A niche for peripheral T cells in the thymus

Peripheral T cell re-entry into the thymus under pathological conditions

Whether mature peripheral T cells can migrate back into the thymus has been a subject of debate. In 1992, following the study of Rouse and co-workers¹³, Sprent and co-workers¹⁴ showed that the majority of thymic cells in scid mice injected earlier with 200 million normal peripheral T cells were derived from donor TCR⁺ T cells. By performing parabiosis between lymphopenic pre-TCR (pT)-deficient and normal mice, we confirmed that mature peripheral T cells can migrate back into a lymphopenic thymus and represent there a significant fraction of the thymic SP cells.^{15, 16} Interestingly, as found in pT-deficient mice, a high SP/DP ratio is observed in other mouse models displaying a partial block in T cell production and concomitant peripheral lymphopenia. This is the case for mice deficient for genes involved in T cell development,¹⁷ mice in which genes encoding molecules expressed by thymic epithelial cells supporting T cell maturation have been deleted¹⁸ and mouse models of T cell lymphopenia.^{16, 19} The SP/DP ratio is also elevated in the cervical thymi of normal mice.^{1, 20} In all these cases, rather than being the result of increased survival, increased efficiency of positive selection or by cell division of SP cells, a considerable fraction of SP cells may actually represent mature recirculating T cells.

The presence of mature T cells migrating back into the thymus can also be observed under pathologic conditions. For instance, in myasthenia gravis, autoreactive T cells specific for the acetylcholine receptor are found in high numbers in the thymus. Here, they cooperate with B cells to mount an autoimmune response characterized by germinal center formation inside the thymus.²¹

Peripheral T cell re-entry into the thymus of normal animals

As T cell migration into the thymus of normal animals occurs at a very low frequency, there are only few published studies on this phenomenon.^{12, 22, 23, 24, 25} Recently, several groups reinvestigated this question using new animal models. To get insights into recent thymic emigrant biology, Fink and co-workers took advantage of RAG2p-GFP transgenic mice generated in M. Nussenzweig's laboratory.²⁶ In these mice, the green fluorescent protein (GFP) is expressed under the control of the Rag2 gene promoter. The Rag proteins are transiently expressed by thymocytes at the DN stage and again at the DP stage; however, as the half-life of the GFP protein is long, in these mice, recent thymic emigrants are GFP⁺ and only progressively lose GFP expression following their migration into the peripheral lymphoid organs. Using these mice, it was revealed that recent thymic emigrants continue to mature in the periphery to become normal peripheral T cell residents as they increase Qa2 and decrease heat stable antigen (HSA) expression.⁸ Surprisingly, in the thymus, some GFP^- cells were observed within the SP fraction and some of them were characterized as mature T cells that apparently had migrated back into the thymus from the periphery. Analysis of thymi from young adult Rag2p-GFP mice revealed that around 5% of SP cells in the thymus were recirculating peripheral lymphocytes, a proportion that increases with age.^{4, 27} Using different protocols, such as adoptive cell transfer and in vivo CFSE labeling, we estimated that, on average, about 1% of thymic SP cells are recirculating peripheral T cells in normal animals.²⁸ As we used approaches in which peripheral cells were positively identified, it may be possible that we underestimated the phenomenon if a proportion of peripheral T cells in the thymus have a slow turnover. In contrast, the use of RAG2p-GFP mice might lead to an overestimate of peripheral T cell recirculation due to an early loss of GFP reporter expression in newly generated SP thymocytes before emigration from the thymus. The overestimate may have been further augmented because and T cells had not been discriminated among SP cells.²⁷ However, taking this into consideration, we estimate that irrespective of thymus size, genotype or experimental setting, the thymus can accommodate about 10⁵ recirculating peripheral T cells. Thus, there is a small niche for peripheral T cells in the thymus.^{28, 29}

Interestingly, more CD4⁺ than CD8⁺ peripheral T cells were found within the thymus. Confirming our earlier report²⁸ and as shown in Figure 1b, the CD4/CD8 ratio of recirculating CFSE⁺CD3⁺ T cells in the thymus of normal mice was 6.52.2%, whereas the ratio was 1.80.1% for the same CFSE⁺CD3⁺ population in the spleen (n=3). Among CD4⁺ SP T cells, we detected an increased proportion of CD25⁺ cells in the thymus (23.43.3% of the thymic CFSE⁺CD4⁺ cells were CD25^+, whereas there were only 6.30.4% in the spleen). Thus, regulatory T cells recirculate very efficiently into the thymus. This is particularly interesting because regulatory T cells are not as motile as naive T cells when considering their migration within secondary lymphoid organs as determined by parabiosis experiments.³⁰

Localization and migration of peripheral T cells in the thymus

Depending on their cortical or medullary location, TECs express different keratin isoforms, cytokeratin-18 or cytokeratin-5, respectively. Using the difference in cytokeratin-5 and cytokeratin-18 expression to delineate the cortico-medullary junction, we confirmed that recirculating T cells were found predominantly in the medulla (Figure 2).^{14, 28} Thus, the SP T cell pool that resides in the medulla represents a mixed population constituting newly generated cortex-derived thymocytes and recirculating peripheral T cells. As for the thymus discussed here, the bone marrow, as well as being the site of B cell lymphopoiesis, has long been known as a site of mature B lymphocyte residence, in particular of memory B cells. Recently, Sapoznikov and colleagues provided a direct dissection of the mature B cell niche in the bone marrow.³¹ Using bone marrow chimera and ablation experiments, they demonstrated that a resident bone marrow DC population was responsible for the maintenance of mature B cells at this site. The function of the DCs was to produce macrophage 'migration inhibitory factor', thereby providing survival signals to the mature B cells. Since 'B lymphocyte-activating factor' is the main factor mediating B cell survival in the spleen, these data point toward a major difference between B cell survival in the spleen and that of mature B cells in the bone marrow. Studies characterizing the essential properties of the peripheral T cell niche in the thymus remain to be performed.

Figure 2.

Mature peripheral T cells recirculating into the thymus are present predominantly within the medulla but a fraction of cells is present in the cortex. Three-color confocal microscopy demonstrates the presence of recirculating donor GFP⁺ cells in the medulla (a) and the cortex (b).²⁸ Methods: CD4⁺ lymph node T cells from GFP transgenic mice were purified by cell sorting and adoptively transferred into B6.pTko recipients. After 7 days, thymic sections were prepared and processed for immunofluorescent histology. Antibodies to cytokeratin-5 and cytokeratin-18 identify medullary epithelial cells and cortical epithelial cells, respectively.

Full figure and legend (304K)

As shown here (Figure 1c), splenic B and T cells migrate preferentially into the bone marrow and thymus, respectively (that is, into the organs where they initiated their development). It is likely that they respond specifically to local chemoattractants, but to date, the nature of the molecules that induce peripheral T cell migration into the thymus in general, and into the medulla in particular, remains to be characterized. The CCR7 ligands, CCL19 and CCL21, are highly concentrated in the medulla, attracting developing cortical thymocytes to this site.^{2, 3} As T cell entry into lymph nodes requires CCR7,^{32, 33} one attractive possibility is that CCL19 and CCL21 may also play a role in T cell re-entry into the thymus.

Functional consequences of peripheral T cells in the thymus

T cells are essential for the differentiation and organization of a normal thymic stroma

Lympho-stromal interactions are required for thymocyte differentiation and proliferation, but the integrity of the thymic stroma itself also depends on the presence of developing thymocytes. This relationship is commonly referred to as 'thymic crosstalk' or 'interdependence'. Terhorst and co-workers showed that in the absence of thymocytes beyond the DN3 stage, the architecture of the cortex was abnormal.³⁴ The growth of mTEC and the maintenance of medulla integrity are also dependent on the presence of T cells because the thymic medullary epithelium fails to mature and organize in those scid mice completely devoid of TCR⁺ cells whereas it does mature in 'leaky' scid mice that contain some TCR⁺ cells or in scid mice that had been transplanted with normal bone marrow as a source of progenitor T cells.³⁵ Importantly, Surh et al. directly demonstrated that recirculating peripheral T cells can contribute to the proper organization of the medulla by injecting lymph node derived T cells into scid mice, resulting in the regeneration of mTEC.¹⁴ Recently, further intricacies of 'thymic crosstalk' between T cells and mTEC came from comparisons between normal thymi and those of (i) RANKL-deficient (Tnfsf11ko) mice, (ii) Tnfsf11ko.CD40ko mice or (iii) mice in which class II expression was expressed by all TECs or (iv) mice where class II expression was restricted to cTEC.^{36, 37, 38} It was concluded that mTEC cellularity and the expression of autoimmune regulator gene (AIRE), essential for the expression of a large set of tissue-specific antigens by mTEC,^{5, 6} were reduced when RANK and/or CD40 on stromal cells could not be engaged.^{36, 37} Their ligands, RANKL and CD40L, are expressed by thymocytes that had been undergoing positive selection, and the retroviral expression of RANKL in TCRko cells that cannot undergo positive (or negative) selection led to the same conclusion.³⁷ Somewhat differently, Irla et al.³⁸ attribute the development of normal numbers of mature AIRE-expressing mTEC to the presence of CD4⁺ SP cells and the interaction of their TCR with class II MHC molecules expressed by mTEC. Furthermore, ongoing negative selection was an essential component. Somewhat contradictory to this finding, an increased frequency of mTECs, activating expression from the targeted AIRE alleles, was found in AIRE-deficient mice, where negative selection would be expected as being reduced compared with the normal situation.³⁹ It will be interesting to determine whether peripheral T cells recirculating back into the thymus have the capacity not only to regenerate mTECs, but also to induce AIRE expression.

Recirculating T cells can mediate negative selection

While positive selection allows for the selection of T cells expressing a TCR able to interact with self-MHC molecules, negative selection eliminates, by apoptosis, those thymocytes that bear a TCR with high affinity for MHC molecules loaded with self peptides. If autoreactive cells are not removed, their presence could lead to the development of autoimmunity upon antigen encounter in the periphery. In the medulla, negative selection is mediated by mTEC and thymic DCs (Figure 3), but upon recirculation into the thymus, cells from the periphery can also be involved. As shown recently, adoptive transfer of OVA-loaded DCs resulted in the clonal deletion of OVA-reactive thymocytes as the DCs entered the thymus.⁴⁰ Importantly, in this study, the authors also showed that when OVA was expressed extrathymically in cardiomyocytes, some heart-derived DCs took up and presented OVA intrathymically in a tolerogenic fashion, such that thymocytes with OVA specificity (OT2 TCR transgenic cells) were deleted. Thus, by migrating into the thymus, peripheral DCs can provide peripheral (self-)antigens for the elimination of the corresponding self-reactive thymocytes.

Figure 3.

Recirculating mature T cells are present within the medulla and represent a sizeable fraction of SP thymocytes in lymphopenic animals; some are present within the cortex where they can complement cTECs in thymocyte-positive selection. '?' depicts the unknown migratory routes whithin the thymic compartments of peripheral T cells re-entering the thymus. The color reproduction of this figure is available on the html full text version of the manuscript.

Full figure and legend (195K)

Searching for the physiological significance of mature T cell re-entry into the thymus, early experiments from the group of Sprent⁴¹ revealed that adoptively transferred purified B and T cells can provide endogenous superantigens (Mtv) to the thymus, resulting in central tolerance. Thus lymphocyte re-entry into the thymus can be instrumental in achieving negative selection of T cells reactive to antigens expressed by the recirculating cells. At least for T cells, actual presentation of the superantigen must have occurred on other cells, as superantigen reactivity requires its presentation by class II MHC molecules. Also, it should be noted that some of these experiments were performed in neonates known to be more permeable to lymphocyte re-entry into the thymus. Nevertheless, the above conclusions were also confirmed for adults by Iacomini and co-workers²⁹. They demonstrated the potential role of transferred allogenic T cells in skin-graft tolerance. The tolerogenic role of the transferred T cells correlated with their ability to re-enter the thymus, which was more effective when these cells had been expanded in the periphery. In summary, these data demonstrate that under normal physiological conditions, different subsets of leukocytes can re-enter the thymus to deliver antigens to which tolerance is induced.

Peripheral T lymphocytes re-entering the thymus can support positive selection

Can peripheral T cells recirculating into the thymus be functional in positive selection? Supporting evidences for such a scenario were recently provided by Li et al.⁴² and Choi et al.⁴³: they generated mice in which T cells, including thymocytes, were the only MHC class II-expressing cells.^{42, 43} In these mice, thymic positive selection into CD4⁺ SP T cells was as efficient as in control mice, although the functionality of the selected CD4⁺ cells was later questioned.⁴⁴ Also, as not all species show MHC class II expression on T cells, it is as yet unclear whether this pathway is of major physiologic relevance. Notably, using several TCR transgenic lines in which positive selection on TEC occurs with a high efficiency, positive selection was relatively inefficient on T cell-expressed class II MHC molecules.⁴² Therefore, the selection on thymocyte-expressed MHC class II molecules may function only for a subset of cells bearing certain TCRs. This may depend on the repertoire of peptides presented by MHC class II molecules on T cells versus those presented on cTECs. Thus, there may exist a population of T cells with 'innate' TCRs that are more susceptible to such selection processes. However, one may also note that in the above studies it has not been excluded that the resulting CD4⁺ SP T cells may in fact be selected on class I MHC molecules. Widespread MHC class II expression by thymocytes may have rescued cells selected on MHC class I molecules, but expressing a mismatched CD4 coreceptor.⁴⁵ Here, this process of 'misdirection' would be enhanced by the possibility of CD4 coreceptor interaction with a multiplicity of MHC class II ligand molecules.

As mentioned earlier, most recirculating peripheral T cells are located within the thymic medulla; however, some cells are also present within the cortex (Figure 2).^{14, 28} As the latter is the location where thymocyte positive selection occurs, we investigated whether mature peripheral T cells recirculating into the thymus could mediate thymocyte positive selection. To test this hypothesis, we created mice in which thymocyte development was arrested at the DP stage by breeding the OT1 TCR transgene onto a K^b MHC class I molecule-deficient background, K^b being essential for positive selection of the OT1 TCR. To exclude selection by endogenous TCR rearrangements, the mice were also bred to be Rag2 deficient, resulting in B6.Rag2ko.OT1tg.K^bko mice.^{46, 47} These mice then served as recipients for purified peripheral T cells expressing K^b, that is by injecting P14 TCR transgenic Rag2ko T cells.⁴⁸ We observed, first, that P14 T cells survived in the B6.Rag2ko.OT1tg.K^bko recipients. This was expected as cells in the recipient mice express D^b, the restriction element of the P14 TCR, the presence of which will be essential for P14 cell survival as peripheral T cells depend on the interaction of their TCR with restricting MHC molecules. In fact, P14 cells showed slow homeostatic proliferation in these recipients.⁴⁹ Second, we found that P14 T cells efficiently entered the thymus. Finally, we observed the development of OT1 CD8⁺ SP cells that downregulated CD24 expression. This was dependent on the expression of K^b on donor P14 T cells. Thus, peripheral P14 T cells migrating into the thymus provided the K^b molecules necessary for the positive selection of host CD8⁺ OT1 thymocytes²⁸ (Figure 3).

