtv-8-mediated deletion, we analyzed V
5 Tg and B6 mice aged 4–42 wk and calculated the absolute numbers of splenic CD4+Foxp3+ and CD4+Foxp3– cells. In contrast to B6 mice, the number of CD4+Foxp3– cells decreases in V
5 Tg mice with age (Fig. 2C), whereas the number of CD4+Foxp3+ cells in V
5 Tg mice remains relatively stable over the same period (Fig. 2D). This suggests that Foxp3+ and Foxp3– cells respond differently to chronic peripheral stimulation, such that cells lacking Foxp3 are deleted or forced to revise their TCR while, in contrast, CD4+V
5+ cells expressing Foxp3 do not undergo either of these fates because their number and TCR
remain unchanged. Moreover, our data indicate that TCR revision does not generate Treg. The fraction of Treg among peripheral CD4+ T cells increases with age (19). The reasons for this increase are unknown but could involve processes similar to those described above, such that some Foxp3– T cells are deleted from the repertoire due to reactivity to self-Ag, whereas Foxp3+ T cells are spared from deletion to increase in prevalence over time. Chronic superantigen stimulation does not alter the phenotype or function of Treg
Having shown that Treg are spared from superantigen-mediated deletion and TCR revision, we analyzed whether chronic recognition of superantigen alters Treg phenotype or function. Sorted CD25+CD4+ T cells from old V
5 Tg and from young and old B6 mice exhibited a comparable degree of suppression; thus, chronic superantigen recognition does not influence Treg suppressive capacity (Fig. 2E). Moreover, phenotyping V
5+ or V
5– CD4+Foxp3+ T cells isolated from spleens of 45-wk-old B6 mice for expression levels of ICOS, GITR, CD25, CD44, CD62L, and CD69 did not reveal significant differences, although a slight increase in the fraction of CD103+ cells among V
5+Foxp3+CD4+ T cells was noted (data not shown).
V
5 Tg mice have an elevated frequency of Th17 T cells that carry revised TCR
We observed that PMA/Ionomycin-stimulated (or anti-CD3 plus anti-CD28-stimulated) CD4+ splenocytes from aged V
5 Tg mice contain a much higher frequency of IL-17 producers than wild-type B6 mice (Fig. 3A and data not shown). Both young and aged V
5 Tg mice contain IL-17-producing T cells at 10-fold higher frequencies than do age-matched B6 mice, and these Th17 lineage cells express normal levels of IL-17 but no IFN-
(data not shown). Thus,
1% of total CD4+ T cells in young V
5 Tg mice and
3.5% of CD4+ T cells in old V
5 Tg mice make IL-17 (Fig. 3B). Interestingly, although CD4+IL-17+ T cells from young and old V
5 Tg express similar surface levels of pan-TCR
(Fig. 3C, left panel), they differ in their levels of surface V
5. Thus, IL-17+ T cells in young V
5 Tg mice express V
5, whereas in old V
5 Tg mice, most IL-17-producing cells are V
5– (Fig. 3C, right panel). These data show that while young V
5 Tg mice already contain an elevated frequency of V
5+ Th17 T cells, these are outnumbered in older mice by cells that have a revised V
5– TCR. Considering that, on average, 25% of total CD4+ T cells in 18-to 42-wk-old V
5 Tg mice are V
5– (Fig. 2B) and that
3% Th17 cells exist among total CD4+ T cells (Fig. 3B), this means that 12% of revised T cells are Th17 lineage cells. Compared with the average 0.4–0.8% IL-17+ of total CD4+ T cells in naive B6 mice (Fig. 3B), this is about a 20-fold bias toward the Th17 lineage.
|
5 repertoire is intrinsically prone to give rise to Th17 T cells. The 40-wk-old V
5 Tg Mtv– mouse has a frequency of IL-17+ T cells similar to that observed in age-matched B6 mice (Fig. 3, compare B with D). In contrast,
14% of V
5–CD4+ T cells in the Mtv-8+ animal produce IL-17 (Fig. 3D). Thus, the expression of V
5 by itself does not intrinsically bias cells toward the Th17 lineage but, instead, Mtv-8 promotes TCR revision and differentiation into IL-17+CD4+ T cells.
In our model, it is possible that chronic superantigen stimulation expands an already existing population of Th17 T cells and renders them highly susceptible to TCR revision. Alternatively, superantigen-driven TCR revision may bias cells toward Th17 differentiation, perhaps through the generation of autoreactive TCR. In fact, previous work suggests that self-Ag recognition can support the differentiation of CD4+ T cells into Th17 cells (20, 21). For unknown reasons, T cells in OT-II TCR
(V
2V
5) Tg B6 mice do not undergo TCR revision despite the fact that this mouse line expresses both V
5 and Mtv8. Interestingly, the frequency of CD4+IL-17+ T cells in these mice is comparable to that in B6 mice (data not shown), suggesting a correlation between elevated numbers of Th17 cells and TCR revision, not solely the expression of both V
5 and Mtv-8.
It is of interest that while Th17 cells play an important role in conferring protection from Citrobacter rodentium (18), V
5 Tg mice are unusually sensitive to this pathogen (22), possibly due to an overzealous Th17 response. In addition to conferring protection from some pathogens, the Th17 population has also been shown to be an important component of autoimmunity. Our work establishes a relationship between TCR revision and the generation of Th17 cells, whereas other reports suggest a connection between TCR revision and autoimmunity (15, 17).
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 This work was supported by the Howard Hughes Medical Institute, National Institutes of Health Grants AI19335 (to M.J.B.) and AG13078 (to P.J.F.), and a fellowship from the German Academic Exchange Service (DAAD) (to D.Z.). ![]()
2 Address correspondence and reprint requests to Dr. Pamela J. Fink, Department of Immunology, University of Washington, Box 357650, Seattle, WA 98195. E-mail address: pfink{at}u.washington.edu ![]()
3 Abbreviations used in this paper: Treg, regulatory T cell; B6, C57BL/6; Mtv, mammary tumor virus; Tg, transgenic; V
5, V
5 TCR
-chain. ![]()
Received for publication August 6, 2007. Accepted for publication August 23, 2007.
| References |
|---|
|
|
|---|
5+ cells. J. Immunol. 152: 4270-4281. [Abstract]
5 among CD4+ and CD8+ T cells. J. Exp. Med. 176: 1733-1738.
locus secondary rearrangement: role in tolerance induction. J. Immunol. 168: 3259-3265.
repertoire. J. Immunol. 165: 6902-6907. This article has been cited by other articles:
![]() |
M. Mahic, S. Yaqub, T. Bryn, K. Henjum, D. M. Eide, K. M. Torgersen, E. M. Aandahl, and K. Tasken Differentiation of naive CD4+ T cells into CD4+CD25+FOXP3+ regulatory T cells by continuous antigen stimulation J. Leukoc. Biol., May 1, 2008; 83(5): 1111 - 1117. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |