|
|
||||||||


* Laboratory of Mammalian Genes and Development, National Institute of Child Health and Human Development, and
Laboratory of Molecular and Cellular Immunology, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD 20892
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
gene rearrangement and subsequently express pre-TCR complexes on their surface. Pre-TCR-mediated signals induce the transition of DN thymocytes to the CD4+CD8+ (double-positive (DP)) stage. Coincident with this developmental transition, thymocytes move from the SCR back into the cortex (1, 2). DP thymocytes that productively rearrange TCR
begin to express the mature 
TCR. Signals transduced by the 
TCR regulate the process of thymocyte selection and induce cells that have survived selection to transition to the CD4+CD8 or CD4CD8+ single-positive (SP) stage. During this process, thymocytes move to the inner cortex and eventually migrate into the medulla as they undergo terminal differentiation. Although this trafficking pattern now appears to be fairly well defined, it remains unclear how movement of thymocyte subsets is regulated during development. Recent data suggest that chemokines may play a critical role in orchestrating trafficking of thymocytes into, through, and out of the thymus (1, 3, 4). Chemokines are a group of small (814 kDa), structurally related molecules that regulate trafficking of leukocytes through interaction with a subset of seven-transmembrane, G protein-coupled receptors (3, 4, 5, 6). Multiple chemokines are expressed in the thymus, and maturation of thymocytes through various stages of development is associated with dramatic changes in chemokine expression patterns and chemokine responsiveness (7, 8). These data suggest that chemokines and their receptors may function to regulate cell migration in the thymus, but little information is available concerning the roles of individual chemokines or chemokine receptors. CCR9 mediates chemotaxis in response to the chemokine CCL25, its only known ligand. The expression of CCL25 was initially detected in medullary dendritic cells in the thymus (9), but recent experiments indicate that CCL25 is also expressed by epithelial cells in both cortex and medulla (10). CCR9 expression is highly regulated during T cell development. Most immature DN thymocytes express no or low levels of CCR9 on their surface (11, 12). However, CCR9 transcripts are present in DN thymocytes, and CCR9 transcription is induced by pre-TCR signaling (13). The majority of DP thymocytes express high levels of CCR9, but CCR9 surface expression is down-regulated as DP thymocytes transition to the SP stage. Interestingly, CD3highCD69+ DP thymocytes preferentially respond to CCL25, and CCL25 responsiveness has been shown to be augmented by TCR cross-linking (11). These observations suggest that CCR9 may be involved in localizing DP thymocytes to the cortex and/or regulating the migration of DP cells into the medulla as they transition to the SP stage. Most mature SP thymocytes down-regulate CCR9 surface expression and lose CCL25 responsiveness before their emigration from the thymus (3, 8, 11, 12). Consequently, it has been proposed that CCL25 may function to retain SP cells in the thymus until they are fully mature and ready for export (8).
To investigate the role of CCR9 during T cell development, we generated CCR9 transgenic mice (CCR9-Tg) in which surface expression of CCR9 was maintained on SP thymocytes and mature T cells. The phenotype of these mice indicates that regulated expression of CCR9 at the DN to DP transition stage is critical for normal T cell development, but that down-regulation of CCR9 on SP thymocytes is not essential for thymic emigration
| Materials and Methods |
|---|
|
|
|---|
The human CD2-CCR9 transgene was generated by substituting murine CCR9 coding sequences (PstI-NdeI fragment; GenBank accession no. MMU132336) for the
cDNA sequences in construct
-CT108 (14). Previous studies have shown that the human CD2 promoter/enhancer directs the expression of transgenes in mice to the T lineage. Transgene expression begins at the DN stage, continues throughout thymocyte development, and is maintained on mature peripheral T cells (14). Two CCR9-Tg founder lines (CCR9-Tg #1 and CCR9-Tg #2) were generated by zygote injection. CCR9 surface expression was quantified by flow cytometric analysis. TCR-V
8.1-Tg (V
8-Tg), Rag-1/, Bcl-2-Tg, and AND TCR-Tg mice were obtained from The Jackson Laboratory.
Cell preparation and flow cytometry
Cell preparation and flow cytometry was performed as previously described (11), using a FACSCalibur and CellQuest software (BD Biosciences) or an LSR II and FACSDiva software (BD Biosciences). Analysis was performed using FlowJo software (Tree Star). The following mAbs were purchased from BD Biosciences: anti-CD3, anti-CD4, anti-CD8
, anti-CD25, anti-CD44, anti-B220, anti-CD19, anti-DX5, anti-TCR
, and anti-TCR
. The biotinylated polyclonal anti-CCR9 Ab has been previously described (11). CyChrome-conjugated streptavidin was also purchased from BD Biosciences. Cell cycle analysis was performed as described previously (15). Apoptosis was detected by staining with annexin V (BD Biosciences) and 7-aminoactinomycin D according to the manufacturers instructions. DN (CD3CD4CD8B220) thymocytes were isolated to
95% purity by negative selection using biotinylated anti-CD3, anti-CD4, anti-CD8, and anti-B220 mAbs and streptavidin beads (Miltenyi Biotec) according to the manufacturers instructions.
Confocal microscopy
Thymus sections were prepared, stained, and analyzed as previously reported (16), except that the following mAbs were used: FITC-conjugated anti-CD25 or anti-pan-cytokeratin (Sigma-Aldrich), Alexa Fluor 555-conjugated anti-CD4, and Cy5-labeled anti-CD8
(Caltag Laboratories).
Chemotaxis assays
Chemotaxis assays were performed as previously described (11). Murine CXCL12 and CCL21 were obtained from PeproTech, and CCL25 was obtained from R&D Systems.
Western blot analysis
Total thymocyte lysates (derived from 2 x 106 cells) were separated on SDS-PAGE gel and transferred onto nitrocellulose membranes. Membranes were probed with anti-Bcl-xL mAb (BD Biosciences) or anti-actin mAb (Sigma-Aldrich). Bound Abs were detected using HRP-conjugated secondary reagents via chemiluminescence.
Bone marrow adoptive transfer experiments
Bone marrow adoptive transfers were performed as previously described (17).
Statistical analysis
Data were analyzed using Students t test.
| Results |
|---|
|
|
|---|
In wild-type mice, CCR9 is expressed at high levels on DP thymocytes, but is down-regulated during later stages of thymocyte development. All CD4 SP peripheral T cells and most CD8 SP peripheral T cells are CCR9 (Fig. 1A) and are unresponsive to CCR9 ligand (CCL25) (3, 8, 11, 12). This pattern of expression has led to speculation that CCR9 may function to retain cells in the thymus until they have completed maturation, and that down-regulation of CCR9 facilitates the export of mature thymocytes (8). To test this hypothesis, we generated Tg mice in which CCR9 surface expression was maintained on mature thymocytes and peripheral T cells by placing CCR9 under the control of the human CD2 promoter and enhancer (14). In the present study, two independently derived founder lines (CCR9-Tg #1 and CCR9-Tg #2) were analyzed. Because thymocyte development was affected to different degrees in CCR9-Tg #1 and #2 mice (see below), we initially confined our analysis to founder line #1, which exhibited the mildest developmental impairment.
|
2-fold relative to that in wild-type mice (Fig. 1D). In wild-type mice, more than half of all CD4 SP thymocytes are CCR9+ (Fig. 2A); however, the mature subset, which includes cells that are candidates for emigration from the thymus, are CD69CD62Lhigh, CCR9 and unresponsive to CCL25 (8, 11, 12). The ratio of immature (CD69+CD62Llow) to mature (CD69CD62Lhigh) CD4 SP thymocytes was similar in wild-type and CCR9-Tg #1 mice (Fig. 1C). In wild-type mice, only immature CD69+CD62Llow CD4-SP thymocytes migrated to CCL25 (Figs. 1C; 3and 8), whereas in CCR9-Tg #1 mice, both immature and mature CD4 SP thymocytes migrated to CCL25 (Fig. 1C).
|
|
|
Premature onset of CCR9 surface expression affects T cell development
Thymocyte cellularity (especially numbers of DP thymocytes) was reduced in adult mice from both CCR9-Tg founder lines (CCR9-Tg #1 and CCR9-Tg #2), and there was a corresponding decrease in the number of CD4 SP and CD8 SP thymocytes and peripheral T cells in these mice (Fig. 4). Interestingly, the reduction in thymocyte and T cell numbers was much more severe in CCR9-Tg #2 mice than in CCR9-Tg #1 mice (Fig. 4). This difference in phenotype could not be explained by differences in CCR9 surface expression at the DP or SP stage, because CCR9 levels on DP and SP thymocytes were similar in CCR9-Tg #1 and CCR9-Tg #2 mice (Fig. 2A). The number of DN thymocytes was slightly decreased in CCR9-Tg #1 mice and was decreased significantly in CCR9-Tg #2 mice (Fig. 4), suggesting a defect in development at the DN or DN to DP transition stage.
|
CD25+CD44+ (DN2)
CD25+CD44low/ (DN3)
CD25CD44low/ (DN4) (20). Analysis of DN subsets by CD44 vs CD25 staining indicated that there was a partial block of development at the DN3 to DN4 transition in CCR9-Tg #2 mice, but not in CCR9-Tg #1 mice (revealed by an increased percentage of DN3 cells and a decreased percentage of DN4 cells in CCR9-Tg #2 mice; Fig. 2B). The partial arrest in development of DN thymocytes in CCR9-Tg #2 mice did not appear to be due to reduced cell survival, because the percentage of apoptotic DN thymocytes was similar in CCR9-Tg #2 and wild-type mice (data not shown). Therefore, we next examined DN thymocytes in CCR9-Tg #2 mice for a defect in cell proliferation. As DN3 thymocytes transition to the DN4 stage, they down-regulate CD25, and transitional CD25lowCD44low/ DN cells are the first to enter cell cycle after
-selection (21). Proliferation of both CD25low and CD25 DN thymocytes was significantly decreased in CCR9-Tg #2 mice, but not in CCR9-Tg #1 mice, compared with corresponding wild-type cells (Table I). In addition, DP thymocytes in CCR9-Tg mice expressed low (CCR9-Tg #1) or intermediate (CCR9-Tg #2) levels of CD25 on their surface (Fig. 2C). Because CD25 is normally down-regulated as DN thymocytes proliferate and transition to the DP stage, the presence of CD25+ DP thymocytes is consistent with reduced proliferation of DN cells in CCR9-Tg mice.
|
75% of DN4 cells in CCR9-Tg #2 mice expressed CCR9 on their surface (Fig. 2B). CCR9 was also detected on
10% of DN3 cells in CCR9-Tg #1 mice and on
20% of DN3 thymocytes in CCR9-Tg #2 mice (Fig. 2B). To determine whether CCR9 surface expression correlated with chemokine responsiveness, we next analyzed the ability of DN thymocytes from CCR9-Tg mice to migrate to CCL25. Indeed, the percentage of responding DN2/3 and DN4 cells in CCR9-Tg #1 mice was increased 2- to 3-fold compared with that in wild-type mice, and the percentage in CCR9-Tg #2 mice was increased 3- to 5-fold compared with that in wild-type mice (Fig. 2D). DP thymocytes in CCR9-Tg mice are functionally immature
To analyze the developmental defects in CCR9-Tg mice, we generated bone marrow chimeras using a 1/1 mixture of wild-type and CCR9-Tg #2 donor cells. As shown in Fig. 5A, total DN thymocytes in the recipient mice were composed of approximately equal proportions of wild-type and donor-derived cells, yet <5% of DP thymocytes were derived from CCR9-Tg #2 donors. The low number of DP thymocytes in CCR9-Tg #2 mice could potentially be explained by the proliferation defect in precursor DN thymocytes (Table I). However, when we examined DP thymocyte survival by culturing thymocytes in vitro, we found that DP thymocytes from both newborn and adult CCR9-Tg #2 mice exhibited markedly reduced survival compared with wild-type thymocytes (Fig. 5C). DP thymocytes from adult CCR9-Tg #1 mice also exhibited reduced survival compared with wild-type mice, but the survival defect was much less than that observed in CCR9-Tg #2 mice (Fig. 5C). Because Bcl-xL has been shown to be important for protecting DP thymocytes from spontaneous cell death (22, 23, 24), we examined Bcl-xL protein expression in thymocytes from CCR9-Tg #2 mice. Fig. 5B shows that thymocytes from CCR9-Tg #2 mice express reduced levels of Bcl-xL compared with wild-type mice. Introduction of a Bcl-2 transgene (Figs. 5C and 6) or addition of the caspase inhibitor zVAD during culture (data not shown) restored survival of CCR9-Tg #2 DP thymocytes to wild-type levels, suggesting that the susceptibility of DP thymocytes from CCR9-Tg mice to apoptosis can be attributed to reduced Bcl-xL expression. However, the number of DP thymocytes in Bcl-2/CCR9-Tg #2 mice was still markedly reduced compared with that in wild-type littermates, and DP cells in Bcl-2/CCR9-Tg #2 mice remained CD25+, indicating that the Bcl-2 transgene was unable to rescue the proliferation defect in CCR9-Tg #2 mice (Fig. 6C).
|
|
|
Previous studies have shown that signals that mimic those delivered by pre-TCR induce CCR9 expression in DN thymocytes from Rag/ mice (12, 13). Moreover, in DP thymocytes, CCL25 responsiveness is markedly enhanced by TCR cross-linking (11). These findings suggest that pre-TCR signaling could regulate both CCR9 expression and CCL25 responsiveness in DN thymocytes. To determine whether CCR9 expression is dependent upon pre-TCR signaling in CCR9-Tg mice, we mated the CCR9-Tg #2 transgene onto the Rag-1/ background, where T cell development is arrested at the DN3 stage. Interestingly, although CCR9 surface expression was detected on DN3 cells in Rag-1+/+ CCR9-Tg #2 mice (Fig. 2B), DN thymocytes from Rag-1/ CCR9-Tg #2 mice (which are predominantly DN3, but do not express pre-TCR) were surface CCR9 (Fig. 8A) and did not migrate to CCL25 (Fig. 8B).
Pre-TCR signaling can be mimicked in Rag/ mice by injection of anti-CD3 mAb, which binds to clonotype-independent CD3 surface complexes on DN thymocytes and induces proliferation and transition to the DP stage. Two days after injection of anti-CD3 mAb, CCR9 surface expression could be observed on thymocytes in Rag-1/ CCR9-Tg #2 mice, which still consisted almost entirely of DN cells (Fig. 8A). Moreover, the percentage of Rag-1/ CCR9-Tg #2 thymocytes that could migrate to CCL25 was increased
5-fold by day 2 (Fig. 8B). In contrast, CCR9 expression was undetectable on thymocytes from Rag-1/ mice 2 days after injection of anti-CD3 mAb, and these cells remained unresponsive to CCL25 (Fig. 8B). Thymocytes from Rag-1/ mice were still CCR9 on day 3, even though they included a large percentage of DP thymocytes (Fig. 8A). By day 5, the majority of thymocytes in both Rag-1/ CCR9-Tg #2 and Rag-1/ mice were DP, and most of these cells were CCR9+ (Fig. 8A). These data demonstrate that in both wild-type and CCR9-Tg mice, CCR9 surface expression requires pre-TCR signaling. In wild-type mice, detectable levels of CCR9 were not expressed on the cell surface until they began their transition to or reached the DP stage. However, in Rag-1/ CCR9-Tg mice, CCR9 expression was detected on the surface of DN thymocytes within 2 days after anti-CD3 mAb injection, and these cells could migrate to CCL25. These results predict that altering the timing of pre-TCR expression should affect the timing of CCR9 surface expression. To test this, we introduced a rearranged TCR
-chain transgene (V
8) onto the wild-type and CCR9-Tg #2 background by mating. The percentage of CCR9+ DN4 thymocytes was increased from <5% in wild-type mice to
12% in V
8-Tg mice (Figs. 2B and 9D). Likewise, the percentage of CCR9+ DN3 thymocytes was increased from
20% in CCR9-Tg #2 mice to
50% in V
8/CCR9-Tg #2 mice (Figs. 2B and 9D). Introduction of the V
8 transgene into wild-type mice induced efficient transition of DN3 cells to the DN4 stage (Fig. 9D). In contrast, the V
8 transgene markedly exacerbated the developmental block in CCR9-Tg #2 mice, as evidenced by a reduction in thymocyte numbers, an increased percentage of DN3 cells, and a decreased percentage of DN4 cells in V
8/CCR9-Tg #2 mice compared with CCR9-Tg #2 mice (Fig. 9). CD25 surface expression was also higher on DP thymocytes in V
8/CCR9-Tg #2 mice compared with CCR9-Tg #2 mice (Fig. 9C), indicating that that the proliferative defect was worsened in CCR9-Tg #2 mice by introduction of the V
8 transgene.
|
The observation that premature expression of CCR9 correlates with CCL25 responsiveness (Figs. 2C and 8B) raised the possibility that localization of DN thymocytes within the thymus may be altered in CCR9-Tg mice. H&E staining of thymus sections from CCR9-Tg #1 and CCR9-Tg #2 mice revealed small medullary areas consistent with the reduction in SP thymocyte numbers; however, corticomedullary junctions could still be distinguished. We next performed confocal microscopy to determine the localization of CD4CD8CD25+ cells within the thymus by three-color staining for CD25, CD4, and CD8. Serial sections were also stained with anti-cytokeratin Ab to distinguish the cortical and medullary zones (data not shown). In wild-type mice, CD25+ DN thymocytes were located primarily in the SCR of the thymus (Fig. 10). In CCR9-Tg #1 mice, CD25+ DN cells were also located predominantly in the SCR, although significant numbers of CD25+ DN cells were detected within the outer cortex (Fig. 10). More strikingly, analysis of CCR9-Tg #2 thymus sections revealed that CD25+ DN thymocytes were scattered throughout the entire cortex (Fig. 10). Examination under high magnification confirmed that the majority of these CD25high cells did not express CD4 and/or CD8 and were therefore DN thymocytes (data not shown). Because of the small thymus size in CCR9-Tg #2 mice, the number of CD25+ DN thymocytes appeared to be increased; however, the number of CD25+ DN cells was, in fact, similar to or slightly reduced compared with that in wild-type mice (Figs. 2B and 4). The localization of DP and SP thymocytes did not appear to be altered in CCR9-Tg mice, because these cells were contained predominantly within the cortex and medulla, respectively.
|
| Discussion |
|---|
|
|
|---|
protein-mediated receptor signaling (26). The restricted expression patterns of chemokines and chemokine receptors in the thymus also suggested that these molecules could function to control thymocyte colonization, migration, and export. Nonetheless, the phenotypes of individual chemokine receptor-deficient mice have to date failed to reveal an obligate requirement for specific receptors in regulating thymocyte emigration. Recently, it was shown that CCR7 does play an important role in regulating the emigration of mature SP cells from the neonatal thymus; however, a defect in thymocyte export was not observed in adult CCR7/ mice (27). Immature (CD69+CDC62Llow) CD4 SP thymocytes migrate in response to CCL25 and CCL22, whereas mature (CD69CD62L+) CD4 SP thymocytes and peripheral CD4 SP T cells do not (3, 8), suggesting that CCL25 and/or CCL22 could function as retention factors for SP thymocytes until they reach terminal differentiation. Interestingly, individual CCR9-deficient and CCR4-deficient mice exhibit no overt defects in thymocyte development or emigration (17, 28, 29), indicating that these and perhaps other chemokines may be functionally redundant. In this study we avoided the potential problem of functional redundancy by generating Tg mice in which CCR9 expression is retained on mature SP thymocytes. We found that forced expression of CCR9 does not alter the ratio of immature to mature CD4 SP cells in the thymus, although these cells and peripheral CD4 SP T cells in CCR9-Tg mice respond to CCL25 (Figs. 1 and 2). Thus, CCR9 is not sufficient to retain SP thymocytes in the thymus, and conversely, loss of CCR9 expression is not essential for thymocyte emigration. It remains to be determined whether other chemokine receptors play important and perhaps redundant roles in regulating thymocyte emigration. However, it is worth noting that recent studies have shown that another G protein-coupled receptor, sphingosine-1-phosphate receptor 1, is critical for promoting the export of SP cells from the postnatal thymus (30, 31). Thus, sphingosine-1-phosphate appears to perform an important function in regulating thymocyte emigration in postnatal mice. Our results demonstrate that the timing of CCR9 surface expression (and CCL25 responsiveness) is important for normal thymocyte differentiation. By our method of analysis, surface expression of CCR9 on wild-type thymocytes is first detectable on a small percentage of DN4 thymocytes, but CCR9 is not highly expressed until the DP stage (Fig. 2) (11). Similar results were reported by another group (12). A third group recently found that CCR9 is expressed at low levels on DN2 and DN3 thymocytes and is highly expressed by the time thymocytes reach the DN4 stage (32). Perhaps more relevant to the present study is the determination of CCL25 responsiveness. In that regard, we (Fig. 2D) (11) and others (10, 12) found that a small percentage (<5%) of immature DN thymocytes (consisting of primarily DN2/3 and DN4 cells) in wild-type mice migrate to CCL25. Thus, some DN2/3 and DN4 thymocytes do express CCR9, although at levels too low to detect by our analysis. Nevertheless, in wild-type mice, a high percentage of CCL25-responding cells is observed only at the DP stage (10, 11). Importantly, in CCR9-Tg mice, both CCR9 surface expression and CCL25 responsiveness were increased significantly in the DN2/3 and DN4 thymocyte subsets, and this correlated with a defect in development (Fig. 2, B and D).
CCR9 transcription is low in Rag/ thymocytes, which consist primarily of DN2/3 cells that are pre-TCR (13). However, CCR9 transcription can be induced in Rag/ thymocytes by anti-CD3 mAb stimulation, which mimics pre-TCR signaling (13). Consistent with this, we (Fig. 8) (11) and others (12) found that Rag/ thymocytes are surface CCR9 and unresponsive to CCL25, but that CCR9 surface expression is induced by anti-CD3 mAb stimulation. We also observed that CCR9 surface expression was detected at an earlier stage of development in V
8-Tg mice relative to wild-type mice (Fig. 9). Taken together, these data are consistent with the idea that both CCR9 transcription and CCR9 surface expression are induced by pre-TCR signaling. CCR9 surface expression was also tightly linked to pre-TCR signaling in CCR9-Tg mice. Rag-1/ CCR9-Tg #2 thymocytes were CCR9, but in response to anti-CD3 mAb stimulation, they expressed high levels of CCR9 on their surface and at an earlier stage relative to corresponding Rag-1/ thymocytes (Fig. 8). Likewise, a higher percentage of DN3 cells in V
8/CCR9-Tg #2 expressed CCR9 compared with (non-V
8-Tg) CCR9-Tg #2 mice (Figs. 2B and 9D). Notably, this shift in the onset/intensity of CCR9 surface expression in V
8/CCR9-Tg #2 mice was associated with a marked worsening of the developmental defects compared with (non-V
8-Tg) CCR9-Tg #2 mice (Fig. 9).
Premature expression of CCR9 resulted in a partial block in T cell development at the DN to DP transition point, characterized by decreased proliferation of CD25low DN and CD25 DN thymocytes and the generation of CD25+ DP thymocytes (Table I and Fig. 2). More strikingly, thymocytes that reached the DP stage in CCR9-Tg mice were highly susceptible to spontaneous cell death (Fig. 5). Again, it is significant that the severity of the CCR9-Tg phenotype correlated with the timing/intensity of CCR9 surface expression on immature DN thymocytes. T cell development was impaired to a greater extent in mice from the CCR9-Tg #2 founder line than in those from the CCR9-Tg #1 founder line (Figs. 2 and 4). A distinguishing difference between thymocytes in these mice is that the former contained a higher percentage of CCR9+ DN3 and DN4 cells than the latter (Fig. 2, B and D).
A clue to the cause of the developmental impairment in CCR9-Tg mice may be the apparent abnormal localization of immature DN thymocytes in the thymus. In CCR9-Tg mice, CD25+ DN thymocytes (which prematurely express CCR9) were scattered throughout the cortex rather than confined primarily to the SCR, and the localization defect was clearly worse in CCR9-Tg #2 mice than in CCR9-Tg #1 mice (Fig. 10). Thus, one explanation for our results is that premature expression of CCR9 alters the localization of CD25+ DN thymocytes within the thymus, and this impairs their subsequent development. During normal development, the linking of CCR9 expression to pre-TCR signaling could ensure that thymocytes receive migration signals only after they have proliferated and begun to transition to the DP stage. However, in CCR9-Tg mice, the early/enhanced expression of CCR9 could result in a premature migration signal that is delivered before DN thymocytes have completed their normal maturation sequence.
Because DN thymocytes undergo a complex pattern of migration after their entry into the thymus, it is unclear how early expression of CCR9 might impact their migration. For example, early expression of CCR9 could cause CD25+ DN cells to be retained in the cortex, thus preventing them from ever reaching the SCR. Alternatively, early expression of CCR9 could induce premature migration of CD25+ DN cells out of the SCR. Because a high percentage of CD25+ DN cells have been shown to be in the outer cortex migrating toward the SCR (2), we favor the former explanation. The majority of cells within the SCR are either transitioning to or have reached the DP stage, and these cells are rapidly proliferating (33). Stromal cells in the SCR produce cytokines and growth factors as well as sonic hedgehog, and sonic hedgehog signaling has been suggested to regulate the differentiation of thymocytes from the DN to the DP stage (34). Thus, retention or possible mislocalization of CD25+ DN cells to regions outside the SCR could place them beyond the reach of these regionally expressed factors. Moreover, stromal cells within the SCR may provide specific and important T cell contacts that are not available in other regions of the cortex. Nevertheless, it should be noted that although our findings suggest that CD25+ DN thymocytes may be mislocalized in CCR9-Tg mice, we cannot presently rule out the possibility that the survival and/or proliferation of CD25+ DN thymocytes within the SCR is selectively impaired.
Benz et al. (35) recently demonstrated that in CCR9/ mice, CD25+ cells are dispersed throughout the cortex rather than confined primarily to the SCR of the thymus; yet several groups have shown that thymocyte development is essentially normal in CCR9/ mice (17, 28, 35). On the basis of these findings, it was concluded that migration of CD25+ DN cells to the SCR was not essential for normal T cell development (35). However, in that study the abnormal localization of CD25+ DN cells was most prominent in 2-wk-old CCR9/ mice, a stage when the thymus contains a mixture of cells derived from both fetal and adult precursors and therefore may not accurately reflect the requirements for either fetal or adult T cell development. In addition, some CD25+ DN cells are still found within the SCR in CCR9/ mice (35), and it is not known whether the CD25+ DN cells that lie outside the SCR are capable of giving rise to DP thymocytes. In fact, in competitive bone marrow reconstitution experiments, CCR9/-derived progenitor cells are disadvantaged in the presence of wild-type progenitor cells, and this can be observed at the DN3 stage (17). Alternatively, it is possible that it is not essential that DN2/3 cells reach the SCR, but that subcapsular localization is important at later stages of development (e.g., DN4, transitioning, and early DP thymocytes), and that this step is defective in CCR9-Tg, but not CCR9/, mice. It is intriguing that the same localization defect is observed in both CCR9/ and CCR9-Tg mice. This is consistent with our finding (Fig. 2D) and those of others (10, 12) that some CD25+ DN cells are responsive to CCL25. These results also suggest that the migration of CD25+ DN cells to CCL25 in the thymus is more complex than simply movement through a chemokine gradient. CCR9 may promote the movement of DN cells toward the SCR by facilitating transient contacts with cortical epithelial cells rather than by determining directionality (which could, instead, be mediated by CXCR4 and/or CCR7) (31, 36). Thus, in CCR9/ mice, the loss of these contacts might impair the migration of cells toward the SCR, whereas in CCR9-Tg mice, increased expression of and signaling by CCR9 could stabilize these cell contacts, preventing the further migration of differentiating cells to the SCR.
In conclusion, this study demonstrates that down-regulation of CCR9 is not essential for export of SP cells from the thymus, but, instead, reveals a potential function for CCR9 in early T cell development. As data accumulate from studies of chemokines and their receptors, it has become apparent that the regulation of intrathymic trafficking is more complex than was originally thought. Additional studies should assist in resolving the role of CCR9 as well as other CCRs in this process.
| Acknowledgments |
|---|
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 Address correspondence and reprint requests to Dr. Paul E. Love, Laboratory of Mammalian Genes and Development, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, MD 20892. E-mail address: lovep{at}mail.nih.gov ![]()
2 Abbreviations used in this paper: SCR, subcapsular region; CD4 SP, CD4+CD8 (single positive); CD8 SP CD4CD8+; DN, CD4CD8 (double negative); DP, CD4+CD8+ (double positive); Tg, transgenic. ![]()
Received for publication May 26, 2005. Accepted for publication October 13, 2005.
| References |
|---|
|
|
|---|

TCR+ thymocytes preferentially respond to CCL25. J. Immunol. 168: 134-142.
-chemokine receptor is modulated in thymocyte differentiation and is selectively maintained in CD8+ T cells from secondary lymphoid organs. Blood 97: 850-857. 
T-cell development in mice deficient in the common cytokine receptor-chain: the requirement for Bcl-2 differs depending on the TCR/MHC affinity. J. Immunol. 162: 782-790. 
+ gut intraepithelial lymphocytes. Blood 98: 2626-2632. This article has been cited by other articles:
![]() |
M. Svensson, J. Marsal, H. Uronen-Hansson, M. Cheng, W. Jenkinson, C. Cilio, S. E. W. Jacobsen, E. Sitnicka, G. Anderson, and W. W. Agace Involvement of CCR9 at multiple stages of adult T lymphopoiesis J. Leukoc. Biol., January 1, 2008; 83(1): 156 - 164. [Abstract] [Full Text] [PDF] |
||||
![]() |
M.-A. Wurbel, M. Malissen, D. Guy-Grand, B. Malissen, and J. J. Campbell Impaired Accumulation of Antigen-Specific CD8 Lymphocytes in Chemokine CCL25-Deficient Intestinal Epithelium and Lamina Propria J. Immunol., June 15, 2007; 178(12): 7598 - 7606. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||