In the experimental system presented above, the intrathymic presence of approximately 10⁵ K^b-expressing T cells generated 2 10⁴–2 10⁵ mature V5⁺ CD8⁺ CD24^lo OT1 SP T cells. Thus, the efficiency of positive selection was low compared with that in normal mice where 10⁷ CD4⁺ SP and 2 10⁶ CD8⁺ SP mature T cells are present per thymus or in B6.Rag2ko.OT1tg mice that contain 5–7 10⁶ V5⁺ CD8⁺ SP mature T cells per thymus. However, in the normal situation, the selecting ligand is expressed at higher levels on approximately 50 000 thymic epithelial cells that are presumably able to mediate efficient thymocyte positive selection by contacting several thymocytes simultaneously. Another aspect to consider here is the possible lack of complete integrity of the thymic stroma in B6.Rag2ko.OT1tg.K^bko mice. In these mice, the absence of crosstalk between mature SP and TEC may affect the capacity of the latter to efficiently support the ongoing selection on T cells and this may not be compensated upon the transfer of limited numbers of peripheral P14 T cells.⁵⁰

Future prospects

As presented above, there are now multiple lines of evidence that recirculating T cells can be found, albeit at low frequency, in the thymus and that T cells can complement cTEC to support thymocyte development. It remains to be determined by which mechanisms these functions are mediated. Considering our experimental system, in which the MHC molecules essential for positive selection are brought into the thymus by T cells, there are two major possibilities. Either T cells are indeed able to directly support thymocyte positive selection by presenting MHC molecules, that is positive selection occurs by cell–cell interaction between recirculating T cells and immature thymocytes. Here cTECs would only act as bystander cells in trans, most likely being responsible for the creation and maintenance of an environment suitable for positive selection. Alternatively, recirculating T cells may provide MHC molecule-containing membrane material only, which may be picked up and presented by cTEC to allow for 'conventional' positive selection by these cells (thus in cis).

It will be challenging to differentiate between these possibilities and, furthermore, to determine whether thymocytes or recirculating T cells are the unique cell population in the thymus able to mediate positive selection, apart from cTEC. With respect to the latter, several authors have observed that in various chimeric mice, MHC restriction can be toward MHC molecules expressed on hematopoietic cells (for a few examples, see Martinic et al.⁵¹ and Zinkernagel et al.⁵² and references therein). However, it is impossible to exclude that T cells may only appear as selected on hematopoietic cells because their TCR is cross-reactive,⁵³ allowing both for 'conventional' positive selection on cTEC-expressed MHC molecules and functional activity toward hematopoietic cell-expressed allogenic MHC. Furthermore, when using TCR transgenic mice to study selection requirements, one possible obstacle is that additional TCRs can be formed by recombination of the endogenous TCR loci. Cells with dual specificity for cTEC and hematopoietic cell-expressed MHC molecules may thus arise because more than one TCR is present. Therefore, it was important to verify that positive selection could indeed occur on hematopoietic cells using TCR monoclonal cells to reconstitute irradiated recipients lacking the restriction element.²⁸ Curiously, selection on hematopoietic cells has in the past been more frequently demonstrated for class I MHC-restricted cells, that is for CD8⁺ T cells. Thus, there may be a dichotomy for CD4⁺ and CD8⁺ cells.⁵⁴ Apart from this, only a fraction of the TCRs normally selected on cTEC may actually be selectable on hematopoietic cells. Interestingly, regarding class II MHC molecule restricted T cells, one careful study showed that positive selection on hematopoietic cells does not occur as class II MHC molecule expression on TEC was essential to derive CD4⁺ SP T cell functionality, despite the ample presence of class II MHC molecules on hematopoietic cells (antigen-presenting cells) within the thymus. Unfortunately, CD4⁺ SP T cell numbers in these mice were not reported.⁵⁴ Their result raises the additional question whether only T cells, but no other hematopoietic cells, may mediate positive selection. Although in the latter study, among hematopoietic cells, only professional antigen-presenting cells in the thymus expressed class II MHC molecules, and these did not mediate positive selection, transgene-driven class II MHC molecule expression specifically in T cells resulted in efficient positive selection.^{42, 43} Here one would like to see results using this model in combination with a defined, fixed TCR specificity to exclude other possibilities on how the selected cells may arise (see above). Contrary to the specific role of T cell-expressed MHC molecules, or expression on hematopoietic cells, efficient positive selection on fibroblast-expressed class I MHC molecules has been noted.⁵⁵ Yet, the functionality of fibroblasts in positive selection was challenged later,⁵⁶ but this study used the HY TCR that is not selected on hematopoietic cells.⁵⁷

There are further situations in which cTEC can be replaced by other cells, supporting some form of positive selection. Extrathymic T cell development can occur in the lymph nodes of Oncostatin M transgenic mice,⁵⁸ or in the gut of athymic nude mice.⁵⁹ T cell development can even be recapitulated in vitro, using the stromal cell line OP9, when transduced to express the Notch ligand Delta-like-1.⁶⁰ In vivo, T cell differentiation toward CD4⁺ and CD8⁺ SP cells can take place in the absence of a thymus in the bone marrow and in secondary lymphoid organs once Delta-like-1 or Delta-like-4 is expressed ectopically in hematopoietic cells.⁶¹ One obstacle to the observed positive selection on OP9-DL1 toward CD4⁺ SP cells in vitro is that it occurs despite the absence of class II MHC molecule expression in the culture system. Thus, selection using OP9-DL1 or hematopoietic cells expressing Notch ligands may not reflect normal physiologic processes. However, these data make the point that hematopoietic cells can participate in T cell selection and, thus, thymus recirculating T cells may do similarly.

Recently, the surprising finding of an additional subunit of the proteasome complex, specifically expressed in cTEC, was reported.⁶² This finding may turn out to provide a major piece to the puzzle of how T cells can be selected on, and at the same time be tolerant toward, self-MHC/self-peptide complexes. Although the 'standard proteasome' and the 'immunoproteasome' constitutively expressed in DCs and cytokine-induced in infected tissues (containing 1i (LMP2), 2i (MECL1) and 5i (LMP7) subunits) both will produce widely available peptide species throughout the tissues, the newly described 'thymoproteasome' contains a novel 5 subunit (5t) having weaker chymotrypsin activity. The nature of the peptides generated by this thymoproteasome is still unknown; however, their presence on cTECs may be of particular relevance for positive selection.⁶³ Indeed, as determined in 5t-deficient mice, the thymoproteasome appears to be essential for the generation of a large fraction of CD8⁺ SP cells. A similar situation exists for CD4⁺ SP thymocytes in which a significant fraction of positive selection may depend on cathepsin L-generated peptides generated in cTEC and presented on MHC class II molecules.⁶⁴ However, other explanations for the paucity of CD4⁺ SP cells in cathepsin L-deficient mice have been put forward.⁶⁵

In any case, the presentation of unique peptides by cTEC makes it possible that thymocyte selection operates as proposed by the 'peptide model' where the net result of thymocyte selection is achieved by positive selection on cTEC-specific peptides, which are (partially) different from the self-peptides against which cells get negatively selected on by mTECs and DCs.⁶⁶ At a glance, this model leaves out selection on hematopoietic cells. However, as there were still reasonable numbers of CD8⁺ SP cells in 5t-deficient mice, this may indicate that these CD8⁺ SP cells are selected through another pathway and this could include selection on hematopoietic cells. In this context, it is interesting to speculate on the MHC-bound peptides that recirculating T cells will transport into the thymus. Some of these are probably derived from the variable regions of the TCR chains and may differ from those present on developing thymocytes in situ, as peripheral competition continuously reshapes the T cell repertoire and activated T cells preferentially re-enter the thymus. Given that in the past immunization with TCR CDR-derived peptides ameliorated some autoimmune phenomena,^{67, 68, 69} one may implicate idiotypic models in which some thymocytes are selected on CDR-derived peptides from recirculating T cells.

As mentioned earlier, an alternative scenario to explain the mechanism of T cell-mediated positive selection is that recirculating T cells may simply provide MHC class I molecule-containing membrane fragments to cTEC and that this indirect mode of MHC molecule presentation on cTEC suffices for positive selection (selection on trogocytosed MHC molecules). However, positive selection of thymocytes on trogocytosed material may only be amenable for high-affinity TCRs, as only limited amounts of material will be transferred. This model gives a plausible explanation for the finding that only presumed high-affinity TCRs can be selected when MHC molecules are provided by hematopoietic cells in trans, or provided by just any cell such as fibroblasts, whereas a low-affinity TCR (HY) may fail to be selected under such conditions.^{28, 55, 56, 70}

In this review, we present studies on T cell recirculation back into the thymus and discuss the thymic niche for these cells. We provide evidence that the relationship between the thymus and other lymphoid organs cannot be considered as unidirectional (Figure 3). Upon migration into the thymus, peripheral T cells introduce MHC molecules loaded with self-peptides. Being of peripheral origin and potentially pathogen derived, the nature of these peptides should be characterized in detail and related to those presented on MHC molecules of TEC. Peptides introduced into the thymus by migrating T cells may influence the selection of SP thymocytes, leaving an imprint on their TCR repertoire or function. In practical terms, recirculation of lymphocytes to the thymus could be envisioned as a group of Trojan horses in that they could be used to introduce antigens necessary for both positive and negative selection to aid in shaping the T cell repertoire.

References

1. Rodewald HR. Thymus organogenesis. Annu Rev Immunol 2008; 26: 355–388. | Article | PubMed | ChemPort |
2. Kwan J, Killeen N. CCR7 directs the migration of thymocytes into the thymic medulla. J Immunol 2004; 172: 3999–4007. | PubMed | ISI | ChemPort |
3. Ueno T, Saito F, Gray DH, Kuse S, Hieshima K, Nakano H et al. CCR7 signals are essential for cortex–medulla migration of developing thymocytes. J Exp Med 2004; 200: 493–505. | Article | PubMed | ISI | ChemPort |
4. McCaughtry TM, Wilken MS, Hogquist KA. Thymic emigration revisited. J Exp Med 2007; 204: 2513–2520. | Article | PubMed | ChemPort |
5. Mathis D, Benoist C. Back to central tolerance. Immunity 2004; 20: 509–516. | Article | PubMed | ISI | ChemPort |
6. Kyewski B, Klein L. A central role for central tolerance. Annu Rev Immunol 2006; 24: 571–606. | Article | PubMed | ISI | ChemPort |
7. Tian T, Zhang J, Gao L, Qian XP, Chen WF. Heterogeneity within medullary-type TCRalphabeta(+)CD3(+)CD4(-)CD8(+) thymocytes in normal mouse thymus. Int Immunol 2001; 13: 313–320. | Article | PubMed | ChemPort |
8. Boursalian TE, Golob J, Soper DM, Cooper CJ, Fink PJ. Continued maturation of thymic emigrants in the periphery. Nat Immunol 2004; 5: 418–425. | Article | PubMed | ISI | ChemPort |
9. Weinreich MA, Hogquist KA. Thymic emigration: when and how T cells leave home. J Immunol 2008; 181: 2265–2270. | PubMed | ChemPort |
10. Matloubian M, Lo CG, Cinamon G, Lesneski MJ, Xu Y, Brinkmann V et al. Lymphocyte egress from thymus and peripheral lymphoid organs is dependent on S1P receptor 1. Nature 2004; 427: 355–360. | Article | PubMed | ISI | ChemPort |
11. Bonasio R, von Andrian UH. Generation, migration and function of circulating dendritic cells. Curr Opin Immunol 2006; 18: 503–511. | Article | PubMed | ChemPort |
12. Dumont FJ, Barrois R, Jacobson EB. Migration of peripheral T and B cells into the thymus of aging (NZB SJL)F1 female mice. Cell Immunol 1984; 83: 292–301. | Article | PubMed | ChemPort |
13. Michie SA, Rouse RV. Traffic of mature lymphocytes into the mouse thymus. Thymus 1989; 13: 141–148. | PubMed | ISI | ChemPort |
14. Surh CD, Ernst B, Sprent J. Growth of epithelial cells in the thymic medulla is under the control of mature T cells. J Exp Med 1992; 176: 611–616. | Article | PubMed | ISI | ChemPort |
15. Fehling HJ, Krotkova A, Saint-Ruf C, von Boehmer H. Crucial role of the pre-T-cell receptor alpha gene in development of alpha beta but not gamma delta T cells. Nature 1995; 375: 795–798. | Article | PubMed | ISI | ChemPort |
16. Bosco N, Agenes F, Rolink AG, Ceredig R. Peripheral T cell lymphopenia and concomitant enrichment in naturally arising regulatory T cells: the case of the pre-Talpha gene-deleted mouse. J Immunol 2006; 177: 5014–5023. | PubMed | ISI | ChemPort |
17. Webb LM, Vigorito E, Wymann MP, Hirsch E, Turner M. Cutting edge: T cell development requires the combined activities of the p110gamma and p110delta catalytic isoforms of phosphatidylinositol 3-kinase. J Immunol 2005; 175: 2783–2787. | PubMed | ChemPort |
18. Munoz JJ, Alfaro D, Garcia-Ceca J, Alonso CL, Jimenez E, Zapata A. Thymic alterations in EphA4-deficient mice. J Immunol 2006; 177: 804–813. | PubMed | ChemPort |
19. Almeida AR, Borghans JA, Freitas AA. T cell homeostasis: thymus regeneration and peripheral T cell restoration in mice with a reduced fraction of competent precursors. J Exp Med 2001; 194: 591–599. | Article | PubMed | ISI | ChemPort |
20. Terszowski G, Muller SM, Bleul CC, Blum C, Schirmbeck R, Reimann J et al. Evidence for a functional second thymus in mice. Science 2006; 312: 284–287. | Article | PubMed | ChemPort |
21. Levinson AI, Song D, Gaulton G, Zheng Y. The intrathymic pathogenesis of myasthenia gravis. Clin Dev Immunol 2004; 11: 215–220. | Article | PubMed |
22. Naparstek Y, Holoshitz J, Eisenstein S, Reshef T, Rappaport S, Chemke J et al. Effector T lymphocyte line cells migrate to the thymus and persist there. Nature 1982; 300: 262–264. | Article | PubMed | ISI | ChemPort |
23. Agus DB, Surh CD, Sprent J. Reentry of T cells to the adult thymus is restricted to activated T cells. J Exp Med 1991; 173: 1039–1046. | Article | PubMed | ISI | ChemPort |
24. Michie SA, Kirkpatrick EA, Rouse RV. Rare peripheral T cells migrate to and persist in normal mouse thymus. J Exp Med 1988; 168: 1929–1934. | Article | PubMed | ISI | ChemPort |
25. Michie SA, Rouse RV. Study of murine T cell migration using the Thy-1 allotypic marker. Demonstration of antigen-specific homing to lymph node germinal centers. Transplantation 1988; 46: 98–104. | Article | PubMed | ChemPort |
26. Yu W, Nagaoka H, Jankovic M, Misulovin Z, Suh H, Rolink A et al. Continued RAG expression in late stages of B cell development and no apparent re-induction after immunization. Nature 1999; 400: 682–687. | Article | PubMed | ISI | ChemPort |
27. Hale JS, Boursalian TE, Turk GL, Fink PJ. Thymic output in aged mice. Proc Natl Acad Sci USA 2006; 103: 8447–8452. | Article | PubMed | ChemPort |
28. Kirberg J, Bosco N, Deloulme JC, Ceredig R, Agenes F. Peripheral T lymphocytes recirculating back into the thymus can mediate thymocyte positive selection. J Immunol 2008; 181: 1207–1214. | PubMed | ChemPort |
29. Tian C, Bagley J, Iacomini J. Homeostatic expansion permits T cells to re-enter the thymus and deliver antigen in a tolerogenic fashion. Am J Transplant 2007; 7: 1934–1941. | Article | PubMed | ChemPort |
30. Agenes F, Bosco N, Mascarell L, Fritah S, Ceredig R. Differential expression of regulator of G-protein signalling transcripts and in vivo migration of CD4+ naive and regulatory T cells. Immunology 2005; 115: 179–188. | Article | PubMed | ChemPort |
31. Sapoznikov A, Pewzner-Jung Y, Kalchenko V, Krauthgamer R, Shachar I, Jung S. Perivascular clusters of dendritic cells provide critical survival signals to B cells in bone marrow niches. Nat Immunol 2008; 9: 388–395. | Article | PubMed | ChemPort |
32. Forster R, Schubel A, Breitfeld D, Kremmer E, Renner-Muller I, Wolf E et al. CCR7 coordinates the primary immune response by establishing functional microenvironments in secondary lymphoid organs. Cell 1999; 99: 23–33. | Article | PubMed | ISI | ChemPort |
33. Gunn MD, Kyuwa S, Tam C, Kakiuchi T, Matsuzawa A, Williams LT et al. Mice lacking expression of secondary lymphoid organ chemokine have defects in lymphocyte homing and dendritic cell localization. J Exp Med 1999; 189: 451–460. | Article | PubMed | ISI | ChemPort |
34. Hollander GA, Wang B, Nichogiannopoulou A, Platenburg PP, van Ewijk W, Burakoff SJ et al. Developmental control point in induction of thymic cortex regulated by a subpopulation of prothymocytes. Nature 1995; 373: 350–353. | Article | PubMed | ISI | ChemPort |
35. Shores EW, Van Ewijk W, Singer A. Disorganization and restoration of thymic medullary epithelial cells in T cell receptor-negative scid mice: evidence that receptor-bearing lymphocytes influence maturation of the thymic microenvironment. Eur J Immunol 1991; 21: 1657–1661. | Article | PubMed | ISI | ChemPort |
36. Akiyama T, Shimo Y, Yanai H, Qin J, Ohshima D, Maruyama Y et al. The tumor necrosis factor family receptors RANK and CD40 cooperatively establish the thymic medullary microenvironment and self-tolerance. Immunity 2008; 29: 423–437. | Article | PubMed | ChemPort |
37. Hikosaka Y, Nitta T, Ohigashi I, Yano K, Ishimaru N, Hayashi Y et al. The cytokine RANKL produced by positively selected thymocytes fosters medullary thymic epithelial cells that express autoimmune regulator. Immunity 2008; 29: 438–450. | Article | PubMed | ChemPort |
38. Irla M, Hugues S, Gill J, Nitta T, Hikosaka Y, Williams IR et al. Autoantigen-specific interactions with CD4+ thymocytes control mature medullary thymic epithelial cell cellularity. Immunity 2008; 29: 451–463. | Article | PubMed | ChemPort |
39. Dooley J, Erickson M, Farr AG. Alterations of the medullary epithelial compartment in the Aire-deficient thymus: implications for programs of thymic epithelial differentiation. J Immunol 2008; 181: 5225–5232. | PubMed | ChemPort |
40. Bonasio R, Scimone ML, Schaerli P, Grabie N, Lichtman AH, von Andrian UH. Clonal deletion of thymocytes by circulating dendritic cells homing to the thymus. Nat Immunol 2006; 7: 1092–1100. | Article | PubMed | ChemPort |
41. Webb SR, Sprent J. Induction of neonatal tolerance to Mlsa antigens by CD8+ T cells. Science 1990; 248: 1643–1646. | Article | PubMed | ChemPort |
42. Li W, Kim MG, Gourley TS, McCarthy BP, Sant'Angelo DB, Chang CH. An alternate pathway for CD4 T cell development: thymocyte-expressed MHC class II selects a distinct T cell population. Immunity 2005; 23: 375–386. | Article | PubMed | ISI | ChemPort |
43. Choi EY, Jung KC, Park HJ, Chung DH, Song JS, Yang SD et al. Thymocyte–thymocyte interaction for efficient positive selection and maturation of CD4 T cells. Immunity 2005; 23: 387–396. | Article | PubMed | ISI | ChemPort |
44. Li W, Sofi MH, Yeh N, Sehra S, McCarthy BP, Patel DR et al. Thymic selection pathway regulates the effector function of CD4 T cells. J Exp Med 2007; 204: 2145–2157. | PubMed | ChemPort |
45. Ge Q, Holler PD, Mahajan VS, Nuygen T, Eisen HN, Chen J. Development of CD4+ T cells expressing a nominally MHC class I-restricted T cell receptor by two different mechanisms. Proc Natl Acad Sci USA 2006; 103: 1822–1827. | Article | PubMed | ChemPort |
46. Clarke SR, Barnden M, Kurts C, Carbone FR, Miller JF, Heath WR. Characterization of the ovalbumin-specific TCR transgenic line OT-I: MHC elements for positive and negative selection. Immunol Cell Biol 2000; 78: 110–117. | Article | PubMed | ISI | ChemPort |
47. Hogquist KA, Jameson SC, Heath WR, Howard JL, Bevan MJ, Carbone FR. T cell receptor antagonist peptides induce positive selection. Cell 1994; 76: 17–27. | Article | PubMed | ISI | ChemPort |
48. Pircher H, Burki K, Lang R, Hengartner H, Zinkernagel RM. Tolerance induction in double specific T-cell receptor transgenic mice varies with antigen. Nature 1989; 342: 559–561. | Article | PubMed | ISI | ChemPort |
49. Agenes F, Dangy JP, Kirberg J. T cell receptor contact to restricting MHC molecules is a prerequisite for peripheral inter-clonal T cell competition. J Exp Med 2008; 205: 2735–2743. | Article | PubMed | ChemPort |
50. Zamisch M, Moore-Scott B, Su DM, Lucas PJ, Manley N, Richie ER. Ontogeny and regulation of IL-7-expressing thymic epithelial cells. J Immunol 2005; 174: 60–67. | PubMed | ISI | ChemPort |
51. Martinic MM, Rulicke T, Althage A, Odermatt B, Hochli M, Lamarre A et al. Efficient T cell repertoire selection in tetraparental chimeric mice independent of thymic epithelial MHC. Proc Natl Acad Sci USA 2003; 100: 1861–1866. | Article | PubMed | ChemPort |
52. Zinkernagel RM, Althage A. On the role of thymic epithelium vs bone marrow-derived cells in repertoire selection of T cells. Proc Natl Acad Sci USA 1999; 96: 8092–8097. | Article | PubMed | ChemPort |
53. Eshima K, Suzuki H, Shinohara N. Cross-positive selection of thymocytes expressing a single TCR by multiple major histocompatibility complex molecules of both classes: implications for CD4+ versus CD8+ lineage commitment. J Immunol 2006; 176: 1628–1636. | PubMed | ChemPort |
54. Martinic MM, van den Broek MF, Rulicke T, Huber C, Odermatt B, Reith W et al. Functional CD8+ but not CD4+ T cell responses develop independent of thymic epithelial MHC. Proc Natl Acad Sci USA 2006; 103: 14435–14440. | Article | PubMed | ChemPort |
55. Pawlowski T, Elliott JD, Loh DY, Staerz UD. Positive selection of T lymphocytes on fibroblasts. Nature 1993; 364: 642–645. | Article | PubMed | ISI | ChemPort |
56. Lilic M, Santori FR, Neilson EG, Frey AB, Vukmanovic S. The role of fibroblasts in thymocyte-positive selection. J Immunol 2002; 169: 4945–4950. | PubMed |
57. Kisielow P, Teh HS, Bluthmann H, von Boehmer H. Positive selection of antigen-specific T cells in thymus by restricting MHC molecules. Nature 1988; 335: 730–733. | Article | PubMed | ISI | ChemPort |
58. Terra R, Labrecque N, Perreault C. Thymic and extrathymic T cell development pathways follow different rules. J Immunol 2002; 169: 684–692. | PubMed | ChemPort |
59. Guy-Grand D, Azogui O, Celli S, Darche S, Nussenzweig MC, Kourilsky P et al. Extrathymic T cell lymphopoiesis: ontogeny and contribution to gut intraepithelial lymphocytes in athymic and euthymic mice. J Exp Med 2003; 197: 333–341. | Article | PubMed | ISI | ChemPort |
60. de Pooter R, Zuniga-Pflucker JC. T-cell potential and development in vitro: the OP9-DL1 approach. Curr Opin Immunol 2007; 19: 163–168. | Article | PubMed | ChemPort |
61. de La Coste A, Six E, Fazilleau N, Mascarell L, Legrand N, Mailhe MP et al. In vivo and in absence of a thymus, the enforced expression of the Notch ligands delta-1 or delta-4 promotes T cell development with specific unique effects. J Immunol 2005; 174: 2730–2737. | PubMed | ISI | ChemPort |
62. Murata S, Sasaki K, Kishimoto T, Niwa S, Hayashi H, Takahama Y et al. Regulation of CD8+ T cell development by thymus-specific proteasomes. Science 2007; 316: 1349–1353. | Article | PubMed | ISI | ChemPort |
63. Murata S, Takahama Y, Tanaka K. Thymoproteasome: probable role in generating positively selecting peptides. Curr Opin Immunol 2008; 20: 192–196. | Article | PubMed | ChemPort |
64. Honey K, Nakagawa T, Peters C, Rudensky A. Cathepsin L regulates CD4+ T cell selection independently of its effect on invariant chain: a role in the generation of positively selecting peptide ligands. J Exp Med 2002; 195: 1349–1358. | Article | PubMed | ISI | ChemPort |
65. Lombardi G, Burzyn D, Mundinano J, Berguer P, Bekinschtein P, Costa H et al. Cathepsin-L influences the expression of extracellular matrix in lymphoid organs and plays a role in the regulation of thymic output and of peripheral T cell number. J Immunol 2005; 174: 7022–7032. | PubMed | ChemPort |
66. Marrack P, Kappler J. The T cell receptor. Science 1987; 238: 1073–1079. | Article | PubMed | ISI | ChemPort |
67. Howell MD, Winters ST, Olee T, Powell HC, Carlo DJ, Brostoff SW. Vaccination against experimental allergic encephalomyelitis with T cell receptor peptides. Science 1989; 246: 668–670. | Article | PubMed | ISI | ChemPort |
68. Kumar V, Sercarz EE. The involvement of T cell receptor peptide-specific regulatory CD4+ T cells in recovery from antigen-induced autoimmune disease. J Exp Med 1993; 178: 909–916. | Article | PubMed | ISI | ChemPort |
69. Vandenbark AA, Hashim G, Offner H. Immunization with a synthetic T-cell receptor V-region peptide protects against experimental autoimmune encephalomyelitis. Nature 1989; 341: 541–544. | Article | PubMed | ISI | ChemPort |
70. Zerrahn J, Volkmann A, Coles MC, Held W, Lemonnier FA, Raulet DH. Class I MHC molecules on hematopoietic cells can support intrathymic positive selection of T cell receptor transgenic T cells. Proc Natl Acad Sci USA 1999; 96: 11470–11475. | Article | PubMed | ChemPort |

Acknowledgements

We express our gratitude to JC Deloulme (INSERM U836, Grenoble, France) for his indispensable help with confocal microscopy and to S Vincent, K Leclerc-Desaulniers, L Thibault and K Huot for excellent animal care (CHUM Montréal, Canada). F Agenes is greatly indebted to Dr RP Sekaly for hosting him at the INSERM U743 (Montréal, Canada). J Kirberg is grateful to Dr H Acha-Orbea for hosting him at the Department of Biochemistry, University of Lausanne, Switzerland.